chatgpt-md-converter 0.3.5__py3-none-any.whl → 0.3.7__py3-none-any.whl

chatgpt_md_converter/extractors.py

@@ -2,16 +2,43 @@ import re
 
 
 def ensure_closing_delimiters(text: str) -> str:
-    """
-    Ensures that if an opening ` or ``` is found without a matching closing delimiter,
-    the missing delimiter is appended to the end of the text.
-    """
-    # For triple backticks
-    if text.count("```") % 2 != 0:
+    """Append missing closing backtick delimiters."""
+
+    code_block_re = re.compile(
+        r"(?P<fence>`{3,})(?P<lang>\w+)?\n?[\s\S]*?(?<=\n)?(?P=fence)",
+        flags=re.DOTALL,
+    )
+
+    # Remove complete code blocks from consideration so inner backticks
+    # don't affect delimiter balancing.
+    cleaned = code_block_re.sub("", text)
+
+    # Detect unclosed fences by tracking opening fence lengths.
+    stack = []
+    for line in cleaned.splitlines():
+        m = re.match(r"^(?P<fence>`{3,})(?P<lang>\w+)?$", line.strip())
+        if not m:
+            continue
+        fence = m.group("fence")
+        if stack and fence == stack[-1]:
+            stack.pop()
+        else:
+            stack.append(fence)
+
+    if stack:
+        text += "\n" + stack[-1]
+
+    cleaned_inline = code_block_re.sub("", text)
+
+    # Balance triple backticks that are not part of a complete fence.
+    if cleaned_inline.count("```") % 2 != 0:
         text += "```"
-    # For single backticks
-    if text.count("`") % 2 != 0:
+
+    # Balance single backticks outside fenced blocks.
+    cleaned_inline = code_block_re.sub("", text)
+    if cleaned_inline.count("`") % 2 != 0:
         text += "`"
+
     return text
 
 
@@ -25,8 +52,8 @@ def extract_and_convert_code_blocks(text: str):
     code_blocks = {}
 
     def replacer(match):
-        language = match.group(1) if match.group(1) else ""
-        code_content = match.group(3)
+        language = match.group("lang") if match.group("lang") else ""
+        code_content = match.group("code")
 
         # Properly escape HTML entities in code content
         escaped_content = (
@@ -44,8 +71,14 @@ def extract_and_convert_code_blocks(text: str):
         return (placeholder, html_code_block)
 
     modified_text = text
-    for match in re.finditer(r"```(\w*)?(\n)?(.*?)```", text, flags=re.DOTALL):
-        placeholder, html_code_block = replacer(match)
+    code_block_pattern = re.compile(
+        r"(?P<fence>`{3,})(?P<lang>\w+)?\n?(?P<code>[\s\S]*?)(?<=\n)?(?P=fence)",
+        flags=re.DOTALL,
+    )
+    for match in code_block_pattern.finditer(text):
+        placeholder, html_code_block = replacer(
+            match
+        )
         code_blocks[placeholder] = html_code_block
         modified_text = modified_text.replace(match.group(0), placeholder, 1)
 

chatgpt_md_converter/html_splitter.py

@@ -0,0 +1,114 @@
+import re
+from html.parser import HTMLParser
+
+MAX_LENGTH = 4096
+
+
+class HTMLTagTracker(HTMLParser):
+    def __init__(self):
+        super().__init__()
+        self.open_tags = []
+
+    def handle_starttag(self, tag, attrs):
+        # saving tags
+        if tag in ("b", "i", "u", "s", "code", "pre", "a", "span", "blockquote"):
+            self.open_tags.append((tag, attrs))
+
+    def handle_endtag(self, tag):
+        for i in range(len(self.open_tags) - 1, -1, -1):
+            if self.open_tags[i][0] == tag:
+                del self.open_tags[i]
+                break
+
+    def get_open_tags_html(self):
+        parts = []
+        for tag, attrs in self.open_tags:
+            attr_str = ""
+            if attrs:
+                attr_str = " " + " ".join(f'{k}="{v}"' for k, v in attrs)
+            parts.append(f"<{tag}{attr_str}>")
+        return "".join(parts)
+
+    def get_closing_tags_html(self):
+        return "".join(f"</{tag}>" for tag, _ in reversed(self.open_tags))
+
+
+def split_pre_block(pre_block: str) -> list[str]:
+    # language-aware: <pre><code class="language-python">...</code></pre>
+    match = re.match(r"<pre><code(.*?)>(.*)</code></pre>", pre_block, re.DOTALL)
+    if match:
+        attr, content = match.groups()
+        lines = content.splitlines(keepends=True)
+        chunks, buf = [], ""
+        for line in lines:
+            if len(buf) + len(line) + len('<pre><code></code></pre>') > MAX_LENGTH:
+                chunks.append(f"<pre><code{attr}>{buf}</code></pre>")
+                buf = ""
+            buf += line
+        if buf:
+            chunks.append(f"<pre><code{attr}>{buf}</code></pre>")
+        return chunks
+    else:
+        # regular <pre>...</pre>
+        inner = pre_block[5:-6]
+        lines = inner.splitlines(keepends=True)
+        chunks, buf = [], ""
+        for line in lines:
+            if len(buf) + len(line) + len('<pre></pre>') > MAX_LENGTH:
+                chunks.append(f"<pre>{buf}</pre>")
+                buf = ""
+            buf += line
+        if buf:
+            chunks.append(f"<pre>{buf}</pre>")
+        return chunks
+
+
+def split_html_for_telegram(text: str) -> list[str]:
+    chunks = []
+    pattern = re.compile(r"(<pre>.*?</pre>|<pre><code.*?</code></pre>)", re.DOTALL)
+    parts = pattern.split(text)
+
+    for part in parts:
+        if not part:
+            continue
+        if part.startswith("<pre>") or part.startswith("<pre><code"):
+            pre_chunks = split_pre_block(part)
+            chunks.extend(pre_chunks)
+        else:
+            # breaking down regular HTML
+            tracker = HTMLTagTracker()
+            current = ""
+            blocks = re.split(r"(\n\s*\n|<br\s*/?>|\n)", part)
+            for block in blocks:
+                prospective = current + block
+                if len(prospective) > MAX_LENGTH:
+                    tracker.feed(current)
+                    open_tags = tracker.get_open_tags_html()
+                    close_tags = tracker.get_closing_tags_html()
+                    chunks.append(open_tags + current + close_tags)
+                    current = block
+                    tracker = HTMLTagTracker()
+                else:
+                    current = prospective
+            if current.strip():
+                tracker.feed(current)
+                open_tags = tracker.get_open_tags_html()
+                close_tags = tracker.get_closing_tags_html()
+                chunks.append(open_tags + current + close_tags)
+
+    # post-unification: combine chunks if they don't exceed the limit in total
+    merged_chunks = []
+    buf = ""
+    for chunk in chunks:
+        # chunk = chunk.lstrip("\n")  # removing leading line breaks
+
+        if len(buf) + len(chunk) <= MAX_LENGTH:
+            buf += chunk
+        else:
+            if buf:
+                merged_chunks.append(buf)
+            buf = chunk
+    if buf:
+        merged_chunks.append(buf)
+
+    return merged_chunks

{chatgpt_md_converter-0.3.5.dist-info → chatgpt_md_converter-0.3.7.dist-info}/METADATA

@@ -1,8 +1,8 @@
 Metadata-Version: 2.4
 Name: chatgpt_md_converter
-Version: 0.3.5
+Version: 0.3.7
 Summary: A package for converting markdown to HTML for chat Telegram bots
-Home-page: https://github.com/Latand/formatter-chatgpt-telegram
+Home-page: https://github.com/botfather-dev/formatter-chatgpt-telegram
 Author: Kostiantyn Kriuchkov
 Author-email: latand666@gmail.com
 Classifier: Programming Language :: Python :: 3

