doc-page-extractor 0.1.0__tar.gz → 0.1.1__tar.gz

{doc_page_extractor-0.1.0 → doc_page_extractor-0.1.1}/PKG-INFO

@@ -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 → doc_page_extractor-0.1.1}/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-0.1.0 → doc_page_extractor-0.1.1}/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-0.1.0 → doc_page_extractor-0.1.1}/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 → doc_page_extractor-0.1.1}/doc_page_extractor.egg-info/PKG-INFO

@@ -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 → doc_page_extractor-0.1.1}/setup.py

@@ -5,7 +5,7 @@ if "doc_page_extractor.struct_eqtable" not in find_packages():
 
 setup(
   name="doc-page-extractor",
-  version="0.1.0",
+  version="0.1.1",
   author="Tao Zeyu",
   author_email="i@taozeyu.com",
   url="https://github.com/Moskize91/doc-page-extractor",