doc-page-extractor 0.0.7__py3-none-any.whl → 0.0.9__py3-none-any.whl

doc_page_extractor/clipper.py

@@ -73,6 +73,9 @@ def _size_and_wrapper(rect: Rectangle):
     else:
       widths.append(distance)
 
+  if len(widths) == 0 and len(heights) == 0:
+    return 0.0, 0.0, 0.0, 0.0
+
   width: float = sum(widths) / len(widths)
   height: float = sum(heights) / len(heights)
   max_width: float = width

doc_page_extractor/extractor.py

@@ -137,6 +137,10 @@ class DocExtractor:
   def _find_matched_layout(self, fragment: OCRFragment, layouts: list[Layout]) -> Layout | None:
     fragment_area = fragment.rect.area
     primary_layouts: list[(Layout, float)] = []
+
+    if fragment_area == 0.0:
+      return None
+
     for layout in layouts:
       area = intersection_area(fragment.rect, layout.rect)
       if area / fragment_area > 0.85:
@@ -169,6 +173,9 @@ class DocExtractor:
       fragment.order = i
 
   def _order_fragments_by_ai(self, width: int, height: int, layouts: list[Layout]):
+    if width == 0 or height == 0:
+      return
+
     layout_model = self._get_layout()
     boxes: list[list[int]] = []
     steps: float = 1000.0 # max value of layoutreader
@@ -268,6 +275,9 @@ class DocExtractor:
     width = right - left
     height = bottom - top
 
+    if width == 0 or height == 0:
+      return
+
     for _left, _top, _right, _bottom in boxes:
       yield (
         (_left - left) / width,

doc_page_extractor/ocr.py

@@ -63,16 +63,21 @@ class OCR:
       text, rank = res
       if is_space_text(text):
         continue
+
+      rect = Rectangle(
+        lt=(box[0][0], box[0][1]),
+        rt=(box[1][0], box[1][1]),
+        rb=(box[2][0], box[2][1]),
+        lb=(box[3][0], box[3][1]),
+      )
+      if rect.area == 0.0:
+        continue
+
       yield OCRFragment(
         order=index,
         text=text,
         rank=rank,
-        rect=Rectangle(
-          lt=(box[0][0], box[0][1]),
-          rt=(box[1][0], box[1][1]),
-          rb=(box[2][0], box[2][1]),
-          lb=(box[3][0], box[3][1]),
-        ),
+        rect=rect,
       )
       index += 1
 

doc_page_extractor/overlap.py

@@ -88,6 +88,9 @@ def regroup_lines(origin_fragments: list[OCRFragment]) -> list[OCRFragment]:
         x2 = max(x2, x)
         y2 = max(y2, y)
 
+    if proto_texts_len == 0:
+      continue
+
     fragments.append(OCRFragment(
       order=min_order,
       text=" ".join(texts),
@@ -112,11 +115,14 @@ def _split_fragments_into_groups(fragments: list[OCRFragment]) -> Generator[list
     height = y2 - y1
     median = (y1 + y2) / 2.0
 
+    if height == 0:
+      continue
+
     if len(group) > 0:
       next_mean_median = (sum_median + median) / (len(group) + 1)
       next_mean_height = (sum_height + height) / (len(group) + 1)
-      deviation_rate = abs(median - next_mean_median) / next_mean_height
 
+      deviation_rate = abs(median - next_mean_median) / next_mean_height
       if deviation_rate > max_deviation_rate:
         yield group
         group = []
@@ -141,7 +147,12 @@ def overlap_rate(polygon1: Polygon, polygon2: Polygon) -> float:
   else:
     overlay_width, overlay_height = _polygon_size(intersection)
     polygon2_width, polygon2_height = _polygon_size(polygon2)
-    return (overlay_width / polygon2_width + overlay_height / polygon2_height) / 2.0
+    if polygon2_width == 0.0 or polygon2_height == 0.0:
+      return 0.0
+    return (
+      overlay_width / polygon2_width +
+      overlay_height / polygon2_height
+    ) / 2.0
 
 def _polygon_size(polygon: Polygon) -> tuple[float, float]:
   x1: float = float("inf")

doc_page_extractor/plot.py

@@ -1,20 +1,73 @@
 from typing import Iterable
 from PIL import ImageDraw
+from PIL.ImageFont import load_default, FreeTypeFont
 from PIL.Image import Image
 from .types import Layout, LayoutClass
+from .rectangle import Point
 
 _FRAGMENT_COLOR = (0x49, 0xCF, 0xCB) # Light Green
+_Color = tuple[int, int, int]
 
