merge from trunk - r17500 to HEAD
authorCampbell Barton <ideasman42@gmail.com>
Wed, 10 Dec 2008 09:45:48 +0000 (09:45 +0000)
committerCampbell Barton <ideasman42@gmail.com>
Wed, 10 Dec 2008 09:45:48 +0000 (09:45 +0000)
17 files changed:
release/scripts/export_m3g.py
release/scripts/image_auto_layout.py
release/scripts/import_lightwave_motion.py
release/scripts/scripttemplate_gamelogic.py
source/blender/blenkernel/BKE_blender.h
source/blender/blenkernel/BKE_brush.h
source/blender/blenkernel/BKE_customdata.h
source/blender/blenkernel/intern/brush.c
source/blender/blenkernel/intern/customdata.c
source/blender/blenloader/intern/readfile.c
source/blender/include/butspace.h
source/blender/makesdna/DNA_customdata_types.h
source/blender/makesdna/DNA_scene_types.h
source/blender/src/buttons_editing.c
source/blender/src/drawview.c
source/blender/src/imagepaint.c
source/blender/src/space.c

index c74e7acbcd37fff72e9e7b6abf6b4b88ddb89e65..86ac03cc407a97134c3a32d41122f097cd7d5bdd 100644 (file)
@@ -11,7 +11,7 @@ Tooltip: 'Export to M3G'
 #
 # Source: http://www.nelson-games.de/bl2m3g/source
 #
-# $Id$
+# $Id: m3g_export.py,v 0.1 2005/04/19 12:25 gerhardv Exp gerhardv $
 #
 # Author: Gerhard Völkl
 #
index c6f97a25434f9ae00da9091f528ca3e3003b433d..19ee396c3b1d2671cbe1fa78ff0008a2c36c54cc 100644 (file)
@@ -265,8 +265,8 @@ def consolidate_mesh_images(mesh_list, scn, PREF_IMAGE_PATH, PREF_IMAGE_SIZE, PR
        render_context.setRenderWinSize(100)
        render_context.setImageType(Render.PNG)
        render_context.enableExtensions(True) 
-       render_context.enablePremultiply() # No alpha needed.
-       render_context.enableRGBAColor()
+       render_context.enableSky() # No alpha needed.
+       render_context.enableRGBColor()
        render_context.threads = 2
        
        #Render.EnableDispView() # Broken??
@@ -275,9 +275,8 @@ def consolidate_mesh_images(mesh_list, scn, PREF_IMAGE_PATH, PREF_IMAGE_SIZE, PR
        render_mat= B.Material.New()
        render_mat.mode |= B.Material.Modes.SHADELESS
        render_mat.mode |= B.Material.Modes.TEXFACE
-       render_mat.mode |= B.Material.Modes.ZTRANSP
-       render_mat.setAlpha(0.0)
-               
+       
+       
        render_me= B.Mesh.New()
        render_me.verts.extend([Vector(0,0,0)]) # Stupid, dummy vert, preverts errors. when assigning UV's/
        render_ob= B.Object.New('Mesh')
index 20c87dfd5c633761db1b14fa5e8234d077f6bf16..c242a9f6bd3ee9507de7848c48ee0d06e056fcc2 100644 (file)
@@ -22,7 +22,7 @@ Be sure to set the framerate correctly
 
 """
 
-# $Id$
+# $Id: export_lightwave_motion.py 9924 2007-01-27 02:15:14Z campbellbarton $
 # --------------------------------------------------------------------------
 # ***** BEGIN GPL LICENSE BLOCK *****
 #
index 7184d7e424fcf3e2344660c2b2e1661d1b7fb5ba..1709525115500c5ee716f70a19b679376d3d937d 100644 (file)
@@ -83,8 +83,7 @@ def main():
                        own.life += ob.life
                        ob.life = 0
        print own.life
-       """
-
+       """\r
 main()
 '''
 
index f392d57521cb9e94285f7b4c376b07863ff3395d..a6334e665d1076a438ed9853f473e20175599520 100644 (file)
@@ -41,7 +41,7 @@ struct ListBase;
 struct MemFile;
 
 #define BLENDER_VERSION                        248
-#define BLENDER_SUBVERSION             1
+#define BLENDER_SUBVERSION             2
 
 #define BLENDER_MINVERSION             245
 #define BLENDER_MINSUBVERSION  15
index f1f4653f092233bb194db32c059c26ebdf7e8c1b..398d203709f0f1c5693b1d5e1362937854beda90 100644 (file)
@@ -53,6 +53,7 @@ int brush_clone_image_delete(struct Brush *brush);
 
 /* sampling */
 float brush_sample_falloff(struct Brush *brush, float dist);
+float brush_sample_falloff_noalpha(struct Brush *brush, float dist);
 void brush_sample_tex(struct Brush *brush, float *xy, float *rgba);
 void brush_imbuf_new(struct Brush *brush, short flt, short texfalloff, int size,
        struct ImBuf **imbuf);
index c84b690bc49d419764b5c5d46aeba2db9a1e4121..10791968f7958abc98dcbfbad0851f0df796dbc3 100644 (file)
@@ -198,8 +198,12 @@ int CustomData_get_layer_index(const struct CustomData *data, int type);
 int CustomData_get_named_layer_index(const struct CustomData *data, int type, char *name);
 int CustomData_get_active_layer_index(const struct CustomData *data, int type);
 int CustomData_get_render_layer_index(const struct CustomData *data, int type);
+int CustomData_get_clone_layer_index(const struct CustomData *data, int type);
+int CustomData_get_mask_layer_index(const struct CustomData *data, int type);
 int CustomData_get_active_layer(const struct CustomData *data, int type);
 int CustomData_get_render_layer(const struct CustomData *data, int type);
+int CustomData_get_clone_layer(const struct CustomData *data, int type);
+int CustomData_get_mask_layer(const struct CustomData *data, int type);
 
 /* copies the data from source to the data element at index in the first
  * layer of type
@@ -227,10 +231,14 @@ void *CustomData_set_layer_n(const struct CustomData *data, int type, int n, voi
 /* sets the nth layer of type as active */
 void CustomData_set_layer_active(struct CustomData *data, int type, int n);
 void CustomData_set_layer_render(struct CustomData *data, int type, int n);
+void CustomData_set_layer_clone(struct CustomData *data, int type, int n);
+void CustomData_set_layer_mask(struct CustomData *data, int type, int n);
 
 /* same as above but works with an index from CustomData_get_layer_index */
 void CustomData_set_layer_active_index(struct CustomData *data, int type, int n);
 void CustomData_set_layer_render_index(struct CustomData *data, int type, int n);
+void CustomData_set_layer_clone_index(struct CustomData *data, int type, int n);
+void CustomData_set_layer_mask_index(struct CustomData *data, int type, int n);
 
 /* adds flag to the layer flags */
 void CustomData_set_layer_flag(struct CustomData *data, int type, int flag);
index 1c53af97dbb86a2f52eb0e54170463a53f89a9fa..021f76fd2f135df7ea68c8dd2f1c01bbbddf949e 100644 (file)
@@ -328,6 +328,23 @@ float brush_sample_falloff(Brush *brush, float dist)
                return 0.0f;
 }
 
+float brush_sample_falloff_noalpha(Brush *brush, float dist)
+{
+       float outer, inner;
+
+       outer = brush->size >> 1;
+       inner = outer*brush->innerradius;
+
+       if (dist <= inner) {
+               return 1.0f;
+       }
+       else if ((dist < outer) && (inner < outer)) {
+               return 1.0f - sqrt((dist - inner)/(outer - inner));
+       }
+       else 
+               return 0.0f;
+}
+
 void brush_sample_tex(Brush *brush, float *xy, float *rgba)
 {
        MTex *mtex= brush->mtex[brush->texact];
index e93266c85f301a422762309e707830267e5d2f3e..05271aa59a7a2ebb645d55d50d43c123af7875c8 100644 (file)
@@ -606,7 +606,7 @@ void CustomData_merge(const struct CustomData *source, struct CustomData *dest,
 {
        const LayerTypeInfo *typeInfo;
        CustomDataLayer *layer, *newlayer;
-       int i, type, number = 0, lasttype = -1, lastactive = 0, lastrender = 0;
+       int i, type, number = 0, lasttype = -1, lastactive = 0, lastrender = 0, lastclone = 0, lastmask = 0;
 
        for(i = 0; i < source->totlayer; ++i) {
                layer = &source->layers[i];
@@ -618,6 +618,8 @@ void CustomData_merge(const struct CustomData *source, struct CustomData *dest,
                        number = 0;
                        lastactive = layer->active;
                        lastrender = layer->active_rnd;
+                       lastclone = layer->active_clone;
+                       lastmask = layer->active_mask;
                        lasttype = type;
                }
                else
@@ -637,6 +639,8 @@ void CustomData_merge(const struct CustomData *source, struct CustomData *dest,
                if(newlayer) {
                        newlayer->active = lastactive;
                        newlayer->active_rnd = lastrender;
+                       newlayer->active_clone = lastclone;
+                       newlayer->active_mask = lastmask;
                }
        }
 }
@@ -736,6 +740,28 @@ int CustomData_get_render_layer_index(const CustomData *data, int type)
        return -1;
 }
 
+int CustomData_get_clone_layer_index(const CustomData *data, int type)
+{
+       int i;
+
+       for(i=0; i < data->totlayer; ++i)
+               if(data->layers[i].type == type)
+                       return i + data->layers[i].active_clone;
+
+       return -1;
+}
+
+int CustomData_get_mask_layer_index(const CustomData *data, int type)
+{
+       int i;
+
+       for(i=0; i < data->totlayer; ++i)
+               if(data->layers[i].type == type)
+                       return i + data->layers[i].active_mask;
+
+       return -1;
+}
+
 int CustomData_get_active_layer(const CustomData *data, int type)
 {
        int i;
@@ -758,6 +784,27 @@ int CustomData_get_render_layer(const CustomData *data, int type)
        return -1;
 }
 
+int CustomData_get_clone_layer(const CustomData *data, int type)
+{
+       int i;
+
+       for(i=0; i < data->totlayer; ++i)
+               if(data->layers[i].type == type)
+                       return data->layers[i].active_clone;
+
+       return -1;
+}
+
+int CustomData_get_mask_layer(const CustomData *data, int type)
+{
+       int i;
+
+       for(i=0; i < data->totlayer; ++i)
+               if(data->layers[i].type == type)
+                       return data->layers[i].active_mask;
+
+       return -1;
+}
 
 void CustomData_set_layer_active(CustomData *data, int type, int n)
 {
@@ -777,6 +824,24 @@ void CustomData_set_layer_render(CustomData *data, int type, int n)
                        data->layers[i].active_rnd = n;
 }
 
+void CustomData_set_layer_clone(CustomData *data, int type, int n)
+{
+       int i;
+
+       for(i=0; i < data->totlayer; ++i)
+               if(data->layers[i].type == type)
+                       data->layers[i].active_clone = n;
+}
+
+void CustomData_set_layer_mask(CustomData *data, int type, int n)
+{
+       int i;
+
+       for(i=0; i < data->totlayer; ++i)
+               if(data->layers[i].type == type)
+                       data->layers[i].active_mask = n;
+}
+
 /* for using with an index from CustomData_get_active_layer_index and CustomData_get_render_layer_index */
 void CustomData_set_layer_active_index(CustomData *data, int type, int n)
 {
@@ -796,6 +861,23 @@ void CustomData_set_layer_render_index(CustomData *data, int type, int n)
                        data->layers[i].active_rnd = n-i;
 }
 
+void CustomData_set_layer_clone_index(CustomData *data, int type, int n)
+{
+       int i;
+
+       for(i=0; i < data->totlayer; ++i)
+               if(data->layers[i].type == type)
+                       data->layers[i].active_clone = n-i;
+}
+
+void CustomData_set_layer_mask_index(CustomData *data, int type, int n)
+{
+       int i;
+
+       for(i=0; i < data->totlayer; ++i)
+               if(data->layers[i].type == type)
+                       data->layers[i].active_mask = n-i;
+}
 
 void CustomData_set_layer_flag(struct CustomData *data, int type, int flag)
 {
@@ -882,9 +964,13 @@ static CustomDataLayer *customData_add_layer__internal(CustomData *data,
        if(index > 0 && data->layers[index-1].type == type) {
                data->layers[index].active = data->layers[index-1].active;
                data->layers[index].active_rnd = data->layers[index-1].active_rnd;
+               data->layers[index].active_clone = data->layers[index-1].active_clone;
+               data->layers[index].active_mask = data->layers[index-1].active_mask;
        } else {
                data->layers[index].active = 0;
                data->layers[index].active_rnd = 0;
+               data->layers[index].active_clone = 0;
+               data->layers[index].active_mask = 0;
        }
        
        customData_update_offsets(data);
@@ -944,6 +1030,8 @@ int CustomData_free_layer(CustomData *data, int type, int totelem, int index)
                        for (; i < data->totlayer && data->layers[i].type == type; i++) {
                                data->layers[i].active--;
                                data->layers[i].active_rnd--;
+                               data->layers[i].active_clone--;
+                               data->layers[i].active_mask--;
                        }
        }
 
index 423d050c8622d5b67929e6097a5f4e4e69f790a4..91170563c1146137cf3140499cc7c686d6e86b91 100644 (file)
@@ -8007,6 +8007,17 @@ static void do_versions(FileData *fd, Library *lib, Main *main)
                }
        }
        
+       if (main->versionfile < 248 || (main->versionfile == 248 && main->subversionfile < 2)) {
+               Scene *sce;
+               
+               /* Note, these will need to be added for painting */
+               for (sce= main->scene.first; sce; sce= sce->id.next) {
+                       sce->toolsettings->imapaint.seam_bleed = 2;
+                       sce->toolsettings->imapaint.normal_angle = 80;
+               }
+       }
+       
+       
        /* WATCH IT!!!: pointers from libdata have not been converted yet here! */
        /* WATCH IT 2!: Userdef struct init has to be in src/usiblender.c! */
 
index 7019a3f6959f0b457e1546b3bdb0bc67fcf40c62..868d0c3e6c0db4a6d29ef8995f8f91f6bda54576 100644 (file)
@@ -453,6 +453,8 @@ void curvemap_buttons(struct uiBlock *block, struct CurveMapping *cumap, char la
 
 #define B_GEN_SKELETON         2085
 #define B_RETARGET_SKELETON    2086
+#define B_SETTFACE_CLONE       2087
+#define B_SETTFACE_MASK                2088
 
 /* *********************** */
 #define B_VGROUPBUTS           2100
index 6c098e220bbbd316e6f9579461eacbcadfa789a0..e6b18641d2afce1def093227e7db1202cf448bd5 100644 (file)
@@ -37,6 +37,8 @@ typedef struct CustomDataLayer {
        int flag;       /* general purpose flag */
        int active;     /* number of the active layer of this type */
        int active_rnd; /* number of the layer to render*/
+       int active_clone; /* number of the layer to render*/
+       int active_mask; /* number of the layer to render*/
        char pad[4];
        char name[32];  /* layer name */
        void *data;     /* layer data */
index c1c164f136f4427708581acfaa031be5be1af20a..8b264602feed4abaacae0db67547bc07e4428584 100644 (file)
@@ -345,7 +345,9 @@ typedef struct TimeMarker {
 typedef struct ImagePaintSettings {
        struct Brush *brush;
        short flag, tool;
-       int pad3;
+       
+       /* for projection painting only - todo - use flags */
+       short seam_bleed,normal_angle;
 } ImagePaintSettings;
 
 typedef struct ParticleBrushData {
@@ -799,6 +801,15 @@ typedef struct Scene {
 #define IMAGEPAINT_DRAW_TOOL                   2
 #define IMAGEPAINT_DRAW_TOOL_DRAWING   4
 
+/* projection painting only */
+#define IMAGEPAINT_PROJECT_DISABLE             8       /* Non projection 3D painting */
+#define IMAGEPAINT_PROJECT_XRAY                        16
+#define IMAGEPAINT_PROJECT_BACKFACE            32
+#define IMAGEPAINT_PROJECT_FLAT                        64
+#define IMAGEPAINT_PROJECT_LAYER_CLONE 128
+#define IMAGEPAINT_PROJECT_LAYER_MASK  256
+#define IMAGEPAINT_PROJECT_LAYER_MASK_INV      512
+
 /* toolsettings->uvcalc_flag */
 #define UVCALC_FILLHOLES                       1
 #define UVCALC_NO_ASPECT_CORRECT       2       /* would call this UVCALC_ASPECT_CORRECT, except it should be default with old file */
index edf4660b71e3fadcb23b678e6e16f71fce1bfb30..65b992682de16134a624d6a0a84996524a33e615 100644 (file)
 
 static float editbutweight= 1.0;
 float editbutvweight= 1;
-static int actmcol= 0, acttface= 0, acttface_rnd = 0, actmcol_rnd = 0;
+static int actmcol= 0, acttface= 0, acttface_rnd = 0, acttface_clone = 0, acttface_mask = 0, actmcol_rnd = 0;
 
 extern ListBase editNurb;
 
@@ -722,10 +722,10 @@ static void delete_customdata_layer(void *data1, void *data2)
        Mesh *me= (Mesh*)data1;
        CustomData *data= (G.obedit)? &G.editMesh->fdata: &me->fdata;
        CustomDataLayer *layer= (CustomDataLayer*)data2;
-       void *actlayerdata, *rndlayerdata, *layerdata=layer->data;
+       void *actlayerdata, *rndlayerdata, *clonelayerdata, *masklayerdata, *layerdata=layer->data;
        int type= layer->type;
        int index= CustomData_get_layer_index(data, type);
-       int i, actindex, rndindex;
+       int i, actindex, rndindex, cloneindex, maskindex;
        
        /*ok, deleting a non-active layer needs to preserve the active layer indices.
          to do this, we store a pointer to the .data member of both layer and the active layer,
@@ -736,6 +736,8 @@ static void delete_customdata_layer(void *data1, void *data2)
          layer. */
        actlayerdata = data->layers[CustomData_get_active_layer_index(data, type)].data;
        rndlayerdata = data->layers[CustomData_get_render_layer_index(data, type)].data;
+       clonelayerdata = data->layers[CustomData_get_clone_layer_index(data, type)].data;
+       masklayerdata = data->layers[CustomData_get_mask_layer_index(data, type)].data;
        CustomData_set_layer_active(data, type, layer - &data->layers[index]);
 
        /* Multires is handled seperately because the display data is separate
@@ -787,6 +789,33 @@ static void delete_customdata_layer(void *data1, void *data2)
                CustomData_set_layer_render(data, type, rndindex);
        }
        
+       if (clonelayerdata != layerdata) {
+               /*find index. . .*/
+               cloneindex = CustomData_get_layer_index(data, type);
+               for (i=cloneindex; i<data->totlayer; i++) {
+                       if (data->layers[i].data == clonelayerdata) {
+                               cloneindex = i - cloneindex;
+                               break;
+                       }
+               }
+               
+               /*set index. . .*/
+               CustomData_set_layer_clone(data, type, cloneindex);
+       }
+       
+       if (masklayerdata != layerdata) {
+               /*find index. . .*/
+               maskindex = CustomData_get_layer_index(data, type);
+               for (i=maskindex; i<data->totlayer; i++) {
+                       if (data->layers[i].data == masklayerdata) {
+                               maskindex = i - maskindex;
+                               break;
+                       }
+               }
+               
+               /*set index. . .*/
+               CustomData_set_layer_mask(data, type, maskindex);
+       }
        
        DAG_object_flush_update(G.scene, OBACT, OB_RECALC_DATA);
        
@@ -802,9 +831,9 @@ static void delete_customdata_layer(void *data1, void *data2)
 
 static int customdata_buttons(
        uiBlock *block, Mesh *me, CustomData *data,
-       int type, int *activep, int *renderp,
-       int setevt, int setevt_rnd, int newevt,
-       char *label, char *shortlabel, char *browsetip, char *browsetip_rnd,
+       int type, int *activep, int *renderp, int *clonep, int *maskp,
+       int setevt, int setevt_rnd, int setevt_clone, int setevt_mask, int newevt,
+       char *label, char *shortlabel, char *browsetip, char *browsetip_rnd, char *browsetip_clone, char *browsetip_mask,
        char *newtip, char *deltip, int x, int y)
 {
        CustomDataLayer *layer;
@@ -828,12 +857,27 @@ static int customdata_buttons(
                layer= &data->layers[i];
 
                if(layer->type == type) {
+                       int xi = 0;
                        *activep= layer->active + 1;
                        *renderp= layer->active_rnd + 1;
+                       if (clonep) *clonep= layer->active_clone + 1;
+                       if (maskp) *maskp= layer->active_mask + 1;
+                       
                        
                        uiDefIconButI(block, ROW, setevt, ICON_VIEW3D, x,y,25,19, activep, 1.0, count, 0, 0, browsetip);
                        uiDefIconButI(block, ROW, setevt_rnd, ICON_SCENE, x+25,y,25,19, renderp, 1.0, count, 0, 0, browsetip_rnd);
-                       but=uiDefBut(block, TEX, setevt, "", x+50,y,145,19, layer->name, 0.0, 31.0, 0, 0, label);
+                       
+                       if (clonep) {
+                               uiDefIconButI(block, ROW, setevt_clone, ICON_TEXTURE, x+50,y,25,19, clonep, 1.0, count, 0, 0, browsetip_clone);
+                               xi += 25;
+                       }
+                       
+                       if (maskp) {
+                               uiDefIconButI(block, ROW, setevt_mask, ICON_PAINT, x+50+xi,y,25,19, maskp, 1.0, count, 0, 0, browsetip_mask);
+                               xi += 25;
+                       }
+                       
+                       but=uiDefBut(block, TEX, setevt, "", x+50+xi,y,145-xi,19, layer->name, 0.0, 31.0, 0, 0, label);
                        uiButSetFunc(but, verify_customdata_name_func, data, layer);
                        but= uiDefIconBut(block, BUT, B_NOP, VICON_X, x+195,y,25,19, NULL, 0.0, 0.0, 0.0, 0.0, deltip);
                        uiButSetFunc(but, delete_customdata_layer, me, layer);
@@ -902,14 +946,14 @@ static void editing_panel_mesh_type(Object *ob, Mesh *me)
        uiBlockEndAlign(block);
 
        fdata= (G.obedit)? &G.editMesh->fdata: &me->fdata;
-       yco= customdata_buttons(block, me, fdata, CD_MTFACE, &acttface, &acttface_rnd,
-               B_SETTFACE, B_SETTFACE_RND, B_NEWTFACE, "UV Texture", "UV Texture:",
-               "Set active UV texture", "Set rendering UV texture", "Creates a new UV texture layer",
+       yco= customdata_buttons(block, me, fdata, CD_MTFACE, &acttface, &acttface_rnd, (G.f & G_TEXTUREPAINT ? &acttface_clone : NULL), (G.f & G_TEXTUREPAINT ? &acttface_mask : NULL),
+               B_SETTFACE, B_SETTFACE_RND, B_SETTFACE_CLONE, B_SETTFACE_MASK, B_NEWTFACE, "UV Texture", "UV Texture:",
+               "Set active UV texture", "Set rendering UV texture", "Set the layer used for texturepaint cloning", "Set the texture paint stencil layer", "Creates a new UV texture layer",
                "Removes the current UV texture layer", 190, 130);
 
-       yco= customdata_buttons(block, me, fdata, CD_MCOL, &actmcol, &actmcol_rnd,
-               B_SETMCOL, B_SETMCOL_RND, B_NEWMCOL, "Vertex Color", "Vertex Color:",
-               "Sets active vertex color layer", "Sets rendering vertex color layer", "Creates a new vertex color layer",
+       yco= customdata_buttons(block, me, fdata, CD_MCOL, &actmcol, &actmcol_rnd, NULL, NULL,
+               B_SETMCOL, B_SETMCOL_RND, B_NOP, B_NOP, B_NEWMCOL, "Vertex Color", "Vertex Color:",
+               "Sets active vertex color layer", "Sets rendering vertex color layer", "", "", "Creates a new vertex color layer",
                "Removes the current vertex color layer", 190, yco-5);
 
        if(yco < 0)
@@ -4964,7 +5008,22 @@ void do_meshbuts(unsigned short event)
                                allqueue(REDRAWBUTSEDIT, 0);
                        }
                        break;
-                       
+               case B_SETTFACE_CLONE:
+                       if (G.obedit || me) {
+                               CustomData *fdata= (G.obedit)? &em->fdata: &me->fdata;
+                               CustomData_set_layer_clone(fdata, CD_MTFACE, acttface_clone-1);
+                               BIF_undo_push("Set Clone UV Texture");
+                               allqueue(REDRAWBUTSEDIT, 0);
+                       }
+                       break;
+               case B_SETTFACE_MASK:
+                       if (G.obedit || me) {
+                               CustomData *fdata= (G.obedit)? &em->fdata: &me->fdata;
+                               CustomData_set_layer_mask(fdata, CD_MTFACE, acttface_mask-1);
+                               BIF_undo_push("Set Mask UV Texture");
+                               allqueue(REDRAWBUTSEDIT, 0);
+                       }
+                       break;
                case B_FLIPNORM:
                        if(G.obedit) {
                                flip_editnormals();
@@ -6279,20 +6338,23 @@ void brush_buttons(uiBlock *block, short sima,
        ToolSettings *settings= G.scene->toolsettings;
        Brush *brush= settings->imapaint.brush;
        ID *id;
-       int yco, xco, butw;
+       int yco, xco, butw, but_idx;
 
        short *menupoin = sima ? &(G.sima->menunr) : &(G.buts->menunr);
+       short do_project = settings->imapaint.flag & IMAGEPAINT_PROJECT_DISABLE ? 0:1;
        
        yco= 160;
 
        butw = sima ? 80 : 106;
        
        uiBlockBeginAlign(block);
-       uiDefButS(block, ROW, evt_change, "Draw",               0,              yco,butw,19, &settings->imapaint.tool, 7.0, PAINT_TOOL_DRAW, 0, 0, "Draw brush");
-       uiDefButS(block, ROW, evt_change, "Soften",             butw,   yco,butw,19, &settings->imapaint.tool, 7.0, PAINT_TOOL_SOFTEN, 0, 0, "Soften brush");
-       uiDefButS(block, ROW, evt_change, "Smear",              butw*2, yco,butw,19, &settings->imapaint.tool, 7.0, PAINT_TOOL_SMEAR, 0, 0, "Smear brush");
-       if (sima)
-               uiDefButS(block, ROW, evt_change, "Clone",      butw*3, yco,butw,19, &settings->imapaint.tool, 7.0, PAINT_TOOL_CLONE, 0, 0, "Clone brush, use RMB to drag source image");
+       but_idx = 0;
+       uiDefButS(block, ROW, evt_change, "Draw",               butw*(but_idx++),yco,butw,19, &settings->imapaint.tool, 7.0, PAINT_TOOL_DRAW, 0, 0, "Draw brush");
+       if (sima || do_project==0)
+               uiDefButS(block, ROW, evt_change, "Soften",     butw*(but_idx++),       yco,butw,19, &settings->imapaint.tool, 7.0, PAINT_TOOL_SOFTEN, 0, 0, "Soften brush");
+       uiDefButS(block, ROW, evt_change, "Smear",              butw*(but_idx++),       yco,butw,19, &settings->imapaint.tool, 7.0, PAINT_TOOL_SMEAR, 0, 0, "Smear brush");
+       if (sima || do_project)
+               uiDefButS(block, ROW, evt_change, "Clone",      butw*(but_idx++),       yco,butw,19, &settings->imapaint.tool, 7.0, PAINT_TOOL_CLONE, 0, 0, "Clone brush, use RMB to drag source image");
        
        uiBlockEndAlign(block);
        yco -= 30;
@@ -6307,16 +6369,44 @@ void brush_buttons(uiBlock *block, short sima,
                butw= 320-(xco+10);
 
