chatgpt-md-converter 0.1.2__tar.gz → 0.3.0__tar.gz

{chatgpt_md_converter-0.1.2 → chatgpt_md_converter-0.3.0}/PKG-INFO

@@ -1,6 +1,6 @@
-Metadata-Version: 2.1
+Metadata-Version: 2.4
 Name: chatgpt_md_converter
-Version: 0.1.2
+Version: 0.3.0
 Summary: A package for converting markdown to HTML for chat Telegram bots
 Home-page: https://github.com/Latand/formatter-chatgpt-telegram
 Author: Kostiantyn Kriuchkov
@@ -11,6 +11,15 @@ Classifier: Operating System :: OS Independent
 Requires-Python: >=3.8
 Description-Content-Type: text/markdown
 License-File: LICENSE
+Dynamic: author
+Dynamic: author-email
+Dynamic: classifier
+Dynamic: description
+Dynamic: description-content-type
+Dynamic: home-page
+Dynamic: license-file
+Dynamic: requires-python
+Dynamic: summary
 
 # ChatGPT Markdown to Telegram HTML Parser
 

{chatgpt_md_converter-0.1.2 → chatgpt_md_converter-0.3.0}/chatgpt_md_converter/converters.py

@@ -19,4 +19,9 @@ def split_by_tag(out_text: str, md_tag: str, html_tag: str) -> str:
         r"(?<!\w){}(.*?){}(?!\w)".format(re.escape(md_tag), re.escape(md_tag)),
         re.DOTALL,
     )
+
+    # Special handling for the tg-spoiler tag
+    if html_tag == 'span class="tg-spoiler"':
+        return tag_pattern.sub(r'<span class="tg-spoiler">\1</span>', out_text)
+
     return tag_pattern.sub(r"<{}>\1</{}>".format(html_tag, html_tag), out_text)

chatgpt_md_converter-0.3.0/chatgpt_md_converter/formatters.py

@@ -0,0 +1,68 @@
+def combine_blockquotes(text: str) -> str:
+    """
+    Combines multiline blockquotes into a single blockquote while keeping the \n characters.
+    Supports both regular blockquotes (>) and expandable blockquotes (**>).
+    """
+    lines = text.split("\n")
+    combined_lines = []
+    blockquote_lines = []
+    in_blockquote = False
+    is_expandable = False
+
+    for line in lines:
+        if line.startswith("**>"):
+            # Expandable blockquote
+            in_blockquote = True
+            is_expandable = True
+            blockquote_lines.append(line[3:].strip())
+        elif line.startswith(">"):
+            # Regular blockquote
+            if not in_blockquote:
+                # This is a new blockquote
+                in_blockquote = True
+                is_expandable = False
+            blockquote_lines.append(line[1:].strip())
+        else:
+            if in_blockquote:
+                # End of blockquote, combine the lines
+                if is_expandable:
+                    combined_lines.append(
+                        "<blockquote expandable>"
+                        + "\n".join(blockquote_lines)
+                        + "</blockquote>"
+                    )
+                else:
+                    combined_lines.append(
+                        "<blockquote>" + "\n".join(blockquote_lines) + "</blockquote>"
+                    )
+                blockquote_lines = []
+                in_blockquote = False
+                is_expandable = False
+            combined_lines.append(line)
+
+    if in_blockquote:
+        # Handle the case where the file ends with a blockquote
+        if is_expandable:
+            combined_lines.append(
+                "<blockquote expandable>"
+                + "\n".join(blockquote_lines)
+                + "</blockquote>"
+            )
+        else:
+            combined_lines.append(
+                "<blockquote>" + "\n".join(blockquote_lines) + "</blockquote>"
+            )
+
+    return "\n".join(combined_lines)
+
+
+def fix_asterisk_equations(text: str) -> str:
+    """
+    Replaces numeric expressions with '*' in them with '×'
+    to avoid accidental italic formatting.
+    e.g. '6*8' -> '6×8', '6 * 8' -> '6×8'
+    """
+    import re
+
+    eq_pattern = re.compile(r"(\d+)\s*\*\s*(\d+)")
+    return eq_pattern.sub(r"\1×\2", text)

chatgpt_md_converter-0.3.0/chatgpt_md_converter/helpers.py

@@ -0,0 +1,27 @@
+def remove_blockquote_escaping(output: str) -> str:
+    """
+    Removes the escaping from blockquote tags, including expandable blockquotes.
+    """
+    # Regular blockquotes
+    output = output.replace("&lt;blockquote&gt;", "<blockquote>").replace(
+        "&lt;/blockquote&gt;", "</blockquote>"
+    )
+
+    # Expandable blockquotes
+    output = output.replace(
+        "&lt;blockquote expandable&gt;", "<blockquote expandable>"
+    ).replace("&lt;/blockquote&gt;", "</blockquote>")
+
+    return output
+
+
+def remove_spoiler_escaping(output: str) -> str:
+    """
+    Ensures spoiler tags are correctly formatted (rather than being escaped).
+    """
+    # Fix any incorrectly escaped spoiler tags
+    output = output.replace(
+        '&lt;span class="tg-spoiler"&gt;', '<span class="tg-spoiler">'
+    )
+    output = output.replace("&lt;/span&gt;", "</span>")
+    return output

