Fixes

Address bug 3207: Multiple sub-paths in a path were not correctly handled when "closepath" commands were present. When encountering "closepath", this fix generates a line that connects the current point with the first point of the current sub-path - however only if these two points are different. The "firstpoint" value is being set exclusively by the "moveto" command.
pymupdf · Mar 6, 2024 · 4ddcaa3 · 4ddcaa3
1 parent 2bfccf6
commit 4ddcaa3
Show file tree

Hide file tree

Showing 7 changed files with 186 additions and 80 deletions.
diff --git a/docs/page.rst b/docs/page.rst
@@ -511,16 +511,18 @@ In a nutshell, this is what you can do with PyMuPDF:
 
       |history_end|
 
-   .. method:: apply_redactions(images=PDF_REDACT_IMAGE_PIXELS)
+   .. method:: apply_redactions(images=PDF_REDACT_IMAGE_PIXELS, graphics=PDF_REDACT_LINE_ART_IF_TOUCHED)
 
-      PDF only: Remove all **text content** contained in any redaction rectangle.
+      PDF only: Remove all **content** contained in any redaction rectangle on the page.
 
       **This method applies and then deletes all redactions from the page.**
 
-      :arg int images: How to redact overlapping images. The default (2) blanks out overlapping pixels. *PDF_REDACT_IMAGE_NONE* (0) ignores, and *PDF_REDACT_IMAGE_REMOVE* (1) completely removes all overlapping images.
+      :arg int images: How to redact overlapping images. The default (2) blanks out overlapping pixels. `PDF_REDACT_IMAGE_NONE` (0) ignores, and `PDF_REDACT_IMAGE_REMOVE` (1) completely removes images overlapping any redaction annotation. Option `PDF_REDACT_IMAGE_REMOVE_UNLESS_INVISIBLE` (3) only removes images that are actually visible.
 
+      :arg int graphics: How to redact overlapping vector graphics (also called "line art" or "drawings"). The default (2) removes any overlapping vector graphics. `PDF_REDACT_LINE_ART_NONE` (0) ignores, and `PDF_REDACT_LINE_ART_IF_COVERED` (1) removes graphics fully contained in a redaction annotation.
 
-      :returns: *True* if at least one redaction annotation has been processed, *False* otherwise.
+
+      :returns: `True` if at least one redaction annotation has been processed, `False` otherwise.
 
       .. note::
          * Text contained in a redaction rectangle will be **physically** removed from the page (assuming :meth:`Document.save` with a suitable garbage option) and will no longer appear in e.g. text extractions or anywhere else. All redaction annotations will also be removed. Other annotations are unaffected.
@@ -546,6 +548,7 @@ In a nutshell, this is what you can do with PyMuPDF:
       * New in v1.16.11
       * Changed in v1.16.12: The previous *mark* parameter is gone. Instead, the respective rectangles are filled with the individual *fill* color of each redaction annotation. If a *text* was given in the annotation, then :meth:`insert_textbox` is invoked to insert it, using parameters provided with the redaction.
       * Changed in v1.18.0: added option for handling images that overlap redaction areas.
+      * Changed in v1.23.27: added option for removing graphics as well.
 