                uiDefButS(block, MENU, evt_nop, "Mix %x0|Add %x1|Subtract %x2|Multiply %x3|Lighten %x4|Darken %x5|Erase Alpha %x6|Add Alpha %x7", xco+10,yco,butw,19, &brush->blend, 0, 0, 0, 0, "Blending method for applying brushes");
-
-               uiDefButBitS(block, TOG|BIT, BRUSH_TORUS, evt_change, "Wrap",   xco+10,yco-25,butw,19, &brush->flag, 0, 0, 0, 0, "Enables torus wrapping");
-
+       
                uiBlockBeginAlign(block);
-               uiDefButBitS(block, TOG|BIT, BRUSH_AIRBRUSH, evt_change, "Airbrush",    xco+10,yco-50,butw,19, &brush->flag, 0, 0, 0, 0, "Keep applying paint effect while holding mouse (spray)");
-               uiDefButF(block, NUM, evt_nop, "Rate ", xco+10,yco-70,butw,19, &brush->rate, 0.01, 1.0, 0, 0, "Number of paints per second for Airbrush");
+               uiDefButBitS(block, TOG|BIT, BRUSH_AIRBRUSH, evt_change, "Airbrush",    xco+10,yco-25,butw/2,19, &brush->flag, 0, 0, 0, 0, "Keep applying paint effect while holding mouse (spray)");
+               uiDefButF(block, NUM, evt_nop, "", xco+10 + butw/2,yco-25,butw/2,19, &brush->rate, 0.01, 1.0, 0, 0, "Number of paints per second for Airbrush");
                uiBlockEndAlign(block);
-
-               yco -= 25;
-
+               
+               if (sima) {
+                       uiDefButBitS(block, TOG|BIT, BRUSH_TORUS, evt_change, "Wrap",   xco+10,yco-45,butw,19, &brush->flag, 0, 0, 0, 0, "Enables torus wrapping");
+                       yco -= 25;
+               }
+               else {
+                       yco -= 25;
+                       uiBlockBeginAlign(block);
+                       uiDefButBitS(block, TOGN|BIT, IMAGEPAINT_PROJECT_DISABLE, B_REDR, "Project Paint",      xco+10,yco-25,butw,19, &settings->imapaint.flag, 0, 0, 0, 0, "Use projection painting for improved consistency in the brush strokes");
+                       
+                       if ((settings->imapaint.flag & IMAGEPAINT_PROJECT_DISABLE)==0) {
+                               
+                               
+                               /* Projection Painting */
+                               
+                               uiDefButBitS(block, TOGN|BIT, IMAGEPAINT_PROJECT_XRAY, B_NOP, "Occlude",        xco+10,yco-45,butw/2,19, &settings->imapaint.flag, 0, 0, 0, 0, "Only paint onto the faces directly under the brush (slower)");
+                               uiDefButBitS(block, TOGN|BIT, IMAGEPAINT_PROJECT_BACKFACE, B_NOP, "Cull",       xco+10+butw/2,yco-45,butw/2,19, &settings->imapaint.flag, 0, 0, 0, 0, "Ignore faces pointing away from the view (faster)");
+                               
+                               uiDefButBitS(block, TOGN|BIT, IMAGEPAINT_PROJECT_FLAT, B_NOP, "Normal", xco+10,yco-65,butw/2,19, &settings->imapaint.flag, 0, 0, 0, 0, "Paint most on faces pointing towards the view");
+                               uiDefButS(block, NUM, B_NOP, "", xco+10 +(butw/2),yco-65,butw/2,19, &settings->imapaint.normal_angle, 10.0, 90.0, 0, 0, "Paint most on faces pointing towards the view acording to this angle)");
+                               
+                               uiDefButS(block, NUM, B_NOP, "Bleed: ", xco+10,yco-85,butw,19, &settings->imapaint.seam_bleed, 0.0, 8.0, 0, 0, "Extend paint beyond the faces UVs to reduce seams (in pixels, slower)");
+                               uiBlockEndAlign(block);
+                               
+                               uiBlockBeginAlign(block);
+                               uiDefButBitS(block, TOG|BIT, IMAGEPAINT_PROJECT_LAYER_MASK, B_NOP, "Stencil Layer",     xco+10,yco-110,butw-30,19, &settings->imapaint.flag, 0, 0, 0, 0, "Set the mask layer from the UV layer buttons");
+                               uiDefButBitS(block, TOG|BIT, IMAGEPAINT_PROJECT_LAYER_MASK_INV, B_NOP, "Inv",   xco+10 + butw-30,yco-110,30,19, &settings->imapaint.flag, 0, 0, 0, 0, "Invert the mask");
+                               uiBlockEndAlign(block);
+                               
+                       }
+                       uiBlockEndAlign(block);
+               }
+               
                uiBlockBeginAlign(block);
                uiDefButF(block, COL, B_VPCOLSLI, "",                                   0,yco,200,19, brush->rgb, 0, 0, 0, 0, "");
                uiDefButF(block, NUMSLI, evt_nop, "Opacity ",           0,yco-20,180,19, &brush->alpha, 0.0, 1.0, 0, 0, "The amount of pressure on the brush");
index de3e464060d00b8e955d49b1c9aa8a732ba6eede..0942d5d1cdb79402109518ef28aee898728a0f10 100644 (file)
@@ -3309,7 +3309,12 @@ void drawview3dspace(ScrArea *sa, void *spacedata)
                        fdrawXORcirc((float)c[0], (float)c[1], (float)pset->brush[pset->brushtype].size);
                }
        }
-
+       if(!G.obedit && OBACT && G.f&G_TEXTUREPAINT && area_is_active_area(v3d->area) && G.scene->toolsettings->imapaint.brush){
+               short c[2];
+               getmouseco_areawin(c);
+               fdrawXORcirc((float)c[0], (float)c[1], (float)G.scene->toolsettings->imapaint.brush->size/2);
+       }
+       
        if(v3d->persp>1) drawviewborder();
        if(v3d->flag2 & V3D_FLYMODE) drawviewborder_flymode();
        
index 149dbf1d0268151f6f1453486260848fe86a1995..2e48942189f1bbab30f403778a96ba5c575ac811 100644 (file)
@@ -23,9 +23,9 @@
  * The Original Code is Copyright (C) 2001-2002 by NaN Holding BV.
  * All rights reserved.
  *
- * The Original Code is: all of this file.
+ * The Original Code is: some of this file.
  *
- * Contributor(s): Jens Ole Wund (bjornmose)
+ * Contributor(s): Jens Ole Wund (bjornmose), Campbell Barton (ideasman42)
  *
  * ***** END GPL LICENSE BLOCK *****
  */
 #include "BLI_arithb.h"
 #include "BLI_blenlib.h"
 #include "BLI_dynstr.h"
+#include "BLI_linklist.h"
+#include "BLI_memarena.h"
 #include "PIL_time.h"
+#include "BLI_threads.h"
 
 #include "IMB_imbuf.h"
 #include "IMB_imbuf_types.h"
@@ -62,6 +65,7 @@
 #include "DNA_space_types.h"
 #include "DNA_userdef_types.h"
 #include "DNA_view3d_types.h"
+#include "DNA_gpencil_types.h"
 
 #include "BKE_brush.h"
 #include "BKE_global.h"
@@ -70,6 +74,7 @@
 #include "BKE_mesh.h"
 #include "BKE_node.h"
 #include "BKE_utildefines.h"
+#include "BKE_DerivedMesh.h"
 
 #include "BIF_interface.h"
 #include "BIF_mywindow.h"
 #include "BIF_space.h"
 #include "BIF_toolbox.h"
 
+#include "BIF_editview.h" /* only for mouse_cursor - could remove this later */
+
 #include "BSE_drawipo.h"
 #include "BSE_node.h"
 #include "BSE_trans_types.h"
 #include "BSE_view.h"
+#include "BSE_drawview.h" /* view3d_test_clipping */
 
 #include "BDR_imagepaint.h"
 #include "BDR_vpaint.h"
-
+#include "BDR_gpencil.h"
 #include "GPU_draw.h"
 
 #include "GHOST_Types.h"
 
 /* Defines and Structs */
 
-#define IMAPAINT_CHAR_TO_FLOAT(c) (c/255.0f)
+#define IMAPAINT_CHAR_TO_FLOAT(c) ((c)/255.0f)
 
-#define IMAPAINT_FLOAT_RGB_TO_CHAR(c, f) { c[0]=FTOCHAR(f[0]); \
-       c[1]=FTOCHAR(f[1]); c[2]=FTOCHAR(f[2]); }
-#define IMAPAINT_CHAR_RGB_TO_FLOAT(f, c) { f[0]=IMAPAINT_CHAR_TO_FLOAT(c[0]); \
-       f[1]=IMAPAINT_CHAR_TO_FLOAT(c[1]); f[2]=IMAPAINT_CHAR_TO_FLOAT(c[2]); }
+#define IMAPAINT_FLOAT_RGB_TO_CHAR(c, f) { (c)[0]=FTOCHAR((f)[0]); (c)[1]=FTOCHAR((f)[1]); (c)[2]=FTOCHAR((f)[2]); }
+#define IMAPAINT_FLOAT_RGBA_TO_CHAR(c, f) { (c)[0]=FTOCHAR((f)[0]); (c)[1]=FTOCHAR((f)[1]); (c)[2]=FTOCHAR((f)[2]); (c)[3]=FTOCHAR((f)[3]); }
+
+#define IMAPAINT_CHAR_RGB_TO_FLOAT(f, c) { (f)[0]=IMAPAINT_CHAR_TO_FLOAT((c)[0]); (f)[1]=IMAPAINT_CHAR_TO_FLOAT((c)[1]); (f)[2]=IMAPAINT_CHAR_TO_FLOAT((c)[2]); }
+#define IMAPAINT_CHAR_RGBA_TO_FLOAT(f, c) { (f)[0]=IMAPAINT_CHAR_TO_FLOAT((c)[0]); (f)[1]=IMAPAINT_CHAR_TO_FLOAT((c)[1]); (f)[2]=IMAPAINT_CHAR_TO_FLOAT((c)[2]); (f)[3]=IMAPAINT_CHAR_TO_FLOAT((c)[3]); }
 #define IMAPAINT_FLOAT_RGB_COPY(a, b) VECCOPY(a, b)
 
 #define IMAPAINT_TILE_BITS                     6
 
 #define MAXUNDONAME    64
 
+static void imapaint_image_update(Image *image, ImBuf *ibuf, short texpaint);
+
+
 typedef struct ImagePaintState {
        Brush *brush;
        short tool, blend;
@@ -126,6 +138,192 @@ typedef struct ImagePaintState {
        float uv[2];
 } ImagePaintState;
 
+typedef struct ImagePaintPartialRedraw {
+       int x1, y1, x2, y2;
+       int enabled;
+} ImagePaintPartialRedraw;
+
+
+/* ProjectionPaint defines */
+
+/* approx the number of buckets to have under the brush,
+ * used with the brush size to set the ps->buckets_x and ps->buckets_y value.
+ * 
+ * When 3 - a brush should have ~9 buckets under it at once
+ * ...this helps for threading while painting as well as
+ * avoiding initializing pixels that wont touch the brush */
+#define PROJ_BUCKET_BRUSH_DIV 4
+
+#define PROJ_BUCKET_RECT_MIN 4
+#define PROJ_BUCKET_RECT_MAX 256
+
+#define PROJ_BOUNDBOX_DIV 8
+#define PROJ_BOUNDBOX_SQUARED  (PROJ_BOUNDBOX_DIV * PROJ_BOUNDBOX_DIV)
+
+//#define PROJ_DEBUG_PAINT 1
+//#define PROJ_DEBUG_NOSEAMBLEED 1
+//#define PROJ_DEBUG_PRINT_CLIP 1
+#define PROJ_DEBUG_WINCLIP 1
+
+/* projectFaceSeamFlags options */
+//#define PROJ_FACE_IGNORE     1<<0    /* When the face is hidden, backfacing or occluded */
+//#define PROJ_FACE_INIT       1<<1    /* When we have initialized the faces data */
+#define PROJ_FACE_SEAM1        1<<0    /* If this face has a seam on any of its edges */
+#define PROJ_FACE_SEAM2        1<<1
+#define PROJ_FACE_SEAM3        1<<2
+#define PROJ_FACE_SEAM4        1<<3
+
+#define PROJ_FACE_NOSEAM1      1<<4
+#define PROJ_FACE_NOSEAM2      1<<5
+#define PROJ_FACE_NOSEAM3      1<<6
+#define PROJ_FACE_NOSEAM4      1<<7
+
+/* a slightly scaled down face is used to get fake 3D location for edge pixels in the seams
+ * as this number approaches  1.0f the likelihood increases of float precision errors where
+ * it is occluded by an adjacent face */
+#define PROJ_FACE_SCALE_SEAM   0.99f
+
+#define PROJ_BUCKET_NULL               0
+#define PROJ_BUCKET_INIT               1<<0
+// #define PROJ_BUCKET_CLONE_INIT      1<<1
+
+/* used for testing doubles, if a point is on a line etc */
+#define PROJ_GEOM_TOLERANCE 0.0002f
+
+/* vert flags */
+#define PROJ_VERT_CULL 1
+
+#define PI_80_DEG ((M_PI_2 / 9) * 8)
+
+/* This is mainly a convenience struct used so we can keep an array of images we use
+ * Thir imbufs, etc, in 1 array, When using threads this array is copied for each thread
+ * because 'partRedrawRect' and 'touch' values would not be thread safe */
+typedef struct ProjPaintImage {
+       Image *ima;
+       ImBuf *ibuf;
+       ImagePaintPartialRedraw *partRedrawRect;
+       struct UndoTile **undoRect; /* only used to build undo tiles after painting */
+       int touch;
+} ProjPaintImage;
+
+/* Main projection painting struct passed to all projection painting functions */
+typedef struct ProjPaintState {
+       Brush *brush;
+       short tool, blend;
+       Object *ob;
+       /* end similarities with ImagePaintState */
+       
+       DerivedMesh    *dm;
+       int                     dm_totface;
+       int                     dm_totvert;
+       
+       MVert              *dm_mvert;
+       MFace              *dm_mface;
+       MTFace             *dm_mtface;
+       MTFace             *dm_mtface_clone;    /* other UV layer, use for cloning between layers */
+       MTFace             *dm_mtface_mask;
+       
+       /* projection painting only */
+       MemArena *arena_mt[BLENDER_MAX_THREADS];/* for multithreading, the first item is sometimes used for non threaded cases too */
+       LinkNode **bucketRect;                          /* screen sized 2D array, each pixel has a linked list of ProjPixel's */
+       LinkNode **bucketFaces;                         /* bucketRect aligned array linkList of faces overlapping each bucket */
+       unsigned char *bucketFlags;                                     /* store if the bucks have been initialized  */
+#ifndef PROJ_DEBUG_NOSEAMBLEED
+       char *faceSeamFlags;                            /* store info about faces, if they are initialized etc*/
+       float (*faceSeamUVs)[4][2];                     /* expanded UVs for faces to use as seams */
+       LinkNode **vertFaces;                           /* Only needed for when seam_bleed_px is enabled, use to find UV seams */
+       char *vertFlags;                                        /* store options per vert, now only store if the vert is pointing away from the view */
+#endif
+       int buckets_x;                                          /* The size of the bucket grid, the grid span's screenMin/screenMax so you can paint outsize the screen or with 2 brushes at once */
+       int buckets_y;
+       
+       ProjPaintImage *projImages;
+       
+       int image_tot;                          /* size of projectImages array */
+       
+       float (*screenCoords)[4];       /* verts projected into floating point screen space */
+       
+       float screenMin[2];                     /* 2D bounds for mesh verts on the screen's plane (screenspace) */
+       float screenMax[2]; 
+       float screen_width;                     /* Calculated from screenMin & screenMax */
+       float screen_height;
+       
+       /* options for projection painting */
+       int do_layer_clone;
+       int do_layer_mask;
+       int do_layer_mask_inv;
+       
+       short do_occlude;                       /* Use raytraced occlusion? - ortherwise will paint right through to the back*/
+       short do_backfacecull;  /* ignore faces with normals pointing away, skips a lot of raycasts if your normals are correctly flipped */
+       short do_mask_normal;                   /* mask out pixels based on their normals */
+       float normal_angle; /* what angle to mask at*/
+       
+       short is_ortho;
+       short is_airbrush;                                      /* only to avoid using (ps.brush->flag & BRUSH_AIRBRUSH) */
+       short is_texbrush;                                      /* only to avoid running  */
+#ifndef PROJ_DEBUG_NOSEAMBLEED
+       float seam_bleed_px;
+#endif
+       /* clone vars */
+       float cloneOffset[2];
+       
+       float projectMat[4][4];         /* Projection matrix, use for getting screen coords */
+       float viewMat[4][4];
+       float viewDir[3];                       /* View vector, use for do_backfacecull and for ray casting with an ortho viewport  */
+       float viewPos[3];                       /* View location in object relative 3D space, so can compare to verts  */
+       float clipsta, clipend;
+       
+       /* threads */
+       int thread_tot;
+       int bucketMin[2];
+       int bucketMax[2];
+       int context_bucket_x, context_bucket_y; /* must lock threads while accessing these */
+} ProjPaintState;
+
+typedef union pixelPointer
+{
+       float *f_pt;                    /* float buffer */
+       unsigned int *uint_pt; /* 2 ways to access a char buffer */
+       unsigned char *ch_pt;
+} PixelPointer;
+
+typedef union pixelStore
+{
+       unsigned char ch[4];
+       unsigned int uint;
+       float f[4];
+} PixelStore;
+
+typedef struct ProjPixel {
+       float projCoSS[2]; /* the floating point screen projection of this pixel */
+       
+       /* Only used when the airbrush is disabled.
+        * Store the max mask value to avoid painting over an area with a lower opacity
+        * with an advantage that we can avoid touching the pixel at all, if the 
+        * new mask value is lower then mask_max */
+       unsigned short mask_max;
+       
+       /* for various reasons we may want to mask out painting onto this pixel */
+       unsigned short mask;
+       
+       short x_px, y_px;
+       
+       PixelStore origColor;
+       PixelStore newColor;
+       PixelPointer pixel;
+       
+       short image_index; /* if anyone wants to paint onto more then 32768 images they can bite me */
+       unsigned char bb_cell_index;
+} ProjPixel;
+
+typedef struct ProjPixelClone {
+       struct ProjPixel __pp;
+       PixelStore clonepx;
+} ProjPixelClone;
+
+/* Finish projection painting structs */
+
+
 typedef struct UndoTile {
        struct UndoTile *next, *prev;
        ID id;
@@ -142,11 +340,6 @@ typedef struct UndoElem {
        ListBase tiles;
 } UndoElem;
 
-typedef struct ImagePaintPartialRedraw {
-       int x1, y1, x2, y2;
-       int enabled;
-} ImagePaintPartialRedraw;
-
 static ListBase undobase = {NULL, NULL};
 static UndoElem *curundo = NULL;
 static ImagePaintPartialRedraw imapaintpartial = {0, 0, 0, 0, 0};
@@ -161,14 +354,42 @@ static void undo_copy_tile(UndoTile *tile, ImBuf *tmpibuf, ImBuf *ibuf, int rest
        IMB_rectcpy(tmpibuf, ibuf, 0, 0, tile->x*IMAPAINT_TILE_SIZE,
                tile->y*IMAPAINT_TILE_SIZE, IMAPAINT_TILE_SIZE, IMAPAINT_TILE_SIZE);
 
-       if(ibuf->rect_float) SWAP(void*, tmpibuf->rect_float, tile->rect)
-       else SWAP(void*, tmpibuf->rect, tile->rect)
+       if(ibuf->rect_float) {
+               SWAP(void*, tmpibuf->rect_float, tile->rect);
+       } else {
+               SWAP(void*, tmpibuf->rect, tile->rect);
+       }
        
        if(restore)
                IMB_rectcpy(ibuf, tmpibuf, tile->x*IMAPAINT_TILE_SIZE,
                        tile->y*IMAPAINT_TILE_SIZE, 0, 0, IMAPAINT_TILE_SIZE, IMAPAINT_TILE_SIZE);
 }
 
+static UndoTile *undo_init_tile(ID *id, ImBuf *ibuf, ImBuf **tmpibuf, int x_tile, int y_tile)
+{
+       UndoTile *tile;
+       int allocsize;
+       
+       if (*tmpibuf==NULL)
+               *tmpibuf = IMB_allocImBuf(IMAPAINT_TILE_SIZE, IMAPAINT_TILE_SIZE, 32, IB_rectfloat|IB_rect, 0);
+       
+       tile= MEM_callocN(sizeof(UndoTile), "ImaUndoTile");
+       tile->id= *id;
+       tile->x= x_tile;
+       tile->y= y_tile;
+
+       allocsize= IMAPAINT_TILE_SIZE*IMAPAINT_TILE_SIZE*4;
+       allocsize *= (ibuf->rect_float)? sizeof(float): sizeof(char);
+       tile->rect= MEM_mapallocN(allocsize, "ImaUndoRect");
+
+       undo_copy_tile(tile, *tmpibuf, ibuf, 0);
+       curundo->undosize += allocsize;
+
+       BLI_addtail(&curundo->tiles, tile);
+       
+       return tile;
+}
+
 static void undo_restore(UndoElem *undo)
 {
        Image *ima = NULL;
@@ -190,100 +411,3497 @@ static void undo_restore(UndoElem *undo)
                                        break;
                }
 
