chatgpt-md-converter 0.3.5__tar.gz → 0.3.7__tar.gz

{chatgpt_md_converter-0.3.5 → chatgpt_md_converter-0.3.7}/PKG-INFO

@@ -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-0.3.5 → chatgpt_md_converter-0.3.7}/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-0.3.7/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 → chatgpt_md_converter-0.3.7}/chatgpt_md_converter.egg-info/PKG-INFO

@@ -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-0.3.5 → chatgpt_md_converter-0.3.7}/chatgpt_md_converter.egg-info/SOURCES.txt

@@ -5,9 +5,11 @@ chatgpt_md_converter/converters.py
 chatgpt_md_converter/extractors.py
 chatgpt_md_converter/formatters.py
 chatgpt_md_converter/helpers.py
+chatgpt_md_converter/html_splitter.py
 chatgpt_md_converter/telegram_formatter.py
 chatgpt_md_converter.egg-info/PKG-INFO
 chatgpt_md_converter.egg-info/SOURCES.txt
 chatgpt_md_converter.egg-info/dependency_links.txt
 chatgpt_md_converter.egg-info/top_level.txt
-tests/test_parser.py
+tests/test_parser.py
+tests/test_splitter.py

{chatgpt_md_converter-0.3.5 → chatgpt_md_converter-0.3.7}/setup.py

@@ -2,13 +2,13 @@ from setuptools import setup
 
 setup(
     name="chatgpt_md_converter",
-    version="0.3.5",
+    version="0.3.7",
     author="Kostiantyn Kriuchkov",
     author_email="latand666@gmail.com",
     description="A package for converting markdown to HTML for chat Telegram bots",
     long_description=open("README.MD").read(),
     long_description_content_type="text/markdown",
-    url="https://github.com/Latand/formatter-chatgpt-telegram",
+    url="https://github.com/botfather-dev/formatter-chatgpt-telegram",
     classifiers=[
         "Programming Language :: Python :: 3",
         "License :: OSI Approved :: MIT License",

{chatgpt_md_converter-0.3.5 → chatgpt_md_converter-0.3.7}/tests/test_parser.py

@@ -730,3 +730,188 @@ def test_ukrainian_text_with_inline_code():
     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}"
+
+
+def test_nested_code_fence_quadruple():
+    input_text = """````markdown
+```python
+def hello_world():
+    print("Hello, World!")
+```
+````"""
+    expected_output = (
+        "<pre><code class=\"language-markdown\">```python\n"
+        "def hello_world():\n    print(\"Hello, World!\")\n```\n</code></pre>"
+    )
+    output = telegram_format(input_text)
+    def show_output():
+      print(f"Expected was: \n\n{expected_output}\n\n")
+      print(f"output was: \n\n{output}")
+    assert output == expected_output, show_output()
+
+
+def test_nested_code_fence_quadruple_no_lang():
+    input_text = """````
+```python
+print('hi')
+```
+````"""
+    expected_output = (
+        "<pre><code>```python\nprint('hi')\n```\n</code></pre>"
+    )
+    output = telegram_format(input_text)
+    def show_output():
+      print(f"Expected was: \n\n{expected_output}\n\n")
+      print(f"output was: \n\n{output}")
+    assert output == expected_output, show_output()
+
+
+def test_nested_code_fence_five_backticks():
+    input_text = """`````markdown
+````python
+print(1)
+````
+`````"""
+    expected_output = (
+        "<pre><code class=\"language-markdown\">````python\nprint(1)\n````\n</code></pre>"
+    )
+    output = telegram_format(input_text)
+    def show_output():
+      print(f"Expected was: \n\n{expected_output}\n\n")
+      print(f"output was: \n\n{output}")
+    assert output == expected_output, show_output()
+
+
+def test_nested_code_fence_five_backticks_with_inner_triple():
+    input_text = """`````markdown
+````python
+print("hello world ```")
+````
+`````"""
+    expected_output = (
+        "<pre><code class=\"language-markdown\">````python\n"
+        "print(\"hello world ```\")\n````\n</code></pre>"
+    )
+    output = telegram_format(input_text)
+    def show_output():
+      print(f"Expected was: \n\n{expected_output}\n\n")
