doc-page-extractor 0.1.0__py3-none-any.whl → 0.1.1__py3-none-any.whl

doc_page_extractor/extractor.py

@@ -116,12 +116,13 @@ class DocExtractor:
         lb=(x1, y2),
         rb=(x2, y2),
       )
-      if cls == LayoutClass.TABLE:
-        yield TableLayout(cls=cls, rect=rect, fragments=[], parsed=None)
-      elif cls == LayoutClass.ISOLATE_FORMULA:
-        yield FormulaLayout(cls=cls, rect=rect, fragments=[], latex=None)
-      else:
-        yield PlainLayout(cls=cls, rect=rect, fragments=[])
+      if rect.is_valid:
+        if cls == LayoutClass.TABLE:
+          yield TableLayout(cls=cls, rect=rect, fragments=[], parsed=None)
+        elif cls == LayoutClass.ISOLATE_FORMULA:
+          yield FormulaLayout(cls=cls, rect=rect, fragments=[], latex=None)
+        else:
+          yield PlainLayout(cls=cls, rect=rect, fragments=[])
 
   def _layouts_matched_by_fragments(self, fragments: list[OCRFragment], layouts: list[Layout]):
     layouts_group = self._split_layouts_by_group(layouts)

doc_page_extractor/ocr.py

@@ -69,7 +69,7 @@ class OCR:
         rb=(box[2][0], box[2][1]),
         lb=(box[3][0], box[3][1]),
       )
-      if rect.area == 0.0:
+      if not rect.is_valid or rect.area == 0.0:
         continue
 
       yield OCRFragment(

doc_page_extractor/rectangle.py

@@ -19,6 +19,10 @@ class Rectangle:
     yield self.rb
     yield self.rt
 
+  @property
+  def is_valid(self) -> bool:
+    return Polygon(self).is_valid
+
   @property
   def segments(self) -> Generator[tuple[Point, Point], None, None]:
     yield (self.lt, self.lb)
@@ -60,6 +64,8 @@ class Rectangle:
 def intersection_area(rect1: Rectangle, rect2: Rectangle) -> float:
   poly1 = Polygon(rect1)
   poly2 = Polygon(rect2)
+  if not poly1.is_valid or not poly2.is_valid:
+    return 0.0
   intersection = poly1.intersection(poly2)
   if intersection.is_empty:
     return 0.0

{doc_page_extractor-0.1.0.dist-info → doc_page_extractor-0.1.1.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: doc-page-extractor
-Version: 0.1.0
+Version: 0.1.1
 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

{doc_page_extractor-0.1.0.dist-info → doc_page_extractor-0.1.1.dist-info}/RECORD

@@ -1,16 +1,16 @@
 doc_page_extractor/__init__.py,sha256=9rWKSMTgzP7Xv15zA4upsyPaR8S8JeNpMyhWElRCW0M,311
 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=njLl8VdOgm-noWPDYTfjIAUU1giNc-wLvCSR1pHkfS8,7267
+doc_page_extractor/extractor.py,sha256=V0S9Nn65lKpkz8DnTUMcsAzUJGwSQvupIBTiVqzLpJ8,7303
 doc_page_extractor/latex.py,sha256=W_zAcksNRuru-WjCq4CSn07s_SWrDhikadJSy_Cg3Do,1954
 doc_page_extractor/layout_order.py,sha256=NwMzTPr4xsriz4slCwqwhw2-vrMu-qfwtcFsDu8d1yM,7426
 doc_page_extractor/layoutreader.py,sha256=BdC4oPbtpXoLmYhjuSFrKn6SNoT2zWw_gi95sGAUwrk,4031
-doc_page_extractor/ocr.py,sha256=KJ5PqtBa4_n8LAfMLGApUVNPUS1DBEwVKcC-zck283I,5161
+doc_page_extractor/ocr.py,sha256=hQhT9bdsJmWESqt1FODCoE19wfOroM8uHZiFoZZrkQU,5182
 doc_page_extractor/ocr_corrector.py,sha256=RfRA1jESEuqC8_a2kUEvHblT_B4xBjE0OApLMl1JiRg,3917
 doc_page_extractor/overlap.py,sha256=z1DF4_2OPvauDHwmz1SC1WosULkE84HKaRfNEgexPzc,5337
 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/rectangle.py,sha256=yeW6srdrsxaJg1eb3nn8oxtY0sfgeBk3hMiuJGaRXwY,1678
 doc_page_extractor/rotation.py,sha256=QCZ-HqfDxIhnQw8KRHki2myj6-UusvNY7Mpjsu-wI-4,4334
 doc_page_extractor/table.py,sha256=AWymTRbOet55uImW8QJqb90Qs_v2V2U1mZv0U6rSz3c,1891
 doc_page_extractor/types.py,sha256=7blT8YNKrOsc4qQdAhM7J7MEQjFcBwE0QV8-lipZBeQ,1305
@@ -35,10 +35,10 @@ doc_page_extractor/struct_eqtable/internvl/internvl_lmdeploy.py,sha256=ACHxFntxS
 doc_page_extractor/struct_eqtable/pix2s/__init__.py,sha256=cXRo4eg6u1-TXktZ8rQf0HIzLmmScIwYQhbxMKl-MyA,76
 doc_page_extractor/struct_eqtable/pix2s/pix2s.py,sha256=fCNve8PNeJ3-AWJIhSeGtp-mYKoMXfW0CIpszkQnAaA,2535
 doc_page_extractor/struct_eqtable/pix2s/pix2s_trt.py,sha256=zSGw45JhWdZ3iuJel5Chsy-NzsOHx9QyPQIUAzzCjFE,43880
-doc_page_extractor-0.1.0.dist-info/licenses/LICENSE,sha256=TfPDBt3ar0uv_f9cqCDMZ5rIzW3CY8anRRd4PkL6ejs,34522
+doc_page_extractor-0.1.1.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-0.1.0.dist-info/METADATA,sha256=8AM05x2gY75j70rD2HNi5zyaKE1okNJG7Pw5iLuIlnQ,2436
-doc_page_extractor-0.1.0.dist-info/WHEEL,sha256=CmyFI0kx5cdEMTLiONQRbGQwjIoR1aIYB7eCAQ4KPJ0,91
-doc_page_extractor-0.1.0.dist-info/top_level.txt,sha256=ErNybD_lBzAmw8mVBAK4htsAH_hp14jioZVex-tUqvM,25
-doc_page_extractor-0.1.0.dist-info/RECORD,,
+doc_page_extractor-0.1.1.dist-info/METADATA,sha256=5bQtvYgjNghsYER1zmc19_6BH4JrgrLj9_KxUmnLnHc,2436
+doc_page_extractor-0.1.1.dist-info/WHEEL,sha256=CmyFI0kx5cdEMTLiONQRbGQwjIoR1aIYB7eCAQ4KPJ0,91
+doc_page_extractor-0.1.1.dist-info/top_level.txt,sha256=ErNybD_lBzAmw8mVBAK4htsAH_hp14jioZVex-tUqvM,25
+doc_page_extractor-0.1.1.dist-info/RECORD,,