почистил ссылки в сипиське источников (там почему-то стояли гиперссылки)

linker.py

@@ -0,0 +1,192 @@
+#!/usr/bin/env python3
+
+import re
+import shutil
+import tempfile
+import xml.etree.ElementTree as ET
+
+from pathlib import Path
+from zipfile import ZipFile
+
+
+FIRST_NEW_SOURCE = 8
+
+NS = {
+    "w": "http://schemas.openxmlformats.org/wordprocessingml/2006/main"
+}
+
+
+def normalize_url(url):
+    url = url.strip()
+
+    if "#:~:text=" in url:
+        url = url.split("#:~:text=", 1)[0]
+
+    url = url.rstrip("/")
+
+    return url
+
+
+def build_url_mapping(root):
+    """
+    Первый проход.
+
+    Собираем все уникальные URL и
+    присваиваем им номера.
+    """
+
+    url_to_index = {}
+    next_index = FIRST_NEW_SOURCE
+
+    for node in root.findall(".//w:instrText", NS):
+
+        text = "".join(node.itertext())
+
+        match = re.search(
+            r'HYPERLINK\s+"([^"]+)"',
+            text
+        )
+
+        if not match:
+            continue
+
+        url = normalize_url(match.group(1))
+
+        if url in url_to_index:
+            continue
+
+        url_to_index[url] = next_index
+        next_index += 1
+
+    return url_to_index
+
+
+def patch_document(root, url_to_index):
+    """
+    Второй проход.
+
+    Находим:
+        HYPERLINK "..."
+
+    Затем ближайший следующий
+        <w:t>[старый номер]</w:t>
+
+    И меняем его на новый.
+    """
+
+    current_url = None
+    waiting_for_visible_text = False
+
+    for elem in root.iter():
+
+        tag = elem.tag
+
+        if tag.endswith("instrText"):
+
+            text = "".join(elem.itertext())
+
+            match = re.search(
+                r'HYPERLINK\s+"([^"]+)"',
+                text
+            )
+
+            if match:
+                current_url = normalize_url(
+                    match.group(1)
+                )
+
+                waiting_for_visible_text = True
+
+            continue
+
+        if (
+            waiting_for_visible_text
+            and tag.endswith("t")
+            and elem.text
+            and re.match(r"\[\d+\]", elem.text.strip())
+        ):
+            index = url_to_index[current_url]
+
+            old = elem.text
+            elem.text = f"[{index}]"
+
+            print(
+                f"{old} -> {elem.text} : {current_url}"
+            )
+
+            waiting_for_visible_text = False
+            current_url = None
+
+
+def main():
+    input_docx = Path("Отчет.docx")
+
+    output_docx = Path(
+        input_docx.stem + "_fixed.docx"
+    )
+
+    with tempfile.TemporaryDirectory() as tmp:
+
+        tmp = Path(tmp)
+
+        with ZipFile(input_docx) as archive:
+            archive.extractall(tmp)
+
+        document_xml = (
+            tmp /
+            "word" /
+            "document.xml"
+        )
+
+        tree = ET.parse(document_xml)
+
+        root = tree.getroot()
+
+        url_to_index = build_url_mapping(root)
+
+        print("\nURL -> INDEX\n")
+
+        for url, index in url_to_index.items():
+            print(
+                f"[{index}] {url}"
+            )
+
+        print(
+            f"\nFound {len(url_to_index)} unique sources\n"
+        )
+
+        patch_document(
+            root,
+            url_to_index
+        )
+
+        tree.write(
+            document_xml,
+            encoding="utf-8",
+            xml_declaration=True
+        )
+
+        with ZipFile(
+            output_docx,
+            "w"
+        ) as out_zip:
+
+            for file in tmp.rglob("*"):
+
+                if file.is_dir():
+                    continue
+
+                out_zip.write(
+                    file,
+                    file.relative_to(tmp)
+                )
+
+    print(
+        f"\nSaved: {output_docx}"
+    )
+
+
+if __name__ == "__main__":
+    main()
+
+