chatgpt-md-converter 0.3.3__tar.gz → 0.3.5__tar.gz

{chatgpt_md_converter-0.3.3 → chatgpt_md_converter-0.3.5}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: chatgpt_md_converter
-Version: 0.3.3
+Version: 0.3.5
 Summary: A package for converting markdown to HTML for chat Telegram bots
 Home-page: https://github.com/Latand/formatter-chatgpt-telegram
 Author: Kostiantyn Kriuchkov
@@ -128,7 +128,7 @@ Feel free to contribute to this project by submitting pull requests or opening i
 > **Note**:
 > Since standard Markdown doesn't include Telegram-specific features like spoilers (`||text||`) and expandable blockquotes (`**> text`), you'll need to explicitly instruct LLMs to use these formats. Here's a suggested prompt addition to include in your system message or initial instructions:
 
-```
+````
 When formatting your responses for Telegram, please use these special formatting conventions:
 
 1. For content that should be hidden as a spoiler (revealed only when users click):
@@ -144,11 +144,11 @@ When formatting your responses for Telegram, please use these special formatting
 
 3. Continue using standard markdown for other formatting:
    - **bold text**
-   - _italic text_ or _italic text_
-   - **underlined text**
+   - *italic text*
+   - __underlined text__
    - ~~strikethrough~~
    - `inline code`
-   - `code blocks`
+   - ```code blocks```
    - [link text](URL)
 
 Apply spoilers for:
@@ -165,6 +165,6 @@ Use expandable blockquotes for:
 - Optional reading
 - Technical details
 - Additional context not needed by all users
-```
+````
 
 You can add this prompt to your system message when initializing your ChatGPT interactions to ensure the model properly formats content for optimal display in Telegram.

{chatgpt_md_converter-0.3.3 → chatgpt_md_converter-0.3.5}/chatgpt_md_converter/extractors.py

@@ -27,13 +27,19 @@ def extract_and_convert_code_blocks(text: str):
     def replacer(match):
         language = match.group(1) if match.group(1) else ""
         code_content = match.group(3)
+
+        # Properly escape HTML entities in code content
+        escaped_content = (
+            code_content.replace("&", "&amp;").replace("<", "&lt;").replace(">", "&gt;")
+        )
+
         placeholder = f"CODEBLOCKPLACEHOLDER{len(placeholders)}"
         placeholders.append(placeholder)
         if not language:
-            html_code_block = f"<pre><code>{code_content}</code></pre>"
+            html_code_block = f"<pre><code>{escaped_content}</code></pre>"
         else:
             html_code_block = (
-                f'<pre><code class="language-{language}">{code_content}</code></pre>'
+                f'<pre><code class="language-{language}">{escaped_content}</code></pre>'
             )
         return (placeholder, html_code_block)
 

{chatgpt_md_converter-0.3.3 → chatgpt_md_converter-0.3.5}/chatgpt_md_converter/telegram_formatter.py

@@ -35,18 +35,14 @@ def telegram_format(text: str) -> str:
     # Step 0: Combine blockquotes
     text = combine_blockquotes(text)
 
-    # Step 1: Convert HTML reserved symbols
-    text = convert_html_chars(text)
-
-    # Step 2: Extract and convert triple-backtick code blocks first
+    # Step 1: Extract and convert triple-backtick code blocks first
     output, triple_code_blocks = extract_and_convert_code_blocks(text)
 
-    # Step 2.5: Extract inline code snippets (single backticks) so they won't be parsed as italics, etc.
+    # Step 2: Extract inline code snippets
     output, inline_code_snippets = extract_inline_code_snippets(output)
 
-    # Step 3: Escape HTML special characters in the output text (for non-code parts)
-    # We do NOT want to escape what's inside placeholders here, only what's outside code placeholders.
-    output = output.replace("<", "&lt;").replace(">", "&gt;")
+    # Step 3: Convert HTML reserved symbols in the text (not in code blocks)
+    output = convert_html_chars(output)
 
     # Convert headings (H1-H6)
     output = re.sub(r"^(#{1,6})\s+(.+)$", r"<b>\2</b>", output, flags=re.MULTILINE)
@@ -80,20 +76,21 @@ def telegram_format(text: str) -> str:
     link_pattern = r"(?:!?)\[((?:[^\[\]]|\[.*?\])*)\]\(([^)]+)\)"
     output = re.sub(link_pattern, r'<a href="\2">\1</a>', output)
 