       |history_end|
 

diff --git a/src/__init__.py b/src/__init__.py
@@ -18839,10 +18839,23 @@ def closepath(self, ctx):    # trace_close().
                 if jm_checkrect(self.dev):
                     #log(f'end1: {self.dev.pathdict=}')
                     return
-            #log('setting self.dev.pathdict[ "closePath"] to true')
-            self.dev.pathdict[ "closePath"] = True
             self.dev.linecount = 0   # reset # of consec. lines
-            #log(f'end2: {self.dev.pathdict=}')
+
+            if self.dev.havemove:
+                if self.dev.lastpoint != self.dev.firstpoint:
+                    item = ("l", JM_py_from_point(self.dev.lastpoint),
+                                 JM_py_from_point(self.dev.firstpoint))
+                    self.dev.pathdict[dictkey_items].append(item)
+                    self.dev.lastpoint = self.dev.firstpoint
+                self.dev.pathdict["closePath"] = False
+
+            else:
+                #log('setting self.dev.pathdict[ "closePath"] to true')
+                self.dev.pathdict[ "closePath"] = True
+                #log(f'end2: {self.dev.pathdict=}')
+
+            self.dev.havemove = 0
+
         except Exception:
             if g_exceptions_verbose:    exception_info()
             raise
@@ -18900,7 +18913,9 @@ def moveto(self, ctx, x, y):   # trace_moveto().
             log(f'self.dev.pathdict:')
             for n, v in self.dev.pathdict.items():
                 log( '    {type(n)=} {len(n)=} {n!r} {n}: {v!r}: {v}')
+
         #log(f'Walker(): {type(self.dev.pathdict)=} {self.dev.pathdict=}')
+
         try:
             #log( '{=dev.ctm type(dev.ctm)}')
             self.dev.lastpoint = mupdf.fz_transform_point(
@@ -18914,6 +18929,8 @@ def moveto(self, ctx, x, y):   # trace_moveto().
                         self.dev.lastpoint.x,
                         self.dev.lastpoint.y,
                         )
+            self.dev.firstpoint = self.dev.lastpoint
+            self.dev.havemove = 1
             self.dev.linecount = 0  # reset # of consec. lines
         except Exception:
             if g_exceptions_verbose:    exception_info()
@@ -19282,6 +19299,8 @@ def __init__(self, out, clips, method):
         self.ctm = mupdf.FzMatrix()
         self.rot = mupdf.FzMatrix()
         self.lastpoint = mupdf.FzPoint()
+        self.firstpoint = mupdf.FzPoint()
+        self.havemove = 0
         self.pathrect = mupdf.FzRect()
         self.pathfactor = 0
         self.linecount = 0

diff --git a/src/extra.i b/src/extra.i
@@ -1957,6 +1957,8 @@ struct jm_lineart_device
     fz_matrix ptm = {};
     fz_matrix rot = {};
     fz_point lastpoint = {};
+    fz_point firstpoint = {};
+    int havemove = 0;
     fz_rect pathrect = {};
     int clips = {};
     int linecount = {};
@@ -2830,6 +2832,8 @@ trace_moveto(fz_context *ctx, void *dev_, float x, float y)
                 dev->lastpoint.y
                 );
     }
+    dev->firstpoint = dev->lastpoint;
+    dev->havemove = 1;
     dev->linecount = 0;  // reset # of consec. lines
 }
 
@@ -2887,8 +2891,22 @@ trace_close(fz_context *ctx, void *dev_)
             return;
         }
     }
-    DICT_SETITEMSTR_DROP(dev->pathdict, "closePath", JM_BOOL(1));
     dev->linecount = 0;  // reset # of consec. lines
+	if (dev->havemove) {
+		if (dev->firstpoint.x != dev->lastpoint.x || dev->firstpoint.y != dev->lastpoint.y) {
+			PyObject *list = PyTuple_New(3);
+			PyTuple_SET_ITEM(list, 0, PyUnicode_FromString("l"));
+			PyTuple_SET_ITEM(list, 1, JM_py_from_point(dev->lastpoint));
+			PyTuple_SET_ITEM(list, 2, JM_py_from_point(dev->firstpoint));
+			dev->lastpoint = dev->firstpoint;
+			PyObject *items = PyDict_GetItem(dev->pathdict, dictkey_items);
+			LIST_APPEND_DROP(items, list);
+		}
+		dev->havemove = 0;
+		DICT_SETITEMSTR_DROP(dev->pathdict, "closePath", JM_BOOL(0));
+	} else {
+		DICT_SETITEMSTR_DROP(dev->pathdict, "closePath", JM_BOOL(1));
+	}
 }
 