-               ibuf= BKE_image_get_ibuf(ima, NULL);
+               ibuf= BKE_image_get_ibuf(ima, NULL);
+
+               if (!ima || !ibuf || !(ibuf->rect || ibuf->rect_float))
+                       continue;
+
+               undo_copy_tile(tile, tmpibuf, ibuf, 1);
+
+               GPU_free_image(ima); /* force OpenGL reload */
+               if(ibuf->rect_float)
+                       imb_freerectImBuf(ibuf); /* force recreate of char rect */
+       }
+
+       IMB_freeImBuf(tmpibuf);
+}
+
+static void undo_free(UndoElem *undo)
+{
+       UndoTile *tile;
+
+       for(tile=undo->tiles.first; tile; tile=tile->next)
+               MEM_freeN(tile->rect);
+       BLI_freelistN(&undo->tiles);
+}
+
+static void undo_imagepaint_push_begin(char *name)
+{
+       UndoElem *uel;
+       int nr;
+       
+       /* Undo push is split up in begin and end, the reason is that as painting
+        * happens more tiles are added to the list, and at the very end we know
+        * how much memory the undo used to remove old undo elements */
+
+       /* remove all undos after (also when curundo==NULL) */
+       while(undobase.last != curundo) {
+               uel= undobase.last;
+               undo_free(uel);
+               BLI_freelinkN(&undobase, uel);
+       }
+       
+       /* make new */
+       curundo= uel= MEM_callocN(sizeof(UndoElem), "undo file");
+       BLI_addtail(&undobase, uel);
+
+       /* name can be a dynamic string */
+       strncpy(uel->name, name, MAXUNDONAME-1);
+       
+       /* limit amount to the maximum amount*/
+       nr= 0;
+       uel= undobase.last;
+       while(uel) {
+               nr++;
+               if(nr==U.undosteps) break;
+               uel= uel->prev;
+       }
+       if(uel) {
+               while(undobase.first!=uel) {
+                       UndoElem *first= undobase.first;
+                       undo_free(first);
+                       BLI_freelinkN(&undobase, first);
+               }
+       }
+}
+
+static void undo_imagepaint_push_end()
+{
+       UndoElem *uel;
+       unsigned long totmem, maxmem;
+
+       if(U.undomemory != 0) {
+               /* limit to maximum memory (afterwards, we can't know in advance) */
+               totmem= 0;
+               maxmem= ((unsigned long)U.undomemory)*1024*1024;
+
+               uel= undobase.last;
+               while(uel) {
+                       totmem+= uel->undosize;
+                       if(totmem>maxmem) break;
+                       uel= uel->prev;
+               }
+
+               if(uel) {
+                       while(undobase.first!=uel) {
+                               UndoElem *first= undobase.first;
+                               undo_free(first);
+                               BLI_freelinkN(&undobase, first);
+                       }
+               }
+       }
+}
+
+/* fast projection bucket array lookup, use the safe version for bound checking  */
+static int project_bucket_offset(const ProjPaintState *ps, const float projCoSS[2])
+{
+       /* If we were not dealing with screenspace 2D coords we could simple do...
+        * ps->bucketRect[x + (y*ps->buckets_y)] */
+       
+       /* please explain?
+        * projCoSS[0] - ps->screenMin[0]       : zero origin
+        * ... / ps->screen_width                               : range from 0.0 to 1.0
+        * ... * ps->buckets_x          : use as a bucket index
+        *
+        * Second multiplication does similar but for vertical offset
+        */
+       return  (       (int)(((projCoSS[0] - ps->screenMin[0]) / ps->screen_width)  * ps->buckets_x)) + 
+               (       (       (int)(((projCoSS[1] - ps->screenMin[1])  / ps->screen_height) * ps->buckets_y)) * ps->buckets_x);
+}
+
+static int project_bucket_offset_safe(const ProjPaintState *ps, const float projCoSS[2])
+{
+       int bucket_index = project_bucket_offset(ps, projCoSS);
+       
+       if (bucket_index < 0 || bucket_index >= ps->buckets_x*ps->buckets_y) {  
+               return -1;
+       }
+       else {
+               return bucket_index;
+       }
+}
+
+#define SIDE_OF_LINE(pa, pb, pp)       ((pa[0]-pp[0])*(pb[1]-pp[1]))-((pb[0]-pp[0])*(pa[1]-pp[1]))
+
+static float AreaSignedF2Dfl(float *v1, float *v2, float *v3)
+{
+   return (float)(0.5f*((v1[0]-v2[0])*(v2[1]-v3[1]) +
+(v1[1]-v2[1])*(v3[0]-v2[0])));
+}
+
+static void BarycentricWeights2f(float v1[2], float v2[2], float v3[2], float pt[2], float w[3])
+{
+   float wtot_inv, wtot, wsign[3];
+
+   wsign[0] = AreaSignedF2Dfl(v2, v3, pt);
+   wsign[1] = AreaSignedF2Dfl(v3, v1, pt);
+   wsign[2] = AreaSignedF2Dfl(v1, v2, pt);
+   wtot = wsign[0]+wsign[1]+wsign[2];
+
+   if (fabs(wtot) > 0.0f) {
+       wtot_inv = 1.0f/wtot;
+
+       w[0] = wsign[0]*wtot_inv;
+       w[1] = wsign[1]*wtot_inv;
+       w[2] = wsign[2]*wtot_inv;
+   }
+   else /* dummy values for zero area face */
+       w[0] = w[1] = w[2] = 1.0f/3.0f;
+}
+
+/* still use 2D X,Y space but this works for verts transformed by a perspective matrix, using their 4th component as a weight */
+
+static void BarycentricWeightsPersp2f(float v1[4], float v2[4], float v3[4], float pt[2], float w[3])
+{
+       float persp_tot, persp_tot_inv;
+       BarycentricWeights2f(v1, v2, v3, pt, w);
+       
+       w[0] /= v1[3];
+       w[1] /= v2[3];
+       w[2] /= v3[3];
+       
+       persp_tot = w[0]+w[1]+w[2];
+       if (persp_tot > 0.0f) {
+               persp_tot_inv = 1.0f / persp_tot;
+               w[0] *= persp_tot_inv;
+               w[1] *= persp_tot_inv;
+               w[2] *= persp_tot_inv;
+       }
+       else {
+               w[0] = w[1] = w[2] = 1.0f/3.0f; /* dummy values for zero area face */
+       }
+}
+
+static void VecWeightf(float p[3], const float v1[3], const float v2[3], const float v3[3], const float w[3])
+{
+       p[0] = v1[0]*w[0] + v2[0]*w[1] + v3[0]*w[2];
+       p[1] = v1[1]*w[0] + v2[1]*w[1] + v3[1]*w[2];
+       p[2] = v1[2]*w[0] + v2[2]*w[1] + v3[2]*w[2];
+}
+
+static void Vec2Weightf(float p[2], const float v1[2], const float v2[2], const float v3[2], const float w[3])
+{
+       p[0] = v1[0]*w[0] + v2[0]*w[1] + v3[0]*w[2];
+       p[1] = v1[1]*w[0] + v2[1]*w[1] + v3[1]*w[2];
+}
+
+static float tri_depth_2d(float v1[3], float v2[3], float v3[3], float pt[2], float w[3])
+{
+       BarycentricWeights2f(v1, v2, v3, pt, w);
+       return (v1[2]*w[0]) + (v2[2]*w[1]) + (v3[2]*w[2]);
+}
+
+
+/* Return the top-most face index that the screen space coord 'pt' touches (or -1) */
+static int project_paint_PickFace(const ProjPaintState *ps, float pt[2], float w[3], int *side)
+{
+       LinkNode *node;
+       float w_tmp[3];
+       float *v1, *v2, *v3, *v4;
+       int bucket_index;
+       int face_index;
+       int best_side = -1;
+       int best_face_index = -1;
+       float z_depth_best = MAXFLOAT, z_depth;
+       MFace *mf;
+       
+       bucket_index = project_bucket_offset_safe(ps, pt);
+       if (bucket_index==-1)
+               return -1;
+       
+       
+       
+       /* we could return 0 for 1 face buckets, as long as this function assumes
+        * that the point its testing is only every originated from an existing face */
+       
+       for (node= ps->bucketFaces[bucket_index]; node; node= node->next) {
+               face_index = GET_INT_FROM_POINTER(node->link);
+               mf= ps->dm_mface + face_index;
+               
+               v1= ps->screenCoords[mf->v1];
+               v2= ps->screenCoords[mf->v2];
+               v3= ps->screenCoords[mf->v3];
+               
+               if (IsectPT2Df(pt, v1, v2, v3)) {
+                       z_depth= tri_depth_2d(v1, v2, v3, pt, w_tmp);
+                       if (z_depth < z_depth_best) {
+                               best_face_index = face_index;
+                               best_side = 0;
+                               z_depth_best = z_depth;
+                               VECCOPY(w, w_tmp);
+                       }
+               }
+               else if (mf->v4) {
+                       v4= ps->screenCoords[mf->v4];
+                       
+                       if (IsectPT2Df(pt, v1, v3, v4)) {
+                               z_depth= tri_depth_2d(v1, v3, v4, pt, w_tmp);
+                               if (z_depth < z_depth_best) {
+                                       best_face_index = face_index;
+                                       best_side= 1;
+                                       z_depth_best = z_depth;
+                                       VECCOPY(w, w_tmp);
+                               }
+                       }
+               }
+       }
+       
+       *side = best_side;
+       return best_face_index; /* will be -1 or a valid face */
+}
+
+/* Converts a uv coord into a pixel location wrapping if the uv is outside 0-1 range */
+static void uvco_to_wrapped_pxco(float uv[2], int ibuf_x, int ibuf_y, float *x, float *y)
+{
+       /* use */
+       *x = (float)fmod(uv[0], 1.0f);
+       *y = (float)fmod(uv[1], 1.0f);
+       
+       if (*x < 0.0f) *x += 1.0f;
+       if (*y < 0.0f) *y += 1.0f;
+       
+       *x = *x * ibuf_x - 0.5f;
+       *y = *y * ibuf_y - 0.5f;
+}
+
+/* Set the top-most face color that the screen space coord 'pt' touches (or return 0 if none touch) */
+static int project_paint_PickColor(const ProjPaintState *ps, float pt[2], float *rgba_fp, unsigned char *rgba, const int interp)
+{
+       float w[3], uv[2];
+       int side;
+       int face_index;
+       MTFace *tf;
+       ImBuf *ibuf;
+       int xi, yi;
+       
+       
+       face_index = project_paint_PickFace(ps, pt, w, &side);
+       
+       if (face_index == -1)
+               return 0;
+       
+       tf = ps->dm_mtface + face_index;
+       
+       if (side == 0) {
+               Vec2Weightf(uv, tf->uv[0], tf->uv[1], tf->uv[2], w);
+       }
+       else { /* QUAD */
+               Vec2Weightf(uv, tf->uv[0], tf->uv[2], tf->uv[3], w);
+       }
+       
+       ibuf = BKE_image_get_ibuf((Image *)tf->tpage, NULL); /* TODO - this may be slow, the only way around it is to have an ibuf index per face */
+       
+
+       
+       if (interp) {
+               float x, y;
+               uvco_to_wrapped_pxco(uv, ibuf->x, ibuf->y, &x, &y);
+               
+               if (ibuf->rect_float) {
+                       if (rgba_fp) {
+                               bilinear_interpolation_color(ibuf, NULL, rgba_fp, x, y);
+                       }
+                       else {
+                               float rgba_tmp_f[4];
+                               bilinear_interpolation_color(ibuf, NULL, rgba_tmp_f, x, y);
+                               IMAPAINT_FLOAT_RGBA_TO_CHAR(rgba, rgba_tmp_f);
+                       }
+               }
+               else {
+                       if (rgba) {
+                               bilinear_interpolation_color(ibuf, rgba, NULL, x, y);
+                       }
+                       else {
+                               unsigned char rgba_tmp[4];
+                               bilinear_interpolation_color(ibuf, rgba_tmp, NULL, x, y);
+                               IMAPAINT_CHAR_RGBA_TO_FLOAT(rgba_fp, rgba_tmp);
+                       }
+               }
+       }
+       else {
+               xi = (uv[0]*ibuf->x) + 0.5f;
+               yi = (uv[1]*ibuf->y) + 0.5f;
+               
+               //if (xi<0 || xi>=ibuf->x  ||  yi<0 || yi>=ibuf->y) return 0;
+               
+               /* wrap */
+               xi = ((int)(uv[0]*ibuf->x)) % ibuf->x;
+               if (xi<0) xi += ibuf->x;
+               yi = ((int)(uv[1]*ibuf->y)) % ibuf->y;
+               if (yi<0) yi += ibuf->y;
+               
+               
+               if (rgba) {
+                       if (ibuf->rect_float) {
+                               float *rgba_tmp_fp = ibuf->rect_float + (xi + yi * ibuf->x * 4);
+                               IMAPAINT_FLOAT_RGBA_TO_CHAR(rgba, rgba_tmp_fp);
+                       }
+                       else {
+                               *((unsigned int *)rgba) = *(unsigned int *)(((char *)ibuf->rect) + ((xi + yi * ibuf->x) * 4));
+                       }
+               }
+               
+               if (rgba_fp) {
+                       if (ibuf->rect_float) {
+                               QUATCOPY(rgba_fp, ((float *)ibuf->rect_float + ((xi + yi * ibuf->x) * 4)));
+                       }
+                       else {
+                               char *tmp_ch= ((char *)ibuf->rect) + ((xi + yi * ibuf->x) * 4);
+                               IMAPAINT_CHAR_RGBA_TO_FLOAT(rgba_fp, tmp_ch);
+                       }
+               }
+       }
+       return 1;
+}
+
+/* Check if 'pt' is infront of the 3 verts on the Z axis (used for screenspace occlusuion test)
+ * return...
+ *  0  : no occlusion
+ * -1  : no occlusion but 2D intersection is true (avoid testing the other half of a quad)
+ *  1  : occluded */
+
+static int project_paint_occlude_ptv(float pt[3], float v1[3], float v2[3], float v3[3])
+{
+       /* if all are behind us, return false */
+       if(v1[2] > pt[2] && v2[2] > pt[2] && v3[2] > pt[2])
+               return 0;
+               
+       /* do a 2D point in try intersection */
+       if (!IsectPT2Df(pt, v1, v2, v3))
+               return 0; /* we know there is  */
+       
+
+       /* From here on we know there IS an intersection */
+       /* if ALL of the verts are infront of us then we know it intersects ? */
+       if(v1[2] < pt[2] && v2[2] < pt[2] && v3[2] < pt[2]) {
+               return 1;
+       }
+       else {
+               float w[3];
+               /* we intersect? - find the exact depth at the point of intersection */
+               if (tri_depth_2d(v1, v2, v3, pt, w) < pt[2]) {
+                       return 1; /* This point is occluded by another face */
+               }
+       }
+       return -1;
+}
+
+
+static int project_paint_occlude_ptv_clip(
+               const ProjPaintState *ps, const MFace *mf,
+               float pt[3], float v1[3], float v2[3], float v3[3],
+               const int side )
+{
+       float w[3], wco[3];
+       
+       /* if all are behind us, return false */
+       if(v1[2] > pt[2] && v2[2] > pt[2] && v3[2] > pt[2])
+               return 0;
+               
+       /* do a 2D point in try intersection */
+       if (!IsectPT2Df(pt, v1, v2, v3))
+               return 0; /* we know there is  */
+       
+       /* we intersect? - find the exact depth at the point of intersection */
+       if (tri_depth_2d(v1, v2, v3, pt, w) > pt[2])
+               return -1;
+       
+       if (side)       VecWeightf(wco, ps->dm_mvert[mf->v1].co, ps->dm_mvert[mf->v3].co, ps->dm_mvert[mf->v4].co, w);
+       else            VecWeightf(wco, ps->dm_mvert[mf->v1].co, ps->dm_mvert[mf->v2].co, ps->dm_mvert[mf->v3].co, w);
+       
+       Mat4MulVecfl(ps->ob->obmat, wco);
+       if(!view3d_test_clipping(G.vd, wco)) {
+               return 1;
+       }
+       
+       return -1;
+}
+
+
+/* Check if a screenspace location is occluded by any other faces
+ * check, pixelScreenCo must be in screenspace, its Z-Depth only needs to be used for comparison
+ * and dosn't need to be correct in relation to X and Y coords (this is the case in perspective view) */
+static int project_bucket_point_occluded(const ProjPaintState *ps, LinkNode *bucketFace, const int orig_face, float pixelScreenCo[4])
+{
+       MFace *mf;
+       int face_index;
+       int isect_ret;
+       
+       /* we could return 0 for 1 face buckets, as long as this function assumes
+        * that the point its testing is only every originated from an existing face */
+       
+       if(G.vd->flag & V3D_CLIPPING) {
+               for (; bucketFace; bucketFace = bucketFace->next) {
+                       face_index = (int)bucketFace->link;
+                       
+                       if (orig_face != face_index) {
+                               mf = ps->dm_mface + face_index;
+                                       isect_ret = project_paint_occlude_ptv_clip(ps, mf, pixelScreenCo, ps->screenCoords[mf->v1], ps->screenCoords[mf->v2], ps->screenCoords[mf->v3], 0);
+                                       
+                                       /* Note, if isect_ret==-1 then we dont want to test the other side of the quad */
+                                       if (isect_ret==0 && mf->v4) {
+                                               isect_ret = project_paint_occlude_ptv_clip(ps, mf, pixelScreenCo, ps->screenCoords[mf->v1], ps->screenCoords[mf->v3], ps->screenCoords[mf->v4], 1);
+                                       }
+                               if (isect_ret==1) {
+                                       /* TODO - we may want to cache the first hit,
+                                        * it is not possible to swap the face order in the list anymore */
+                                       return 1; 
+                               }
+                       }
+               }
+       }
+       else {
+               for (; bucketFace; bucketFace = bucketFace->next) {
+                       face_index = (int)bucketFace->link;
+                       
+                       if (orig_face != face_index) {
+                               mf = ps->dm_mface + face_index;
+                                       isect_ret = project_paint_occlude_ptv(pixelScreenCo, ps->screenCoords[mf->v1], ps->screenCoords[mf->v2], ps->screenCoords[mf->v3]);
+                                       
+                                       /* Note, if isect_ret==-1 then we dont want to test the other side of the quad */
+                                       if (isect_ret==0 && mf->v4) {
+                                               isect_ret = project_paint_occlude_ptv(pixelScreenCo, ps->screenCoords[mf->v1], ps->screenCoords[mf->v3], ps->screenCoords[mf->v4]);
+                                       }
+                               if (isect_ret==1) {
+                                       /* TODO - we may want to cache the first hit,
+                                        * it is not possible to swap the face order in the list anymore */
+                                       return 1; 
+                               }
+                       }
+               }
+       }
+       return 0;
+}
+
+/* basic line intersection, could move to arithb.c, 2 points with a horiz line
+ * 1 for an intersection, 2 if the first point is aligned, 3 if the second point is aligned */
+#define ISECT_TRUE 1
+#define ISECT_TRUE_P1 2
+#define ISECT_TRUE_P2 3
+static int line_isect_y(const float p1[2], const float p2[2], const float y_level, float *x_isect)
+{
+       float y_diff;
+       
+       if (y_level==p1[1]) { /* are we touching the first point? - no interpolation needed */
+               *x_isect = p1[0];
+               return ISECT_TRUE_P1;
+       }
+       if (y_level==p2[1]) { /* are we touching the second point? - no interpolation needed */
+               *x_isect = p2[0];
+               return ISECT_TRUE_P2;
+       }
+       
+       y_diff= fabs(p1[1]-p2[1]); /* yuck, horizontal line, we cant do much here */
+       
+       if (y_diff < 0.000001f) {
+               *x_isect = (p1[0]+p2[0]) * 0.5f;
+               return ISECT_TRUE;              
+       }
+       
+       if (p1[1] > y_level && p2[1] < y_level) {
+               *x_isect = (p2[0]*(p1[1]-y_level) + p1[0]*(y_level-p2[1])) / y_diff;  /*(p1[1]-p2[1]);*/
+               return ISECT_TRUE;
+       }
+       else if (p1[1] < y_level && p2[1] > y_level) {
+               *x_isect = (p2[0]*(y_level-p1[1]) + p1[0]*(p2[1]-y_level)) / y_diff;  /*(p2[1]-p1[1]);*/
+               return ISECT_TRUE;
+       }
+       else {
+               return 0;
+       }
+}
+
+static int line_isect_x(const float p1[2], const float p2[2], const float x_level, float *y_isect)
+{
+       float x_diff;
+       
+       if (x_level==p1[0]) { /* are we touching the first point? - no interpolation needed */
+               *y_isect = p1[1];
+               return ISECT_TRUE_P1;
+       }
+       if (x_level==p2[0]) { /* are we touching the second point? - no interpolation needed */
+               *y_isect = p2[1];
+               return ISECT_TRUE_P2;
+       }
+       
+       x_diff= fabs(p1[0]-p2[0]); /* yuck, horizontal line, we cant do much here */
+       
+       if (x_diff < 0.000001) { /* yuck, vertical line, we cant do much here */
+               *y_isect = (p1[0]+p2[0]) * 0.5f;
+               return ISECT_TRUE;              
+       }
+       
+       if (p1[0] > x_level && p2[0] < x_level) {
+               *y_isect = (p2[1]*(p1[0]-x_level) + p1[1]*(x_level-p2[0])) / x_diff; /*(p1[0]-p2[0]);*/
+               return ISECT_TRUE;
+       }
+       else if (p1[0] < x_level && p2[0] > x_level) {
+               *y_isect = (p2[1]*(x_level-p1[0]) + p1[1]*(p2[0]-x_level)) / x_diff; /*(p2[0]-p1[0]);*/
+               return ISECT_TRUE;
+       }
+       else {
+               return 0;
+       }
+}
+
+/* simple func use for comparing UV locations to check if there are seams.
+ * Its possible this gives incorrect results, when the UVs for 1 face go into the next 
+ * tile, but do not do this for the adjacent face, it could return a false positive.
+ * This is so unlikely that Id not worry about it. */
+static int cmp_uv(const float vec2a[2], const float vec2b[2])
+{
+       /* if the UV's are not between 0.0 and 1.0 */
+       float xa = (float)fmod(vec2a[0], 1.0f);
+       float ya = (float)fmod(vec2a[1], 1.0f);
+       
+       float xb = (float)fmod(vec2b[0], 1.0f);
+       float yb = (float)fmod(vec2b[1], 1.0f); 
+       
+       if (xa < 0.0f) xa += 1.0f;
+       if (ya < 0.0f) ya += 1.0f;
+       
+       if (xb < 0.0f) xb += 1.0f;
+       if (yb < 0.0f) yb += 1.0f;
+       
+       return ((fabs(xa-xb) < PROJ_GEOM_TOLERANCE) && (fabs(ya-yb) < PROJ_GEOM_TOLERANCE)) ? 1:0;
+}
+
+
+/* set min_px and max_px to the image space bounds of the UV coords 
+ * return zero if there is no area in the returned rectangle */
+static int pixel_bounds_uv(
+               const float uv1[2], const float uv2[2], const float uv3[2], const float uv4[2],
+               rcti *bounds_px,
+               const int ibuf_x, const int ibuf_y,
+               int is_quad
+) {
+       float min_uv[2], max_uv[2]; /* UV bounds */
+       
+       INIT_MINMAX2(min_uv, max_uv);
+       
+       DO_MINMAX2(uv1, min_uv, max_uv);
+       DO_MINMAX2(uv2, min_uv, max_uv);
+       DO_MINMAX2(uv3, min_uv, max_uv);
+       if (is_quad)
+               DO_MINMAX2(uv4, min_uv, max_uv);
+       
+       bounds_px->xmin = (int)(ibuf_x * min_uv[0]);
+       bounds_px->ymin = (int)(ibuf_y * min_uv[1]);
+       
+       bounds_px->xmax = (int)(ibuf_x * max_uv[0]) +1;
+       bounds_px->ymax = (int)(ibuf_y * max_uv[1]) +1;
+       
+       /*printf("%d %d %d %d \n", min_px[0], min_px[1], max_px[0], max_px[1]);*/
+       
+       /* face uses no UV area when quantized to pixels? */
+       return (bounds_px->xmin == bounds_px->xmax || bounds_px->ymin == bounds_px->ymax) ? 0 : 1;
+}
+
+static int pixel_bounds_array(float (* uv)[2], rcti *bounds_px, const int ibuf_x, const int ibuf_y, int tot)
+{
+       float min_uv[2], max_uv[2]; /* UV bounds */
+       
+       if (tot==0) {
+               return 0;
+       }
+       
+       INIT_MINMAX2(min_uv, max_uv);
+       
+       while (tot--) {
+               DO_MINMAX2((*uv), min_uv, max_uv);
+               uv++;
+       }
+       
+       bounds_px->xmin = (int)(ibuf_x * min_uv[0]);
+       bounds_px->ymin = (int)(ibuf_y * min_uv[1]);
+       
+       bounds_px->xmax = (int)(ibuf_x * max_uv[0]) +1;
+       bounds_px->ymax = (int)(ibuf_y * max_uv[1]) +1;
+       
+       /*printf("%d %d %d %d \n", min_px[0], min_px[1], max_px[0], max_px[1]);*/
+       
+       /* face uses no UV area when quantized to pixels? */
+       return (bounds_px->xmin == bounds_px->xmax || bounds_px->ymin == bounds_px->ymax) ? 0 : 1;
+}
+
+#ifndef PROJ_DEBUG_NOSEAMBLEED
+
+/* This function returns 1 if this face has a seam along the 2 face-vert indicies
+ * 'orig_i1_fidx' and 'orig_i2_fidx' */
+static int check_seam(const ProjPaintState *ps, const int orig_face, const int orig_i1_fidx, const int orig_i2_fidx, int *other_face, int *orig_fidx)
+{
+       LinkNode *node;
+       int face_index;
+       int i1, i2;
+       int i1_fidx = -1, i2_fidx = -1; /* index in face */
+       MFace *mf;
+       MTFace *tf;
+       const MFace *orig_mf = ps->dm_mface + orig_face;  
+       const MTFace *orig_tf = ps->dm_mtface + orig_face;
+       
+       /* vert indicies from face vert order indicies */
+       i1 = (*(&orig_mf->v1 + orig_i1_fidx));
+       i2 = (*(&orig_mf->v1 + orig_i2_fidx));
+       
+       for (node = ps->vertFaces[i1]; node; node = node->next) {
+               face_index = (int)node->link;
+               if (face_index != orig_face) {
+                       mf = ps->dm_mface + face_index;
+                       /* could check if the 2 faces images match here,
+                        * but then there wouldn't be a way to return the opposite face's info */
+                       
+                       
+                       /* We need to know the order of the verts in the adjacent face 
+                        * set the i1_fidx and i2_fidx to (0,1,2,3) */
+                       if              (mf->v1==i1)                    i1_fidx = 0;
+                       else if (mf->v2==i1)                    i1_fidx = 1;
+                       else if (mf->v3==i1)                    i1_fidx = 2;
+                       else if (mf->v4 && mf->v4==i1)  i1_fidx = 3;
+                       
+                       if              (mf->v1==i2)                    i2_fidx = 0;
+                       else if (mf->v2==i2)                    i2_fidx = 1;
+                       else if (mf->v3==i2)                    i2_fidx = 2;
+                       else if (mf->v4 && mf->v4==i2)  i2_fidx = 3;
+                       
+                       /* Only need to check if 'i2_fidx' is valid because we know i1_fidx is the same vert on both faces */
+                       if (i2_fidx != -1) {
+                               /* This IS an adjacent face!, now lets check if the UVs are ok */
+                               tf = ps->dm_mtface + face_index;
+                               
+                               /* set up the other face */
+                               *other_face = face_index;
+                               *orig_fidx = (i1_fidx < i2_fidx) ? i1_fidx : i2_fidx;
+                               
+                               /* first test if they have the same image */
+                               if (    (orig_tf->tpage == tf->tpage) &&
+                                               cmp_uv(orig_tf->uv[orig_i1_fidx], tf->uv[i1_fidx]) &&
+                                               cmp_uv(orig_tf->uv[orig_i2_fidx], tf->uv[i2_fidx]) )
+                               {
+                                       // printf("SEAM (NONE)\n");
+                                       return 0;
+                                       
+                               }
+                               else {
+                                       // printf("SEAM (UV GAP)\n");
+                                       return 1;
+                               }
+                       }
+               }
+       }
+       // printf("SEAM (NO FACE)\n");
+       *other_face = -1;
+       return 1;
+}
+
+/* TODO - move to arithb.c */
+/* Converts an angle to a length that can be used for maintaining an even margin around UV's */
+static float angleToLength(float angle)
+{
+       // already accounted for
+       if (angle < 0.000001f) {
+               return 1.0f;
+       }
+       else {
+               return fabs(1.0f / cos(angle * (M_PI/180.0f)));
+       }
+}
+
+/* Calculate outset UV's, this is not the same as simply scaling the UVs,
+ * since the outset coords are a margin that keep an even distance from the original UV's,
+ * note that the image aspect is taken into account */
+static void uv_image_outset(float (*orig_uv)[2], float (*outset_uv)[2], const float scaler, const int ibuf_x, const int ibuf_y, const int is_quad)
+{
+       float a1, a2, a3, a4=0.0f;
+       float puv[4][2]; /* pixelspace uv's */
+       float no1[2], no2[2], no3[2], no4[2]; /* normals */
+       float dir1[2], dir2[2], dir3[2], dir4[2];
+       float ibuf_x_inv = 1.0f / (float)ibuf_x; 
+       float ibuf_y_inv = 1.0f / (float)ibuf_y; 
+       
+       /* make UV's in pixel space so we can */
+       puv[0][0] = orig_uv[0][0] * ibuf_x;
+       puv[0][1] = orig_uv[0][1] * ibuf_y;
+       
+       puv[1][0] = orig_uv[1][0] * ibuf_x;
+       puv[1][1] = orig_uv[1][1] * ibuf_y;
+       
+       puv[2][0] = orig_uv[2][0] * ibuf_x;
+       puv[2][1] = orig_uv[2][1] * ibuf_y;
+       
+       if (is_quad) {
+               puv[3][0] = orig_uv[3][0] * ibuf_x;
+               puv[3][1] = orig_uv[3][1] * ibuf_y;
+       }
+       
+       /* face edge directions */
+       Vec2Subf(dir1, puv[1], puv[0]);
+       Vec2Subf(dir2, puv[2], puv[1]);
+       Normalize2(dir1);
+       Normalize2(dir2);
+       
+       if (is_quad) {
+               Vec2Subf(dir3, puv[3], puv[2]);
+               Vec2Subf(dir4, puv[0], puv[3]);
+               Normalize2(dir3);
+               Normalize2(dir4);
+       }
+       else {
+               Vec2Subf(dir3, puv[0], puv[2]);
+               Normalize2(dir3);
+       }
+       
+       if (is_quad) {
+               a1 = angleToLength(NormalizedVecAngle2_2D(dir4, dir1));
+               a2 = angleToLength(NormalizedVecAngle2_2D(dir1, dir2));
+               a3 = angleToLength(NormalizedVecAngle2_2D(dir2, dir3));
+               a4 = angleToLength(NormalizedVecAngle2_2D(dir3, dir4));
+       }
+       else {
+               a1 = angleToLength(NormalizedVecAngle2_2D(dir3, dir1));
+               a2 = angleToLength(NormalizedVecAngle2_2D(dir1, dir2));
+               a3 = angleToLength(NormalizedVecAngle2_2D(dir2, dir3));
+       }
+       
+       if (is_quad) {
+               Vec2Subf(no1, dir4, dir1);
+               Vec2Subf(no2, dir1, dir2);
+               Vec2Subf(no3, dir2, dir3);
+               Vec2Subf(no4, dir3, dir4);
+               Normalize2(no1);
+               Normalize2(no2);
+               Normalize2(no3);
+               Normalize2(no4);
+               Vec2Mulf(no1, a1*scaler);
+               Vec2Mulf(no2, a2*scaler);
+               Vec2Mulf(no3, a3*scaler);
+               Vec2Mulf(no4, a4*scaler);
+               Vec2Addf(outset_uv[0], puv[0], no1);
+               Vec2Addf(outset_uv[1], puv[1], no2);
+               Vec2Addf(outset_uv[2], puv[2], no3);
+               Vec2Addf(outset_uv[3], puv[3], no4);
+               outset_uv[0][0] *= ibuf_x_inv;
+               outset_uv[0][1] *= ibuf_y_inv;
+               
+               outset_uv[1][0] *= ibuf_x_inv;
+               outset_uv[1][1] *= ibuf_y_inv;
+               
+               outset_uv[2][0] *= ibuf_x_inv;
+               outset_uv[2][1] *= ibuf_y_inv;
+               
+               outset_uv[3][0] *= ibuf_x_inv;
+               outset_uv[3][1] *= ibuf_y_inv;
+       }
+       else {
+               Vec2Subf(no1, dir3, dir1);
+               Vec2Subf(no2, dir1, dir2);
+               Vec2Subf(no3, dir2, dir3);
+               Normalize2(no1);
+               Normalize2(no2);
+               Normalize2(no3);
+               Vec2Mulf(no1, a1*scaler);
+               Vec2Mulf(no2, a2*scaler);
+               Vec2Mulf(no3, a3*scaler);
+               Vec2Addf(outset_uv[0], puv[0], no1);
+               Vec2Addf(outset_uv[1], puv[1], no2);
+               Vec2Addf(outset_uv[2], puv[2], no3);
+               outset_uv[0][0] *= ibuf_x_inv;
+               outset_uv[0][1] *= ibuf_y_inv;
+               
+               outset_uv[1][0] *= ibuf_x_inv;
+               outset_uv[1][1] *= ibuf_y_inv;
+               
+               outset_uv[2][0] *= ibuf_x_inv;
+               outset_uv[2][1] *= ibuf_y_inv;
+       }
+}
+
+/* 
+ * Be tricky with flags, first 4 bits are PROJ_FACE_SEAM1 to 4, last 4 bits are PROJ_FACE_NOSEAM1 to 4
+ * 1<<i - where i is (0-3) 
+ * 
+ * If we're multithreadng, make sure threads are locked when this is called
+ */
+static void project_face_seams_init(const ProjPaintState *ps, const int face_index, const int is_quad)
+{
+       int other_face, other_fidx; /* vars for the other face, we also set its flag */
+       int fidx1 = is_quad ? 3 : 2;
+       int fidx2 = 0; /* next fidx in the face (0,1,2,3) -> (1,2,3,0) or (0,1,2) -> (1,2,0) for a tri */
+       
+       do {
+               if ((ps->faceSeamFlags[face_index] & (1<<fidx1|16<<fidx1)) == 0) {
+                       if (check_seam(ps, face_index, fidx1, fidx2, &other_face, &other_fidx)) {
+                               ps->faceSeamFlags[face_index] |= 1<<fidx1;
+                               if (other_face != -1)
+                                       ps->faceSeamFlags[other_face] |= 1<<other_fidx;
+                       }
+                       else {
+                               ps->faceSeamFlags[face_index] |= 16<<fidx1;
+                               if (other_face != -1)
+                                       ps->faceSeamFlags[other_face] |= 16<<other_fidx; /* second 4 bits for disabled */
+                       }
+               }
+               
+               fidx2 = fidx1;
+       } while (fidx1--);
+}
+#endif // PROJ_DEBUG_NOSEAMBLEED
+
+
+/* TODO - move to arithb.c */
+
+/* little sister we only need to know lambda */
+static float lambda_cp_line2(const float p[2], const float l1[2], const float l2[2])
+{
+       float h[2], u[2];
+       
+       u[0] = l2[0] - l1[0];
+       u[1] = l2[1] - l1[1];
+
+       h[0] = p[0] - l1[0];
+       h[1] = p[1] - l1[1];
+       
+       return(Inp2f(u, h)/Inp2f(u, u));
+}
+
+
+/* Converts a UV location to a 3D screenspace location
+ * Takes a 'uv' and 3 UV coords, and sets the values of pixelScreenCo
+ * 
+ * This is used for finding a pixels location in screenspace for painting */
+static void screen_px_from_ortho(
+               float uv[2],
+               float v1co[3], float v2co[3], float v3co[3], /* Screenspace coords */
+               float uv1co[2], float uv2co[2], float uv3co[2],
+               float pixelScreenCo[4],
+               float w[3])
+{
+       BarycentricWeights2f(uv1co, uv2co, uv3co, uv, w);
+       VecWeightf(pixelScreenCo, v1co, v2co, v3co, w);
+}
+
+/* same as screen_px_from_ortho except we need to take into account
+ * the perspective W coord for each vert */
+static void screen_px_from_persp(
+               float uv[2],
+               float v1co[3], float v2co[3], float v3co[3], /* screenspace coords */
+               float uv1co[2], float uv2co[2], float uv3co[2],
+               float pixelScreenCo[4],
+               float w[3])
+{
+
+       float wtot_inv, wtot;
+       BarycentricWeights2f(uv1co, uv2co, uv3co, uv, w);
+       
+       /* re-weight from the 4th coord of each screen vert */
+       w[0] *= v1co[3];
+       w[1] *= v2co[3];
+       w[2] *= v3co[3];
+       
+       wtot = w[0]+w[1]+w[2];
+       
+       if (wtot > 0.0f) {
+               wtot_inv = 1.0f / wtot;
+               w[0] *= wtot_inv;
+               w[1] *= wtot_inv;
+               w[2] *= wtot_inv;
+       }
+       else {
+               w[0] = w[1] = w[2] = 1.0/3.0; /* dummy values for zero area face */
+       }
+       /* done re-weighting */
+       
+       VecWeightf(pixelScreenCo, v1co, v2co, v3co, w);
+}
+
+static void project_face_pixel(const MTFace *tf_other, ImBuf *ibuf_other, const float w[3], int side, unsigned char rgba_ub[4], float rgba_f[4])
+{
+       float *uvCo1, *uvCo2, *uvCo3;
+       float uv_other[2], x, y;
+       
+       uvCo1 =  (float *)tf_other->uv[0];
+       if (side==1) {
+               uvCo2 =  (float *)tf_other->uv[2];
+               uvCo3 =  (float *)tf_other->uv[3];
+       }
+       else {
+               uvCo2 =  (float *)tf_other->uv[1];
+               uvCo3 =  (float *)tf_other->uv[2];
+       }
+       
+       Vec2Weightf(uv_other, uvCo1, uvCo2, uvCo3, w);
+       
+       /* use */
+       uvco_to_wrapped_pxco(uv_other, ibuf_other->x, ibuf_other->y, &x, &y);
+       
+       
+       if (ibuf_other->rect_float) { /* from float to float */
+               bilinear_interpolation_color(ibuf_other, NULL, rgba_f, x, y);
+       }
+       else { /* from char to float */
+               bilinear_interpolation_color(ibuf_other, rgba_ub, NULL, x, y);
+       }
+               
+}
+
+/* run this outside project_paint_uvpixel_init since pixels with mask 0 dont need init */
+float project_paint_uvpixel_mask(
+               const ProjPaintState *ps,
+               const int face_index,
+               const int side,
+               const float w[3])
+{
+       float mask, mask_angle;
+       
+       /* Image Mask */
+       if (ps->do_layer_mask) {
+               /* another UV layers image is masking this one's */
+               ImBuf *ibuf_other;
+               const MTFace *tf_other = ps->dm_mtface_mask + face_index;
+               
+               if (tf_other->tpage && (ibuf_other = BKE_image_get_ibuf((Image *)tf_other->tpage, NULL))) {
+                       /* BKE_image_get_ibuf - TODO - this may be slow */
+                       unsigned char rgba_ub[4];
+                       float rgba_f[4];
+                       
+                       project_face_pixel(tf_other, ibuf_other, w, side, rgba_ub, rgba_f);
+                       
+                       if (ibuf_other->rect_float) { /* from float to float */
+                               mask = ((rgba_f[0]+rgba_f[1]+rgba_f[2])/3.0f) * rgba_f[3];
+                       }
+                       else { /* from char to float */
+                               mask = ((rgba_ub[0]+rgba_ub[1]+rgba_ub[2])/(256*3.0f)) * (rgba_ub[3]/256.0f);
+                       }
+                       
+                       if (!ps->do_layer_mask_inv) /* matching the gimps layer mask black/white rules, white==full opacity */
+                               mask = (1.0f - mask);
+
+                       if (mask == 0.0f) {
+                               return 0.0f;
+                       }
+               }
+               else {
+                       return 0.0f;
+               }
+       } else {
+               mask = 1.0f;
+       }
+       
+       /* calculate mask */
+       if (ps->do_mask_normal) {
+               MFace *mf = ps->dm_mface + face_index;
+               short *no1, *no2, *no3;
+               float no[3], angle;
+               no1 = ps->dm_mvert[mf->v1].no;
+               if (side==1) {
+                       no2 = ps->dm_mvert[mf->v3].no;
+                       no3 = ps->dm_mvert[mf->v4].no;
+               }
+               else {
+                       no2 = ps->dm_mvert[mf->v2].no;
+                       no3 = ps->dm_mvert[mf->v3].no;
+               }
+               
+               no[0] = w[0]*no1[0] + w[1]*no2[0] + w[2]*no3[0];
+               no[1] = w[0]*no1[1] + w[1]*no2[1] + w[2]*no3[1];
+               no[2] = w[0]*no1[2] + w[1]*no2[2] + w[2]*no3[2];
+               Normalize(no);
+               
+               /* now we can use the normal as a mask */
+               if (ps->is_ortho) {
+                       angle = NormalizedVecAngle2((float *)ps->viewDir, no);
+               }
+               else {
+                       /* Annoying but for the perspective view we need to get the pixels location in 3D space :/ */
+                       float viewDirPersp[3];
+                       float *co1, *co2, *co3;
+                       co1 = ps->dm_mvert[mf->v1].co;
+                       if (side==1) {
+                               co2 = ps->dm_mvert[mf->v3].co;
+                               co3 = ps->dm_mvert[mf->v4].co;
+                       }
+                       else {
+                               co2 = ps->dm_mvert[mf->v2].co;
+                               co3 = ps->dm_mvert[mf->v3].co;
+                       }
+                       viewDirPersp[0] = (ps->viewPos[0] - (w[0]*co1[0] + w[1]*co2[0] + w[2]*co3[0]));
+                       viewDirPersp[1] = (ps->viewPos[1] - (w[0]*co1[1] + w[1]*co2[1] + w[2]*co3[1]));
+                       viewDirPersp[2] = (ps->viewPos[2] - (w[0]*co1[2] + w[1]*co2[2] + w[2]*co3[2]));
+                       Normalize(viewDirPersp);
+                       
+                       angle = NormalizedVecAngle2(viewDirPersp, no);
+               }
+               
+               if (angle >= (M_PI_2 / 90) * ps->normal_angle) {
+                       return 0.0f;
+               }
+               else {
+                       float mask_no;
+#if 0
+                       mask *= 1.0f - (angle / PI_80_DEG); /* map angle to 1.0-facing us, 0.0 right angles to the view direction */
+#endif
+                       
+                       /* trickier method that clips the normal so its more useful */
+                       /* M_PI_2 is 90d, we want 80 though */
+                       mask_angle = ((M_PI_2 / 90) * ps->normal_angle);
+
+                       /*printf("normal_angle : %f \n" ,ps->normal_angle);
+                       printf("mask_angle : %f \n" ,mask_angle);*/
+
+                       mask_no = (angle / mask_angle); /* map angle to 1.0-facing us, 0.0 right angles to the view direction */
+                       mask_no = (1.0f - (mask_no * mask_no * mask_no)) * 1.4f;
+                       if (mask_no > 1.0f) {
+                               mask_no = 1.0f;
+                       }
+                       
+                       mask *= mask_no;
+               }
+       }
+       
+       // This only works when the opacity dosnt change while paintnig, stylus pressure messes with this
+       // so dont use it.
+       // if (ps->is_airbrush==0) mask *= ps->brush->alpha;
+       
+       return mask;
+}
+
+/* run this function when we know a bucket's, face's pixel can be initialized,
+ * return the ProjPixel which is added to 'ps->bucketRect[bucket_index]' */
+static ProjPixel *project_paint_uvpixel_init(
+               const ProjPaintState *ps,
+               MemArena *arena,
+               const ImBuf *ibuf,
+               short x_px, short y_px,
+               const float mask,
+               const int face_index,
+               const int image_index,
+               const float pixelScreenCo[4],
+               const int side,
+               const float w[3])
+{
+       ProjPixel *projPixel;
+       short size;
+       
+       /* wrap pixel location */
+       x_px = x_px % ibuf->x;
+       if (x_px<0) x_px += ibuf->x;
+       y_px = y_px % ibuf->y;
+       if (y_px<0) y_px += ibuf->y;
+       
+       if (ps->tool==PAINT_TOOL_CLONE) {
+               size = sizeof(ProjPixelClone);
+       }
+       else if (ps->tool==PAINT_TOOL_SMEAR) {
+               size = sizeof(ProjPixelClone);
+       }
+       else {
+               size = sizeof(ProjPixel);
+       }
+       
+       projPixel = (ProjPixel *)BLI_memarena_alloc(arena, size);
+       //memset(projPixel, 0, size);
+       
+       if (ibuf->rect_float) {
+               projPixel->pixel.f_pt = (float *)ibuf->rect_float + ((x_px + y_px * ibuf->x) * 4);
+               projPixel->origColor.f[0] = projPixel->newColor.f[0] = projPixel->pixel.f_pt[0];  
+               projPixel->origColor.f[1] = projPixel->newColor.f[1] = projPixel->pixel.f_pt[1];  
+               projPixel->origColor.f[2] = projPixel->newColor.f[2] = projPixel->pixel.f_pt[2];  
+               projPixel->origColor.f[3] = projPixel->newColor.f[3] = projPixel->pixel.f_pt[3];  
+       }
+       else {
+               projPixel->pixel.ch_pt = ((unsigned char *)ibuf->rect + ((x_px + y_px * ibuf->x) * 4));
+               projPixel->origColor.uint = projPixel->newColor.uint = *projPixel->pixel.uint_pt;
+       }
+       
+       /* screenspace unclamped, we could keep its z and w values but dont need them at the moment */
+       VECCOPY2D(projPixel->projCoSS, pixelScreenCo);
+       
+       projPixel->x_px = x_px;
+       projPixel->y_px = y_px;
+       
+       projPixel->mask = (unsigned short)(mask * 65535);
+       projPixel->mask_max = 0;
+       
+       /* which bounding box cell are we in?, needed for undo */
+       projPixel->bb_cell_index = ((int)(((float)x_px/(float)ibuf->x) * PROJ_BOUNDBOX_DIV)) + ((int)(((float)y_px/(float)ibuf->y) * PROJ_BOUNDBOX_DIV)) * PROJ_BOUNDBOX_DIV ;
+       
+       /* done with view3d_project_float inline */
+       if (ps->tool==PAINT_TOOL_CLONE) {
+               if (ps->dm_mtface_clone) {
+                       ImBuf *ibuf_other;
+                       const MTFace *tf_other = ps->dm_mtface_clone + face_index;
+                       
+                       if (tf_other->tpage && (ibuf_other = BKE_image_get_ibuf((Image *)tf_other->tpage, NULL))) {
+                               /* BKE_image_get_ibuf - TODO - this may be slow */
+                               
+                               if (ibuf->rect_float) {
+                                       if (ibuf_other->rect_float) { /* from float to float */
+                                               project_face_pixel(tf_other, ibuf_other, w, side, NULL, ((ProjPixelClone *)projPixel)->clonepx.f);
+                                       }
+                                       else { /* from char to float */
+                                               unsigned char rgba_ub[4];
+                                               project_face_pixel(tf_other, ibuf_other, w, side, rgba_ub, NULL);
+                                               IMAPAINT_CHAR_RGBA_TO_FLOAT(((ProjPixelClone *)projPixel)->clonepx.f, rgba_ub);
+                                       }
+                               }
+                               else {
+                                       if (ibuf_other->rect_float) { /* float to char */
+                                               float rgba[4];
+                                               project_face_pixel(tf_other, ibuf_other, w, side, NULL, rgba);
+                                               IMAPAINT_FLOAT_RGBA_TO_CHAR(((ProjPixelClone *)projPixel)->clonepx.ch, rgba)
+                                       }
+                                       else { /* char to char */
+                                               project_face_pixel(tf_other, ibuf_other, w, side, ((ProjPixelClone *)projPixel)->clonepx.ch, NULL);
+                                       }
+                               }
+                       }
+                       else {
+                               if (ibuf->rect_float) {
+                                       ((ProjPixelClone *)projPixel)->clonepx.f[3] = 0;
+                               }
+                               else {
+                                       ((ProjPixelClone *)projPixel)->clonepx.ch[3] = 0;
+                               }
+                       }
+                       
+               }
+               else {
+                       float co[2];
+                       Vec2Subf(co, projPixel->projCoSS, (float *)ps->cloneOffset);
+                       
+                       /* no need to initialize the bucket, we're only checking buckets faces and for this
+                        * the faces are alredy initialized in project_paint_delayed_face_init(...) */
+                       if (ibuf->rect_float) {
+                               if (!project_paint_PickColor(ps, co, ((ProjPixelClone *)projPixel)->clonepx.f, NULL, 1)) {
+                                       ((ProjPixelClone *)projPixel)->clonepx.f[3] = 0; /* zero alpha - ignore */
+                               }
+                       }
+                       else {
+                               if (!project_paint_PickColor(ps, co, NULL, ((ProjPixelClone *)projPixel)->clonepx.ch, 1)) {
+                                       ((ProjPixelClone *)projPixel)->clonepx.ch[3] = 0; /* zero alpha - ignore */
+                               }
+                       }
+               }
+       }
+       
+#ifdef PROJ_DEBUG_PAINT
+       if (ibuf->rect_float)   projPixel->pixel.f_pt[0] = 0;
+       else                                    projPixel->pixel.ch_pt[0] = 0;
+#endif
+       projPixel->image_index = image_index;
+       
+       return projPixel;
+}
+
+static int line_clip_rect2f(
+               rctf *rect,
+               const float l1[2], const float l2[2],
+               float l1_clip[2], float l2_clip[2])
+{
+       /* first account for horizontal, then vertical lines */
+       /* horiz */
+       if (fabs(l1[1]-l2[1]) < PROJ_GEOM_TOLERANCE) {
+               /* is the line out of range on its Y axis? */
+               if (l1[1] < rect->ymin || l1[1] > rect->ymax) {
+                       return 0;
+               }
+               /* line is out of range on its X axis */
+               if ((l1[0] < rect->xmin && l2[0] < rect->xmin) || (l1[0] > rect->xmax && l2[0] > rect->xmax)) {
+                       return 0;
+               }
+               
+               
+               if (fabs(l1[0]-l2[0]) < PROJ_GEOM_TOLERANCE) { /* this is a single point  (or close to)*/
+                       if (BLI_in_rctf(rect, l1[0], l1[1])) {
+                               VECCOPY2D(l1_clip, l1);
+                               VECCOPY2D(l2_clip, l2);
+                               return 1;
+                       }
+                       else {
+                               return 0;
+                       }
+               }
+               
+               VECCOPY2D(l1_clip, l1);
+               VECCOPY2D(l2_clip, l2);
+               CLAMP(l1_clip[0], rect->xmin, rect->xmax);
+               CLAMP(l2_clip[0], rect->xmin, rect->xmax);
+               return 1;
+       }
+       else if (fabs(l1[0]-l2[0]) < PROJ_GEOM_TOLERANCE) {
+               /* is the line out of range on its X axis? */
+               if (l1[0] < rect->xmin || l1[0] > rect->xmax) {
+                       return 0;
+               }
+               
+               /* line is out of range on its Y axis */
+               if ((l1[1] < rect->ymin && l2[1] < rect->ymin) || (l1[1] > rect->ymax && l2[1] > rect->ymax)) {
