feat(fonts): map italic font to kai

- add LXGWWenKai-Regular.ttf to the assets list
- update fontmap to include kai font and use it for italic characters

chore(project): bump version to 0.0.1a18

- update project version in pyproject.toml

pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "yadt"
-version = "0.0.1a17"
+version = "0.0.1a18"
 description = "Yet Another Document Translator"
 license = "AGPL-3.0"
 readme = "README.md"

yadt/document_il/utils/fontmap.py

@@ -26,11 +26,17 @@ def __init__(self, translation_config: TranslationConfig):
         self.base_font = pymupdf.Font(fontfile=self.base_font_path)
         self.fallback_font = pymupdf.Font(fontfile=self.fallback_font_path)
 
+        self.kai_font_path = get_cache_file_path("LXGWWenKai-Regular.ttf")
+        self.kai_font = pymupdf.Font(fontfile=self.kai_font_path)
+
         self.base_font.font_id = "base"
         self.fallback_font.font_id = "fallback"
+        self.kai_font.font_id = "kai"
 
     def map(self, original_font: PdfFont, char_unicode: str):
         char_unicode = ord(char_unicode)
+        if original_font.italic and self.kai_font.has_glyph(char_unicode):
+            return self.kai_font
         for k, font in self.fonts.items():
             if not font.has_glyph(char_unicode):
                 continue
@@ -51,6 +57,7 @@ def add_font(self, doc_zh: pymupdf.Document, il: il_version_1.Document):
         font_list = [
             ("base", self.base_font_path),
             ("fallback", self.fallback_font_path),
+            ("kai", self.kai_font_path),
         ]
         font_list.extend(
             [

yadt/main.py

@@ -180,6 +180,11 @@ def download_font_assets():
             "SourceHanSansSC-Bold.ttf",
             "https://github.com/iizyd/SourceHanSansCN-TTF-Min"
             "/raw/refs/heads/main/source-file/ttf/SourceHanSansSC-Bold.ttf"
+        ),
+        (
+            'LXGWWenKai-Regular.ttf',
+            'https://github.com/lxgw/LxgwWenKai'
+            '/raw/refs/heads/main/fonts/TTF/LXGWWenKai-Regular.ttf'
         )
     ]
     for name, url in assets: