Merge pull request #132 from funstory-ai/ft/better-font-download

fix: Fixed watermark becoming bold

babeldoc/__init__.py

@@ -1 +1 @@
-__version__ = "0.1.20"
+__version__ = "0.1.21"

babeldoc/assets/embedding_assets_metadata.py

@@ -370,6 +370,7 @@
         "GoNotoKurrent-Regular.ttf",
         "GoNotoKurrent-Bold.ttf",
     ],
+    "base": ["SourceHanSansCN-Regular.ttf"],
 }
 
 HK_FONT_FAMILY = {
@@ -388,6 +389,7 @@
         "GoNotoKurrent-Regular.ttf",
         "GoNotoKurrent-Bold.ttf",
     ],
+    "base": ["SourceHanSansCN-Regular.ttf"],
 }
 
 TW_FONT_FAMILY = {
@@ -406,6 +408,7 @@
         "GoNotoKurrent-Regular.ttf",
         "GoNotoKurrent-Bold.ttf",
     ],
+    "base": ["SourceHanSansCN-Regular.ttf"],
 }
 
 ALL_FONT_FAMILY = {
@@ -431,7 +434,7 @@ def verify_font_family(font_family: str | dict):
     if isinstance(font_family, str):
         font_family = ALL_FONT_FAMILY[font_family]
     for k in font_family:
-        if k not in ["script", "normal", "fallback"]:
+        if k not in ["script", "normal", "fallback", "base"]:
             raise ValueError(f"Invalid font family: {font_family}")
         for font_file_name in font_family[k]:
             if font_file_name not in EMBEDDING_FONT_METADATA:

babeldoc/const.py

@@ -2,7 +2,7 @@
 import subprocess
 from pathlib import Path
 
-__version__ = "0.1.20"
+__version__ = "0.1.21"
 
 CACHE_FOLDER = Path.home() / ".cache" / "babeldoc"
 

babeldoc/document_il/utils/fontmap.py

@@ -23,6 +23,7 @@ def __init__(self, translation_config: TranslationConfig):
             "normal",
             "script",
             "fallback",
+            "base",
         ):
             self.font_file_names.extend(font_family[k])
 
@@ -35,6 +36,7 @@ def __init__(self, translation_config: TranslationConfig):
             self.fonts[font_file_name] = pymupdf.Font(fontfile=str(font_path))
             self.fontid2fontpath[font_file_name] = font_path
             self.fonts[font_file_name].font_id = font_file_name
+            self.fonts[font_file_name].font_path = font_path
             self.fonts[font_file_name].ascent_fontmap = font_metadata["ascent"]
             self.fonts[font_file_name].descent_fontmap = font_metadata["descent"]
             self.fonts[font_file_name].encoding_length = font_metadata[
@@ -44,8 +46,8 @@ def __init__(self, translation_config: TranslationConfig):
         self.normal_font_ids: list[str] = font_family["normal"]
         self.script_font_ids: list[str] = font_family["script"]
         self.fallback_font_ids: list[str] = font_family["fallback"]
-
-        self.fontid2fontpath["base"] = self.fontid2fontpath[self.normal_font_ids[0]]
+        self.base_font_ids: list[str] = font_family["base"]
+        self.fontid2fontpath["base"] = self.fontid2fontpath[font_family["base"][0]]
 
         self.fontid2font: dict[str, pymupdf.Font] = {
             f.font_id: f for f in self.fonts.values()
@@ -56,7 +58,7 @@ def __init__(self, translation_config: TranslationConfig):
                 font.char_lengths,
             )
 
-        self.fontid2font["base"] = self.fontid2font[self.normal_font_ids[0]]
+        self.fontid2font["base"] = self.fontid2font[self.base_font_ids[0]]
 
         self.normal_fonts: list[pymupdf.Font] = [
             self.fontid2font[font_id] for font_id in self.normal_font_ids
@@ -68,12 +70,13 @@ def __init__(self, translation_config: TranslationConfig):
             self.fontid2font[font_id] for font_id in self.fallback_font_ids
         ]
 
-        self.base_font = self.normal_fonts[0]
+        self.base_font = self.fontid2font["base"]
 
         self.type2font: dict[str, list[pymupdf.Font]] = {
             "normal": self.normal_fonts,
             "script": self.script_fonts,
             "fallback": self.fallback_fonts,
+            "base": [self.base_font],
         }
 
         self.has_char = functools.lru_cache(maxsize=10240, typed=True)(self.has_char)
@@ -167,6 +170,8 @@ def add_font(self, doc_zh: pymupdf.Document, il: il_version_1.Document):
             xreflen - 1 + len(font_list) + len(il.page) + len(font_list),
         ) as pbar:
             for font in font_list:
+                if font[0] in font_id:
+                    continue
                 font_id[font[0]] = doc_zh[0].insert_font(font[0], font[1])
                 pbar.advance(1)
             for xref in range(1, xreflen):

babeldoc/main.py

@@ -23,7 +23,7 @@
 from babeldoc.translation_config import TranslationConfig
 
 logger = logging.getLogger(__name__)
-__version__ = "0.1.20"
+__version__ = "0.1.21"
 
 
 def create_parser():

pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "BabelDOC"
-version = "0.1.20"
+version = "0.1.21"
 description = "Yet Another Document Translator"
 license = "AGPL-3.0"
 readme = "README.md"
@@ -132,7 +132,7 @@ pythonpath = [".", "src"]
 testpaths = ["tests"]
 
 [bumpver]
-current_version = "0.1.20"
+current_version = "0.1.21"
 version_pattern = "MAJOR.MINOR.PATCH[.PYTAGNUM]"
 
 [bumpver.file_patterns]