+                       return 0;
+               }
+               
+               if (fabs(l1[1]-l2[1]) < PROJ_GEOM_TOLERANCE) { /* this is a single point  (or close to)*/
+                       if (BLI_in_rctf(rect, l1[0], l1[1])) {
+                               VECCOPY2D(l1_clip, l1);
+                               VECCOPY2D(l2_clip, l2);
+                               return 1;
+                       }
+                       else {
+                               return 0;
+                       }
+               }
+               
+               VECCOPY2D(l1_clip, l1);
+               VECCOPY2D(l2_clip, l2);
+               CLAMP(l1_clip[1], rect->ymin, rect->ymax);
+               CLAMP(l2_clip[1], rect->ymin, rect->ymax);
+               return 1;
+       }
+       else {
+               float isect;
+               short ok1 = 0;
+               short ok2 = 0;
+               
+               /* Done with vertical lines */
+               
+               /* are either of the points inside the rectangle ? */
+               if (BLI_in_rctf(rect, l1[0], l1[1])) {
+                       VECCOPY2D(l1_clip, l1);
+                       ok1 = 1;
+               }
+               
+               if (BLI_in_rctf(rect, l2[0], l2[1])) {
+                       VECCOPY2D(l2_clip, l2);
+                       ok2 = 1;
+               }
+               
+               /* line inside rect */
+               if (ok1 && ok2) return 1;
+               
+               /* top/bottom */
+               if (line_isect_y(l1, l2, rect->ymin, &isect) && (isect >= rect->xmin) && (isect <= rect->xmax)) {
+                       if (l1[1] < l2[1]) { /* line 1 is outside */
+                               l1_clip[0] = isect;
+                               l1_clip[1] = rect->ymin;
+                               ok1 = 1;
+                       }
+                       else {
+                               l2_clip[0] = isect;
+                               l2_clip[1] = rect->ymin;
+                               ok2 = 2;
+                       }
+               }
+               
+               if (ok1 && ok2) return 1;
+               
+               if (line_isect_y(l1, l2, rect->ymax, &isect) && (isect >= rect->xmin) && (isect <= rect->xmax)) {
+                       if (l1[1] > l2[1]) { /* line 1 is outside */
+                               l1_clip[0] = isect;
+                               l1_clip[1] = rect->ymax;
+                               ok1 = 1;
+                       }
+                       else {
+                               l2_clip[0] = isect;
+                               l2_clip[1] = rect->ymax;
+                               ok2 = 2;
+                       }
+               }
+               
+               if (ok1 && ok2) return 1;
+               
+               /* left/right */
+               if (line_isect_x(l1, l2, rect->xmin, &isect) && (isect >= rect->ymin) && (isect <= rect->ymax)) {
+                       if (l1[0] < l2[0]) { /* line 1 is outside */
+                               l1_clip[0] = rect->xmin;
+                               l1_clip[1] = isect;
+                               ok1 = 1;
+                       }
+                       else {
+                               l2_clip[0] = rect->xmin;
+                               l2_clip[1] = isect;
+                               ok2 = 2;
+                       }
+               }
+       
+               if (ok1 && ok2) return 1;
+               
+               if (line_isect_x(l1, l2, rect->xmax, &isect) && (isect >= rect->ymin) && (isect <= rect->ymax)) {
+                       if (l1[0] > l2[0]) { /* line 1 is outside */
+                               l1_clip[0] = rect->xmax;
+                               l1_clip[1] = isect;
+                               ok1 = 1;
+                       }
+                       else {
+                               l2_clip[0] = rect->xmax;
+                               l2_clip[1] = isect;
+                               ok2 = 2;
+                       }
+               }
+               
+               if (ok1 && ok2) {
+                       return 1;
+               }
+               else {
+                       return 0;
+               }
+       }
+}
+
+
+
+/* scale the quad & tri about its center
+ * scaling by PROJ_FACE_SCALE_SEAM (0.99x) is used for getting fake UV pixel coords that are on the
+ * edge of the face but slightly inside it occlusion tests dont return hits on adjacent faces */
+static void scale_quad(float insetCos[4][3], float *origCos[4], const float inset)
+{
+       float cent[3];
+       cent[0] = (origCos[0][0] + origCos[1][0] + origCos[2][0] + origCos[3][0]) / 4.0f;
+       cent[1] = (origCos[0][1] + origCos[1][1] + origCos[2][1] + origCos[3][1]) / 4.0f;
+       cent[2] = (origCos[0][2] + origCos[1][2] + origCos[2][2] + origCos[3][2]) / 4.0f;
+       
+       VecSubf(insetCos[0], origCos[0], cent);
+       VecSubf(insetCos[1], origCos[1], cent);
+       VecSubf(insetCos[2], origCos[2], cent);
+       VecSubf(insetCos[3], origCos[3], cent);
+       
+       VecMulf(insetCos[0], inset);
+       VecMulf(insetCos[1], inset);
+       VecMulf(insetCos[2], inset);
+       VecMulf(insetCos[3], inset);
+       
+       VecAddf(insetCos[0], insetCos[0], cent);
+       VecAddf(insetCos[1], insetCos[1], cent);
+       VecAddf(insetCos[2], insetCos[2], cent);
+       VecAddf(insetCos[3], insetCos[3], cent);
+}
+
+
+static void scale_tri(float insetCos[4][3], float *origCos[4], const float inset)
+{
+       float cent[3];
+       cent[0] = (origCos[0][0] + origCos[1][0] + origCos[2][0]) / 3.0f;
+       cent[1] = (origCos[0][1] + origCos[1][1] + origCos[2][1]) / 3.0f;
+       cent[2] = (origCos[0][2] + origCos[1][2] + origCos[2][2]) / 3.0f;
+       
+       VecSubf(insetCos[0], origCos[0], cent);
+       VecSubf(insetCos[1], origCos[1], cent);
+       VecSubf(insetCos[2], origCos[2], cent);
+       
+       VecMulf(insetCos[0], inset);
+       VecMulf(insetCos[1], inset);
+       VecMulf(insetCos[2], inset);
+       
+       VecAddf(insetCos[0], insetCos[0], cent);
+       VecAddf(insetCos[1], insetCos[1], cent);
+       VecAddf(insetCos[2], insetCos[2], cent);
+}
+
+
+static float Vec2Lenf_nosqrt(const float *v1, const float *v2)
+{
+       float x, y;
+
+       x = v1[0]-v2[0];
+       y = v1[1]-v2[1];
+       return x*x+y*y;
+}
+
+static float Vec2Lenf_nosqrt_other(const float *v1, const float v2_1, const float v2_2)
+{
+       float x, y;
+
+       x = v1[0]-v2_1;
+       y = v1[1]-v2_2;
+       return x*x+y*y;
+}
+
+/* note, use a squared value so we can use Vec2Lenf_nosqrt
+ * be sure that you have done a bounds check first or this may fail */
+/* only give bucket_bounds as an arg because we need it elsewhere */
+static int project_bucket_isect_circle(const int bucket_x, const int bucket_y, const float cent[2], const float radius_squared, rctf *bucket_bounds)
+{
+        
+       /* Would normally to a simple intersection test, however we know the bounds of these 2 alredy intersect 
+        * so we only need to test if the center is inside the vertical or horizontal bounds on either axis,
+        * this is even less work then an intersection test
+        * 
+       if (BLI_in_rctf(bucket_bounds, cent[0], cent[1]))
+               return 1;
+        */
+       
+       if((bucket_bounds->xmin <= cent[0] && bucket_bounds->xmax >= cent[0]) || (bucket_bounds->ymin <= cent[1] && bucket_bounds->ymax >= cent[1]) ) {
+          return 1;
+       }
+       
+       /* out of bounds left */
+       if (cent[0] < bucket_bounds->xmin) {
+               /* lower left out of radius test */
+               if (cent[1] < bucket_bounds->ymin) {
+                       return (Vec2Lenf_nosqrt_other(cent, bucket_bounds->xmin, bucket_bounds->ymin) < radius_squared) ? 1 : 0;
+               } 
+               /* top left test */
+               else if (cent[1] > bucket_bounds->ymax) {
+                       return (Vec2Lenf_nosqrt_other(cent, bucket_bounds->xmin, bucket_bounds->ymax) < radius_squared) ? 1 : 0;
+               }
+       }
+       else if (cent[0] > bucket_bounds->xmax) {
+               /* lower right out of radius test */
+               if (cent[1] < bucket_bounds->ymin) {
+                       return (Vec2Lenf_nosqrt_other(cent, bucket_bounds->xmax, bucket_bounds->ymin) < radius_squared) ? 1 : 0;
+               } 
+               /* top right test */
+               else if (cent[1] > bucket_bounds->ymax) {
+                       return (Vec2Lenf_nosqrt_other(cent, bucket_bounds->xmax, bucket_bounds->ymax) < radius_squared) ? 1 : 0;
+               }
+       }
+       
+       return 0;
+}
+
+
+
+/* Note for rect_to_uvspace_ortho() and rect_to_uvspace_persp()
+ * in ortho view this function gives good results when bucket_bounds are outside the triangle
+ * however in some cases, perspective view will mess up with faces that have minimal screenspace area (viewed from the side)
+ * 
+ * for this reason its not relyable in this case so we'll use the Simple Barycentric' funcs that only account for points inside the triangle.
+ * however switching back to this for ortho is always an option */
+
+static void rect_to_uvspace_ortho(
+               rctf *bucket_bounds,
+               float *v1coSS, float *v2coSS, float *v3coSS,
+               float *uv1co, float *uv2co, float *uv3co,
+               float bucket_bounds_uv[4][2],
+               const int flip)
+{
+       float uv[2];
+       float w[3];
+       
+       /* get the UV space bounding box */
+       uv[0] = bucket_bounds->xmax;
+       uv[1] = bucket_bounds->ymin;
+       BarycentricWeights2f(v1coSS, v2coSS, v3coSS, uv, w);    
+       Vec2Weightf(bucket_bounds_uv[flip?3:0], uv1co, uv2co, uv3co, w);
+
+       //uv[0] = bucket_bounds->xmax; // set above
+       uv[1] = bucket_bounds->ymax;
+       BarycentricWeights2f(v1coSS, v2coSS, v3coSS, uv, w);
+       Vec2Weightf(bucket_bounds_uv[flip?2:1], uv1co, uv2co, uv3co, w);
+
+       uv[0] = bucket_bounds->xmin;
+       //uv[1] = bucket_bounds->ymax; // set above
+       BarycentricWeights2f(v1coSS, v2coSS, v3coSS, uv, w);
+       Vec2Weightf(bucket_bounds_uv[flip?1:2], uv1co, uv2co, uv3co, w);
+
+       //uv[0] = bucket_bounds->xmin; // set above
+       uv[1] = bucket_bounds->ymin;
+       BarycentricWeights2f(v1coSS, v2coSS, v3coSS, uv, w);
+       Vec2Weightf(bucket_bounds_uv[flip?0:3], uv1co, uv2co, uv3co, w);
+}
+
+/* same as above but use BarycentricWeightsPersp2f */
+static void rect_to_uvspace_persp(
+               rctf *bucket_bounds,
+               float *v1coSS, float *v2coSS, float *v3coSS,
+               float *uv1co, float *uv2co, float *uv3co,
+               float bucket_bounds_uv[4][2],
+               const int flip
+       )
+{
+       float uv[2];
+       float w[3];
+       
+       /* get the UV space bounding box */
+       uv[0] = bucket_bounds->xmax;
+       uv[1] = bucket_bounds->ymin;
+       BarycentricWeightsPersp2f(v1coSS, v2coSS, v3coSS, uv, w);       
+       Vec2Weightf(bucket_bounds_uv[flip?3:0], uv1co, uv2co, uv3co, w);
+
+       //uv[0] = bucket_bounds->xmax; // set above
+       uv[1] = bucket_bounds->ymax;
+       BarycentricWeightsPersp2f(v1coSS, v2coSS, v3coSS, uv, w);
+       Vec2Weightf(bucket_bounds_uv[flip?2:1], uv1co, uv2co, uv3co, w);
+
+       uv[0] = bucket_bounds->xmin;
+       //uv[1] = bucket_bounds->ymax; // set above
+       BarycentricWeightsPersp2f(v1coSS, v2coSS, v3coSS, uv, w);
+       Vec2Weightf(bucket_bounds_uv[flip?1:2], uv1co, uv2co, uv3co, w);
+
+       //uv[0] = bucket_bounds->xmin; // set above
+       uv[1] = bucket_bounds->ymin;
+       BarycentricWeightsPersp2f(v1coSS, v2coSS, v3coSS, uv, w);
+       Vec2Weightf(bucket_bounds_uv[flip?0:3], uv1co, uv2co, uv3co, w);
+}
+
+/* This works as we need it to but we can save a few steps and not use it */
+
+static float angle_2d_clockwise(const float p1[2], const float p2[2], const float p3[2])
+{
+       float v1[2], v2[2];
+       
+       v1[0] = p1[0]-p2[0];    v1[1] = p1[1]-p2[1];
+       v2[0] = p3[0]-p2[0];    v2[1] = p3[1]-p2[1];
+       
+       return -atan2(v1[0]*v2[1] - v1[1]*v2[0], v1[0]*v2[0]+v1[1]*v2[1]);
+}
+
+#define ISECT_1 (1)
+#define ISECT_2 (1<<1)
+#define ISECT_3 (1<<2)
+#define ISECT_4 (1<<3)
+#define ISECT_ALL3 ((1<<3)-1)
+#define ISECT_ALL4 ((1<<4)-1)
+
+/* limit must be a fraction over 1.0f */
+static int IsectPT2Df_limit(float pt[2], float v1[2], float v2[2], float v3[2], float limit)
+{
+       return ((AreaF2Dfl(pt,v1,v2) + AreaF2Dfl(pt,v2,v3) + AreaF2Dfl(pt,v3,v1)) / (AreaF2Dfl(v1,v2,v3))) < limit;
+}
+
+/* Clip the face by a bucket and set the uv-space bucket_bounds_uv
+ * so we have the clipped UV's to do pixel intersection tests with 
+ * */
+static int float_z_sort_flip(const void *p1, const void *p2) {
+       return (((float *)p1)[2] < ((float *)p2)[2] ? 1:-1);
+}
+
+static int float_z_sort(const void *p1, const void *p2) {
+       return (((float *)p1)[2] < ((float *)p2)[2] ?-1:1);
+}
+
+static void project_bucket_clip_face(
+               const int is_ortho,
+               rctf *bucket_bounds,
+               float *v1coSS, float *v2coSS, float *v3coSS,
+               float *uv1co, float *uv2co, float *uv3co,
+               float bucket_bounds_uv[8][2],
+               int *tot)
+{
+       int inside_bucket_flag = 0;
+       int inside_face_flag = 0;
+       const int flip = ((SIDE_OF_LINE(v1coSS, v2coSS, v3coSS) > 0.0f) != (SIDE_OF_LINE(uv1co, uv2co, uv3co) > 0.0f));
+       
+       float bucket_bounds_ss[4][2];
+       float w[3];
+
+       /* get the UV space bounding box */
+       inside_bucket_flag |= BLI_in_rctf(bucket_bounds, v1coSS[0], v1coSS[1]);
+       inside_bucket_flag |= BLI_in_rctf(bucket_bounds, v2coSS[0], v2coSS[1])          << 1;
+       inside_bucket_flag |= BLI_in_rctf(bucket_bounds, v3coSS[0], v3coSS[1])          << 2;
+       
+       if (inside_bucket_flag == ISECT_ALL3) {
+               /* all screenspace points are inside the bucket bounding box, this means we dont need to clip and can simply return the UVs */
+               if (flip) { /* facing the back? */
+                       VECCOPY2D(bucket_bounds_uv[0], uv3co);
+                       VECCOPY2D(bucket_bounds_uv[1], uv2co);
+                       VECCOPY2D(bucket_bounds_uv[2], uv1co);
+               }
+               else {
+                       VECCOPY2D(bucket_bounds_uv[0], uv1co);
+                       VECCOPY2D(bucket_bounds_uv[1], uv2co);
+                       VECCOPY2D(bucket_bounds_uv[2], uv3co);
+               }
+               
+               *tot = 3; 
+               return;
+       }
+       
+       /* get the UV space bounding box */
+       /* use IsectPT2Df_limit here so we catch points are are touching the tri edge (or a small fraction over) */
+       bucket_bounds_ss[0][0] = bucket_bounds->xmax;
+       bucket_bounds_ss[0][1] = bucket_bounds->ymin;
+       inside_face_flag |= (IsectPT2Df_limit(bucket_bounds_ss[0], v1coSS, v2coSS, v3coSS, 1+PROJ_GEOM_TOLERANCE) ? ISECT_1 : 0);
+       
+       bucket_bounds_ss[1][0] = bucket_bounds->xmax;
+       bucket_bounds_ss[1][1] = bucket_bounds->ymax;
+       inside_face_flag |= (IsectPT2Df_limit(bucket_bounds_ss[1], v1coSS, v2coSS, v3coSS, 1+PROJ_GEOM_TOLERANCE) ? ISECT_2 : 0);
+
+       bucket_bounds_ss[2][0] = bucket_bounds->xmin;
+       bucket_bounds_ss[2][1] = bucket_bounds->ymax;
+       inside_face_flag |= (IsectPT2Df_limit(bucket_bounds_ss[2], v1coSS, v2coSS, v3coSS, 1+PROJ_GEOM_TOLERANCE) ? ISECT_3 : 0);
+
+       bucket_bounds_ss[3][0] = bucket_bounds->xmin;
+       bucket_bounds_ss[3][1] = bucket_bounds->ymin;
+       inside_face_flag |= (IsectPT2Df_limit(bucket_bounds_ss[3], v1coSS, v2coSS, v3coSS, 1+PROJ_GEOM_TOLERANCE) ? ISECT_4 : 0);
+       
+       if (inside_face_flag == ISECT_ALL4) {
+               /* bucket is totally inside the screenspace face, we can safely use weights */
+               
+               if (is_ortho)   rect_to_uvspace_ortho(bucket_bounds, v1coSS, v2coSS, v3coSS, uv1co, uv2co, uv3co, bucket_bounds_uv, flip);
+               else                    rect_to_uvspace_persp(bucket_bounds, v1coSS, v2coSS, v3coSS, uv1co, uv2co, uv3co, bucket_bounds_uv, flip);
+               
+               *tot = 4;
+               return;
+       }
+       else {
+               /* The Complicated Case! 
+                * 
+                * The 2 cases above are where the face is inside the bucket or the bucket is inside the face.
+                * 
+                * we need to make a convex polyline from the intersection between the screenspace face
+                * and the bucket bounds.
+                * 
+                * There are a number of ways this could be done, currently it just collects all intersecting verts,
+                * and line intersections,  then sorts them clockwise, this is a lot easier then evaluating the geometry to
+                * do a correct clipping on both shapes. */
+               
+               
+               /* add a bunch of points, we know must make up the convex hull which is the clipped rect and triangle */
+               
+               
+               
+               /* Maximum possible 6 intersections when using a rectangle and triangle */
+               float isectVCosSS[8][3]; /* The 3rd float is used to store angle for qsort(), NOT as a Z location */
+               float v1_clipSS[2], v2_clipSS[2];
+               
+               /* calc center*/
+               float cent[2] = {0.0f, 0.0f};
+               /*float up[2] = {0.0f, 1.0f};*/
+               int i;
+               short doubles;
+               
+               (*tot) = 0;
+               
+               if (inside_face_flag & ISECT_1) { VECCOPY2D(isectVCosSS[*tot], bucket_bounds_ss[0]); (*tot)++; }
+               if (inside_face_flag & ISECT_2) { VECCOPY2D(isectVCosSS[*tot], bucket_bounds_ss[1]); (*tot)++; }
+               if (inside_face_flag & ISECT_3) { VECCOPY2D(isectVCosSS[*tot], bucket_bounds_ss[2]); (*tot)++; }
+               if (inside_face_flag & ISECT_4) { VECCOPY2D(isectVCosSS[*tot], bucket_bounds_ss[3]); (*tot)++; }
+               
+               if (inside_bucket_flag & ISECT_1) {     VECCOPY2D(isectVCosSS[*tot], v1coSS); (*tot)++; }
+               if (inside_bucket_flag & ISECT_2) {     VECCOPY2D(isectVCosSS[*tot], v2coSS); (*tot)++; }
+               if (inside_bucket_flag & ISECT_3) {     VECCOPY2D(isectVCosSS[*tot], v3coSS); (*tot)++; }
+               
+               if ((inside_bucket_flag & (ISECT_1|ISECT_2)) != (ISECT_1|ISECT_2)) {
+                       if (line_clip_rect2f(bucket_bounds, v1coSS, v2coSS, v1_clipSS, v2_clipSS)) {
+                               if ((inside_bucket_flag & ISECT_1)==0) { VECCOPY2D(isectVCosSS[*tot], v1_clipSS); (*tot)++; }
+                               if ((inside_bucket_flag & ISECT_2)==0) { VECCOPY2D(isectVCosSS[*tot], v2_clipSS); (*tot)++; }
+                       }
+               }
+               
+               if ((inside_bucket_flag & (ISECT_2|ISECT_3)) != (ISECT_2|ISECT_3)) {
+                       if (line_clip_rect2f(bucket_bounds, v2coSS, v3coSS, v1_clipSS, v2_clipSS)) {
+                               if ((inside_bucket_flag & ISECT_2)==0) { VECCOPY2D(isectVCosSS[*tot], v1_clipSS); (*tot)++; }
+                               if ((inside_bucket_flag & ISECT_3)==0) { VECCOPY2D(isectVCosSS[*tot], v2_clipSS); (*tot)++; }
+                       }
+               }       
+               
+               if ((inside_bucket_flag & (ISECT_3|ISECT_1)) != (ISECT_3|ISECT_1)) {
+                       if (line_clip_rect2f(bucket_bounds, v3coSS, v1coSS, v1_clipSS, v2_clipSS)) {
+                               if ((inside_bucket_flag & ISECT_3)==0) { VECCOPY2D(isectVCosSS[*tot], v1_clipSS); (*tot)++; }
+                               if ((inside_bucket_flag & ISECT_1)==0) { VECCOPY2D(isectVCosSS[*tot], v2_clipSS); (*tot)++; }
+                       }
+               }
+               
+               
+               if ((*tot) < 3) { /* no intersections to speak of */
+                       *tot = 0;
+               }
+       
+               /* now we have all points we need, collect their angles and sort them clockwise */
+               
+               for(i=0; i<(*tot); i++) {
+                       cent[0] += isectVCosSS[i][0];
+                       cent[1] += isectVCosSS[i][1];
+               }
+               cent[0] = cent[0] / (float)(*tot);
+               cent[1] = cent[1] / (float)(*tot);
+               
+               
+               
+               /* Collect angles for every point around the center point */
+
+               
+#if 0  /* uses a few more cycles then the above loop */
+               for(i=0; i<(*tot); i++) {
+                       isectVCosSS[i][2] = angle_2d_clockwise(up, cent, isectVCosSS[i]);
+               }
+#endif
+
+               v1_clipSS[0] = cent[0]; /* Abuse this var for the loop below */
+               v1_clipSS[1] = cent[1] + 1.0f;
+               
+               for(i=0; i<(*tot); i++) {
+                       v2_clipSS[0] = isectVCosSS[i][0] - cent[0];
+                       v2_clipSS[1] = isectVCosSS[i][1] - cent[1];
+                       isectVCosSS[i][2] = atan2(v1_clipSS[0]*v2_clipSS[1] - v1_clipSS[1]*v2_clipSS[0], v1_clipSS[0]*v2_clipSS[0]+v1_clipSS[1]*v2_clipSS[1]); 
+               }
+               
+               if (flip)       qsort(isectVCosSS, *tot, sizeof(float)*3, float_z_sort_flip);
+               else            qsort(isectVCosSS, *tot, sizeof(float)*3, float_z_sort);
+               
+               
+               /* remove doubles */
+               /* first/last check */
+               if (fabs(isectVCosSS[0][0]-isectVCosSS[(*tot)-1][0]) < PROJ_GEOM_TOLERANCE &&  fabs(isectVCosSS[0][1]-isectVCosSS[(*tot)-1][1]) < PROJ_GEOM_TOLERANCE) {
+                       (*tot)--;
+               }
+               
+               /* its possible there is only a few left after remove doubles */
+               if ((*tot) < 3) {
+                       // printf("removed too many doubles A\n");
+                       *tot = 0;
+                       return;
+               }
+               
+               doubles = TRUE;
+               while (doubles==TRUE) {
+                       doubles = FALSE;
+                       for(i=1; i<(*tot); i++) {
+                               if (fabs(isectVCosSS[i-1][0]-isectVCosSS[i][0]) < PROJ_GEOM_TOLERANCE &&
+                                       fabs(isectVCosSS[i-1][1]-isectVCosSS[i][1]) < PROJ_GEOM_TOLERANCE)
+                               {
+                                       int j;
+                                       for(j=i+1; j<(*tot); j++) {
+                                               isectVCosSS[j-1][0] = isectVCosSS[j][0]; 
+                                               isectVCosSS[j-1][1] = isectVCosSS[j][1]; 
+                                       }
+                                       doubles = TRUE; /* keep looking for more doubles */
+                                       (*tot)--;
+                               }
+                       }
+               }
+               
+               /* its possible there is only a few left after remove doubles */
+               if ((*tot) < 3) {
+                       // printf("removed too many doubles B\n");
+                       *tot = 0;
+                       return;
+               }
+               
+               
+               if (is_ortho) {
+                       for(i=0; i<(*tot); i++) {
+                               BarycentricWeights2f(v1coSS, v2coSS, v3coSS, isectVCosSS[i], w);
+                               Vec2Weightf(bucket_bounds_uv[i], uv1co, uv2co, uv3co, w);
+                       }
+               }
+               else {
+                       for(i=0; i<(*tot); i++) {
+                               BarycentricWeightsPersp2f(v1coSS, v2coSS, v3coSS, isectVCosSS[i], w);
+                               Vec2Weightf(bucket_bounds_uv[i], uv1co, uv2co, uv3co, w);
+                       }
+               }
+       }
+
+#ifdef PROJ_DEBUG_PRINT_CLIP
+       /* include this at the bottom of the above function to debug the output */
+
+       {
+               /* If there are ever any problems, */
+               float test_uv[4][2];
+               int i;
+               if (is_ortho)   rect_to_uvspace_ortho(bucket_bounds, v1coSS, v2coSS, v3coSS, uv1co, uv2co, uv3co, test_uv, flip);
+               else                            rect_to_uvspace_persp(bucket_bounds, v1coSS, v2coSS, v3coSS, uv1co, uv2co, uv3co, test_uv, flip);
+               printf("(  [(%f,%f), (%f,%f), (%f,%f), (%f,%f)], ", test_uv[0][0], test_uv[0][1],   test_uv[1][0], test_uv[1][1],    test_uv[2][0], test_uv[2][1],    test_uv[3][0], test_uv[3][1]);
+               
+               printf("  [(%f,%f), (%f,%f), (%f,%f)], ", uv1co[0], uv1co[1],   uv2co[0], uv2co[1],    uv3co[0], uv3co[1]);
+               
+               printf("[");
+               for (i=0; i < (*tot); i++) {
+                       printf("(%f, %f),", bucket_bounds_uv[i][0], bucket_bounds_uv[i][1]);
+               }
+               printf("]),\\\n");
+       }
+#endif
+}
+
+       /*
+# This script creates faces in a blender scene from printed data above.
+
+project_ls = [
+...(output from above block)...
+]
+from Blender import Scene, Mesh, Window, sys, Mathutils
+
+import bpy
+
+V = Mathutils.Vector
+
+def main():
+       sce = bpy.data.scenes.active
+       
+       for item in project_ls:
+               bb = item[0]
+               uv = item[1]
+               poly = item[2]
+               
+               me = bpy.data.meshes.new()
+               ob = sce.objects.new(me)
+               
+               me.verts.extend([V(bb[0]).resize3D(), V(bb[1]).resize3D(), V(bb[2]).resize3D(), V(bb[3]).resize3D()])
+               me.faces.extend([(0,1,2,3),])
+               me.verts.extend([V(uv[0]).resize3D(), V(uv[1]).resize3D(), V(uv[2]).resize3D()])
+               me.faces.extend([(4,5,6),])
+               
+               vs = [V(p).resize3D() for p in poly]
+               print len(vs)
+               l = len(me.verts)
+               me.verts.extend(vs)
+               
+               i = l
+               while i < len(me.verts):
+                       ii = i+1
+                       if ii==len(me.verts):
+                               ii = l
+                       me.edges.extend([i, ii])
+                       i+=1
+
+if __name__ == '__main__':
+       main()
+ */    
+
+
+#undef ISECT_1
+#undef ISECT_2
+#undef ISECT_3
+#undef ISECT_4
+#undef ISECT_ALL3
+#undef ISECT_ALL4
+
+       
+/* checks if pt is inside a convex 2D polyline, the polyline must be ordered rotating clockwise
+ * otherwise it would have to test for mixed (SIDE_OF_LINE > 0.0f) cases */
+int IsectPoly2Df(const float pt[2], float uv[][2], const int tot)
+{
+       int i;
+       if (SIDE_OF_LINE(uv[tot-1], uv[0], pt) < 0.0f)
+               return 0;
+       
+       for (i=1; i<tot; i++) {
+               if (SIDE_OF_LINE(uv[i-1], uv[i], pt) < 0.0f)
+                       return 0;
+               
+       }
+       
+       return 1;
+}
+
+/* One of the most important function for projectiopn painting, since it selects the pixels to be added into each bucket.
+ * initialize pixels from this face where it intersects with the bucket_index, optionally initialize pixels for removing seams */
+static void project_paint_face_init(const ProjPaintState *ps, const int thread_index, const int bucket_index, const int face_index, const int image_index, rctf *bucket_bounds, const ImBuf *ibuf)
+{
+       /* Projection vars, to get the 3D locations into screen space  */
+       MemArena *arena = ps->arena_mt[thread_index];
+       LinkNode **bucketPixelNodes = ps->bucketRect + bucket_index;
+       LinkNode *bucketFaceNodes = ps->bucketFaces[bucket_index];
+       
+       const MFace *mf = ps->dm_mface + face_index;
+       const MTFace *tf = ps->dm_mtface + face_index;
+       
+       /* UV/pixel seeking data */
+       int x; /* Image X-Pixel */
+       int y;/* Image Y-Pixel */
+       float mask;
+       float uv[2]; /* Image floating point UV - same as x, y but from 0.0-1.0 */
+       
+       int side;
+       float *v1coSS, *v2coSS, *v3coSS; /* vert co screen-space, these will be assigned to mf->v1,2,3 or mf->v1,3,4 */
+       
+       float *vCo[4]; /* vertex screenspace coords */
+       
+       float w[3], wco[3];
+       
+       float *uv1co, *uv2co, *uv3co; /* for convenience only, these will be assigned to tf->uv[0],1,2 or tf->uv[0],2,3 */
+       float pixelScreenCo[4];
+       
+       rcti bounds_px; /* ispace bounds */
+       /* vars for getting uvspace bounds */
+       
+       float tf_uv_pxoffset[4][2]; /* bucket bounds in UV space so we can init pixels only for this face,  */
+       float xhalfpx, yhalfpx;
+       const float ibuf_xf = ibuf->x, ibuf_yf = ibuf->y;
+       
+       int has_x_isect = 0, has_isect = 0; /* for early loop exit */
+       
+       int i1, i2, i3;
+       
+       float uv_clip[8][2];
+       int uv_clip_tot;
+       const short is_ortho = ps->is_ortho;
+       
+       vCo[0] = ps->dm_mvert[mf->v1].co;
+       vCo[1] = ps->dm_mvert[mf->v2].co;
+       vCo[2] = ps->dm_mvert[mf->v3].co;
+       
+       
+       /* Use tf_uv_pxoffset instead of tf->uv so we can offset the UV half a pixel
+        * this is done so we can avoid offseting all the pixels by 0.5 which causes
+        * problems when wrapping negative coords */
+       xhalfpx = 0.5f / ibuf_xf;
+       yhalfpx = 0.5f / ibuf_yf;
+       
+       tf_uv_pxoffset[0][0] = tf->uv[0][0] - xhalfpx;
+       tf_uv_pxoffset[0][1] = tf->uv[0][1] - yhalfpx;
+
+       tf_uv_pxoffset[1][0] = tf->uv[1][0] - xhalfpx;
+       tf_uv_pxoffset[1][1] = tf->uv[1][1] - yhalfpx;
+       
+       tf_uv_pxoffset[2][0] = tf->uv[2][0] - xhalfpx;
+       tf_uv_pxoffset[2][1] = tf->uv[2][1] - yhalfpx;  
+       
+       if (mf->v4) {
+               vCo[3] = ps->dm_mvert[ mf->v4 ].co;
+               
+               tf_uv_pxoffset[3][0] = tf->uv[3][0] - xhalfpx;
+               tf_uv_pxoffset[3][1] = tf->uv[3][1] - yhalfpx;
+               side = 1;
+       }
+       else {
+               side = 0;
+       }
+       
+       do {
+               if (side==1) {
+                       i1=0; i2=2; i3=3;
+               }
+               else {
+                       i1=0; i2=1; i3=2;
+               }
+               
+               uv1co = tf_uv_pxoffset[i1]; // was tf->uv[i1];
+               uv2co = tf_uv_pxoffset[i2]; // was tf->uv[i2];
+               uv3co = tf_uv_pxoffset[i3]; // was tf->uv[i3];
+
+               v1coSS = ps->screenCoords[ (*(&mf->v1 + i1)) ];
+               v2coSS = ps->screenCoords[ (*(&mf->v1 + i2)) ];
+               v3coSS = ps->screenCoords[ (*(&mf->v1 + i3)) ];
+
+
+               /* This funtion gives is a concave polyline in UV space from the clipped quad and tri*/
+               project_bucket_clip_face(
+                               is_ortho, bucket_bounds,
+                               v1coSS, v2coSS, v3coSS,
+                               uv1co, uv2co, uv3co,
+                               uv_clip, &uv_clip_tot
+               );
+               
+               
+               /* sometimes this happens, better just allow for 8 intersectiosn even though there should be max 6 */
+               /*
+               if (uv_clip_tot>6) {
+                       printf("this should never happen! %d\n", uv_clip_tot);
+               }*/
+               
+
+               if (pixel_bounds_array(uv_clip, &bounds_px, ibuf->x, ibuf->y, uv_clip_tot)) {
+                       
+                       /* clip face and */
+                       
+                       has_isect = 0;
+                       for (y = bounds_px.ymin; y < bounds_px.ymax; y++) {
+                               //uv[1] = (((float)y) + 0.5f) / (float)ibuf->y;
+                               uv[1] = (float)y / ibuf_yf; /* use pixel offset UV coords instead */
+                               
+                               has_x_isect = 0;
+                               for (x = bounds_px.xmin; x < bounds_px.xmax; x++) {
+                                       //uv[0] = (((float)x) + 0.5f) / ibuf->x;
+                                       uv[0] = (float)x / ibuf_xf; /* use pixel offset UV coords instead */
+                                       
+                                       if (IsectPoly2Df(uv, uv_clip, uv_clip_tot)) {
+                                               
+                                               has_x_isect = has_isect = 1;
+                                               
+                                               if (is_ortho)   screen_px_from_ortho(uv, v1coSS, v2coSS, v3coSS, uv1co, uv2co, uv3co, pixelScreenCo, w);
+                                               else                    screen_px_from_persp(uv, v1coSS, v2coSS, v3coSS, uv1co, uv2co, uv3co, pixelScreenCo, w);
+                                               
+                                               /* a pitty we need to get the worldspace pixel location here */
+                                               if(G.vd->flag & V3D_CLIPPING) {
+                                                       VecWeightf(wco, ps->dm_mvert[ (*(&mf->v1 + i1)) ].co, ps->dm_mvert[ (*(&mf->v1 + i2)) ].co, ps->dm_mvert[ (*(&mf->v1 + i3)) ].co, w);
+                                                       Mat4MulVecfl(ps->ob->obmat, wco);
+                                                       if(view3d_test_clipping(G.vd, wco)) {
+                                                               continue; /* Watch out that no code below this needs to run */
+                                                       }
+                                               }
+                                               
+                                               /* Is this UV visible from the view? - raytrace */
+                                               /* project_paint_PickFace is less complex, use for testing */
+                                               //if (project_paint_PickFace(ps, pixelScreenCo, w, &side) == face_index) {
+                                               if (ps->do_occlude==0 || !project_bucket_point_occluded(ps, bucketFaceNodes, face_index, pixelScreenCo)) {
+                                                       
+                                                       mask = project_paint_uvpixel_mask(ps, face_index, side, w);
+                                                       
+                                                       if (mask > 0.0f) {
+                                                               BLI_linklist_prepend_arena(
+                                                                       bucketPixelNodes,
+                                                                       project_paint_uvpixel_init(ps, arena, ibuf, x, y, mask, face_index, image_index, pixelScreenCo, side, w),
+                                                                       arena
+                                                               );
+                                                       }
+                                               }
+                                               
+                                       }
+//#if 0
+                                       else if (has_x_isect) {
+                                               /* assuming the face is not a bow-tie - we know we cant intersect again on the X */
+                                               break;
+                                       }
+//#endif
+                               }
+                               
+                               
+#if 0                  /* TODO - investigate why this dosnt work sometimes! it should! */
+                               /* no intersection for this entire row, after some intersection above means we can quit now */
+                               if (has_x_isect==0 && has_isect) { 
+                                       break;
+                               }
+#endif
+                       }
+               }
+       } while(side--);
+
+       
+       
+#ifndef PROJ_DEBUG_NOSEAMBLEED
+       if (ps->seam_bleed_px > 0.0f) {
+               int face_seam_flag;
+               
+               if (ps->thread_tot > 1)
+                       BLI_lock_thread(LOCK_CUSTOM1); /* Other threads could be modifying these vars */
+               
+               face_seam_flag = ps->faceSeamFlags[face_index];
+               
+               /* are any of our edges un-initialized? */
+               if ((face_seam_flag & (PROJ_FACE_SEAM1|PROJ_FACE_NOSEAM1))==0 || 
+                       (face_seam_flag & (PROJ_FACE_SEAM2|PROJ_FACE_NOSEAM2))==0 || 
+                       (face_seam_flag & (PROJ_FACE_SEAM3|PROJ_FACE_NOSEAM3))==0 || 
+                       (face_seam_flag & (PROJ_FACE_SEAM4|PROJ_FACE_NOSEAM4))==0
+               ) {
+                       project_face_seams_init(ps, face_index, mf->v4);
+                       face_seam_flag = ps->faceSeamFlags[face_index];
+                       //printf("seams - %d %d %d %d\n", flag&PROJ_FACE_SEAM1, flag&PROJ_FACE_SEAM2, flag&PROJ_FACE_SEAM3, flag&PROJ_FACE_SEAM4);
+               }
+               
+               if ((face_seam_flag & (PROJ_FACE_SEAM1|PROJ_FACE_SEAM2|PROJ_FACE_SEAM3|PROJ_FACE_SEAM4))==0) {
+                       
+                       if (ps->thread_tot > 1)
+                               BLI_unlock_thread(LOCK_CUSTOM1); /* Other threads could be modifying these vars */
+                       
+               }
+               else {
+                       /* we have a seam - deal with it! */
+                       
+                       /* Now create new UV's for the seam face */
+                       float (*outset_uv)[2] = ps->faceSeamUVs[face_index];
+                       float insetCos[4][3]; /* inset face coords.  NOTE!!! ScreenSace for ortho, Worldspace in prespective view */
+
+                       float *uv_seam_quad[4];
+                       float fac;
+                       float *vCoSS[4]; /* vertex screenspace coords */
+                       
+                       float bucket_clip_edges[2][2]; /* store the screenspace coords of the face, clipped by the bucket's screen aligned rectangle */
+                       float edge_verts_inset_clip[2][3];
+                       int fidx1, fidx2; /* face edge pairs - loop throuh these ((0,1), (1,2), (2,3), (3,0)) or ((0,1), (1,2), (2,0)) for a tri */
+                       
+                       float seam_subsection[4][2];
+                       float fac1, fac2, ftot;
+                       
+                       
+                       if (outset_uv[0][0]==MAXFLOAT) /* first time initialize */
+                               uv_image_outset(tf_uv_pxoffset, outset_uv, ps->seam_bleed_px, ibuf->x, ibuf->y, mf->v4);
+                       
+                       /* ps->faceSeamUVs cant be modified when threading, now this is done we can unlock */
+                       if (ps->thread_tot > 1)
+                               BLI_unlock_thread(LOCK_CUSTOM1); /* Other threads could be modifying these vars */
+                       
+                       vCoSS[0] = ps->screenCoords[mf->v1];
+                       vCoSS[1] = ps->screenCoords[mf->v2];
+                       vCoSS[2] = ps->screenCoords[mf->v3];
+                       if (mf->v4)
+                               vCoSS[3] = ps->screenCoords[ mf->v4 ];
+                       
+                       /* PROJ_FACE_SCALE_SEAM must be slightly less then 1.0f */
+                       if (is_ortho) {
+                               if (mf->v4)     scale_quad(insetCos, vCoSS, PROJ_FACE_SCALE_SEAM);
+                               else            scale_tri(insetCos, vCoSS, PROJ_FACE_SCALE_SEAM);
+                       }
+                       else {
+                               if (mf->v4)     scale_quad(insetCos, vCo, PROJ_FACE_SCALE_SEAM);
+                               else            scale_tri(insetCos, vCo, PROJ_FACE_SCALE_SEAM);
+                       }
+                       
+                       side = 0; /* for triangles this wont need to change */
+                       
+                       for (fidx1 = 0; fidx1 < (mf->v4 ? 4 : 3); fidx1++) {
+                               if (mf->v4)             fidx2 = (fidx1==3) ? 0 : fidx1+1; /* next fidx in the face (0,1,2,3) -> (1,2,3,0) */
+                               else                    fidx2 = (fidx1==2) ? 0 : fidx1+1; /* next fidx in the face (0,1,2) -> (1,2,0) */
+                               
+                               if (    (face_seam_flag & (1<<fidx1)) && /* 1<<fidx1 -> PROJ_FACE_SEAM# */
+                                               line_clip_rect2f(bucket_bounds, vCoSS[fidx1], vCoSS[fidx2], bucket_clip_edges[0], bucket_clip_edges[1])
+                               ) {
+
+                                       ftot = Vec2Lenf(vCoSS[fidx1], vCoSS[fidx2]); /* screenspace edge length */
+                                       
+                                       if (ftot > 0.0f) { /* avoid div by zero */
+                                               if (mf->v4) {
+                                                       if (fidx1==2 || fidx2==2)       side= 1;
+                                                       else                                            side= 0;
+                                               }
+                                               
+                                               fac1 = Vec2Lenf(vCoSS[fidx1], bucket_clip_edges[0]) / ftot;
+                                               fac2 = Vec2Lenf(vCoSS[fidx1], bucket_clip_edges[1]) / ftot;
+                                               
+                                               uv_seam_quad[0] = tf_uv_pxoffset[fidx1];
+                                               uv_seam_quad[1] = tf_uv_pxoffset[fidx2];
+                                               uv_seam_quad[2] = outset_uv[fidx2];
+                                               uv_seam_quad[3] = outset_uv[fidx1];
+                                               
+                                               Vec2Lerpf(seam_subsection[0], uv_seam_quad[0], uv_seam_quad[1], fac1);
+                                               Vec2Lerpf(seam_subsection[1], uv_seam_quad[0], uv_seam_quad[1], fac2);
+                                               
+                                               Vec2Lerpf(seam_subsection[2], uv_seam_quad[3], uv_seam_quad[2], fac2);
+                                               Vec2Lerpf(seam_subsection[3], uv_seam_quad[3], uv_seam_quad[2], fac1);
+                                               
+                                               /* if the bucket_clip_edges values Z values was kept we could avoid this
+                                                * Inset needs to be added so occlusion tests wont hit adjacent faces */
+                                               VecLerpf(edge_verts_inset_clip[0], insetCos[fidx1], insetCos[fidx2], fac1);
+                                               VecLerpf(edge_verts_inset_clip[1], insetCos[fidx1], insetCos[fidx2], fac2);
+                                               
+
+                                               if (pixel_bounds_uv(seam_subsection[0], seam_subsection[1], seam_subsection[2], seam_subsection[3], &bounds_px, ibuf->x, ibuf->y, 1)) {
+                                                       /* bounds between the seam rect and the uvspace bucket pixels */
+                                                       
+                                                       has_isect = 0;
+                                                       for (y = bounds_px.ymin; y < bounds_px.ymax; y++) {
+                                                               // uv[1] = (((float)y) + 0.5f) / (float)ibuf->y;
+                                                               uv[1] = (float)y / ibuf_yf; /* use offset uvs instead */
+                                                               
+                                                               has_x_isect = 0;
+                                                               for (x = bounds_px.xmin; x < bounds_px.xmax; x++) {
+                                                                       //uv[0] = (((float)x) + 0.5f) / (float)ibuf->x;
+                                                                       uv[0] = (float)x / ibuf_xf; /* use offset uvs instead */
+                                                                       
+                                                                       /* test we're inside uvspace bucket and triangle bounds */
+                                                                       if (IsectPQ2Df(uv, seam_subsection[0], seam_subsection[1], seam_subsection[2], seam_subsection[3])) {
+                                                                               
+                                                                               /* We need to find the closest point along the face edge,
+                                                                                * getting the screen_px_from_*** wont work because our actual location
+                                                                                * is not relevent, since we are outside the face, Use VecLerpf to find
+                                                                                * our location on the side of the face's UV */
+                                                                               /*
+                                                                               if (is_ortho)   screen_px_from_ortho(ps, uv, v1co, v2co, v3co, uv1co, uv2co, uv3co, pixelScreenCo);
+                                                                               else                                    screen_px_from_persp(ps, uv, v1co, v2co, v3co, uv1co, uv2co, uv3co, pixelScreenCo);
+                                                                               */
+                                                                               
+                                                                               /* Since this is a seam we need to work out where on the line this pixel is */
+                                                                               //fac = lambda_cp_line2(uv, uv_seam_quad[0], uv_seam_quad[1]);
+                                                                               
+                                                                               fac = lambda_cp_line2(uv, seam_subsection[0], seam_subsection[1]);
+                                                                               if (fac < 0.0f)         { VECCOPY(pixelScreenCo, edge_verts_inset_clip[0]); }
+                                                                               else if (fac > 1.0f)    { VECCOPY(pixelScreenCo, edge_verts_inset_clip[1]); }
+                                                                               else                            { VecLerpf(pixelScreenCo, edge_verts_inset_clip[0], edge_verts_inset_clip[1], fac); }
+                                                                               
+                                                                               if (!is_ortho) {
+                                                                                       pixelScreenCo[3] = 1.0f;
+                                                                                       Mat4MulVec4fl(ps->projectMat, pixelScreenCo);
+                                                                                       pixelScreenCo[0] = (float)(curarea->winx/2.0f)+(curarea->winx/2.0f)*pixelScreenCo[0]/pixelScreenCo[3];  
+                                                                                       pixelScreenCo[1] = (float)(curarea->winy/2.0f)+(curarea->winy/2.0f)*pixelScreenCo[1]/pixelScreenCo[3];
+                                                                                       pixelScreenCo[2] = pixelScreenCo[2]/pixelScreenCo[3]; /* Use the depth for bucket point occlusion */
+                                                                               }
+                                                                               
+                                                                               if (ps->do_occlude==0 || !project_bucket_point_occluded(ps, bucketFaceNodes, face_index, pixelScreenCo)) {
+                                                                                       
+                                                                                       /* Only bother calculating the weights if we intersect */
+                                                                                       if (ps->do_mask_normal || ps->dm_mtface_clone) {
+                                                                                               /* TODO, this is not QUITE correct since UV is not inside the UV's but good enough for seams */
+                                                                                               if (side) {
+                                                                                                       BarycentricWeights2f(tf_uv_pxoffset[0], tf_uv_pxoffset[2], tf_uv_pxoffset[3], uv, w);
+                                                                                               }
+                                                                                               else {
+                                                                                                       BarycentricWeights2f(tf_uv_pxoffset[0], tf_uv_pxoffset[1], tf_uv_pxoffset[2], uv, w);
+                                                                                               }
+                                                                                               
+                                                                                       }
