chatgpt-md-converter 0.1.1__py3-none-any.whl → 0.2.0__py3-none-any.whl

chatgpt_md_converter/converters.py

@@ -16,6 +16,7 @@ def split_by_tag(out_text: str, md_tag: str, html_tag: str) -> str:
     Splits the text by markdown tag and replaces it with the specified HTML tag.
     """
     tag_pattern = re.compile(
-        r"{}(.*?){}".format(re.escape(md_tag), re.escape(md_tag)), re.DOTALL
+        r"(?<!\w){}(.*?){}(?!\w)".format(re.escape(md_tag), re.escape(md_tag)),
+        re.DOTALL,
     )
     return tag_pattern.sub(r"<{}>\1</{}>".format(html_tag, html_tag), out_text)

chatgpt_md_converter/formatters.py

@@ -1,9 +1,6 @@
-import re
-
-
 def combine_blockquotes(text: str) -> str:
     """
-    Combines multiline blockquotes into a single blockquote.
+    Combines multiline blockquotes into a single blockquote while keeping the \n characters.
     """
     lines = text.split("\n")
     combined_lines = []
@@ -17,7 +14,7 @@ def combine_blockquotes(text: str) -> str:
         else:
             if in_blockquote:
                 combined_lines.append(
-                    "<blockquote>" + " ".join(blockquote_lines) + "</blockquote>"
+                    "<blockquote>" + "\n".join(blockquote_lines) + "</blockquote>"
                 )
                 blockquote_lines = []
                 in_blockquote = False
@@ -25,7 +22,18 @@ def combine_blockquotes(text: str) -> str:
 
     if in_blockquote:
         combined_lines.append(
-            "<blockquote>" + " ".join(blockquote_lines) + "</blockquote>"
+            "<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/telegram_formatter.py

@@ -1,56 +1,100 @@
 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
 
 
+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 code blocks first
-    output, code_blocks = extract_and_convert_code_blocks(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
+    # 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;")
 
-    # Inline code
-    output = re.sub(r"`(.*?)`", r"<code>\1</code>", output)
+    # 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)
+
+    # Remove this old inline code replacement — now handled by extract_inline_code_snippets()
+    # output = re.sub(r"`(.*?)`", r"<code>\1</code>", output)
 
     # 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 formatting tags (bold, underline, italic, strikethrough)
-    # and convert them to their respective HTML tags
+    # Process markdown for bold (**), underline (__), strikethrough (~~)
     output = split_by_tag(output, "**", "b")
     output = split_by_tag(output, "__", "u")
-    output = split_by_tag(output, "_", "i")
-    output = split_by_tag(output, "*", "i")
     output = split_by_tag(output, "~~", "s")
 
-    # Remove storage links
-    output = re.sub(r"【[^】]+】", "", output)
+    # 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)
 
-    # Convert links
-    output = re.sub(r"\[(.*?)\]\((.*?)\)", r'<a href="\2">\1</a>', 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 lists
-    output = re.sub(r"^\s*[\-\*] (.+)", r"• \1", output, flags=re.MULTILINE)
+    # Convert Markdown links/images to <a href="">…</a>
+    link_pattern = r"(?:!?)\[((?:[^\[\]]|\[.*?\])*)\]\(([^)]+)\)"
+    output = re.sub(link_pattern, r'<a href="\2">\1</a>', output)
 
-    # Convert headings
-    output = re.sub(r"^\s*#+ (.+)", r"<b>\1</b>", output, flags=re.MULTILINE)
+    # 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 HTML code blocks
-    output = reinsert_code_blocks(output, code_blocks)
+    # 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)
 