chatgpt_md_converter-0.3.0/chatgpt_md_converter/telegram_formatter.py

@@ -0,0 +1,102 @@
+import re
+
+from .converters import convert_html_chars, split_by_tag
+from .extractors import extract_and_convert_code_blocks, reinsert_code_blocks
+from .formatters import combine_blockquotes
+from .helpers import remove_blockquote_escaping, remove_spoiler_escaping
+
+
+def extract_inline_code_snippets(text: str):
+    """
+    Extracts inline code (single-backtick content) from the text,
+    replacing it with placeholders, returning modified text and a dict of placeholders -> code text.
+    This ensures characters like '*' or '_' inside inline code won't be interpreted as Markdown.
+    """
+    placeholders = []
+    code_snippets = {}
+    inline_code_pattern = re.compile(r"`([^`]+)`")
+
+    def replacer(match):
+        snippet = match.group(1)
+        placeholder = f"INLINECODEPLACEHOLDER{len(placeholders)}"
+        placeholders.append(placeholder)
+        code_snippets[placeholder] = snippet
+        return placeholder
+
+    new_text = inline_code_pattern.sub(replacer, text)
+    return new_text, code_snippets
+
+
+def telegram_format(text: str) -> str:
+    """
+    Converts markdown in the provided text to HTML supported by Telegram.
+    """
+
+    # 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
+    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.
+    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;")
+
+    # Convert headings (H1-H6)
+    output = re.sub(r"^(#{1,6})\s+(.+)$", r"<b>\2</b>", output, flags=re.MULTILINE)
+
+    # Convert unordered lists (do this before italic detection so that leading '*' is recognized as bullet)
+    output = re.sub(r"^(\s*)[\-\*]\s+(.+)$", r"\1• \2", output, flags=re.MULTILINE)
+
+    # Nested Bold and Italic
+    output = re.sub(r"\*\*\*(.*?)\*\*\*", r"<b><i>\1</i></b>", output)
+    output = re.sub(r"\_\_\_(.*?)\_\_\_", r"<u><i>\1</i></u>", output)
+
+    # Process markdown for bold (**), underline (__), strikethrough (~~), and spoiler (||)
+    output = split_by_tag(output, "**", "b")
+    output = split_by_tag(output, "__", "u")
+    output = split_by_tag(output, "~~", "s")
+    output = split_by_tag(output, "||", 'span class="tg-spoiler"')
+
+    # Custom approach for single-asterisk italic
+    italic_pattern = re.compile(
+        r"(?<![A-Za-z0-9])\*(?=[^\s])(.*?)(?<!\s)\*(?![A-Za-z0-9])", re.DOTALL
+    )
+    output = italic_pattern.sub(r"<i>\1</i>", output)
+
+    # Process single underscore-based italic
+    output = split_by_tag(output, "_", "i")
+
+    # Remove storage links (Vector storage placeholders like 【4:0†source】)
+    output = re.sub(r"【[^】]+】", "", output)
+
+    # Convert Markdown links/images to <a href="">…</a>
+    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
+    for placeholder, snippet in inline_code_snippets.items():
+        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
+    output = reinsert_code_blocks(output, triple_code_blocks)
+
+    # Step 5: Remove blockquote escaping
+    output = remove_blockquote_escaping(output)
+
+    # Step 6: Remove spoiler tag escaping
+    output = remove_spoiler_escaping(output)
+
+    # Clean up multiple consecutive newlines, but preserve intentional spacing
+    output = re.sub(r"\n{3,}", "\n\n", output)
+
+    return output.strip()

{chatgpt_md_converter-0.1.2 → chatgpt_md_converter-0.3.0}/chatgpt_md_converter.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
-Metadata-Version: 2.1
+Metadata-Version: 2.4
 Name: chatgpt_md_converter
-Version: 0.1.2
+Version: 0.3.0
 Summary: A package for converting markdown to HTML for chat Telegram bots
 Home-page: https://github.com/Latand/formatter-chatgpt-telegram
 Author: Kostiantyn Kriuchkov
@@ -11,6 +11,15 @@ Classifier: Operating System :: OS Independent
 Requires-Python: >=3.8
 Description-Content-Type: text/markdown
 License-File: LICENSE
+Dynamic: author
+Dynamic: author-email
+Dynamic: classifier
+Dynamic: description
+Dynamic: description-content-type
+Dynamic: home-page
+Dynamic: license-file
+Dynamic: requires-python
+Dynamic: summary
 
 # ChatGPT Markdown to Telegram HTML Parser
 

{chatgpt_md_converter-0.1.2 → chatgpt_md_converter-0.3.0}/setup.py

@@ -2,7 +2,7 @@ from setuptools import setup
 
 setup(
     name="chatgpt_md_converter",
-    version="0.1.2",
+    version="0.3.0",
     author="Kostiantyn Kriuchkov",
     author_email="latand666@gmail.com",
     description="A package for converting markdown to HTML for chat Telegram bots",