+      print(f"output was: \n\n{output}")
+    assert output == expected_output, show_output()
+
+
+def test_nested_code_fence_six_backticks():
+    input_text = """``````markdown
+`````python
+print('hi')
+`````
+``````"""
+    expected_output = """<pre><code class=\"language-markdown\">`````python
+print('hi')
+`````
+</code></pre>"""
+    output = telegram_format(input_text)
+    def show_output():
+      print(f"Expected was: \n\n{expected_output}\n\n")
+      print(f"output was: \n\n{output}")
+    assert output == expected_output, show_output()
+
+
+def test_nested_code_fence_plain_text():
+    input_text = """
+````markdown
+```
+hello
+```
+````"""
+    expected_output = """<pre><code class=\"language-markdown\">```
+hello
+```
+</code></pre>"""
+    output = telegram_format(input_text)
+    def show_output():
+      print(f"Expected was: \n\n{expected_output}\n\n")
+      print(f"output was: \n\n{output}")
+    assert output == expected_output, show_output()
+
+
+
+
+
+def test_expensive_nested_code_five_fence_plain_text():
+    input_text = """
+`````markdown
+````
+```python
+print("hello world ```")
+```
+`````"""
+
+    expected_output = """<pre><code class=\"language-markdown\">````
+```python
+print("hello world ```")
+```
+</code></pre>"""
+    output = telegram_format(input_text)
+    def show_output():
+      print(f"Expected was: \n\n{expected_output}\n\n")
+      print(f"output was: \n\n{output}")
+    assert output == expected_output, show_output()
+
+def test_another_expensive_nested_code_five_fence_plain_text():
+    input_text = """`````markdown
+````python
+print("hello world ```"')
+```
+`````"""
+
+    expected_output = """<pre><code class=\"language-markdown\">````python
+print("hello world ```"')
+```
+</code></pre>"""
+    output = telegram_format(input_text)
+    def show_output():
+      print(f"Expected was: \n\n{expected_output}\n\n")
+      print(f"output was: \n\n{output}")
+    assert output == expected_output, show_output()
+
+def test_hard_level_nested_code_five_fence_plain_text():
+    input_text = """`````markdown
+````python
+print("hello world ```"')
+````
+`````
+```python
+print("Some another text")""" # That's where closing the second block of python code is missing.
+
+    expected_output = """<pre><code class="language-markdown">````python
+print("hello world ```"')
+````
+</code></pre>
+<pre><code class="language-python">print("Some another text")
+</code></pre>""" # But the code block is still closed correctly.
+    
+    output = telegram_format(input_text)
+    def show_output():
+      print(f"Expected was: \n\n{expected_output}\n\n")
+      print(f"output was: \n\n{output}")
+    assert output == expected_output, show_output()
+
+def test_hard_level_nested_code_five_fence_plain_text_2():
+    input_text = """`````markdown
+````python
+print("hello world ```"')
+`````
+```python
+print("Some another text")""" # That's where closing the second block of python code is missing.
+
+    expected_output = """<pre><code class="language-markdown">````python
+print("hello world ```"')
+</code></pre>
+<pre><code class="language-python">print("Some another text")
+</code></pre>""" # But the code block is still closed correctly.
+    
+    output = telegram_format(input_text)
+    def show_output():
+      print(f"Expected was: \n\n{expected_output}\n\n")
+      print(f"output was: \n\n{output}")
+    assert output == expected_output, show_output()

chatgpt_md_converter-0.3.7/tests/test_splitter.py

@@ -0,0 +1,436 @@
+from chatgpt_md_converter.html_splitter import split_html_for_telegram
+
+input_text ="""
+Absolutely! Here’s a Markdown-formatted message exceeding 5,000 characters, exploring <b>The History and Impact of Computer Programming</b>. (You can verify the character count using any online tool.)