+                                                                                       
+                                                                                       /* a pitty we need to get the worldspace pixel location here */
+                                                                                       if(G.vd->flag & V3D_CLIPPING) {
+                                                                                               if (side)       VecWeightf(wco, ps->dm_mvert[mf->v1].co, ps->dm_mvert[mf->v3].co, ps->dm_mvert[mf->v4].co, w);
+                                                                                               else            VecWeightf(wco, ps->dm_mvert[mf->v1].co, ps->dm_mvert[mf->v2].co, ps->dm_mvert[mf->v3].co, w);
+                                                                                               
+                                                                                               Mat4MulVecfl(ps->ob->obmat, wco);
+                                                                                               if(view3d_test_clipping(G.vd, wco)) {
+                                                                                                       continue; /* Watch out that no code below this needs to run */
+                                                                                               }
+                                                                                       }
+                                                                                       
+                                                                                       mask = project_paint_uvpixel_mask(ps, face_index, side, w);
+                                                                                       
+                                                                                       if (mask > 0.0f) {
+                                                                                               BLI_linklist_prepend_arena(
+                                                                                                       bucketPixelNodes,
+                                                                                                       project_paint_uvpixel_init(ps, arena, ibuf, x, y, mask, face_index, image_index, pixelScreenCo, side, w),
+                                                                                                       arena
+                                                                                               );
+                                                                                       }
+                                                                                       
+                                                                               }
+                                                                       }
+                                                                       else if (has_x_isect) {
+                                                                               /* assuming the face is not a bow-tie - we know we cant intersect again on the X */
+                                                                               break;
+                                                                       }
+                                                               }
+                                                               
+#if 0                                                  /* TODO - investigate why this dosnt work sometimes! it should! */
+                                                               /* no intersection for this entire row, after some intersection above means we can quit now */
+                                                               if (has_x_isect==0 && has_isect) { 
+                                                                       break;
+                                                               }
+#endif
+                                                       }
+                                               }
+                                       }
+                               }
+                       }
+               }
+       }
+#endif // PROJ_DEBUG_NOSEAMBLEED
+}
+
+
+/* takes floating point screenspace min/max and returns int min/max to be used as indicies for ps->bucketRect, ps->bucketFlags */
+static void project_paint_bucket_bounds(const ProjPaintState *ps, const float min[2], const float max[2], int bucketMin[2], int bucketMax[2])
+{
+       /* divide by bucketWidth & bucketHeight so the bounds are offset in bucket grid units */
+       bucketMin[0] = (int)(((float)(min[0] - ps->screenMin[0]) / ps->screen_width) * ps->buckets_x) + 0.5f; /* these offsets of 0.5 and 1.5 seem odd but they are correct */
+       bucketMin[1] = (int)(((float)(min[1] - ps->screenMin[1]) / ps->screen_height) * ps->buckets_y) + 0.5f;
+       
+       bucketMax[0] = (int)(((float)(max[0] - ps->screenMin[0]) / ps->screen_width) * ps->buckets_x) + 1.5f;
+       bucketMax[1] = (int)(((float)(max[1] - ps->screenMin[1]) / ps->screen_height) * ps->buckets_y) + 1.5f;  
+       
+       /* incase the rect is outside the mesh 2d bounds */
+       CLAMP(bucketMin[0], 0, ps->buckets_x);
+       CLAMP(bucketMin[1], 0, ps->buckets_y);
+       
+       CLAMP(bucketMax[0], 0, ps->buckets_x);
+       CLAMP(bucketMax[1], 0, ps->buckets_y);
+}
+
+/* set bucket_bounds to a screen space-aligned floating point bound-box */
+static void project_bucket_bounds(const ProjPaintState *ps, const int bucket_x, const int bucket_y, rctf *bucket_bounds)
+{
+       bucket_bounds->xmin =   ps->screenMin[0]+((bucket_x)*(ps->screen_width / ps->buckets_x));               /* left */
+       bucket_bounds->xmax =   ps->screenMin[0]+((bucket_x+1)*(ps->screen_width / ps->buckets_x));     /* right */
+       
+       bucket_bounds->ymin =   ps->screenMin[1]+((bucket_y)*(ps->screen_height / ps->buckets_y));              /* bottom */
+       bucket_bounds->ymax =   ps->screenMin[1]+((bucket_y+1)*(ps->screen_height  / ps->buckets_y));   /* top */
+}
+
+/* Fill this bucket with pixels from the faces that intersect it.
+ * 
+ * have bucket_bounds as an argument so we don;t need to give bucket_x/y the rect function needs */
+static void project_bucket_init(const ProjPaintState *ps, const int thread_index, const int bucket_index, rctf *bucket_bounds)
+{
+       LinkNode *node;
+       int face_index, image_index;
+       ImBuf *ibuf = NULL;
+       MTFace *tf;
+       
+       Image *tpage_last = NULL;
+       
+
+       if (ps->image_tot==1) {
+               /* Simple loop, no context switching */
+               ibuf = ps->projImages[0].ibuf;
+               
+               for (node = ps->bucketFaces[bucket_index]; node; node= node->next) { 
+                       project_paint_face_init(ps, thread_index, bucket_index, (int)node->link, 0, bucket_bounds, ibuf);
+               }
+       }
+       else {
+               
+               /* More complicated loop, switch between images */
+               for (node = ps->bucketFaces[bucket_index]; node; node= node->next) {
+                       face_index = (int)node->link;
+                               
+                       /* Image context switching */
+                       tf = ps->dm_mtface+face_index;
+                       if (tpage_last != tf->tpage) {
+                               tpage_last = tf->tpage;
+                               
+                               image_index = -1; /* sanity check */
+                               
+                               for (image_index=0; image_index < ps->image_tot; image_index++) {
+                                       if (ps->projImages[image_index].ima == tpage_last) {
+                                               ibuf = ps->projImages[image_index].ibuf;
+                                               break;
+                                       }
+                               }
+                       }
+                       /* context switching done */
+                       
+                       project_paint_face_init(ps, thread_index, bucket_index, face_index, image_index, bucket_bounds, ibuf);
+                       
+               }
+       }
+       
+       ps->bucketFlags[bucket_index] |= PROJ_BUCKET_INIT;
+}
+
+
+/* We want to know if a bucket and a face overlap in screen-space
+ * 
+ * Note, if this ever returns false positives its not that bad, since a face in the bounding area will have its pixels
+ * calculated when it might not be needed later, (at the moment at least)
+ * obviously it shouldn't have bugs though */
+
+static int project_bucket_face_isect(ProjPaintState *ps, float min[2], float max[2], int bucket_x, int bucket_y, int bucket_index, const MFace *mf)
+{
+       /* TODO - replace this with a tricker method that uses sideofline for all screenCoords's edges against the closest bucket corner */
+       rctf bucket_bounds;
+       float p1[2], p2[2], p3[2], p4[2];
+       float *v, *v1,*v2,*v3,*v4;
+       int fidx;
+       
+       project_bucket_bounds(ps, bucket_x, bucket_y, &bucket_bounds);
+       
+       /* Is one of the faces verts in the bucket bounds? */
+       
+       fidx = mf->v4 ? 3:2;
+       do {
+               v = ps->screenCoords[ (*(&mf->v1 + fidx)) ];
+               if (BLI_in_rctf(&bucket_bounds, v[0], v[1])) {
+                       return 1;
+               }
+       } while (fidx--);
+       
+       v1 = ps->screenCoords[mf->v1];
+       v2 = ps->screenCoords[mf->v2];
+       v3 = ps->screenCoords[mf->v3];
+       if (mf->v4) {
+               v4 = ps->screenCoords[mf->v4];
+       }
+       
+       p1[0] = bucket_bounds.xmin; p1[1] = bucket_bounds.ymin;
+       p2[0] = bucket_bounds.xmin;     p2[1] = bucket_bounds.ymax;
+       p3[0] = bucket_bounds.xmax;     p3[1] = bucket_bounds.ymax;
+       p4[0] = bucket_bounds.xmax;     p4[1] = bucket_bounds.ymin;
+               
+       if (mf->v4) {
+               if(     IsectPQ2Df(p1, v1, v2, v3, v4) || IsectPQ2Df(p2, v1, v2, v3, v4) || IsectPQ2Df(p3, v1, v2, v3, v4) || IsectPQ2Df(p4, v1, v2, v3, v4) ||
+                       /* we can avoid testing v3,v1 because another intersection MUST exist if this intersects */
+                       (IsectLL2Df(p1, p2, v1, v2) || IsectLL2Df(p1, p2, v2, v3) || IsectLL2Df(p1, p2, v3, v4)) ||
+                       (IsectLL2Df(p2, p3, v1, v2) || IsectLL2Df(p2, p3, v2, v3) || IsectLL2Df(p2, p3, v3, v4)) ||
+                       (IsectLL2Df(p3, p4, v1, v2) || IsectLL2Df(p3, p4, v2, v3) || IsectLL2Df(p3, p4, v3, v4)) ||
+                       (IsectLL2Df(p4, p1, v1, v2) || IsectLL2Df(p4, p1, v2, v3) || IsectLL2Df(p4, p1, v3, v4))
+               ) {
+                       return 1;
+               }
+       }
+       else {
+               if(     IsectPT2Df(p1, v1, v2, v3) || IsectPT2Df(p2, v1, v2, v3) || IsectPT2Df(p3, v1, v2, v3) || IsectPT2Df(p4, v1, v2, v3) ||
+                       /* we can avoid testing v3,v1 because another intersection MUST exist if this intersects */
+                       (IsectLL2Df(p1, p2, v1, v2) || IsectLL2Df(p1, p2, v2, v3)) ||
+                       (IsectLL2Df(p2, p3, v1, v2) || IsectLL2Df(p2, p3, v2, v3)) ||
+                       (IsectLL2Df(p3, p4, v1, v2) || IsectLL2Df(p3, p4, v2, v3)) ||
+                       (IsectLL2Df(p4, p1, v1, v2) || IsectLL2Df(p4, p1, v2, v3))
+               ) {
+                       return 1;
+               }
+       }
+
+       return 0;
+}
+
+/* Add faces to the bucket but dont initialize its pixels
+ * TODO - when painting occluded, sort the faces on their min-Z and only add faces that faces that are not occluded */
+static void project_paint_delayed_face_init(ProjPaintState *ps, const MFace *mf, const MTFace *tf, const int face_index)
+{
+       float min[2], max[2], *vCoSS;
+       int bucketMin[2], bucketMax[2]; /* for  ps->bucketRect indexing */
+       int fidx, bucket_x, bucket_y, bucket_index;
+       int has_x_isect = -1, has_isect = 0; /* for early loop exit */
+       MemArena *arena = ps->arena_mt[0]; /* just use the first thread arena since threading has not started yet */
+       
+       INIT_MINMAX2(min, max);
+       
+       fidx = mf->v4 ? 3:2;
+       do {
+               vCoSS = ps->screenCoords[ *(&mf->v1 + fidx) ];
+               DO_MINMAX2(vCoSS, min, max);
+       } while (fidx--);
+       
+       project_paint_bucket_bounds(ps, min, max, bucketMin, bucketMax);
+       
+       for (bucket_y = bucketMin[1]; bucket_y < bucketMax[1]; bucket_y++) {
+               has_x_isect = 0;
+               for (bucket_x = bucketMin[0]; bucket_x < bucketMax[0]; bucket_x++) {
+                       
+                       bucket_index = bucket_x + (bucket_y * ps->buckets_x);
+                       
+                       if (project_bucket_face_isect(ps, min, max, bucket_x, bucket_y, bucket_index, mf)) {
+                               BLI_linklist_prepend_arena(
+                                       &ps->bucketFaces[ bucket_index ],
+                                       SET_INT_IN_POINTER(face_index), /* cast to a pointer to shut up the compiler */
+                                       arena
+                               );
+                               
+                               has_x_isect = has_isect = 1;
+                       }
+                       else if (has_x_isect) {
+                               /* assuming the face is not a bow-tie - we know we cant intersect again on the X */
+                               break;
+                       }
+               }
+               
+               /* no intersection for this entire row, after some intersection above means we can quit now */
+               if (has_x_isect==0 && has_isect) { 
+                       break;
+               }
+       }
+       
+#ifndef PROJ_DEBUG_NOSEAMBLEED
+       if (ps->seam_bleed_px > 0.0f) {
+               if (!mf->v4) {
+                       ps->faceSeamFlags[face_index] |= PROJ_FACE_NOSEAM4; /* so this wont show up as an untagged edge */
+               }
+               **ps->faceSeamUVs[face_index] = MAXFLOAT; /* set as uninitialized */
+       }
+#endif
+}
+
+/* run once per stroke before projection painting */
+static void project_paint_begin(ProjPaintState *ps, short mval[2])
+{      
+       /* Viewport vars */
+       float mat[3][3];
+       
+       float no[3];
+       
+       float (*projScreenCo)[4]; /* Note, we could have 4D vectors are only needed for */
+       float projMargin;
+       /* Image Vars - keep track of images we have used */
+       LinkNode *image_LinkList = NULL;
+       LinkNode *node;
+       
+       ProjPaintImage *projIma;
+       Image *tpage_last = NULL;
+       
+       /* Face vars */
+       MFace *mf;
+       MTFace *tf;
+       
+       int a, i; /* generic looping vars */
+       int image_index = -1, face_index;
+       
+       MemArena *arena; /* at the moment this is just ps->arena_mt[0], but use this to show were not multithreading */
+       
+       /* ---- end defines ---- */
+       
+       /* paint onto the derived mesh */
+       ps->dm = mesh_get_derived_final(ps->ob, get_viewedit_datamask());
+       
+       if ( !CustomData_has_layer( &ps->dm->faceData, CD_MTFACE) ) {
+               ps->dm = NULL;
+               return; 
+       }
+       ps->dm_mvert = ps->dm->getVertArray(ps->dm);
+       ps->dm_mface = ps->dm->getFaceArray(ps->dm);
+       ps->dm_mtface= ps->dm->getFaceDataArray(ps->dm, CD_MTFACE);
+       
+       ps->dm_totvert = ps->dm->getNumVerts(ps->dm);
+       ps->dm_totface = ps->dm->getNumFaces(ps->dm);
+       
+       /* use clone mtface? */
+       
+       
+       /* Note, use the original mesh for getting the clone and mask layer index
+        * this avoids re-generating the derived mesh just to get the new index */
+       if (ps->do_layer_clone) {
+               //int layer_num = CustomData_get_clone_layer(&ps->dm->faceData, CD_MTFACE);
+               int layer_num = CustomData_get_clone_layer(&((Mesh *)ps->ob->data)->fdata, CD_MTFACE);
+               if (layer_num != -1)
+                       ps->dm_mtface_clone = CustomData_get_layer_n(&ps->dm->faceData, CD_MTFACE, layer_num);
+               
+               if (ps->dm_mtface_clone==NULL || ps->dm_mtface_clone==ps->dm_mtface) {
+                       ps->do_layer_clone = 0;
+                       ps->dm_mtface_clone= NULL;
+               }
+       }
+       
+       if (ps->do_layer_mask) {
+               //int layer_num = CustomData_get_mask_layer(&ps->dm->faceData, CD_MTFACE);
+               int layer_num = CustomData_get_mask_layer(&((Mesh *)ps->ob->data)->fdata, CD_MTFACE);
+               if (layer_num != -1)
+                       ps->dm_mtface_mask = CustomData_get_layer_n(&ps->dm->faceData, CD_MTFACE, layer_num);
+               
+               if (ps->dm_mtface_mask==NULL || ps->dm_mtface_mask==ps->dm_mtface) {
+                       ps->do_layer_mask = 0;
+                       ps->dm_mtface_mask = NULL;
+               }
+       }
+       
+
+       
+       ps->viewDir[0] = 0.0f;
+       ps->viewDir[1] = 0.0f;
+       ps->viewDir[2] = 1.0f;
+       
+       view3d_get_object_project_mat(curarea, ps->ob, ps->projectMat, ps->viewMat);
+       
+       /* viewDir - object relative */
+       Mat4Invert(ps->ob->imat, ps->ob->obmat);
+       Mat3CpyMat4(mat, G.vd->viewinv);
+       Mat3MulVecfl(mat, ps->viewDir);
+       Mat3CpyMat4(mat, ps->ob->imat);
+       Mat3MulVecfl(mat, ps->viewDir);
+       Normalize(ps->viewDir);
+       
+       /* viewPos - object relative */
+       VECCOPY(ps->viewPos, G.vd->viewinv[3]);
+       Mat3CpyMat4(mat, ps->ob->imat);
+       Mat3MulVecfl(mat, ps->viewPos);
+       VecAddf(ps->viewPos, ps->viewPos, ps->ob->imat[3]);
+       
+       {       /* only use these for running 'get_view3d_viewplane' */
+               rctf viewplane;
+               
+               ps->is_ortho = get_view3d_viewplane(curarea->winx, curarea->winy, &viewplane, &ps->clipsta, &ps->clipend, NULL);
+               
+               //printf("%f %f\n", ps->clipsta, ps->clipend);
+               if (ps->is_ortho) { /* only needed for ortho */
+                       float fac = 2.0f / (ps->clipend - ps->clipsta);  
+                       ps->clipsta *= fac;
+                       ps->clipend *= fac;
+               }
+               else {
+                       /* TODO - can we even adjust for clip start/end? */
+               }
+               
+       }
+       
+       ps->is_airbrush = (ps->brush->flag & BRUSH_AIRBRUSH) ? 1 : 0;
+       
+       ps->is_texbrush = (ps->brush->mtex[ps->brush->texact] && ps->brush->mtex[ps->brush->texact]->tex) ? 1 : 0;
+
+       
+       /* calculate vert screen coords
+        * run this early so we can calculate the x/y resolution of our bucket rect */
+       INIT_MINMAX2(ps->screenMin, ps->screenMax);
+       
+       ps->screenCoords = MEM_mallocN(sizeof(float) * ps->dm_totvert * 4, "ProjectPaint ScreenVerts");
+       projScreenCo = ps->screenCoords;
+       
+       if (ps->is_ortho) {
+               for(a=0; a < ps->dm_totvert; a++, projScreenCo++) {
+                       VECCOPY((*projScreenCo), ps->dm_mvert[a].co);
+                       Mat4MulVecfl(ps->projectMat, (*projScreenCo));
+                       
+                       /* screen space, not clamped */
+                       (*projScreenCo)[0] = (float)(curarea->winx/2.0f)+(curarea->winx/2.0f)*(*projScreenCo)[0];
+                       (*projScreenCo)[1] = (float)(curarea->winy/2.0f)+(curarea->winy/2.0f)*(*projScreenCo)[1];
+                       DO_MINMAX2((*projScreenCo), ps->screenMin, ps->screenMax);
+               }
+       }
+       else {
+               for(a=0; a < ps->dm_totvert; a++, projScreenCo++) {
+                       VECCOPY((*projScreenCo), ps->dm_mvert[a].co);
+                       (*projScreenCo)[3] = 1.0f;
+
+                       Mat4MulVec4fl(ps->projectMat, (*projScreenCo));
+
+                       
+                       if ((*projScreenCo)[3] > ps->clipsta) {
+                               /* screen space, not clamped */
+                               (*projScreenCo)[0] = (float)(curarea->winx/2.0f)+(curarea->winx/2.0f)*(*projScreenCo)[0]/(*projScreenCo)[3];
+                               (*projScreenCo)[1] = (float)(curarea->winy/2.0f)+(curarea->winy/2.0f)*(*projScreenCo)[1]/(*projScreenCo)[3];
+                               (*projScreenCo)[2] = (*projScreenCo)[2]/(*projScreenCo)[3]; /* Use the depth for bucket point occlusion */
+                               DO_MINMAX2((*projScreenCo), ps->screenMin, ps->screenMax);
+                       }
+                       else {
+                               /* TODO - deal with cases where 1 side of a face goes behind the view ?
+                                * 
+                                * After some research this is actually very tricky, only option is to
+                                * clip the derived mesh before painting, which is a Pain */
+                               (*projScreenCo)[0] = MAXFLOAT;
+                       }
+               }
+       }
+       
+       /* If this border is not added we get artifacts for faces that
+        * have a parallel edge and at the bounds of the the 2D projected verts eg
+        * - a single screen aligned quad */
+       projMargin = (ps->screenMax[0] - ps->screenMin[0]) * 0.000001f;
+       ps->screenMax[0] += projMargin;
+       ps->screenMin[0] -= projMargin;
+       projMargin = (ps->screenMax[1] - ps->screenMin[1]) * 0.000001f;
+       ps->screenMax[1] += projMargin;
+       ps->screenMin[1] -= projMargin;
+       
+#ifdef PROJ_DEBUG_WINCLIP
+       CLAMP(ps->screenMin[0], -ps->brush->size, curarea->winx + ps->brush->size);
+       CLAMP(ps->screenMax[0], -ps->brush->size, curarea->winx + ps->brush->size);
+
+       CLAMP(ps->screenMin[1], -ps->brush->size, curarea->winy + ps->brush->size);
+       CLAMP(ps->screenMax[1], -ps->brush->size, curarea->winy + ps->brush->size);
+#endif
+       
+       /* only for convenience */
+       ps->screen_width  = ps->screenMax[0] - ps->screenMin[0];
+       ps->screen_height = ps->screenMax[1] - ps->screenMin[1];
+       
+       ps->buckets_x = (int)(ps->screen_width / (((float)ps->brush->size) / PROJ_BUCKET_BRUSH_DIV));
+       ps->buckets_y = (int)(ps->screen_height / (((float)ps->brush->size) / PROJ_BUCKET_BRUSH_DIV));
+       
+       printf("\tscreenspace bucket division x:%d y:%d\n", ps->buckets_x, ps->buckets_y);
+       
+       /* really high values could cause problems since it has to allocate a few
+        * (ps->buckets_x*ps->buckets_y) sized arrays  */
+       CLAMP(ps->buckets_x, PROJ_BUCKET_RECT_MIN, PROJ_BUCKET_RECT_MAX);
+       CLAMP(ps->buckets_y, PROJ_BUCKET_RECT_MIN, PROJ_BUCKET_RECT_MAX);
+       
+       ps->bucketRect = (LinkNode **)MEM_callocN(sizeof(LinkNode *) * ps->buckets_x * ps->buckets_y, "paint-bucketRect");
+       ps->bucketFaces= (LinkNode **)MEM_callocN(sizeof(LinkNode *) * ps->buckets_x * ps->buckets_y, "paint-bucketFaces");
+       
+       ps->bucketFlags= (unsigned char *)MEM_callocN(sizeof(char) * ps->buckets_x * ps->buckets_y, "paint-bucketFaces");
+#ifndef PROJ_DEBUG_NOSEAMBLEED
+       if (ps->seam_bleed_px > 0.0f) {
+               ps->vertFaces= (LinkNode **)MEM_callocN(sizeof(LinkNode *) * ps->dm_totvert, "paint-vertFaces");
+               ps->faceSeamFlags = (char *)MEM_callocN(sizeof(char) * ps->dm_totface, "paint-faceSeamFlags");
+               ps->faceSeamUVs= MEM_mallocN(sizeof(float) * ps->dm_totface * 8, "paint-faceSeamUVs");
+       }
+#endif
+       
+       /* Thread stuff
+        * 
+        * very small brushes run a lot slower multithreaded since the advantage with
+        * threads is being able to fill in multiple buckets at once.
+        * Only use threads for bigger brushes. */
+       
+       if (G.scene->r.mode & R_FIXED_THREADS) {
+               ps->thread_tot = G.scene->r.threads;
+       }
+       else {
+               ps->thread_tot = BLI_system_thread_count();
+       }
+       for (a=0; a<ps->thread_tot; a++) {
+               ps->arena_mt[a] = BLI_memarena_new(1<<16);
+       }
+       
+       arena = ps->arena_mt[0]; 
+       
+       if (ps->do_backfacecull && ps->do_mask_normal) {
+               MVert *v = ps->dm_mvert;
+               float viewDirPersp[3];
+               
+               ps->vertFlags = MEM_callocN(sizeof(char) * ps->dm_totvert, "paint-vertFlags");
+               
+               for(a=0; a < ps->dm_totvert; a++, v++) {
+                       no[0] = (float)(v->no[0] / 32767.0f);
+                       no[1] = (float)(v->no[1] / 32767.0f);
+                       no[2] = (float)(v->no[2] / 32767.0f);
+                       
+                       if (ps->is_ortho) {
+                               if (NormalizedVecAngle2(ps->viewDir, no) >= (M_PI_2 / 90) * ps->normal_angle) { /* 1 vert of this face is towards us */
+                                       ps->vertFlags[a] |= PROJ_VERT_CULL;
+                               }
+                       }
+                       else {
+                               VecSubf(viewDirPersp, ps->viewPos, v->co);
+                               Normalize(viewDirPersp);
+                               if (NormalizedVecAngle2(viewDirPersp, no) >= (M_PI_2 / 90) * ps->normal_angle) { /* 1 vert of this face is towards us */
+                                       ps->vertFlags[a] |= PROJ_VERT_CULL;
+                               }
+                       }
+               }
+       }
+       
+       /* setup clone offset */
+       if (ps->tool == PAINT_TOOL_CLONE) {
+               float projCo[4];
+               float *curs= give_cursor();
+               VECCOPY(projCo, curs);
+               Mat4MulVecfl(ps->ob->imat, projCo);
+               
+               projCo[3] = 1.0f;
+               Mat4MulVec4fl(ps->projectMat, projCo);
+               ps->cloneOffset[0] = mval[0] - ((float)(curarea->winx/2.0f)+(curarea->winx/2.0f)*projCo[0]/projCo[3]);
+               ps->cloneOffset[1] = mval[1] - ((float)(curarea->winy/2.0f)+(curarea->winy/2.0f)*projCo[1]/projCo[3]);
+               
+               // printf("%f %f   %f %f %f\n", ps->cloneOffset[0], ps->cloneOffset[1], curs[0], curs[1], curs[2]);
+               
+       }       
+
+       
+       
+       for(face_index = 0, tf = ps->dm_mtface, mf = ps->dm_mface; face_index < ps->dm_totface; mf++, tf++, face_index++) {
+               
+#ifndef PROJ_DEBUG_NOSEAMBLEED
+               /* add face user if we have bleed enabled, set the UV seam flags later */
+               /* annoying but we need to add all faces even ones we never use elsewhere */
+               if (ps->seam_bleed_px > 0.0f) {
+                       BLI_linklist_prepend_arena(&ps->vertFaces[mf->v1], (void *)face_index, arena);
+                       BLI_linklist_prepend_arena(&ps->vertFaces[mf->v2], (void *)face_index, arena);
+                       BLI_linklist_prepend_arena(&ps->vertFaces[mf->v3], (void *)face_index, arena);
+                       if (mf->v4) {
+                               BLI_linklist_prepend_arena(&ps->vertFaces[ mf->v4 ], (void *)face_index, arena);
+                       }
+               }
+#endif
+               
+               if (tf->tpage && ((G.f & G_FACESELECT)==0 || mf->flag & ME_FACE_SEL)) {
+                       
+                       float *v1coSS, *v2coSS, *v3coSS, *v4coSS;
+                       
+                       v1coSS = ps->screenCoords[mf->v1]; 
+                       v2coSS = ps->screenCoords[mf->v2]; 
+                       v3coSS = ps->screenCoords[mf->v3];
+                       if (mf->v4) {
+                               v4coSS = ps->screenCoords[mf->v4]; 
+                       }
+                       
+                       
+                       if (!ps->is_ortho) {
+                               if (    v1coSS[0]==MAXFLOAT ||
+                                               v2coSS[0]==MAXFLOAT ||
+                                               v3coSS[0]==MAXFLOAT ||
+                                               (mf->v4 && v4coSS[0]==MAXFLOAT)
+                               ) {
+                                       continue;
+                               }
+                       }
+                       
+#ifdef PROJ_DEBUG_WINCLIP
+                       /* ignore faces outside the view */
+                       if (
+                                  (v1coSS[0] < ps->screenMin[0] &&
+                                       v2coSS[0] < ps->screenMin[0] &&
+                                       v3coSS[0] < ps->screenMin[0] &&
+                                       (mf->v4 && v4coSS[0] < ps->screenMin[0])) ||
+                                       
+                                  (v1coSS[0] > ps->screenMax[0] &&
+                                       v2coSS[0] > ps->screenMax[0] &&
+                                       v3coSS[0] > ps->screenMax[0] &&
+                                       (mf->v4 && v4coSS[0] > ps->screenMax[0])) ||
+                                       
+                                  (v1coSS[1] < ps->screenMin[1] &&
+                                       v2coSS[1] < ps->screenMin[1] &&
+                                       v3coSS[1] < ps->screenMin[1] &&
+                                       (mf->v4 && v4coSS[1] < ps->screenMin[1])) ||
+                                       
+                                  (v1coSS[1] > ps->screenMax[1] &&
+                                       v2coSS[1] > ps->screenMax[1] &&
+                                       v3coSS[1] > ps->screenMax[1] &&
+                                       (mf->v4 && v4coSS[1] > ps->screenMax[1]))
+                       ) {
+                               continue;
+                       }
+                       
+#endif //PROJ_DEBUG_WINCLIP
+       
+                       
+                       if (ps->do_backfacecull) {
+                               if (ps->do_mask_normal) {
+                                       /* Since we are interpolating the normals of faces, we want to make 
+                                        * sure all the verts are pointing away from the view,
+                                        * not just the face */
+                                       if (    (ps->vertFlags[mf->v1] & PROJ_VERT_CULL) &&
+                                                       (ps->vertFlags[mf->v2] & PROJ_VERT_CULL) &&
+                                                       (ps->vertFlags[mf->v3] & PROJ_VERT_CULL) &&
+                                                       (mf->v4==0 || ps->vertFlags[mf->v4] & PROJ_VERT_CULL)
+                                                       
+                                       ) {
+                                               continue;
+                                       }
+                               }
+                               else {
+                                       if (SIDE_OF_LINE(v1coSS, v2coSS, v3coSS) < 0.0f) {
+                                               continue;
+                                       }
+                                       
+                               }
+                       }
+                       
+                       if (tpage_last != tf->tpage) {
+                               
+                               image_index = BLI_linklist_index(image_LinkList, tf->tpage);
+                               
+                               if (image_index==-1 && BKE_image_get_ibuf((Image *)tf->tpage, NULL)) { /* MemArena dosnt have an append func */
+                                       BLI_linklist_append(&image_LinkList, tf->tpage);
+                                       image_index = ps->image_tot;
+                                       ps->image_tot++;
+                               }
+                               
+                               tpage_last = tf->tpage;
+                       }
+                       
+                       if (image_index != -1) {
+                               /* Initialize the faces screen pixels */
+                               /* Add this to a list to initialize later */
+                               project_paint_delayed_face_init(ps, mf, tf, face_index);
+                       }
+               }
+       }
+       
+       /* build an array of images we use*/
+       projIma = ps->projImages = (ProjPaintImage *)BLI_memarena_alloc(arena, sizeof(ProjPaintImage) * ps->image_tot);
+       
+       for (node= image_LinkList, i=0; node; node= node->next, i++, projIma++) {
+               projIma->ima = node->link;
+               // calloced - projIma->touch = 0;
+               projIma->ibuf = BKE_image_get_ibuf(projIma->ima, NULL);
+               projIma->partRedrawRect =  BLI_memarena_alloc(arena, sizeof(ImagePaintPartialRedraw) * PROJ_BOUNDBOX_SQUARED);
+               // calloced - memset(projIma->partRedrawRect, 0, sizeof(ImagePaintPartialRedraw) * PROJ_BOUNDBOX_SQUARED);
+       }
+       
+       /* we have built the array, discard the linked list */
+       BLI_linklist_free(image_LinkList, NULL);
+}
+
+static void project_paint_end(ProjPaintState *ps)
+{
+       int a;
+       
+       /* build undo data from original pixel colors */
+       if(U.uiflag & USER_GLOBALUNDO) {
+               ProjPixel *projPixel;
+               ImBuf *tmpibuf = NULL, *tmpibuf_float = NULL;
+               LinkNode *pixel_node;
+               UndoTile *tile;
+               MemArena *arena = ps->arena_mt[0]; /* threaded arena re-used for non threaded case */
+                               
+               int bucket_tot = (ps->buckets_x * ps->buckets_y); /* we could get an X/Y but easier to loop through all possible buckets */
+               int bucket_index; 
+               int tile_index;
+               int x_round, y_round;
+               int x_tile, y_tile;
+               int is_float = -1;
+               
+               /* context */
+               ProjPaintImage *last_projIma;
+               int last_image_index = -1;
+               int last_tile_width;
+               
+               for(a=0, last_projIma=ps->projImages; a < ps->image_tot; a++, last_projIma++) {
+                       int size = sizeof(UndoTile **) * IMAPAINT_TILE_NUMBER(last_projIma->ibuf->x) * IMAPAINT_TILE_NUMBER(last_projIma->ibuf->y);
+                       last_projIma->undoRect = (UndoTile **) BLI_memarena_alloc(arena, size);
+                       memset(last_projIma->undoRect, 0, size);
+               }
+               
+               for (bucket_index = 0; bucket_index < bucket_tot; bucket_index++) {
+                       /* loop through all pixels */
+                       for(pixel_node= ps->bucketRect[bucket_index]; pixel_node; pixel_node= pixel_node->next) {
+                       
+                               /* ok we have a pixel, was it modified? */
+                               projPixel = (ProjPixel *)pixel_node->link;
+                               
+                               if (last_image_index != projPixel->image_index) {
+                                       /* set the context */
+                                       last_image_index =      projPixel->image_index;
+                                       last_projIma =          ps->projImages + last_image_index;
+                                       last_tile_width =       IMAPAINT_TILE_NUMBER(last_projIma->ibuf->x);
+                                       is_float =                      last_projIma->ibuf->rect_float ? 1 : 0;
+                               }
+                               
+                               
+                               if (    (is_float == 0 && projPixel->origColor.uint != *projPixel->pixel.uint_pt) || 
+                                                               
+                                               (is_float == 1 && 
+                                               (       projPixel->origColor.f[0] != projPixel->pixel.f_pt[0] || 
+                                                       projPixel->origColor.f[1] != projPixel->pixel.f_pt[1] ||
+                                                       projPixel->origColor.f[2] != projPixel->pixel.f_pt[2] ||
+                                                       projPixel->origColor.f[3] != projPixel->pixel.f_pt[3] ))
+                               ) {
+                                       
+                                       x_tile =  projPixel->x_px >> IMAPAINT_TILE_BITS;
+                                       y_tile =  projPixel->y_px >> IMAPAINT_TILE_BITS;
+                                       
+                                       x_round = x_tile * IMAPAINT_TILE_SIZE;
+                                       y_round = y_tile * IMAPAINT_TILE_SIZE;
+                                       
+                                       tile_index = x_tile + y_tile * last_tile_width;
+                                       
+                                       if (last_projIma->undoRect[tile_index]==NULL) {
+                                               /* add the undo tile from the modified image, then write the original colors back into it */
+                                               tile = last_projIma->undoRect[tile_index] = undo_init_tile(&last_projIma->ima->id, last_projIma->ibuf, is_float ? (&tmpibuf_float):(&tmpibuf) , x_tile, y_tile);
+                                       }
+                                       else {
+                                               tile = last_projIma->undoRect[tile_index];
+                                       }
+                                       
+                                       /* This is a BIT ODD, but overwrite the undo tiles image info with this pixels original color
+                                        * because allocating the tiles allong the way slows down painting */
+                                       
+                                       if (is_float) {
+                                               float *rgba_fp = (float *)tile->rect + (((projPixel->x_px - x_round) + (projPixel->y_px - y_round) * IMAPAINT_TILE_SIZE)) * 4;
+                                               QUATCOPY(rgba_fp, projPixel->origColor.f);
+                                       }
+                                       else {
+                                               ((unsigned int *)tile->rect)[ (projPixel->x_px - x_round) + (projPixel->y_px - y_round) * IMAPAINT_TILE_SIZE ] = projPixel->origColor.uint;
+                                       }
+                               }
+                       }
+               }
+               
+               if (tmpibuf)            IMB_freeImBuf(tmpibuf);
+               if (tmpibuf_float)      IMB_freeImBuf(tmpibuf_float);
+       }
+       /* done calculating undo data */
+       
+       MEM_freeN(ps->screenCoords);
+       MEM_freeN(ps->bucketRect);
+       MEM_freeN(ps->bucketFaces);
+       MEM_freeN(ps->bucketFlags);
+       
+       if (ps->seam_bleed_px > 0.0f) {
+               MEM_freeN(ps->vertFaces);
+               MEM_freeN(ps->faceSeamFlags);
+               MEM_freeN(ps->faceSeamUVs);
+       }
+       
+       if (ps->vertFlags) MEM_freeN(ps->vertFlags);
+       
+       
+       for (a=0; a<ps->thread_tot; a++) {
+               BLI_memarena_free(ps->arena_mt[a]);
+       }
+       
+       ps->dm->release(ps->dm);
+}
+
+
+/* external functions */
+
+/* 1= an undo, -1 is a redo. */
+static void partial_redraw_array_init(ImagePaintPartialRedraw *pr)
+{
+       int tot = PROJ_BOUNDBOX_SQUARED;
+       while (tot--) {
+               pr->x1 = 10000000;
+               pr->y1 = 10000000;
+               
+               pr->x2 = -1;
+               pr->y2 = -1;
+               
+               pr->enabled = 1;
+               
+               pr++;
+       }
+}
+
+
+static int partial_redraw_array_merge(ImagePaintPartialRedraw *pr, ImagePaintPartialRedraw *pr_other, int tot)
+{
+       int touch;
+       while (tot--) {
+               pr->x1 = MIN2(pr->x1, pr_other->x1);
+               pr->y1 = MIN2(pr->y1, pr_other->y1);
+               
+               pr->x2 = MAX2(pr->x2, pr_other->x2);
+               pr->y2 = MAX2(pr->y2, pr_other->y2);
+               
+               if (pr->x2 != -2)
+                       touch = 1;
+               
+               pr++; pr_other++;
+       }
+       
+       return touch;
+}
+
+/* Loop over all images on this mesh and update any we have touched */
+static int project_image_refresh_tagged(ProjPaintState *ps)
+{
+       ImagePaintPartialRedraw *pr;
+       ProjPaintImage *projIma;
+       int a,i;
+       int redraw = 0;
+       
+       
+       for (a=0, projIma=ps->projImages; a < ps->image_tot; a++, projIma++) {
+               if (projIma->touch) {
+                       /* look over each bound cell */
+                       for (i=0; i<PROJ_BOUNDBOX_SQUARED; i++) {
+                               pr = &(projIma->partRedrawRect[i]);
+                               if (pr->x2 != -1) { /* TODO - use 'enabled' ? */
+                                       imapaintpartial = *pr;
+                                       imapaint_image_update(projIma->ima, projIma->ibuf, 1); /*last 1 is for texpaint*/
+                                       redraw = 1;
+                               }
+                       }
+                       
+                       projIma->touch = 0; /* clear for reuse */
+               }
+       }
+       
+       return redraw;
+}
+
+/* run this per painting onto each mouse location */
+static int project_bucket_iter_init(ProjPaintState *ps, const float mval_f[2])
+{
+       float min_brush[2], max_brush[2];
+       float size_half = ((float)ps->brush->size) * 0.5f;
+       
+       /* so we dont have a bucket bounds that is way too small to paint into */
+       // if (size_half < 1.0f) size_half = 1.0f; // this dosnt work yet :/
+       
+       min_brush[0] = mval_f[0] - size_half;
+       min_brush[1] = mval_f[1] - size_half;
+       
+       max_brush[0] = mval_f[0] + size_half;
+       max_brush[1] = mval_f[1] + size_half;
+       
+       /* offset to make this a valid bucket index */
+       project_paint_bucket_bounds(ps, min_brush, max_brush, ps->bucketMin, ps->bucketMax);
+       
+       /* mouse outside the model areas? */
+       if (ps->bucketMin[0]==ps->bucketMax[0] || ps->bucketMin[1]==ps->bucketMax[1]) {
+               return 0;
+       }
+       
+       ps->context_bucket_x = ps->bucketMin[0];
+       ps->context_bucket_y = ps->bucketMin[1];
+       return 1;
+}
+
+static int project_bucket_iter_next(ProjPaintState *ps, int *bucket_index, rctf *bucket_bounds, const float mval[2])
+{
+       if (ps->thread_tot > 1)
+               BLI_lock_thread(LOCK_CUSTOM1);
+       
+       //printf("%d %d \n", ps->context_bucket_x, ps->context_bucket_y);
+       
+       for ( ; ps->context_bucket_y < ps->bucketMax[1]; ps->context_bucket_y++) {
+               for ( ; ps->context_bucket_x < ps->bucketMax[0]; ps->context_bucket_x++) {
+                       
+                       /* use bucket_bounds for project_bucket_isect_circle and project_bucket_init*/
+                       project_bucket_bounds(ps, ps->context_bucket_x, ps->context_bucket_y, bucket_bounds);
+                       
+                       if (project_bucket_isect_circle(ps->context_bucket_x, ps->context_bucket_y, mval, ps->brush->size * ps->brush->size, bucket_bounds)) {
+                               *bucket_index = ps->context_bucket_x + (ps->context_bucket_y * ps->buckets_x);
+                               ps->context_bucket_x++;
+                               
+                               if (ps->thread_tot > 1)
+                                       BLI_unlock_thread(LOCK_CUSTOM1);
+                               
+                               return 1;
+                       }
+               }
+               ps->context_bucket_x = ps->bucketMin[0];
+       }
+       
+       if (ps->thread_tot > 1)
+               BLI_unlock_thread(LOCK_CUSTOM1);
+       return 0;
+}
+
+/* Each thread gets one of these, also used as an argument to pass to project_paint_op */
+typedef struct ProjectHandle {
+       /* args */
+       ProjPaintState *ps;
+       float prevmval[2];
+       float mval[2];
+       
+       /* annoying but we need to have image bounds per thread, then merge into ps->projectPartialRedraws */
+       ProjPaintImage *projImages;     /* array of partial redraws */
+       
+       /* thread settings */
+       int thread_index;
+} ProjectHandle;
+
+static void blend_color_mix(unsigned char *cp, const unsigned char *cp1, const unsigned char *cp2, const int fac)
+{
+       /* this and other blending modes previously used >>8 instead of /255. both
+          are not equivalent (>>8 is /256), and the former results in rounding
+          errors that can turn colors black fast after repeated blending */
+       const int mfac= 255-fac;
+
+       cp[0]= (mfac*cp1[0]+fac*cp2[0])/255;
+       cp[1]= (mfac*cp1[1]+fac*cp2[1])/255;
+       cp[2]= (mfac*cp1[2]+fac*cp2[2])/255;
+}
+
+static void blend_color_mix_float(float *cp, const float *cp1, const float *cp2, const float fac)
+{
+       const float mfac= 1.0-fac;
+       cp[0]= mfac*cp1[0] + fac*cp2[0];
+       cp[1]= mfac*cp1[1] + fac*cp2[1];
+       cp[2]= mfac*cp1[2] + fac*cp2[2];
+}
+
+static void do_projectpaint_clone(ProjPaintState *ps, ProjPixel *projPixel, float *rgba, float alpha, float mask)
+{
+       if (ps->is_airbrush==0 && mask < 1.0f) {
+               projPixel->newColor.uint = IMB_blend_color(projPixel->newColor.uint, ((ProjPixelClone*)projPixel)->clonepx.uint, (int)(alpha*255), ps->blend);
+               blend_color_mix(projPixel->pixel.ch_pt,  projPixel->origColor.ch, projPixel->newColor.ch, (int)(mask*255));
+       }
+       else {
+               *projPixel->pixel.uint_pt = IMB_blend_color(*projPixel->pixel.uint_pt, ((ProjPixelClone*)projPixel)->clonepx.uint, (int)(alpha*mask*255), ps->blend);
+       }
+}
+
+static void do_projectpaint_clone_f(ProjPaintState *ps, ProjPixel *projPixel, float *rgba, float alpha, float mask)
+{
+       if (ps->is_airbrush==0 && mask < 1.0f) {
+               IMB_blend_color_float(projPixel->newColor.f, projPixel->newColor.f, ((ProjPixelClone *)projPixel)->clonepx.f, alpha, ps->blend);
+               blend_color_mix_float(projPixel->pixel.f_pt,  projPixel->origColor.f, projPixel->newColor.f, mask);
+       }
+       else {
+               IMB_blend_color_float(projPixel->pixel.f_pt, projPixel->pixel.f_pt, ((ProjPixelClone *)projPixel)->clonepx.f, alpha*mask, ps->blend);
+       }
+}
+
+/* do_projectpaint_smear*
+ * 
+ * note, mask is used to modify the alpha here, this is not correct since it allows
+ * accumulation of color greater then 'projPixel->mask' however in the case of smear its not 
+ * really that important to be correct as it is with clone and painting 
+ */
+static void do_projectpaint_smear(ProjPaintState *ps, ProjPixel *projPixel, float *rgba, float alpha, float mask, MemArena *smearArena, LinkNode **smearPixels, float co[2])
+{
+       unsigned char rgba_ub[4];
+       
+       if (project_paint_PickColor(ps, co, NULL, rgba_ub, 1)==0)
+               return; 
+       
+       ((ProjPixelClone *)projPixel)->clonepx.uint = IMB_blend_color(*projPixel->pixel.uint_pt, *((unsigned int *)rgba_ub), (int)(alpha*mask*255), ps->blend);
+       BLI_linklist_prepend_arena(smearPixels, (void *)projPixel, smearArena);
+} 
+
+static void do_projectpaint_smear_f(ProjPaintState *ps, ProjPixel *projPixel, float *rgba, float alpha, float mask, MemArena *smearArena, LinkNode **smearPixels_f, float co[2])
+{
+       unsigned char rgba_ub[4];
+       unsigned char rgba_smear[4];
+       
+       if (project_paint_PickColor(ps, co, NULL, rgba_ub, 1)==0)
+               return;
+       
+       IMAPAINT_FLOAT_RGBA_TO_CHAR(rgba_smear, projPixel->pixel.f_pt);
+       ((ProjPixelClone *)projPixel)->clonepx.uint = IMB_blend_color(*((unsigned int *)rgba_smear), *((unsigned int *)rgba_ub), (int)(alpha*mask*255), ps->blend);
+       BLI_linklist_prepend_arena(smearPixels_f, (void *)projPixel, smearArena);
+}
+
+static void do_projectpaint_draw(ProjPaintState *ps, ProjPixel *projPixel, float *rgba, float alpha, float mask)
+{
+       unsigned char rgba_ub[4];
+       
+       if (ps->is_texbrush) {
+               rgba_ub[0] = FTOCHAR(rgba[0] * ps->brush->rgb[0]);
+               rgba_ub[1] = FTOCHAR(rgba[1] * ps->brush->rgb[1]);
+               rgba_ub[2] = FTOCHAR(rgba[2] * ps->brush->rgb[2]);
+               rgba_ub[3] = FTOCHAR(rgba[3]);
+       }
+       else {
+               IMAPAINT_FLOAT_RGB_TO_CHAR(rgba_ub, ps->brush->rgb);
+               rgba_ub[3] = 255;
+       }
+       
+       if (ps->is_airbrush==0 && mask < 1.0f) {
+               projPixel->newColor.uint = IMB_blend_color(projPixel->newColor.uint, *((unsigned int *)rgba_ub), (int)(alpha*255), ps->blend);
+               blend_color_mix(projPixel->pixel.ch_pt,  projPixel->origColor.ch, projPixel->newColor.ch, (int)(mask*255));
+       }
+       else {
+               *projPixel->pixel.uint_pt = IMB_blend_color(*projPixel->pixel.uint_pt, *((unsigned int *)rgba_ub), (int)(alpha*mask*255), ps->blend);
+       }
+}
+
+static void do_projectpaint_draw_f(ProjPaintState *ps, ProjPixel *projPixel, float *rgba, float alpha, float mask) {
+       if (ps->is_texbrush) {
+               rgba[0] *= ps->brush->rgb[0];
+               rgba[1] *= ps->brush->rgb[1];
+               rgba[2] *= ps->brush->rgb[2];
+       }
+       else {
+               VECCOPY(rgba, ps->brush->rgb);
+       }
+       
+       if (ps->is_airbrush==0 && mask < 1.0f) {
+               IMB_blend_color_float(projPixel->newColor.f, projPixel->newColor.f, rgba, alpha, ps->blend);
+               blend_color_mix_float(projPixel->pixel.f_pt,  projPixel->origColor.f, projPixel->newColor.f, mask);
+       }
+       else {
+               IMB_blend_color_float(projPixel->pixel.f_pt, projPixel->pixel.f_pt, rgba, alpha*mask, ps->blend);
+       }
+}
 