-    # Step 3.5: Reinsert inline code snippets, escaping special chars in code content
+    # Step 4: Reinsert inline code snippets, applying HTML escaping to the content
     for placeholder, snippet in inline_code_snippets.items():
+        # Apply HTML escaping to the content of inline code
         escaped_snippet = (
             snippet.replace("&", "&amp;").replace("<", "&lt;").replace(">", "&gt;")
         )
         output = output.replace(placeholder, f"<code>{escaped_snippet}</code>")
 
-    # Step 4: Reinsert the converted triple-backtick code blocks
+    # Step 5: Reinsert the converted triple-backtick code blocks
     output = reinsert_code_blocks(output, triple_code_blocks)
 
-    # Step 5: Remove blockquote escaping
+    # Step 6: Remove blockquote escaping
     output = remove_blockquote_escaping(output)
 
-    # Step 6: Remove spoiler tag escaping
+    # Step 7: Remove spoiler tag escaping
     output = remove_spoiler_escaping(output)
 
     # Clean up multiple consecutive newlines, but preserve intentional spacing

{chatgpt_md_converter-0.3.3 → chatgpt_md_converter-0.3.5}/chatgpt_md_converter.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: chatgpt_md_converter
-Version: 0.3.3
+Version: 0.3.5
 Summary: A package for converting markdown to HTML for chat Telegram bots
 Home-page: https://github.com/Latand/formatter-chatgpt-telegram
 Author: Kostiantyn Kriuchkov
@@ -128,7 +128,7 @@ Feel free to contribute to this project by submitting pull requests or opening i
 > **Note**:
 > Since standard Markdown doesn't include Telegram-specific features like spoilers (`||text||`) and expandable blockquotes (`**> text`), you'll need to explicitly instruct LLMs to use these formats. Here's a suggested prompt addition to include in your system message or initial instructions:
 
-```
+````
 When formatting your responses for Telegram, please use these special formatting conventions:
 
 1. For content that should be hidden as a spoiler (revealed only when users click):
@@ -144,11 +144,11 @@ When formatting your responses for Telegram, please use these special formatting
 
 3. Continue using standard markdown for other formatting:
    - **bold text**
-   - _italic text_ or _italic text_
-   - **underlined text**
+   - *italic text*
+   - __underlined text__
    - ~~strikethrough~~
    - `inline code`
-   - `code blocks`
+   - ```code blocks```
    - [link text](URL)
 
 Apply spoilers for:
@@ -165,6 +165,6 @@ Use expandable blockquotes for:
 - Optional reading
 - Technical details
 - Additional context not needed by all users
-```
+````
 
 You can add this prompt to your system message when initializing your ChatGPT interactions to ensure the model properly formats content for optimal display in Telegram.

{chatgpt_md_converter-0.3.3 → chatgpt_md_converter-0.3.5}/setup.py

@@ -2,7 +2,7 @@ from setuptools import setup
 
 setup(
     name="chatgpt_md_converter",
-    version="0.3.3",
+    version="0.3.5",
     author="Kostiantyn Kriuchkov",
     author_email="latand666@gmail.com",
     description="A package for converting markdown to HTML for chat Telegram bots",

{chatgpt_md_converter-0.3.3 → chatgpt_md_converter-0.3.5}/tests/test_parser.py

@@ -720,3 +720,13 @@ def test_multiple_spoilers():
     expected_output = 'First <span class="tg-spoiler">spoiler</span> and then another <span class="tg-spoiler">spoiler with <i>italic</i></span>'
     output = telegram_format(input_text)
     assert output == expected_output, "Failed handling multiple spoilers"
+
+
+def test_ukrainian_text_with_inline_code():
+    """Test that Ukrainian text with inline code is properly formatted"""
+    input_text = (
+        """звісно, майстре тестування. ой та зрозуміло `<LAUGH>` що ти тут тестуєш."""
+    )
+    expected_output = """звісно, майстре тестування. ой та зрозуміло <code>&lt;LAUGH&gt;</code> що ти тут тестуєш."""
+    output = telegram_format(input_text)
+    assert output == expected_output, f"Output was: {output}"