Detect dotted gridlines for tables

Addresses #3539

We previously did not detect dotted lines used as table gridlines.

If one of width / height is LE edge_min_length and LT the other dimension, the rectangle is treated as a vertical / horizontal line.
We incorrectly used the dimension-specific snap values for this.

We now no longer ignore rectangles if both, width and height are smaller then edge_min_length, but leave this to the snapping and joining algorithms further down the road.

src/table.py

@@ -1946,6 +1946,7 @@ def make_edges(page, clip=None, tset=None, add_lines=None):
     global EDGES
     snap_x = tset.snap_x_tolerance
     snap_y = tset.snap_y_tolerance
+    min_length = tset.edge_min_length
     lines_strict = (
         tset.vertical_strategy == "lines_strict"
         or tset.horizontal_strategy == "lines_strict"
@@ -2117,13 +2118,14 @@ def make_line(p, p1, p2, clip):
                 if line_dict:
                     EDGES.append(line_to_edge(line_dict))
 
-            elif i[0] == "re":  # a rectangle: decompose into 4 lines
-                rect = i[1].normalize()  # rectangle itself
-                # ignore minute rectangles
-                if rect.height <= snap_y and rect.width <= snap_x:
-                    continue
+            elif i[0] == "re":
+                # A rectangle: decompose into 4 lines, but filter out
+                # the ones that simulate a line
+                rect = i[1].normalize()  # normalize the rectangle
 
-                if rect.width <= snap_x:  # simulates a vertical line
+                if (
+                    rect.width <= min_length and rect.width < rect.height
+                ):  # simulates a vertical line
                     x = abs(rect.x1 + rect.x0) / 2  # take middle value for x
                     p1 = Point(x, rect.y0)
                     p2 = Point(x, rect.y1)
@@ -2132,7 +2134,9 @@ def make_line(p, p1, p2, clip):
                         EDGES.append(line_to_edge(line_dict))
                     continue
 
-                if rect.height <= snap_y:  # simulates a horizontal line
+                if (
+                    rect.height <= min_length and rect.height < rect.width
+                ):  # simulates a horizontal line
                     y = abs(rect.y1 + rect.y0) / 2  # take middle value for y
                     p1 = Point(rect.x0, y)
                     p2 = Point(rect.x1, y)

tests/test_tables.py

@@ -176,7 +176,9 @@ def test_2979():
     assert len(lengths) == 1
 
     # test 3001
-    assert pymupdf.TOOLS.set_small_glyph_heights() is False, f'{pymupdf.TOOLS.set_small_glyph_heights()=}'
+    assert (
+        pymupdf.TOOLS.set_small_glyph_heights() is False
+    ), f"{pymupdf.TOOLS.set_small_glyph_heights()=}"
 
 
 def test_3062():
@@ -292,3 +294,17 @@ def test_markdown():
         "|Col15|Col25 Col26||\n\n"
     )
     assert tab.to_markdown() == text
+
+
+def test_dotted_grid():
+    """Confirm dotted lines are detected as gridlines."""
+    filename = os.path.join(scriptdir, "resources", "dotted-gridlines.pdf")
+    doc = pymupdf.open(filename)
+    page = doc[0]
+    tabs = page.find_tables()
+    assert len(tabs.tables) == 3  # must be 3 tables
+    t0, t1, t2 = tabs  # extract them
+    # check that they have expected dimensions
+    assert t0.row_count, t0.col_count == (11, 12)
+    assert t1.row_count, t1.col_count == (25, 11)
+    assert t2.row_count, t2.col_count == (1, 10)