+
+---
+
+<b>The History and Impact of Computer Programming</b>
+
+<i>“The computer was born to solve problems that did not exist before.”</i>  
+— Bill Gates
+
+---
+
+<b>Table of Contents</b>
+
+1. <a href="#introduction">Introduction</a>  
+2. <a href="#ancient-beginnings-from-algorithms-to-machines">Ancient Beginnings: From Algorithms to Machines</a>  
+    • <a href="#al-khwarizmi-and-the-algorithm">Al-Khwarizmi and the Algorithm</a>  
+    • <a href="#the-analytical-engine">The Analytical Engine</a>  
+    • <a href="#punch-cards-and-the-jacquard-loom">Punch Cards and the Jacquard Loom</a>  
+3. <a href="#20th-century-the-birth-of-modern-programming">20th Century: The Birth of Modern Programming</a>  
+    • <a href="#eniac-and-early-programmers">ENIAC and Early Programmers</a>  
+    • <a href="#assembly-language-and-early-high-level-languages">Assembly Language and Early High-level Languages</a>  
+    • <a href="#cobol-fortran-and-the-expansion">COBOL, FORTRAN, and the Expansion</a>  
+4. <a href="#modern-era-languages-paradigms-and-the-internet">Modern Era: Languages, Paradigms, and the Internet</a>  
+    • <a href="#object-oriented-programming">Object-Oriented Programming</a>  
+    • <a href="#internet-and-open-source">Internet and Open Source</a>  
+    • <a href="#mobile-and-cloud-computing">Mobile and Cloud Computing</a>    
+5. <a href="#programmings-societal-impact">Programming’s Societal Impact</a>  
+6. <a href="#ethics-challenges-and-the-future">Ethics, Challenges, and the Future</a>  
+7. <a href="#conclusion">Conclusion</a>  
+8. <a href="#useful-resources">Useful Resources</a>  
+
+---
+
+<b>Introduction</b>
+
+Computer programming is the science and art of giving computers instructions to perform specific tasks. Today, it's impossible to imagine a world without software: from banking systems and mobile applications to traffic lights and airplanes, programming is everywhere.
+
+But how did programming begin, and what has it become today? This document explores the journey of programming, from ancient mathematical roots to the future of artificial intelligence.
+
+---
+
+<b>Ancient Beginnings: From Algorithms to Machines</b>
+
+<b>Al-Khwarizmi and the Algorithm</b>
+
+The term "<b>algorithm</b>" (the foundation of programming) comes from Abu Abdullah Muhammad ibn Musa Al-Khwarizmi, a 9th-century Persian mathematician. His works on systematic procedures laid the groundwork for computational thinking.
+
+<b>The Analytical Engine</b>
+
+In the 19th century, <b>Charles Babbage</b> designed the Analytical Engine, a mechanical general-purpose computer. Though never built in his lifetime, it could—in theory—read instructions from punched cards.
+
+<b>Ada Lovelace</b>, Babbage's collaborator, is often called the first computer programmer. She wrote notes describing algorithms (in essence, programs) for the Analytical Engine to compute Bernoulli numbers.
+
+<blockquote>"That brain of mine is something more than merely mortal; as time will show."
+– Ada Lovelace</blockquote>
+
+<b>Punch Cards and the Jacquard Loom</b>
+
+The concept of programming a machine with punched cards predates computers. <b>Joseph Marie Jacquard</b> invented a loom in 1804 that used punched cards to control patterns in woven fabric—an early example of machine automation.
+
+---
+
+<b>20th Century: The Birth of Modern Programming</b>
+
+<b>ENIAC and Early Programmers</b>
+
+ENIAC (Electronic Numerical Integrator and Computer), completed in 1945, is often cited as the first electronic general-purpose computer.
+
+Early programming was entirely manual and physically laborious—think patch cables and switches!
+
+Notably, many of the earliest programmers were women, such as <b>Kathleen McNulty</b>, <b>Jean Jennings</b>, and <b>Grace Hopper</b>.
+
+<b>Assembly Language and Early High-level Languages</b>
+
+The problem of complexity led to <b>assembly languages</b>, where mnemonics like <code>MOV</code> and <code>ADD</code> replaced binary codes. Programming became more accessible, but code was still hardware-specific.