-    return 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-0.1.1.dist-info → chatgpt_md_converter-0.2.0.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: chatgpt_md_converter
-Version: 0.1.1
+Version: 0.2.0
 Summary: A package for converting markdown to HTML for chat Telegram bots
 Home-page: https://github.com/Latand/formatter-chatgpt-telegram
 Author: Kostiantyn Kriuchkov

chatgpt_md_converter-0.2.0.dist-info/RECORD

@@ -0,0 +1,11 @@
+chatgpt_md_converter/__init__.py,sha256=AfkikySkXsJ8HKQcSlU7B1KBHz54QCGJ7MO5Ka9oWRM,79
+chatgpt_md_converter/converters.py,sha256=nfbKCcYCAYBk_0RQntCVQFQgAlEUWrGtLWULE1wETmU,657
+chatgpt_md_converter/extractors.py,sha256=RNwo57_6jCe-HoX5eCvvZcjSTc2uPax-6QEtXqXA5QQ,1880
+chatgpt_md_converter/formatters.py,sha256=daekV8M-42E3_N1uXx6M4EbZpSToHo8Vt8fl8AP_yyA,1197
+chatgpt_md_converter/helpers.py,sha256=9CtBeMzKYrymECNPl0MXsW0Vscp4A02a64a5z0sVWqE,261
+chatgpt_md_converter/telegram_formatter.py,sha256=MDyC_gkjN7J-LoMxQaJ1awcEQZzcaYFosOdCgDeDkRU,4036
+chatgpt_md_converter-0.2.0.dist-info/LICENSE,sha256=SDr2jeP-s2g4vf17-jdLXrrqA4_mU7L_RtSJlv4Y2mk,1077
+chatgpt_md_converter-0.2.0.dist-info/METADATA,sha256=zm80EZ56yxE7Z3AZkZXIm9CKxPDamouKuqFjS4y1xgU,3086
+chatgpt_md_converter-0.2.0.dist-info/WHEEL,sha256=PZUExdf71Ui_so67QXpySuHtCi3-J3wvF4ORK6k_S8U,91
+chatgpt_md_converter-0.2.0.dist-info/top_level.txt,sha256=T2o7csVtZgr-Pwm83aSUkZn0humJmDFNqW38tRSsNqw,21
+chatgpt_md_converter-0.2.0.dist-info/RECORD,,

{chatgpt_md_converter-0.1.1.dist-info → chatgpt_md_converter-0.2.0.dist-info}/WHEEL

@@ -1,5 +1,5 @@
 Wheel-Version: 1.0
-Generator: bdist_wheel (0.43.0)
+Generator: setuptools (75.6.0)
 Root-Is-Purelib: true
 Tag: py3-none-any
 

chatgpt_md_converter-0.1.1.dist-info/RECORD

@@ -1,11 +0,0 @@
-chatgpt_md_converter/__init__.py,sha256=AfkikySkXsJ8HKQcSlU7B1KBHz54QCGJ7MO5Ka9oWRM,79
-chatgpt_md_converter/converters.py,sha256=-SbsAiMetDZVkC7PrEQrrKlpoagnUycCL1WNBozd7u0,635
-chatgpt_md_converter/extractors.py,sha256=RNwo57_6jCe-HoX5eCvvZcjSTc2uPax-6QEtXqXA5QQ,1880
-chatgpt_md_converter/formatters.py,sha256=T85JwXI7t3PpqAHvkV7FFrmBar6pYRYLVLpET0TeRp0,856
-chatgpt_md_converter/helpers.py,sha256=9CtBeMzKYrymECNPl0MXsW0Vscp4A02a64a5z0sVWqE,261
-chatgpt_md_converter/telegram_formatter.py,sha256=3TrQpuVm1P4Qv1ZMrcBwD7sA2GI-yCDGuTUwUZSlw3E,1896
-chatgpt_md_converter-0.1.1.dist-info/LICENSE,sha256=SDr2jeP-s2g4vf17-jdLXrrqA4_mU7L_RtSJlv4Y2mk,1077
-chatgpt_md_converter-0.1.1.dist-info/METADATA,sha256=635E9EVpEVMP1fpbPAuDl0x_Y9cu3Ye5WfnwdjnWeYc,3086
-chatgpt_md_converter-0.1.1.dist-info/WHEEL,sha256=GJ7t_kWBFywbagK5eo9IoUwLW6oyOeTKmQ-9iHFVNxQ,92
-chatgpt_md_converter-0.1.1.dist-info/top_level.txt,sha256=T2o7csVtZgr-Pwm83aSUkZn0humJmDFNqW38tRSsNqw,21
-chatgpt_md_converter-0.1.1.dist-info/RECORD,,