-def plot(image: Image, layouts: Iterable[Layout]):
+def plot(image: Image, layouts: Iterable[Layout]) -> None:
+  layout_font = load_default(size=35)
+  fragment_font = load_default(size=25)
   draw = ImageDraw.Draw(image, mode="RGBA")
+
+  def _draw_number(position: Point, number: int, font: FreeTypeFont, bold: bool, color: _Color) -> None:
+    nonlocal draw
+    x, y = position
+    text = str(object=number)
+    width = len(text) * font.size
+    offset = round(font.size * 0.15)
+
+    for dx, dy in _generate_delta(bold):
+      draw.text(
+        xy=(x + dx - width - offset, y + dy),
+        text=text,
+        font=font,
+        fill=color,
+      )
+
   for layout in layouts:
-    draw.polygon([p for p in layout.rect], outline=_layout_color(layout), width=5)
+    draw.polygon(
+      xy=[p for p in layout.rect],
+      outline=_layout_color(layout),
+      width=5,
+    )
 
   for layout in layouts:
-    for fragments in layout.fragments:
-      draw.polygon([p for p in fragments.rect], outline=_FRAGMENT_COLOR, width=3)
+    for fragment in layout.fragments:
+      draw.polygon(
+        xy=[p for p in fragment.rect],
+        outline=_FRAGMENT_COLOR,
+        width=3,
+      )
+      _draw_number(
+        position=fragment.rect.lt,
+        number=fragment.order + 1,
+        font=fragment_font,
+        bold=False,
+        color=_FRAGMENT_COLOR,
+      )
+
+  for i, layout in enumerate(layouts):
+    _draw_number(
+      position=layout.rect.lt,
+      number=i + 1,
+      font=layout_font,
+      bold=True,
+      color=_layout_color(layout),
+    )
+
+def _generate_delta(bold: bool):
+  if bold:
+    for dx in range(-1, 2):
+      for dy in range(-1, 2):
+        yield dx, dy
+  else:
+    yield 0, 0
 
-def _layout_color(layout: Layout) -> tuple[int, int, int]:
+def _layout_color(layout: Layout) -> _Color:
   cls = layout.cls
   if cls == LayoutClass.TITLE:
     return (0x0A, 0x12, 0x2C) # Dark

doc_page_extractor/rotation.py

@@ -129,7 +129,9 @@ def _find_median(rotations: list[float]):
     mid2 = rotations[n // 2]
     return (mid1 + mid2) / 2
 
-def _find_mean(rotations: list[float]):
+def _find_mean(rotations: list[float]) -> float:
+  if len(rotations) == 0:
+    return 0.0
   return sum(rotations) / len(rotations)
 
 # rotation is in [0, pi)

{doc_page_extractor-0.0.7.dist-info → doc_page_extractor-0.0.9.dist-info}/METADATA

@@ -1,6 +1,6 @@
-Metadata-Version: 2.2
+Metadata-Version: 2.4
 Name: doc-page-extractor
-Version: 0.0.7
+Version: 0.0.9
 Summary: doc page extractor can identify text and format in images and return structured data.
 Home-page: https://github.com/Moskize91/doc-page-extractor
 Author: Tao Zeyu
@@ -19,6 +19,7 @@ Dynamic: author-email
 Dynamic: description
 Dynamic: description-content-type
 Dynamic: home-page
+Dynamic: license-file
 Dynamic: requires-dist
 Dynamic: summary
 