-               if (!ima || !ibuf || !(ibuf->rect || ibuf->rect_float))
-                       continue;
 
-               undo_copy_tile(tile, tmpibuf, ibuf, 1);
 
-               GPU_free_image(ima); /* force OpenGL reload */
-               if(ibuf->rect_float)
-                       imb_freerectImBuf(ibuf); /* force recreate of char rect */
+/* run this for single and multithreaded painting */
+static void *do_projectpaint_thread(void *ph_v)
+{
+       /* First unpack args from the struct */
+       ProjPaintState *ps =                    ((ProjectHandle *)ph_v)->ps;
+       ProjPaintImage *projImages =    ((ProjectHandle *)ph_v)->projImages;
+       const float *lastpos =                  ((ProjectHandle *)ph_v)->prevmval;
+       const float *pos =                              ((ProjectHandle *)ph_v)->mval;
+       const int thread_index =                ((ProjectHandle *)ph_v)->thread_index;
+       /* Done with args from ProjectHandle */
+
+       LinkNode *node;
+       ProjPixel *projPixel;
+       
+       int last_index = -1;
+       ProjPaintImage *last_projIma;
+       ImagePaintPartialRedraw *last_partial_redraw_cell;
+       
+       float rgba[4], alpha, dist_nosqrt;
+       
+       float brush_size_sqared;
+       float falloff;
+       int bucket_index;
+       int is_floatbuf = 0;
+       const short tool =  ps->tool;
+       rctf bucket_bounds;
+       
+       /* for smear only */
+       float pos_ofs[2];
+       float co[2];
+       float mask = 1.0f; /* airbrush wont use mask */
+       unsigned short mask_short;
+       
+       LinkNode *smearPixels = NULL;
+       LinkNode *smearPixels_f = NULL;
+       MemArena *smearArena = NULL; /* mem arena for this brush projection only */
+       
+       
+       if (tool==PAINT_TOOL_SMEAR) {
+               pos_ofs[0] = pos[0] - lastpos[0];
+               pos_ofs[1] = pos[1] - lastpos[1];
+               
+               smearArena = BLI_memarena_new(1<<16);
        }
+       
+       /* avoid a square root with every dist comparison */
+       brush_size_sqared = ps->brush->size * ps->brush->size; 
+       
+       /* printf("brush bounds %d %d %d %d\n", bucketMin[0], bucketMin[1], bucketMax[0], bucketMax[1]); */
+       
+       while (project_bucket_iter_next(ps, &bucket_index, &bucket_bounds, pos)) {                              
+               
+               /* Check this bucket and its faces are initialized */
+               if (ps->bucketFlags[bucket_index] == PROJ_BUCKET_NULL) {
+                       /* No pixels initialized */
+                       project_bucket_init(ps, thread_index, bucket_index, &bucket_bounds);
+               }
 