 static const fz_path_walker trace_path_walker =
@@ -2915,6 +2933,7 @@ jm_lineart_path(jm_lineart_device *dev, const fz_path *path)
     dev->pathrect = fz_infinite_rect;
     dev->linecount = 0;
     dev->lastpoint = fz_make_point(0, 0);
+    dev->firstpoint = fz_make_point(0, 0);
     if (dev->pathdict) {
         Py_CLEAR(dev->pathdict);
     }

diff --git a/src_classic/helper-devices.i b/src_classic/helper-devices.i
@@ -26,6 +26,8 @@ static fz_matrix trace_device_ptm;  // page transformation matrix
 static fz_matrix trace_device_ctm;  // trace device matrix
 static fz_matrix trace_device_rot;
 static fz_point dev_lastpoint = {0, 0};
+static fz_point dev_firstpoint = {0, 0};
+static int dev_havemove = 0;
 static fz_rect dev_pathrect;
 static float dev_pathfactor = 0;
 static int dev_linecount = 0;
@@ -47,6 +49,8 @@ static void trace_device_reset()
     trace_device_rot = fz_identity;
     dev_lastpoint.x = 0;
     dev_lastpoint.y = 0;
+    dev_firstpoint.x = 0;
+    dev_firstpoint.y = 0;
     dev_pathrect.x0 = 0;
     dev_pathrect.y0 = 0;
     dev_pathrect.x1 = 0;
@@ -218,6 +222,8 @@ trace_moveto(fz_context *ctx, void *dev_, float x, float y)
 		dev_pathrect = fz_make_rect(dev_lastpoint.x, dev_lastpoint.y,
 		                            dev_lastpoint.x, dev_lastpoint.y);
 	}
+	dev_firstpoint = dev_lastpoint;
+	dev_havemove = 1;
 	dev_linecount = 0;  // reset # of consec. lines
 }
 
@@ -272,8 +278,22 @@ trace_close(fz_context *ctx, void *dev_)
 			return;
 		}
 	}
-	DICT_SETITEMSTR_DROP(dev_pathdict, "closePath", JM_BOOL(1));
 	dev_linecount = 0;  // reset # of consec. lines
+	if (dev_havemove) {
+		if (dev_firstpoint.x != dev_lastpoint.x || dev_firstpoint.y != dev_lastpoint.y) {
+			PyObject *list = PyTuple_New(3);
+			PyTuple_SET_ITEM(list, 0, PyUnicode_FromString("l"));
+			PyTuple_SET_ITEM(list, 1, JM_py_from_point(dev_lastpoint));
+			PyTuple_SET_ITEM(list, 2, JM_py_from_point(dev_firstpoint));
+			dev_lastpoint = dev_firstpoint;
+			PyObject *items = PyDict_GetItem(dev_pathdict, dictkey_items);
+			LIST_APPEND_DROP(items, list);
+		}
+		dev_havemove = 0;
+		DICT_SETITEMSTR_DROP(dev_pathdict, "closePath", JM_BOOL(0));
+	} else {
+		DICT_SETITEMSTR_DROP(dev_pathdict, "closePath", JM_BOOL(1));
+	}
 }
 