{doc_page_extractor-0.0.7.dist-info → doc_page_extractor-0.0.9.dist-info}/RECORD

@@ -1,15 +1,15 @@
 doc_page_extractor/__init__.py,sha256=jCf5lo3A9JVDquflYMlvH8nJIs3EjBt8AG5y8mwfS68,210
-doc_page_extractor/clipper.py,sha256=PDafB_9JGwV_dRY7oWe1yf44roSsEaCuEdg3VRMvJNo,3125
+doc_page_extractor/clipper.py,sha256=5S1TI0aqMebwlPv_Ih4Nxpp6MchEjOih-CiZfMWUAhI,3201
 doc_page_extractor/downloader.py,sha256=NbGN9ARnER8-gd4T1uc3W98WMEClVxMrqnShq8HibTw,455
-doc_page_extractor/extractor.py,sha256=D3SLWUAciq8jGU6mlkVwIon-4nHJaYoKpPjGCN_YLhQ,10055
+doc_page_extractor/extractor.py,sha256=RPoqc_UtbHe4BoQvtaJxWuKyHwoDkK_fuYvuyA31p0Y,10200
 doc_page_extractor/layoutreader.py,sha256=BdC4oPbtpXoLmYhjuSFrKn6SNoT2zWw_gi95sGAUwrk,4031
-doc_page_extractor/ocr.py,sha256=6eLUVx6NSuRAwrq8Mc2zYs3yocxpOgUQS_4LIIqywnQ,5147
+doc_page_extractor/ocr.py,sha256=tEalIlEz_80bJN8BXnHWi3UCMoPdJ5ylgkmaBV1dzVs,5201
 doc_page_extractor/ocr_corrector.py,sha256=RfRA1jESEuqC8_a2kUEvHblT_B4xBjE0OApLMl1JiRg,3917
-doc_page_extractor/overlap.py,sha256=9_WbHxbKIbHM6R3ZUP2YG33pZlbLCHgwFb--NF3cCG0,5155
-doc_page_extractor/plot.py,sha256=R8hbmdGjtw2pAH1lJkGc7Qbis4aRaaAkrkEo6WjbqyM,1378
+doc_page_extractor/overlap.py,sha256=Nx7Z6Ih_7SBE-JAjqlMx51PYiciXomPxIOAS37x6fco,5327
+doc_page_extractor/plot.py,sha256=4uibjS_x1SyEyjaJJd0YsBbzkgldDOCct4Ry2cOhdXU,2556
 doc_page_extractor/raw_optimizer.py,sha256=1KghECq_rJwuZZITTLQnGTKYivFKg_qDvMLN9g17sks,2844
 doc_page_extractor/rectangle.py,sha256=Tp__NPiY6JlYwYxejST7BUXhv_bl8tkmDXi4JgHCK6E,1539
-doc_page_extractor/rotation.py,sha256=Dp8rXfgCzHQwqlAbU-uQt-zHC6Jm9KsIjcR6IhFQ5EU,4284
+doc_page_extractor/rotation.py,sha256=QCZ-HqfDxIhnQw8KRHki2myj6-UusvNY7Mpjsu-wI-4,4334
 doc_page_extractor/types.py,sha256=UWghDwajMtEKEYUcOOjr5dM-MWiJ-P-8nWYagouf9ds,631
 doc_page_extractor/utils.py,sha256=3rtIxiTJ7W5yOuY0UHedUJ3G34tPOw0jdHnUdOQ1tWI,207
 doc_page_extractor/onnxocr/__init__.py,sha256=BK4YpX4pU0nRxbcI5f5cbIVfdBEsx4W980QYmpNQaH0,38
@@ -24,10 +24,10 @@ doc_page_extractor/onnxocr/predict_rec.py,sha256=qQrCs5jzCf5PYp-iEKJ53pcx_xRoJdJ
 doc_page_extractor/onnxocr/predict_system.py,sha256=yoqXunAsoboPsWe7qQjvQf2_SMW1T1QMriEoiGdX3BM,2721
 doc_page_extractor/onnxocr/rec_postprocess.py,sha256=qZt5Ripal7z9hniKq5e7azOkD9e6NR1ylWpRpznhweg,29556
 doc_page_extractor/onnxocr/utils.py,sha256=AQoHgQyv-jpPo4BsVzq3r7_ze698EZ-a7LJobm2fwUI,1864
+doc_page_extractor-0.0.9.dist-info/licenses/LICENSE,sha256=TfPDBt3ar0uv_f9cqCDMZ5rIzW3CY8anRRd4PkL6ejs,34522
 tests/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 tests/test_history_bus.py,sha256=WaCUW3U75SESMcLq0f5FKnpVUVRDvmfxLFE7Zo83e48,2517
-doc_page_extractor-0.0.7.dist-info/LICENSE,sha256=TfPDBt3ar0uv_f9cqCDMZ5rIzW3CY8anRRd4PkL6ejs,34522
-doc_page_extractor-0.0.7.dist-info/METADATA,sha256=s-ewJAyPQ1I_fgTee91NN99T42HcAaKFu1MAUhZKqdk,2203
-doc_page_extractor-0.0.7.dist-info/WHEEL,sha256=jB7zZ3N9hIM9adW7qlTAyycLYW9npaWKLRzaoVcLKcM,91
-doc_page_extractor-0.0.7.dist-info/top_level.txt,sha256=ErNybD_lBzAmw8mVBAK4htsAH_hp14jioZVex-tUqvM,25
-doc_page_extractor-0.0.7.dist-info/RECORD,,
+doc_page_extractor-0.0.9.dist-info/METADATA,sha256=vwcUU9DE9rUhpbz1UigpDN3YhNnNSRbHGAyJroHPV7Q,2225
+doc_page_extractor-0.0.9.dist-info/WHEEL,sha256=CmyFI0kx5cdEMTLiONQRbGQwjIoR1aIYB7eCAQ4KPJ0,91
+doc_page_extractor-0.0.9.dist-info/top_level.txt,sha256=ErNybD_lBzAmw8mVBAK4htsAH_hp14jioZVex-tUqvM,25
+doc_page_extractor-0.0.9.dist-info/RECORD,,

{doc_page_extractor-0.0.7.dist-info → doc_page_extractor-0.0.9.dist-info}/WHEEL

@@ -1,5 +1,5 @@
 Wheel-Version: 1.0
-Generator: setuptools (75.8.2)
+Generator: setuptools (78.1.0)
 Root-Is-Purelib: true
 Tag: py3-none-any