+
+The 1950s saw the creation of:
+
+• <b>FORTRAN</b> (FORmula TRANslation) for scientific computation
+• <b>COBOL</b> (COmmon Business-Oriented Language) for business applications
+
+<b>Code Example: Hello World in COBOL</b>
+<pre><code class="language-cobol">IDENTIFICATION DIVISION.
+PROGRAM-ID. HELLO-WORLD.
+PROCEDURE DIVISION.
+    DISPLAY "Hello, World!".
+STOP RUN.
+</code></pre>
+
+<b>COBOL, FORTRAN, and the Expansion</b>
+
+With the advent of high-level languages, programming became less about circuitry and more about solving problems. Standardized languages allowed code to run on multiple machines.
+
+Other languages soon emerged:
+
+• <b>LISP</b> (for AI research)
+• <b>ALGOL</b> (basis for many future languages)
+• <b>BASIC</b> (for beginners and education)
+
+---
+
+<b>Modern Era: Languages, Paradigms, and the Internet</b>
+
+<b>Object-Oriented Programming</b>
+
+The 1970s and 1980s introduced <b>object-oriented programming</b> (OOP), where data and behavior are bundled together. The most influential languages here include:
+
+• <b>Smalltalk</b>: pioneered OOP concepts
+• <b>C++</b>: combined OOP with the efficiency of C
+• <b>Java</b>: “Write Once, Run Anywhere” with the Java Virtual Machine
+
+<b>Code Example: Simple Class in Java</b>
+<pre><code class="language-java">public class HelloWorld {
+    public static void main(String[] args) {
+        System.out.println("Hello, World!");
+    }
+}
+</code></pre>
+
+<b>Internet and Open Source</b>
+
+The rise of the World Wide Web transformed programming. JavaScript, PHP, and Python became staples for Internet-connected software.
+
+<b>Open source</b> projects like Linux, Apache, and MySQL changed collaboration forever—developers worldwide could contribute to shared codebases.
+
+| Year | Technology  | Impact                                  |
+|------|-------------|-----------------------------------------|
+| 1991 | Linux       | Free, open-source operating systems     |
+| 1995 | JavaScript  | Interactive web applications            |
+| 2001 | Wikipedia   | Collaborative knowledge base            |
+
+<b>Mobile and Cloud Computing</b>
+
+Smartphones spawned new languages and frameworks (Swift, Kotlin, React Native).
+
+<b>Cloud computing</b> and <b>APIs</b> mean programs can collaborate on a global scale, in real-time.
+
+---
+
+<b>Programming’s Societal Impact</b>
+
+Programming is reshaping society in profound ways:
+
+• <b>Healthcare</b>: Medical imaging, diagnostics, record management
+• <b>Finance</b>: Online banking, stock trading algorithms
+• <b>Entertainment</b>: Gaming, music streaming, social networks
+• <b>Education</b>: E-learning, interactive simulations, content platforms
+• <b>Transportation</b>: Navigation, ride-sharing apps, autonomous vehicles
+• <b>Science</b>: Processing large datasets, running complex simulations
+
+<blockquote>“Software is eating the world.”
+– Marc Andreessen</blockquote>
+
+<b>Programming Jobs In Demand</b>
+
+<pre><code class="language-mermaid">pie
+    title Programming Job Market (2024)
+    "Web Development" : 31
+    "Data Science" : 22
+    "Mobile Development" : 12
+    "Embedded Systems" : 8
+    "Cybersecurity" : 9
+    "Other": 18
+</code></pre>
+
+<b>Note:</b> Mermaid diagrams require compatible renderers (e.g., GitHub, Obsidian).
+
+---
+
+<b>Ethics, Challenges, and the Future</b>
+
+With great power comes responsibility. Programmers face new challenges:
+
+• <b>Bias in Algorithms:</b> Unintentional biases in data can lead to unfair outcomes (e.g., in hiring software or criminal justice prediction).
+• <b>Privacy:</b> Handling personal data securely is more critical than ever.