 static const fz_path_walker trace_path_walker =

diff --git a/tests/resources/symbols.txt b/tests/resources/symbols.txt
@@ -1,11 +1,12 @@
-[{'closePath': True,
+[{'closePath': False,
   'color': (1.0, 1.0, 1.0),
   'dashes': '[] 0',
   'even_odd': False,
   'fill': (1.0, 0.0, 0.0),
   'fill_opacity': 1.0,
   'items': [('l', (50.0, 50.0), (50.0, 100.0)),
-            ('l', (50.0, 100.0), (100.0, 75.0))],
+            ('l', (50.0, 100.0), (100.0, 75.0)),
+            ('l', (100.0, 75.0), (50.0, 50.0))],
   'layer': '',
   'lineCap': (0, 0, 0),
   'lineJoin': 0.0,
@@ -14,7 +15,7 @@
   'stroke_opacity': 1.0,
   'type': 'fs',
   'width': 1.0},
- {'closePath': True,
+ {'closePath': False,
   'color': (1.0, 1.0, 1.0),
   'dashes': '[] 0',
   'even_odd': False,
@@ -48,7 +49,7 @@
   'stroke_opacity': 1.0,
   'type': 'fs',
   'width': 1.0},
- {'closePath': True,
+ {'closePath': False,
   'color': (0.0, 1.0, 0.0),
   'dashes': '[] 0',
   'even_odd': False,
@@ -66,15 +67,16 @@
   'stroke_opacity': 1.0,
   'type': 'fs',
   'width': 0.30000001192092896},
- {'closePath': True,
+ {'closePath': False,
   'color': (1.0, 1.0, 1.0),
   'dashes': '[] 0',
   'even_odd': False,
   'fill': (1.0, 0.0, 0.0),
   'fill_opacity': 1.0,
   'items': [('l', (75.0, 230.0), (100.0, 255.0)),
             ('l', (100.0, 255.0), (75.0, 280.0)),
-            ('l', (75.0, 280.0), (50.0, 255.0))],
+            ('l', (75.0, 280.0), (50.0, 255.0)),
+            ('l', (50.0, 255.0), (75.0, 230.0))],
   'layer': '',
   'lineCap': (0, 0, 0),
   'lineJoin': 0.0,
@@ -83,7 +85,7 @@
   'stroke_opacity': 1.0,
   'type': 'fs',
   'width': 1.0},
- {'closePath': True,
+ {'closePath': False,
   'color': (1.0, 1.0, 1.0),
   'dashes': '[] 0',
   'even_odd': False,
@@ -117,7 +119,7 @@
   'stroke_opacity': 1.0,
   'type': 'fs',
   'width': 2.0},
- {'closePath': True,
+ {'closePath': False,
   'color': (0.0, 0.0, 0.0),
   'dashes': '[] 0',
   'items': [('c',
@@ -163,7 +165,7 @@
   'stroke_opacity': 1.0,
   'type': 'fs',
   'width': 3.0},
- {'closePath': True,
+ {'closePath': False,
   'even_odd': False,
   'fill': (1.0, 1.0, 0.0),
   'fill_opacity': 1.0,
@@ -191,7 +193,7 @@
   'rect': (50.0, 350.0, 100.0, 400.0),
   'seqno': 13,
   'type': 'f'},
- {'closePath': True,
+ {'closePath': False,
   'even_odd': False,
   'fill': (0.0, 0.0, 0.0),
   'fill_opacity': 1.0,
@@ -317,14 +319,15 @@
   'stroke_opacity': 1.0,
   'type': 'fs',
   'width': 1.0},
- {'closePath': True,
+ {'closePath': False,
   'color': (1.0, 0.0, 0.0),
   'dashes': '[] 0',
   'even_odd': False,
   'fill': (1.0, 0.0, 0.0),
   'fill_opacity': 1.0,
   'items': [('c', (75.0, 485.0), (62.5, 470.0), (50.0, 490.0), (75.0, 510.0)),
-            ('c', (75.0, 485.0), (87.5, 470.0), (100.0, 490.0), (75.0, 510.0))],
+            ('c', (75.0, 485.0), (87.5, 470.0), (100.0, 490.0), (75.0, 510.0)),
+            ('l', (75.0, 510.0), (75.0, 485.0))],
   'layer': '',
   'lineCap': (0, 0, 0),
   'lineJoin': 0.0,
@@ -451,7 +454,7 @@
   'rect': (85.5072021484375, 547.7540283203125, 100.0, 562.2459716796875),
   'seqno': 27,
   'type': 'f'},
- {'closePath': True,
+ {'closePath': False,
   'color': (0.7215690016746521, 0.5254899859428406, 0.04313730075955391),