-       IMB_freeImBuf(tmpibuf);
-}
-
-static void undo_free(UndoElem *undo)
-{
-       UndoTile *tile;
+               for (node = ps->bucketRect[bucket_index]; node; node = node->next) {
+                       
+                       projPixel = (ProjPixel *)node->link;
+                       
+                       /*dist = Vec2Lenf(projPixel->projCoSS, pos);*/ /* correct but uses a sqrt */
+                       dist_nosqrt = Vec2Lenf_nosqrt(projPixel->projCoSS, pos);
+                       
+                       /*if (dist < s->brush->size) {*/ /* correct but uses a sqrt */
+                       if (dist_nosqrt < brush_size_sqared) {
+                               falloff = brush_sample_falloff_noalpha(ps->brush, sqrt(dist_nosqrt));
+                               if (falloff > 0.0f) {
+                                       if (ps->is_texbrush) {
+                                               brush_sample_tex(ps->brush, projPixel->projCoSS, rgba);
+                                               alpha = rgba[3];
+                                       } else {
+                                               alpha = 1.0f;
+                                       }
+                                       
+                                       if (ps->is_airbrush) {
+                                               /* for an aurbrush there is no real mask, so just multiply the alpha by it */
+                                               alpha *= falloff * ps->brush->alpha;
+                                               mask = ((float)projPixel->mask)/65535.0f;
+                                       }
+                                       else {
+                                               /* This brush dosnt accumulate so add some curve to the brushes falloff */
+                                               falloff = 1.0f - falloff;
+                                               falloff = 1.0f - (falloff * falloff);
+                                               
+                                               mask_short = projPixel->mask * (ps->brush->alpha * falloff);
+                                               if (mask_short > projPixel->mask_max) {
+                                                       mask = ((float)mask_short)/65535.0f;
+                                                       projPixel->mask_max = mask_short;
+                                               }
+                                               else {
+                                                       /*mask = ((float)projPixel->mask_max)/65535.0f;*/
+                                                       
+                                                       /* Go onto the next pixel */
+                                                       continue;
+                                               }
+                                       }
+                                       
+                                       if (alpha > 0.0f) {
+                                               
+                                               if (last_index != projPixel->image_index) {
+                                                       last_index = projPixel->image_index;
+                                                       last_projIma = projImages + last_index;
+                                                       
+                                                       last_projIma->touch = 1;
+                                                       is_floatbuf = last_projIma->ibuf->rect_float ? 1 : 0;
+                                               }
+                                               
+                                               last_partial_redraw_cell = last_projIma->partRedrawRect + projPixel->bb_cell_index;
+                                               last_partial_redraw_cell->x1 = MIN2(last_partial_redraw_cell->x1, projPixel->x_px);
+                                               last_partial_redraw_cell->y1 = MIN2(last_partial_redraw_cell->y1, projPixel->y_px);
+                                               
+                                               last_partial_redraw_cell->x2 = MAX2(last_partial_redraw_cell->x2, projPixel->x_px+1);
+                                               last_partial_redraw_cell->y2 = MAX2(last_partial_redraw_cell->y2, projPixel->y_px+1);
+                                               
+                                               
+                                               switch(tool) {
+                                               case PAINT_TOOL_CLONE:
+                                                       if (is_floatbuf) {
+                                                               if (((ProjPixelClone *)projPixel)->clonepx.f[3]) {
+                                                                       do_projectpaint_clone_f(ps, projPixel, rgba, alpha, mask);
+                                                               }
+                                                       }
+                                                       else {
+                                                               if (((ProjPixelClone*)projPixel)->clonepx.ch[3]) { 
+                                                                       do_projectpaint_clone(ps, projPixel, rgba, alpha, mask);
+                                                               }
+                                                       }
+                                                       break;
+                                               case PAINT_TOOL_SMEAR:
+                                                       Vec2Subf(co, projPixel->projCoSS, pos_ofs);
+                                                       
+                                                       if (is_floatbuf)        do_projectpaint_smear_f(ps, projPixel, rgba, alpha, mask, smearArena, &smearPixels_f, co);
+                                                       else                            do_projectpaint_smear(ps, projPixel, rgba, alpha, mask, smearArena, &smearPixels, co);
+                                                       break;
+                                               default:
+                                                       if (is_floatbuf)        do_projectpaint_draw_f(ps, projPixel, rgba, alpha, mask);
+                                                       else                            do_projectpaint_draw(ps, projPixel, rgba, alpha, mask);
+                                                       break;
+                                               }
+                                       }
+                                       /* done painting */
+                               }
+                       }
+               }
+       }
 
