fix small bugs with docx reader such as non-integer sizes in docx sty…

dedoc/readers/docx_reader/properties_extractor.py

@@ -74,7 +74,10 @@ def change_indent(old_properties: BaseProperties, tree: Tag) -> None:
     if not tree.ind:
         return
 
-    attributes = {attribute: 0 for attribute in ["firstLine", "firstLineChars", "hanging", "hangingChars", "start", "startChars", "left"]}
+    attributes = {
+        attribute: 0 for attribute in
+        ["firstLine", "firstLineChars", "hanging", "hangingChars", "start", "startChars", "left"]
+    }
     for attribute in attributes:
         attributes[attribute] = float(tree.ind.get(f"w:{attribute}", 0))
 
@@ -106,7 +109,8 @@ def change_size(old_properties: BaseProperties, tree: Tag) -> None:
     :param tree: BeautifulSoup tree with properties
     """
     if tree.sz:
-        old_properties.size = int(tree.sz.get("w:val", old_properties.size))
+        new_size = float(tree.sz.get("w:val", old_properties.size))
+        old_properties.size = int(new_size)
 
 
 def change_jc(old_properties: BaseProperties, tree: Tag) -> None:
@@ -176,19 +180,19 @@ def change_spacing(old_properties: BaseProperties, tree: Tag) -> None:
 
     if not before_autospacing:
         before_lines = tree.spacing.get("w:beforeLines", False)
-        before_lines = int(before_lines) if before_lines else before_lines
+        before_lines = int(float(before_lines)) if before_lines else before_lines
         if not before_lines:
             before_tag = tree.spacing.get("w:before", False)
-            before = int(before_tag) if before_tag else before
+            before = int(float(before_tag)) if before_tag else before
         else:
             before = before_lines
 
     if not after_autospacing:
         after_lines = tree.spacing.get("w:afterLines", False)
-        after_lines = int(after_lines) if after_lines else after_lines
+        after_lines = int(float(after_lines)) if after_lines else after_lines
         if not after_lines:
             after_tag = tree.spacing.get("w:after", False)
-            after = int(after_tag) if after_tag else after
+            after = int(float(after_tag)) if after_tag else after
         else:
             after = after_lines
 

dedoc/utils/utils.py

@@ -63,6 +63,13 @@ def splitext_(path: str) -> Tuple[str, str]:
     """
     get extensions with several dots
     """
+    if len(path.split()) > 1:
+        first, second = path.rsplit(maxsplit=1)
+        sep = path[len(first)]
+        name, ext = splitext(second)
+        if len(ext) == 0:
+            name, ext = ext, name
+        return first + sep + name, ext
     if len(path.split(".")) > 2:
         return path.split(".")[0], "." + ".".join(path.split(".")[-2:])
     return splitext(path)

tests/api_tests/test_api_format_docx.py

@@ -118,6 +118,16 @@ def test_docx_heading_new(self) -> None:
         data = dict(structure_type="tree", return_format="html")
         _ = self._send_request(file_name, data=data)
 
+    def test_properties_extractor(self) -> None:
+        file_name = "broken_properties.docx"
+        result = self._send_request(file_name, data={})
+        content = result["content"]["structure"]
+        self.assertEqual("FonFfff", get_by_tree_path(content, "0.0")["text"].strip())
+
+    def test_name_with_apostrophe(self) -> None:
+        file_name = "Well. Known -Nik O'Tinn -Ireland 2023- DRAFT.doc"
+        _ = self._send_request(file_name, data={})
+
     def __check_doc_like(self, result: dict) -> None:
         content = result["content"]["structure"]
         self.assertEqual("", get_by_tree_path(content, "0")["text"])

tests/unit_tests/test_module_utils.py

@@ -0,0 +1,24 @@
+import unittest
+
+from dedoc.utils.utils import splitext_
+
+
+class TestUtils(unittest.TestCase):
+
+    def test_splitext_simple_name(self) -> None:
+        name_extension = "name.doc"
+        name, extension = splitext_(name_extension)
+        self.assertEqual("name", name)
+        self.assertEqual(".doc", extension)
+
+    def test_splitext_apostrophe_name(self) -> None:
+        name_extension = "Well. Known -Nik O'Tinn -Ireland 2023- DRAFT.doc"
+        name, extension = splitext_(name_extension)
+        self.assertEqual("Well. Known -Nik O'Tinn -Ireland 2023- DRAFT", name)
+        self.assertEqual(".doc", extension)
+
+    def test_splitext_space_name(self) -> None:
+        name_extension = "some file .doc"
+        name, extension = splitext_(name_extension)
+        self.assertEqual("some file ", name)
+        self.assertEqual(".doc", extension)