+• <b>Safety:</b> In fields like self-driving cars or medical devices, software bugs can have real-world consequences.
+• <b>Sustainability:</b> Software should be efficient, minimizing environmental impact in data centers.
+
+<b>Emerging Trends:</b>
+
+• <b>Artificial Intelligence:</b> Programs that learn, adapt, and sometimes surprise their creators.
+• <b>Quantum Computing:</b> New paradigms for solving currently intractable problems.
+• <b>No-Code/Low-Code:</b> Empowering more people to harness computational power.
+
+---
+
+<b>Conclusion</b>
+
+From mechanical looms to neural networks, programming continues to redefine what’s possible. It’s not just for professional engineers: millions of people use programming as a tool for art, science, business, and personal growth.
+
+<b>Everyone can learn to code.</b> It might change your life—or even the world.
+
+<blockquote>"Any sufficiently advanced technology is indistinguishable from magic."
+— Arthur C. Clarke</blockquote>
+
+---
+
+<b>Useful Resources</b>
+
+• <a href="https://www-cs-faculty.stanford.edu/~knuth/taocp.html">The Art of Computer Programming (Donald Knuth)</a>
+• <a href="https://www.khanacademy.org/computing/computer-programming">Khan Academy Computer Programming</a>
+• <a href="https://www.w3schools.com/">W3Schools Online Tutorials</a>
+• <a href="https://www.freecodecamp.org/">freeCodeCamp</a>
+• <a href="https://stackoverflow.com/">Stack Overflow</a>
+• <a href="https://guides.github.com/">GitHub Guides</a>
+
+---
+
+<i>Thank you for reading! If you’re inspired to begin your coding journey, there has never been a better time to start.</i>
+
+---"""
+
+valid_chunk_1 = """
+Absolutely! Here’s a Markdown-formatted message exceeding 5,000 characters, exploring <b>The History and Impact of Computer Programming</b>. (You can verify the character count using any online tool.)
+
+---
+
+<b>The History and Impact of Computer Programming</b>
+
+<i>“The computer was born to solve problems that did not exist before.”</i>  
+— Bill Gates
+
+---
+
+<b>Table of Contents</b>
+
+1. <a href="#introduction">Introduction</a>  
+2. <a href="#ancient-beginnings-from-algorithms-to-machines">Ancient Beginnings: From Algorithms to Machines</a>  
+    • <a href="#al-khwarizmi-and-the-algorithm">Al-Khwarizmi and the Algorithm</a>  
+    • <a href="#the-analytical-engine">The Analytical Engine</a>  
+    • <a href="#punch-cards-and-the-jacquard-loom">Punch Cards and the Jacquard Loom</a>  
+3. <a href="#20th-century-the-birth-of-modern-programming">20th Century: The Birth of Modern Programming</a>  
+    • <a href="#eniac-and-early-programmers">ENIAC and Early Programmers</a>  
+    • <a href="#assembly-language-and-early-high-level-languages">Assembly Language and Early High-level Languages</a>  
+    • <a href="#cobol-fortran-and-the-expansion">COBOL, FORTRAN, and the Expansion</a>  
+4. <a href="#modern-era-languages-paradigms-and-the-internet">Modern Era: Languages, Paradigms, and the Internet</a>  
+    • <a href="#object-oriented-programming">Object-Oriented Programming</a>  
+    • <a href="#internet-and-open-source">Internet and Open Source</a>  
+    • <a href="#mobile-and-cloud-computing">Mobile and Cloud Computing</a>    
+5. <a href="#programmings-societal-impact">Programming’s Societal Impact</a>  
+6. <a href="#ethics-challenges-and-the-future">Ethics, Challenges, and the Future</a>  
+7. <a href="#conclusion">Conclusion</a>  
+8. <a href="#useful-resources">Useful Resources</a>  
+
+---
+
+<b>Introduction</b>
+
+Computer programming is the science and art of giving computers instructions to perform specific tasks. Today, it's impossible to imagine a world without software: from banking systems and mobile applications to traffic lights and airplanes, programming is everywhere.