   'dashes': '[] 0',
   'even_odd': False,
@@ -461,7 +464,10 @@
              (85.5072021484375, 547.7540283203125),
              (86.30770111083984, 548.553955078125),
              (85.00990295410156, 549.8519897460938),
-             (82.60870361328125, 550.6519775390625))],
+             (82.60870361328125, 550.6519775390625)),
+            ('l',
+             (82.60870361328125, 550.6519775390625),
+             (85.5072021484375, 547.7540283203125))],
   'layer': '',
   'lineCap': (0, 0, 0),
   'lineJoin': 0.0,
@@ -473,7 +479,7 @@
   'stroke_opacity': 1.0,
   'type': 'fs',
   'width': 0.07246380299329758},
- {'closePath': True,
+ {'closePath': False,
   'color': (0.7215690016746521, 0.5254899859428406, 0.04313730075955391),
   'dashes': '[] 0',
   'even_odd': False,
@@ -483,7 +489,10 @@
              (82.60870361328125, 550.6519775390625),
              (87.2510986328125, 553.052978515625),
              (87.2510986328125, 556.947021484375),
-             (82.60870361328125, 559.3480224609375))],
+             (82.60870361328125, 559.3480224609375)),
+            ('l',
+             (82.60870361328125, 559.3480224609375),
+             (82.60870361328125, 550.6519775390625))],
   'layer': '',
   'lineCap': (0, 0, 0),
   'lineJoin': 0.0,
@@ -495,7 +504,7 @@
   'stroke_opacity': 1.0,
   'type': 'fs',
   'width': 0.07246380299329758},
- {'closePath': True,
+ {'closePath': False,
   'color': (0.7215690016746521, 0.5254899859428406, 0.04313730075955391),
   'dashes': '[] 0',
   'even_odd': False,
@@ -505,7 +514,10 @@
              (82.60870361328125, 559.3480224609375),
              (85.00990295410156, 560.1480102539062),
              (86.30770111083984, 561.446044921875),
-             (85.5072021484375, 562.2459716796875))],
+             (85.5072021484375, 562.2459716796875)),
+            ('l',
+             (85.5072021484375, 562.2459716796875),
+             (82.60870361328125, 559.3480224609375))],
   'layer': '',
   'lineCap': (0, 0, 0),
   'lineJoin': 0.0,
@@ -599,15 +611,18 @@
            557.1740112304688),
   'seqno': 37,
   'type': 'f'},
- {'closePath': True,
+ {'closePath': False,
   'color': (1.0, 1.0, 1.0),
   'dashes': '[] 0',
   'items': [('l',
              (58.937198638916016, 548.47802734375),
              (58.937198638916016, 561.52197265625)),
             ('l',
              (61.352699279785156, 548.47802734375),
-             (61.352699279785156, 561.52197265625))],
+             (61.352699279785156, 561.52197265625)),
+            ('l',
+             (61.352699279785156, 561.52197265625),
+             (61.352699279785156, 548.47802734375))],
   'layer': '',
   'lineCap': (0, 0, 0),
   'lineJoin': 0.0,
@@ -619,7 +634,7 @@
   'stroke_opacity': 1.0,
   'type': 's',
   'width': 1.1594200134277344},
- {'closePath': True,
+ {'closePath': False,
   'even_odd': False,
   'fill': (1.0, 1.0, 0.0),
   'fill_opacity': 1.0,
@@ -647,7 +662,7 @@
   'rect': (50.0, 590.0, 100.0, 640.0),
   'seqno': 39,
   'type': 'f'},
- {'closePath': True,
+ {'closePath': False,
   'even_odd': False,
   'fill': (0.0, 0.0, 0.0),
   'fill_opacity': 1.0,

diff --git a/tests/resources/test-3207.pdf b/tests/resources/test-3207.pdf