chatgpt-md-converter 0.4.0b1__tar.gz → 0.4.0b3__tar.gz

{chatgpt_md_converter-0.4.0b1 → chatgpt_md_converter-0.4.0b3}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: chatgpt_md_converter
-Version: 0.4.0b1
+Version: 0.4.0b3
 Summary: A package for converting markdown to HTML for chat Telegram bots
 Home-page: https://github.com/botfather-dev/formatter-chatgpt-telegram
 Author: Kostiantyn Kriuchkov

{chatgpt_md_converter-0.4.0b1 → chatgpt_md_converter-0.4.0b3}/chatgpt_md_converter/telegram_entities/extractors/links.py

@@ -10,7 +10,10 @@ from ..entity import EntityType, TelegramEntity
 _LINK_PATTERN = re.compile(r"!?\[((?:[^\[\]]|\[.*?\])*)\]\(([^)]+)\)")
 
 
-def extract_link_entities(text: str) -> Tuple[str, List[TelegramEntity]]:
+def extract_link_entities(
+    text: str,
+    existing_entities: List[TelegramEntity] | None = None,
+) -> Tuple[str, List[TelegramEntity], List[TelegramEntity]]:
     """
     Extract Markdown links and return plain text with TEXT_LINK entities.
 
@@ -19,14 +22,18 @@ def extract_link_entities(text: str) -> Tuple[str, List[TelegramEntity]]:
 
     Args:
         text: Input text with Markdown links
+        existing_entities: Optional list of entities to adjust offsets for
 
     Returns:
-        Tuple of (text_with_links_replaced, list_of_entities)
+        Tuple of (text_with_links_replaced, link_entities, adjusted_existing_entities)
     """
     entities: List[TelegramEntity] = []
     result_parts: List[str] = []
     last_end = 0
 
+    # Track adjustments: list of (position_in_original, chars_removed)
+    adjustments: List[Tuple[int, int]] = []
+
     for match in _LINK_PATTERN.finditer(text):
         # Add text before this link
         result_parts.append(text[last_end : match.start()])
@@ -38,6 +45,12 @@ def extract_link_entities(text: str) -> Tuple[str, List[TelegramEntity]]:
         link_text = match.group(1)
         url = match.group(2)
 
+        # Calculate how many chars are removed
+        # Original: [text](url) or ![text](url)
+        # New: text
+        chars_removed = len(match.group(0)) - len(link_text)
+        adjustments.append((match.start(), chars_removed))
+
         # Add the link text (without the markdown syntax)
         result_parts.append(link_text)
 
@@ -56,4 +69,23 @@ def extract_link_entities(text: str) -> Tuple[str, List[TelegramEntity]]:
     # Add remaining text
     result_parts.append(text[last_end:])
 
-    return "".join(result_parts), entities
+    # Adjust existing entities
+    adjusted_existing: List[TelegramEntity] = []
+    if existing_entities:
+        for e in existing_entities:
+            new_offset = e.offset
+            # Apply all adjustments that occur before this entity
+            for adj_pos, chars_removed in adjustments:
+                if adj_pos < e.offset:
+                    new_offset -= chars_removed
+            adjusted_existing.append(
+                TelegramEntity(
+                    type=e.type,
+                    offset=new_offset,
+                    length=e.length,
+                    url=e.url,
+                    language=e.language,
+                )
+            )
+
+    return "".join(result_parts), entities, adjusted_existing

{chatgpt_md_converter-0.4.0b1 → chatgpt_md_converter-0.4.0b3}/chatgpt_md_converter/telegram_entities/parser.py

@@ -167,8 +167,8 @@ def parse_entities(text: str) -> Tuple[str, List[TelegramEntity]]:
     text, inline_entities = extract_inline_formatting_entities(text)
     all_entities.extend(inline_entities)
 
-    # Extract links AFTER inline formatting so offsets are correct
-    text, link_entities = extract_link_entities(text)
+    # Extract links AFTER inline formatting, adjusting existing entity offsets
+    text, link_entities, all_entities = extract_link_entities(text, all_entities)
     all_entities.extend(link_entities)
 
     # Phase 4: Restore code placeholders and create entities

{chatgpt_md_converter-0.4.0b1 → chatgpt_md_converter-0.4.0b3}/chatgpt_md_converter/telegram_markdown/renderer.py

@@ -34,6 +34,12 @@ def telegram_format(text: str) -> str:
 
     output = re.sub(r"【[^】]+】", "", output)
 
+    # Handle Telegram custom emoji before generic links
+    # ![emoji](tg://emoji?id=123) -> <tg-emoji emoji-id="123">emoji</tg-emoji>
+    emoji_pattern = r"!\[([^\]]*)\]\(tg://emoji\?id=(\d+)\)"
+    output = re.sub(emoji_pattern, r'<tg-emoji emoji-id="\2">\1</tg-emoji>', output)
+
+    # Handle all links including images (! prefix is stripped for non-emoji images)
     link_pattern = r"(?:!?)\[((?:[^\[\]]|\[.*?\])*)\]\(([^)]+)\)"
     output = re.sub(link_pattern, r'<a href="\2">\1</a>', output)
 

{chatgpt_md_converter-0.4.0b1 → chatgpt_md_converter-0.4.0b3}/chatgpt_md_converter.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: chatgpt_md_converter
-Version: 0.4.0b1
+Version: 0.4.0b3
 Summary: A package for converting markdown to HTML for chat Telegram bots
 Home-page: https://github.com/botfather-dev/formatter-chatgpt-telegram
 Author: Kostiantyn Kriuchkov

{chatgpt_md_converter-0.4.0b1 → chatgpt_md_converter-0.4.0b3}/setup.py

@@ -2,7 +2,7 @@ from setuptools import setup
 
 setup(
     name="chatgpt_md_converter",
-    version="0.4.0b1",
+    version="0.4.0b3",
     author="Kostiantyn Kriuchkov",
     author_email="latand666@gmail.com",
     description="A package for converting markdown to HTML for chat Telegram bots",

{chatgpt_md_converter-0.4.0b1 → chatgpt_md_converter-0.4.0b3}/tests/test_parser.py

@@ -1038,3 +1038,19 @@ def test_inline_code_with_escaped_backtick_trailing_text():
     expected_output = "Escaped \\*asterisks\\* and <code>code with \\</code> backtick`"
     output = telegram_format(input_text)
     assert output == expected_output
+
+
+def test_custom_emoji_conversion():
+    """Test that custom emoji markdown is converted to tg-emoji HTML tag."""
+    input_text = "Hello ![❤️](tg://emoji?id=5226457415154701085) world"
+    expected_output = 'Hello <tg-emoji emoji-id="5226457415154701085">❤️</tg-emoji> world'
+    output = telegram_format(input_text)
+    assert output == expected_output, "Failed converting custom emoji to <tg-emoji> tag"
+
+
+def test_custom_emoji_with_regular_link():
+    """Test that custom emoji and regular links are both handled correctly."""
+    input_text = "Emoji ![👍](tg://emoji?id=5368324170671202286) and [link](https://example.com)"
+    expected_output = 'Emoji <tg-emoji emoji-id="5368324170671202286">👍</tg-emoji> and <a href="https://example.com">link</a>'
+    output = telegram_format(input_text)
+    assert output == expected_output, "Failed handling emoji and link together"