chatgpt_md_converter-0.3.7.dist-info/RECORD

@@ -0,0 +1,12 @@
+chatgpt_md_converter/__init__.py,sha256=AfkikySkXsJ8HKQcSlU7B1KBHz54QCGJ7MO5Ka9oWRM,79
+chatgpt_md_converter/converters.py,sha256=fgebhbhMcIOqnr0xuV04v81RD91FfaGfA0kO417cDqc,831
+chatgpt_md_converter/extractors.py,sha256=uThH9vnjlEwZowCbxvcZreMZUPqUEiuq0nbWva3K-CE,3023
+chatgpt_md_converter/formatters.py,sha256=UbjRG7bLETIGDaFDbFybwW8dKYBMDmgLmIasJiw_j60,2304
+chatgpt_md_converter/helpers.py,sha256=2Nc9_s0HcLq79mBt7Hje19LzbO6z9mUNgayoMyWkIhI,874
+chatgpt_md_converter/html_splitter.py,sha256=8ao4QU5PFDFCHMg8pj5kBqmxSOUO6RfzqQfk4o1F8ms,3897
+chatgpt_md_converter/telegram_formatter.py,sha256=YlWW8JUlXqP_3chz53_kj15o4d2uW0RlVsuJVcCrzic,3872
+chatgpt_md_converter-0.3.7.dist-info/licenses/LICENSE,sha256=SDr2jeP-s2g4vf17-jdLXrrqA4_mU7L_RtSJlv4Y2mk,1077
+chatgpt_md_converter-0.3.7.dist-info/METADATA,sha256=4gweCWqlv3a6pR6FJbf-ycCEToIjCRf2Ohnk5p81bwQ,5792
+chatgpt_md_converter-0.3.7.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
+chatgpt_md_converter-0.3.7.dist-info/top_level.txt,sha256=T2o7csVtZgr-Pwm83aSUkZn0humJmDFNqW38tRSsNqw,21
+chatgpt_md_converter-0.3.7.dist-info/RECORD,,

{chatgpt_md_converter-0.3.5.dist-info → chatgpt_md_converter-0.3.7.dist-info}/WHEEL

@@ -1,5 +1,5 @@
 Wheel-Version: 1.0
-Generator: setuptools (80.0.0)
+Generator: setuptools (80.9.0)
 Root-Is-Purelib: true
 Tag: py3-none-any
 

chatgpt_md_converter-0.3.5.dist-info/RECORD

@@ -1,11 +0,0 @@
-chatgpt_md_converter/__init__.py,sha256=AfkikySkXsJ8HKQcSlU7B1KBHz54QCGJ7MO5Ka9oWRM,79
-chatgpt_md_converter/converters.py,sha256=fgebhbhMcIOqnr0xuV04v81RD91FfaGfA0kO417cDqc,831
-chatgpt_md_converter/extractors.py,sha256=WU38iAG-MANmilqR73gAvxqqXvx4JT8q3xrac_GRXGI,2071
-chatgpt_md_converter/formatters.py,sha256=UbjRG7bLETIGDaFDbFybwW8dKYBMDmgLmIasJiw_j60,2304
-chatgpt_md_converter/helpers.py,sha256=2Nc9_s0HcLq79mBt7Hje19LzbO6z9mUNgayoMyWkIhI,874
-chatgpt_md_converter/telegram_formatter.py,sha256=YlWW8JUlXqP_3chz53_kj15o4d2uW0RlVsuJVcCrzic,3872
-chatgpt_md_converter-0.3.5.dist-info/licenses/LICENSE,sha256=SDr2jeP-s2g4vf17-jdLXrrqA4_mU7L_RtSJlv4Y2mk,1077
-chatgpt_md_converter-0.3.5.dist-info/METADATA,sha256=ly_GTnX933MbdUCZKKQAn3kQrICrRiCWvgaJ1dBgWCc,5785
-chatgpt_md_converter-0.3.5.dist-info/WHEEL,sha256=ck4Vq1_RXyvS4Jt6SI0Vz6fyVs4GWg7AINwpsaGEgPE,91
-chatgpt_md_converter-0.3.5.dist-info/top_level.txt,sha256=T2o7csVtZgr-Pwm83aSUkZn0humJmDFNqW38tRSsNqw,21
-chatgpt_md_converter-0.3.5.dist-info/RECORD,,