-       for(tile=undo->tiles.first; tile; tile=tile->next)
-               MEM_freeN(tile->rect);
-       BLI_freelistN(&undo->tiles);
+       
+       if (tool==PAINT_TOOL_SMEAR) {
+               
+               for (node= smearPixels; node; node= node->next) { /* this wont run for a float image */
+                       projPixel = node->link;
+                       *projPixel->pixel.uint_pt = ((ProjPixelClone *)projPixel)->clonepx.uint;
+               }
+               
+               for (node= smearPixels_f; node; node= node->next) { /* this wont run for a float image */
+                       projPixel = node->link;
+                       IMAPAINT_CHAR_RGBA_TO_FLOAT(projPixel->pixel.f_pt,  ((ProjPixelClone *)projPixel)->clonepx.ch);
+                       node = node->next;
+               }
+               
+               BLI_memarena_free(smearArena);
+       }
+       
+       return NULL;
 }
 
-static void undo_imagepaint_push_begin(char *name)
+static int project_paint_op(void *state, ImBuf *ibufb, float *lastpos, float *pos)
 {
-       UndoElem *uel;
-       int nr;
+       /* First unpack args from the struct */
+       ProjPaintState *ps = (ProjPaintState *)state;
+       int touch_any = 0;      
        
-       /* Undo push is split up in begin and end, the reason is that as painting
-        * happens more tiles are added to the list, and at the very end we know
-        * how much memory the undo used to remove old undo elements */
-
-       /* remove all undos after (also when curundo==NULL) */
-       while(undobase.last != curundo) {
-               uel= undobase.last;
-               undo_free(uel);
-               BLI_freelinkN(&undobase, uel);
+       ProjectHandle handles[BLENDER_MAX_THREADS];
+       ListBase threads;
+       int a,i;
+       
+       if (!project_bucket_iter_init(ps, pos)) {
+               return 0;
        }
        
-       /* make new */
-       curundo= uel= MEM_callocN(sizeof(UndoElem), "undo file");
-       BLI_addtail(&undobase, uel);
-
-       /* name can be a dynamic string */
-       strncpy(uel->name, name, MAXUNDONAME-1);
+       if (ps->thread_tot > 1)
+               BLI_init_threads(&threads, do_projectpaint_thread, ps->thread_tot);
        
-       /* limit amount to the maximum amount*/
-       nr= 0;
-       uel= undobase.last;
-       while(uel) {
-               nr++;
-               if(nr==U.undosteps) break;
-               uel= uel->prev;
+       /* get the threads running */
+       for(a=0; a < ps->thread_tot; a++) {
+               
+               /* set defaults in handles */
+               //memset(&handles[a], 0, sizeof(BakeShade));
+               
+               handles[a].ps = ps;
+               VECCOPY2D(handles[a].mval, pos);
+               VECCOPY2D(handles[a].prevmval, lastpos);
+               
+               /* thread spesific */
+               handles[a].thread_index = a;
+               
+               handles[a].projImages = (ProjPaintImage *)BLI_memarena_alloc(ps->arena_mt[a], ps->image_tot * sizeof(ProjPaintImage));
+               
+               memcpy(handles[a].projImages, ps->projImages, ps->image_tot * sizeof(ProjPaintImage));
+               
+               /* image bounds */
+               for (i=0; i< ps->image_tot; i++) {
+                       handles[a].projImages[i].partRedrawRect = (ImagePaintPartialRedraw *)BLI_memarena_alloc(ps->arena_mt[a], sizeof(ImagePaintPartialRedraw) * PROJ_BOUNDBOX_SQUARED);
+                       memcpy(handles[a].projImages[i].partRedrawRect, ps->projImages[i].partRedrawRect, sizeof(ImagePaintPartialRedraw) * PROJ_BOUNDBOX_SQUARED);                     
+               }
+
+               if (ps->thread_tot > 1)
+                       BLI_insert_thread(&threads, &handles[a]);
        }
-       if(uel) {
-               while(undobase.first!=uel) {
-                       UndoElem *first= undobase.first;
-                       undo_free(first);
-                       BLI_freelinkN(&undobase, first);
+       
+       if (ps->thread_tot > 1) /* wait for everything to be done */
+               BLI_end_threads(&threads);
+       else
+               do_projectpaint_thread(&handles[0]);
+               
+       
+       /* move threaded bounds back into ps->projectPartialRedraws */
+       for(i=0; i < ps->image_tot; i++) {
+               int touch = 0;
+               for(a=0; a < ps->thread_tot; a++) {
+                       touch |= partial_redraw_array_merge(ps->projImages[i].partRedrawRect, handles[a].projImages[i].partRedrawRect, PROJ_BOUNDBOX_SQUARED);
+               }
+               
+               if (touch) {
+                       ps->projImages[i].touch = 1;
+                       touch_any = 1;
                }
        }
+       
+       return touch_any;
 }
 
-static void undo_imagepaint_push_end()
-{
-       UndoElem *uel;
-       unsigned long totmem, maxmem;
 
-       if(U.undomemory != 0) {
-               /* limit to maximum memory (afterwards, we can't know in advance) */
-               totmem= 0;
-               maxmem= ((unsigned long)U.undomemory)*1024*1024;
+static int project_paint_sub_stroke(ProjPaintState *ps, BrushPainter *painter, short *prevmval_i, short *mval_i, double time, float pressure)
+{
+       
+       /* Use mouse coords as floats for projection painting */
+       float pos[2];
+       
+       pos[0] = mval_i[0];
+       pos[1] = mval_i[1];
+       
+       // we may want to use this later 
+       // brush_painter_require_imbuf(painter, ((ibuf->rect_float)? 1: 0), 0, 0);
+       
+       if (brush_painter_paint(painter, project_paint_op, pos, time, pressure, ps)) {
+               return 1;
+       }
+       else return 0;
+}
 
-               uel= undobase.last;
-               while(uel) {
-                       totmem+= uel->undosize;
-                       if(totmem>maxmem) break;
-                       uel= uel->prev;
-               }
 
-               if(uel) {
-                       while(undobase.first!=uel) {
-                               UndoElem *first= undobase.first;
-                               undo_free(first);
-                               BLI_freelinkN(&undobase, first);
+static int project_paint_stroke(ProjPaintState *ps, BrushPainter *painter, short *prevmval_i, short *mval_i, double time, int update, float pressure)
+{
+       int a, redraw = 0;
+       
+       for (a=0; a < ps->image_tot; a++) {
+               partial_redraw_array_init(ps->projImages[a].partRedrawRect);
+       }
+       
+       redraw |= project_paint_sub_stroke(ps, painter, prevmval_i, mval_i, time, pressure);
+       
+       if (update) {
+               if (project_image_refresh_tagged(ps)) {
+                       if (redraw) {
+                               force_draw(0); /* imapaint_redraw just calls this in viewport paint anyway */
+                               /* imapaint_redraw(0, 1, NULL); */
+                               /* imapaint_clear_partial_redraw(); */ /* not needed since we use our own array */
                        }
                }
        }
+       
+       return redraw;
 }
 
-/* external functions */
-
-/* 1= an undo, -1 is a redo. */
 void undo_imagepaint_step(int step)
 {
        UndoElem *undo;
@@ -333,9 +3951,9 @@ static void imapaint_clear_partial_redraw()
 
 static void imapaint_dirty_region(Image *ima, ImBuf *ibuf, int x, int y, int w, int h)
 {
-       ImBuf *tmpibuf;
+       ImBuf *tmpibuf = NULL;
        UndoTile *tile;
-       int srcx= 0, srcy= 0, origx, allocsize;
+       int srcx= 0, srcy= 0, origx;
 
        IMB_rectclip(ibuf, NULL, &x, &y, &srcx, &srcy, &w, &h);
 
@@ -360,9 +3978,6 @@ static void imapaint_dirty_region(Image *ima, ImBuf *ibuf, int x, int y, int w,
        h = ((y + h - 1) >> IMAPAINT_TILE_BITS);
        origx = (x >> IMAPAINT_TILE_BITS);
        y = (y >> IMAPAINT_TILE_BITS);
-
-       tmpibuf= IMB_allocImBuf(IMAPAINT_TILE_SIZE, IMAPAINT_TILE_SIZE, 32,
-                               IB_rectfloat|IB_rect, 0);
        
        for (; y <= h; y++) {
                for (x=origx; x <= w; x++) {
@@ -371,32 +3986,21 @@ static void imapaint_dirty_region(Image *ima, ImBuf *ibuf, int x, int y, int w,
                                        break;
 
                        if(!tile) {
-                               tile= MEM_callocN(sizeof(UndoTile), "ImaUndoTile");
-                               tile->id= ima->id;
-                               tile->x= x;
-                               tile->y= y;
-
-                               allocsize= IMAPAINT_TILE_SIZE*IMAPAINT_TILE_SIZE*4;
-                               allocsize *= (ibuf->rect_float)? sizeof(float): sizeof(char);
-                               tile->rect= MEM_mapallocN(allocsize, "ImaUndoRect");
-
-                               undo_copy_tile(tile, tmpibuf, ibuf, 0);
-                               curundo->undosize += allocsize;
-
-                               BLI_addtail(&curundo->tiles, tile);
+                               undo_init_tile(&ima->id, ibuf, &tmpibuf, x, y);
                        }
                }
        }
 
        ibuf->userflags |= IB_BITMAPDIRTY;
-
-       IMB_freeImBuf(tmpibuf);
+       
+       if (tmpibuf)
+               IMB_freeImBuf(tmpibuf);
 }
 
 static void imapaint_image_update(Image *image, ImBuf *ibuf, short texpaint)
 {
        if(ibuf->rect_float)
-               imb_freerectImBuf(ibuf); /* force recreate of char rect */
+               imb_freerectImBuf(ibuf); /* force recreate of char rect */ /* TODO - should just update a portion from imapaintpartial! */
        if(ibuf->mipmap[0])
                imb_freemipmapImBuf(ibuf);
 
@@ -404,6 +4008,7 @@ static void imapaint_image_update(Image *image, ImBuf *ibuf, short texpaint)
        if(texpaint || G.sima->lock) {
                int w = imapaintpartial.x2 - imapaintpartial.x1;
                int h = imapaintpartial.y2 - imapaintpartial.y1;
+               // printf("%d, %d, \n", w, h);
                GPU_paint_update_image(image, imapaintpartial.x1, imapaintpartial.y1, w, h);
        }
 }
@@ -461,14 +4066,20 @@ static void imapaint_ibuf_get_set_rgb(ImBuf *ibuf, int x, int y, short torus, sh
        if (ibuf->rect_float) {
                float *rrgbf = ibuf->rect_float + (ibuf->x*y + x)*4;
 
-               if (set) IMAPAINT_FLOAT_RGB_COPY(rrgbf, rgb)
-               else IMAPAINT_FLOAT_RGB_COPY(rgb, rrgbf)
+               if (set) {
+                       IMAPAINT_FLOAT_RGB_COPY(rrgbf, rgb);
+               } else {
+                       IMAPAINT_FLOAT_RGB_COPY(rgb, rrgbf);
+               }
        }
        else {
                char *rrgb = (char*)ibuf->rect + (ibuf->x*y + x)*4;
 
-               if (set) IMAPAINT_FLOAT_RGB_TO_CHAR(rrgb, rgb)
-               else IMAPAINT_CHAR_RGB_TO_FLOAT(rgb, rrgb)
+               if (set) {
+                       IMAPAINT_FLOAT_RGB_TO_CHAR(rrgb, rgb)
+               } else {
+                       IMAPAINT_CHAR_RGB_TO_FLOAT(rgb, rrgb)
+               }
        }
 }
 
@@ -569,6 +4180,8 @@ static void imapaint_convert_brushco(ImBuf *ibufb, float *pos, int *ipos)
        ipos[1]= (int)(pos[1] - ibufb->y/2);
 }
 
+/* dosnt run for projection painting
+ * only the old style painting in the 3d view */
 static int imapaint_paint_op(void *state, ImBuf *ibufb, float *lastpos, float *pos)
 {
        ImagePaintState *s= ((ImagePaintState*)state);
@@ -710,6 +4323,7 @@ static int imapaint_paint_sub_stroke(ImagePaintState *s, BrushPainter *painter,
        else return 0;
 }
 
+/* this is only useful for debugging at the moment */
 static void imapaint_paint_stroke(ImagePaintState *s, BrushPainter *painter, short texpaint, short *prevmval, short *mval, double time, float pressure)
 {
        Image *newimage = NULL;
@@ -790,28 +4404,139 @@ static void imapaint_paint_stroke(ImagePaintState *s, BrushPainter *painter, sho
        }
 }
 
+
+static int imapaint_paint_gp_to_stroke(float **points_gp) {
+       bGPdata *gpd;
+       bGPDlayer *gpl;
+       bGPDframe *gpf;
+       bGPDstroke *gps;
+       bGPDspoint *pt;
+       
+       int tot_gp = 0;
+       float *vec_gp;
+       
+
+       
+       gpd = gpencil_data_getactive(NULL);
+       
+       if (gpd==NULL)
+               return 0;
+
+       
+       for (gpl= gpd->layers.first; gpl; gpl= gpl->next) {
+
+               
+               if (gpl->flag & GP_LAYER_HIDE) continue;
+               
+               gpf= gpencil_layer_getframe(gpl, CFRA, 0);
+               if (gpf==NULL)  continue;
+               
+               for (gps= gpf->strokes.first; gps; gps= gps->next) {
+                       //if (gps->flag & GP_STROKE_2DSPACE) {
+                               tot_gp += gps->totpoints;
+                       //}
+               }
+       }
+       
+       if (tot_gp==0)
+               return 0;
+       
+       *points_gp = MEM_mallocN(tot_gp*sizeof(float)*2, "gp_points");
+       vec_gp = *points_gp;
+       
+       printf("%d\n" , tot_gp);
+       
+       for (gpl= gpd->layers.first; gpl; gpl= gpl->next) {
+
+               
+               if (gpl->flag & GP_LAYER_HIDE) continue;
+               
+               gpf= gpencil_layer_getframe(gpl, CFRA, 0);
+               if (gpf==NULL)  continue;
+               
+               for (gps= gpf->strokes.first; gps; gps= gps->next) {
+                       //if (gps->flag & GP_STROKE_2DSPACE) {
+                               int i;
+
+                               /* fill up the array with points */
+                               for (i=0, pt=gps->points; i < gps->totpoints && pt; i++, pt++) {
+                                       //printf("- %f %f\n", pt->x, pt->y);
+                                       
+                                       vec_gp[0] = pt->x;
+                                       vec_gp[1] = pt->y;
+                                       //printf("%f %f\n", vec_gp[0], vec_gp[1]);
+                                       
+                                       vec_gp+=2;
+                               }
+                       //}
+               }
+       }
+       
+       return tot_gp;
+}
+
+extern int view_autodist(float mouse_worldloc[3]); /* view.c */
+
 void imagepaint_paint(short mousebutton, short texpaint)
 {
        ImagePaintState s;
+       ProjPaintState ps;
        BrushPainter *painter;
        ToolSettings *settings= G.scene->toolsettings;
-       short prevmval[2], mval[2];
+       short prevmval[2], mval[2], project = 0;
        double time;
        float pressure;
-
+       int init = 1;
+       
+       /* optional grease pencil stroke path */
+       float *points_gp = NULL;
+       float *vec_gp;
+       int tot_gp = 0, index_gp=0;
+       int stroke_gp = 0;
+       
+       double benchmark_time;
+       
        if(!settings->imapaint.brush)
                return;
-
+       
+       if (texpaint) { /* are we painting in the 3D view ? */
+               if ((settings->imapaint.flag & IMAGEPAINT_PROJECT_DISABLE)==0) {
+                       project= 1;
+               }
+       }
+       
+       
+       if (G.qual & LR_CTRLKEY) {
+               float *curs;
+               mouse_cursor();
+               curs = give_cursor();
+               view_autodist(curs);
+               return;
+       }
+       
+       /* TODO - grease pencil stroke is very basic now and only useful for benchmarking, should make this nicer */
+       if (G.rt==123) {
+               stroke_gp = 1;          
+       }
+       
        /* initialize state */
        memset(&s, 0, sizeof(s));
+       memset(&ps, 0, sizeof(ps));
+       
        s.brush = settings->imapaint.brush;
        s.tool = settings->imapaint.tool;
-       if(texpaint && (s.tool == PAINT_TOOL_CLONE))
+       if(texpaint && (project==0) && (s.tool == PAINT_TOOL_CLONE))
                s.tool = PAINT_TOOL_DRAW;
        s.blend = s.brush->blend;
-
+       
+       if (project) {
+               ps.brush = s.brush;
+               ps.tool = s.tool;
+               ps.blend = s.blend;
+       }
+       
        if(texpaint) {
-               s.ob = OBACT;
+               ps.ob = s.ob = OBACT;
                if (!s.ob || !(s.ob->lay & G.vd->lay)) return;
                s.me = get_mesh(s.ob);
                if (!s.me) return;
@@ -829,56 +4554,134 @@ void imagepaint_paint(short mousebutton, short texpaint)
                        return;
                }
        }
-
+       
+       getmouseco_areawin(mval); /* make sure this runs before project_paint_begin() */
+       
+       /* note, if we have no UVs on the derived mesh, then we must return here */
+       if (project) {
+               /* setup projection painting data */
+               ps.do_backfacecull = (settings->imapaint.flag & IMAGEPAINT_PROJECT_BACKFACE) ? 0 : 1;
+               ps.do_occlude = (settings->imapaint.flag & IMAGEPAINT_PROJECT_XRAY) ? 0 : 1;
+               ps.do_mask_normal = (settings->imapaint.flag & IMAGEPAINT_PROJECT_FLAT) ? 0 : 1;;
+               
+               if (ps.tool == PAINT_TOOL_CLONE)
+                       ps.do_layer_clone = (settings->imapaint.flag & IMAGEPAINT_PROJECT_LAYER_CLONE);
+               
+               ps.do_layer_mask = (settings->imapaint.flag & IMAGEPAINT_PROJECT_LAYER_MASK) ? 1 : 0;
+               ps.do_layer_mask_inv = (settings->imapaint.flag & IMAGEPAINT_PROJECT_LAYER_MASK_INV) ? 1 : 0;
+               
+               
+#ifndef PROJ_DEBUG_NOSEAMBLEED
+               ps.seam_bleed_px = settings->imapaint.seam_bleed; /* pixel num to bleed */
+#endif
+               ps.normal_angle = settings->imapaint.normal_angle;
+               
+               project_paint_begin(&ps, mval);
+               
+               if (ps.dm==NULL) {
+                       return;
+               }
+       }
+       
        settings->imapaint.flag |= IMAGEPAINT_DRAWING;
        undo_imagepaint_push_begin("Image Paint");
 
        /* create painter and paint once */
        painter= brush_painter_new(s.brush);
 
-       getmouseco_areawin(mval);
-
        pressure = get_pressure();
        s.blend = (get_activedevice() == 2)? BRUSH_BLEND_ERASE_ALPHA: s.brush->blend;
        
-       time= PIL_check_seconds_timer();
+       time= benchmark_time = PIL_check_seconds_timer();
        prevmval[0]= mval[0];
        prevmval[1]= mval[1];
-
-       /* special exception here for too high pressure values on first touch in
-          windows for some tablets */
-    if (!((s.brush->flag & (BRUSH_ALPHA_PRESSURE|BRUSH_SIZE_PRESSURE|
-               BRUSH_SPACING_PRESSURE|BRUSH_RAD_PRESSURE)) && (get_activedevice() != 0) && (pressure >= 0.99f)))
-               imapaint_paint_stroke(&s, painter, texpaint, prevmval, mval, time, pressure);
-
+       
+       if (project) {
+               if (stroke_gp) {
+                       tot_gp = imapaint_paint_gp_to_stroke(&points_gp);
+                       vec_gp = points_gp;
+                       prevmval[0]= (short)vec_gp[0];
+                       prevmval[1]= (short)vec_gp[1];
+               }
+       } else {
+               /* special exception here for too high pressure values on first touch in
+                windows for some tablets */ 
+               if (!((s.brush->flag & (BRUSH_ALPHA_PRESSURE|BRUSH_SIZE_PRESSURE|
+                       BRUSH_SPACING_PRESSURE|BRUSH_RAD_PRESSURE)) && (get_activedevice() != 0) && (pressure >= 0.99f)))
+                       imapaint_paint_stroke(&s, painter, texpaint, prevmval, mval, time, pressure);
+       }
        /* paint loop */
        do {
-               getmouseco_areawin(mval);
+               if (stroke_gp) {
+                       /* Stroke grease pencil path */
+                       mval[0]= (short)vec_gp[0];
+                       mval[1]= (short)vec_gp[1];
+                       //printf("%d %d\n", mval[0], mval[1]);
+                       vec_gp+=2;
+                       index_gp++;
+               }
+               else {
+                       getmouseco_areawin(mval);
+               }
 
                pressure = get_pressure();
                s.blend = (get_activedevice() == 2)? BRUSH_BLEND_ERASE_ALPHA: s.brush->blend;
                        
                time= PIL_check_seconds_timer();
 
-               if((mval[0] != prevmval[0]) || (mval[1] != prevmval[1])) {
-                       imapaint_paint_stroke(&s, painter, texpaint, prevmval, mval, time, pressure);
-                       prevmval[0]= mval[0];
-                       prevmval[1]= mval[1];
+               if (project) { /* Projection Painting */
+                       int redraw = 1;
+                       
+                       if (((s.brush->flag & BRUSH_AIRBRUSH) || init)  || ((mval[0] != prevmval[0]) || (mval[1] != prevmval[1]))) {
+                               redraw = project_paint_stroke(&ps, painter, prevmval, mval, time, stroke_gp ? 0 : 1, pressure);
+                               prevmval[0]= mval[0];
+                               prevmval[1]= mval[1];
+                       }
+                       else {
+                               if (stroke_gp==0) {
+                                       BIF_wait_for_statechange();
+                                       
+                                       if (redraw==0) {
+                                               /* Only so the brush outline is redrawn, pitty we need to do this
+                                                * however it wont run when the mouse is still so not too bad */
+                                               force_draw(0);
+                                       }
+                               }
+                       }
+                       
+                       init = 0;
+               }
+               else { 
+                       if((mval[0] != prevmval[0]) || (mval[1] != prevmval[1])) {
+                               imapaint_paint_stroke(&s, painter, texpaint, prevmval, mval, time, pressure);
+                               prevmval[0]= mval[0];
+                               prevmval[1]= mval[1];
+                       }
+                       else if (s.brush->flag & BRUSH_AIRBRUSH)
+                               imapaint_paint_stroke(&s, painter, texpaint, prevmval, mval, time, pressure);
+                       else
+                               BIF_wait_for_statechange();
                }
-               else if (s.brush->flag & BRUSH_AIRBRUSH)
-                       imapaint_paint_stroke(&s, painter, texpaint, prevmval, mval, time, pressure);
-               else
-                       BIF_wait_for_statechange();
-
                /* do mouse checking at the end, so don't check twice, and potentially
                   miss a short tap */
-       } while(get_mbut() & mousebutton);
+       } while((stroke_gp && index_gp < tot_gp) || (stroke_gp==0 && (get_mbut() & mousebutton)));
+       //} while(get_mbut() & mousebutton);
 
        /* clean up */
        settings->imapaint.flag &= ~IMAGEPAINT_DRAWING;
        imapaint_canvas_free(&s);
        brush_painter_free(painter);
 
+       if (project) {
+               project_paint_end(&ps);
+       }
+       
+       if (points_gp)
+               MEM_freeN(points_gp);
+       
+       if (stroke_gp)
+               printf("timed test %f\n", (float)(PIL_check_seconds_timer() - benchmark_time));
+       
        imapaint_redraw(1, texpaint, s.image);
        undo_imagepaint_push_end();
        
index e34624a25fecee7f03c4a14365d171a580a2979b..4267159baee6983e8444542ade0830d10e7905df 100644 (file)
@@ -1307,6 +1307,14 @@ static void winqreadview3dspace(ScrArea *sa, void *spacedata, BWinEvent *evt)
                                        }
                                }
                        }
+                       else if (!G.obedit && OBACT && G.f&G_TEXTUREPAINT){
+                               if(G.scene->toolsettings->imapaint.brush &&
+                                       event!=LEFTMOUSE && event!=RIGHTMOUSE && event!=MIDDLEMOUSE &&
+                                       (event==MOUSEY || event==MOUSEX) && bwin_qtest(sa->win)==0) {
+                                       allqueue(REDRAWVIEW3D, 0);
+                               }
+                       }
+                       
 
                        /* Handle retopo painting */
                        if(retopo_mesh_paint_check()) {