+
+But how did programming begin, and what has it become today? This document explores the journey of programming, from ancient mathematical roots to the future of artificial intelligence.
+
+---
+
+<b>Ancient Beginnings: From Algorithms to Machines</b>
+
+<b>Al-Khwarizmi and the Algorithm</b>
+
+The term "<b>algorithm</b>" (the foundation of programming) comes from Abu Abdullah Muhammad ibn Musa Al-Khwarizmi, a 9th-century Persian mathematician. His works on systematic procedures laid the groundwork for computational thinking.
+
+<b>The Analytical Engine</b>
+
+In the 19th century, <b>Charles Babbage</b> designed the Analytical Engine, a mechanical general-purpose computer. Though never built in his lifetime, it could—in theory—read instructions from punched cards.
+
+<b>Ada Lovelace</b>, Babbage's collaborator, is often called the first computer programmer. She wrote notes describing algorithms (in essence, programs) for the Analytical Engine to compute Bernoulli numbers.
+
+<blockquote>"That brain of mine is something more than merely mortal; as time will show."
+– Ada Lovelace</blockquote>
+
+<b>Punch Cards and the Jacquard Loom</b>
+
+The concept of programming a machine with punched cards predates computers. <b>Joseph Marie Jacquard</b> invented a loom in 1804 that used punched cards to control patterns in woven fabric—an early example of machine automation.
+
+---
+
+<b>20th Century: The Birth of Modern Programming</b>
+
+<b>ENIAC and Early Programmers</b>
+
+ENIAC (Electronic Numerical Integrator and Computer), completed in 1945, is often cited as the first electronic general-purpose computer.
+
+Early programming was entirely manual and physically laborious—think patch cables and switches!
+
+Notably, many of the earliest programmers were women, such as <b>Kathleen McNulty</b>, <b>Jean Jennings</b>, and <b>Grace Hopper</b>.
+
+<b>Assembly Language and Early High-level Languages</b>
+
+"""
+
+valid_chunk_2 = """The problem of complexity led to <b>assembly languages</b>, where mnemonics like <code>MOV</code> and <code>ADD</code> replaced binary codes. Programming became more accessible, but code was still hardware-specific.
+
+The 1950s saw the creation of:
+
+• <b>FORTRAN</b> (FORmula TRANslation) for scientific computation
+• <b>COBOL</b> (COmmon Business-Oriented Language) for business applications
+
+<b>Code Example: Hello World in COBOL</b>
+<pre><code class="language-cobol">IDENTIFICATION DIVISION.
+PROGRAM-ID. HELLO-WORLD.
+PROCEDURE DIVISION.
+    DISPLAY "Hello, World!".
+STOP RUN.
+</code></pre>
+
+<b>COBOL, FORTRAN, and the Expansion</b>
+
+With the advent of high-level languages, programming became less about circuitry and more about solving problems. Standardized languages allowed code to run on multiple machines.
+
+Other languages soon emerged:
+
+• <b>LISP</b> (for AI research)
+• <b>ALGOL</b> (basis for many future languages)
+• <b>BASIC</b> (for beginners and education)
+
+---
+
+<b>Modern Era: Languages, Paradigms, and the Internet</b>
+
+<b>Object-Oriented Programming</b>
+
+The 1970s and 1980s introduced <b>object-oriented programming</b> (OOP), where data and behavior are bundled together. The most influential languages here include:
+
+• <b>Smalltalk</b>: pioneered OOP concepts
+• <b>C++</b>: combined OOP with the efficiency of C
+• <b>Java</b>: “Write Once, Run Anywhere” with the Java Virtual Machine
+
+<b>Code Example: Simple Class in Java</b>
+<pre><code class="language-java">public class HelloWorld {
+    public static void main(String[] args) {
+        System.out.println("Hello, World!");
+    }
+}
+</code></pre>
+
+<b>Internet and Open Source</b>
+
+The rise of the World Wide Web transformed programming. JavaScript, PHP, and Python became staples for Internet-connected software.
+
+<b>Open source</b> projects like Linux, Apache, and MySQL changed collaboration forever—developers worldwide could contribute to shared codebases.
+
+| Year | Technology  | Impact                                  |
+|------|-------------|-----------------------------------------|
+| 1991 | Linux       | Free, open-source operating systems     |
+| 1995 | JavaScript  | Interactive web applications            |
+| 2001 | Wikipedia   | Collaborative knowledge base            |
+
+<b>Mobile and Cloud Computing</b>
+
+Smartphones spawned new languages and frameworks (Swift, Kotlin, React Native).
+
+<b>Cloud computing</b> and <b>APIs</b> mean programs can collaborate on a global scale, in real-time.
+
+---
+
+<b>Programming’s Societal Impact</b>
+
+Programming is reshaping society in profound ways:
+
+• <b>Healthcare</b>: Medical imaging, diagnostics, record management
+• <b>Finance</b>: Online banking, stock trading algorithms
+• <b>Entertainment</b>: Gaming, music streaming, social networks
+• <b>Education</b>: E-learning, interactive simulations, content platforms
+• <b>Transportation</b>: Navigation, ride-sharing apps, autonomous vehicles
+• <b>Science</b>: Processing large datasets, running complex simulations
+
+<blockquote>“Software is eating the world.”
+– Marc Andreessen</blockquote>
+
+<b>Programming Jobs In Demand</b>
+
+<pre><code class="language-mermaid">pie
+    title Programming Job Market (2024)
+    "Web Development" : 31
+    "Data Science" : 22
+    "Mobile Development" : 12
+    "Embedded Systems" : 8
+    "Cybersecurity" : 9
+    "Other": 18
+</code></pre>"""
+
+valid_chunk_3 = """
+
+<b>Note:</b> Mermaid diagrams require compatible renderers (e.g., GitHub, Obsidian).
+
+---
+
+<b>Ethics, Challenges, and the Future</b>
+
+With great power comes responsibility. Programmers face new challenges:
+
+• <b>Bias in Algorithms:</b> Unintentional biases in data can lead to unfair outcomes (e.g., in hiring software or criminal justice prediction).
+• <b>Privacy:</b> Handling personal data securely is more critical than ever.
+• <b>Safety:</b> In fields like self-driving cars or medical devices, software bugs can have real-world consequences.
+• <b>Sustainability:</b> Software should be efficient, minimizing environmental impact in data centers.
+
+<b>Emerging Trends:</b>
+
+• <b>Artificial Intelligence:</b> Programs that learn, adapt, and sometimes surprise their creators.
+• <b>Quantum Computing:</b> New paradigms for solving currently intractable problems.
+• <b>No-Code/Low-Code:</b> Empowering more people to harness computational power.
+
+---
+
+<b>Conclusion</b>
+
+From mechanical looms to neural networks, programming continues to redefine what’s possible. It’s not just for professional engineers: millions of people use programming as a tool for art, science, business, and personal growth.
+
+<b>Everyone can learn to code.</b> It might change your life—or even the world.
+
+<blockquote>"Any sufficiently advanced technology is indistinguishable from magic."
+— Arthur C. Clarke</blockquote>
+
+---
+
+<b>Useful Resources</b>
+
+• <a href="https://www-cs-faculty.stanford.edu/~knuth/taocp.html">The Art of Computer Programming (Donald Knuth)</a>
+• <a href="https://www.khanacademy.org/computing/computer-programming">Khan Academy Computer Programming</a>
+• <a href="https://www.w3schools.com/">W3Schools Online Tutorials</a>
+• <a href="https://www.freecodecamp.org/">freeCodeCamp</a>
+• <a href="https://stackoverflow.com/">Stack Overflow</a>
+• <a href="https://guides.github.com/">GitHub Guides</a>
+
+---
+
+<i>Thank you for reading! If you’re inspired to begin your coding journey, there has never been a better time to start.</i>
+
+---"""
+
+def test_splitter_test():
+  chunks = split_html_for_telegram(input_text)
+  valid_chunks = [valid_chunk_1, valid_chunk_2, valid_chunk_3]
+  assert chunks == valid_chunks