kristovatlas/ascii_to_markdown_tbl.py

## ascii_to_markdown_tbl.py
import sys


def is_separator(line):
    """Check if a line is a horizontal separator (e.g. ├───┼───┤, ╞═══╪═══╡, etc.)"""
    stripped = line.strip()
    # Separator lines are made up of box-drawing border characters and have no letter/digit content
    border_chars = set('─━═┼╪╫┬┴├┤┌┐└┘╔╗╚╝╠╣╦╩╬╞╡╥╨╤╧╟╢+|-=')
    return bool(stripped) and all(c in border_chars or c.isspace() for c in stripped)


def parse_columns(line):
    """Split a │-delimited line into cell values."""
    parts = line.split('│')
    if len(parts) >= 3:
        return [part.strip() for part in parts[1:-1]]
    return None


def main():
    if len(sys.argv) != 2:
        print("Usage: python script.py <input_file>")
        sys.exit(1)

    filename = sys.argv[1]
    with open(filename, 'r') as f:
        lines = f.read().splitlines()

    # Group lines between separators into logical rows
    logical_rows = []
    current_group = []

    for line in lines:
        if is_separator(line):
            if current_group:
                logical_rows.append(current_group)
                current_group = []
        else:
            stripped = line.strip()
            if stripped.startswith('│') and '│' in stripped[1:]:
                current_group.append(line)

    # Don't forget the last group if the table doesn't end with a separator
    if current_group:
        logical_rows.append(current_group)

    if not logical_rows:
        print("No table found in the input file.")
        sys.exit(1)

    # Merge multi-line groups into single rows, joining cell content with spaces
    rows = []
    for group in logical_rows:
        merged = None
        for line in group:
            cols = parse_columns(line)
            if cols is None:
                continue
            if merged is None:
                merged = cols
            else:
                # Pad to same length if needed
                while len(merged) < len(cols):
                    merged.append('')
                while len(cols) < len(merged):
                    cols.append('')
                # Append each cell's continuation text
                for i in range(len(merged)):
                    if cols[i]:
                        merged[i] = (merged[i] + ' ' + cols[i]).strip()
        if merged and any(merged):
            rows.append(merged)

    if not rows:
        print("No table found in the input file.")
        sys.exit(1)

    # First row is headers, rest are data
    headers = rows[0]
    data_rows = rows[1:]

    # Escape any pipes inside cell content so they don't break Markdown
    def escape_cell(cell):
        return cell.replace('|', '\\|')

    md_table = '| ' + ' | '.join(escape_cell(h) for h in headers) + ' |\n'
    md_table += '| ' + ' | '.join(['---'] * len(headers)) + ' |\n'
    for row in data_rows:
        # Pad short rows
        while len(row) < len(headers):
            row.append('')
        md_table += '| ' + ' | '.join(escape_cell(c) for c in row) + ' |\n'

    print(md_table)


if __name__ == "__main__":
    main()

## text_stripper.html
<!DOCTYPE html>
<html lang="en">
<head>
    <meta charset="UTF-8">
    <meta name="viewport" content="width=device-width, initial-scale=1.0">
    <title>Blank Line & Whitespace Remover</title>
    <style>
        * {
            margin: 0;
            padding: 0;
            box-sizing: border-box;
        }

        body {
            font-family: -apple-system, BlinkMacSystemFont, 'Segoe UI', Roboto, Oxygen, Ubuntu, Cantarell, sans-serif;
            padding: 20px;
            background-color: #f5f5f5;
            line-height: 1.6;
        }

        .container {
            max-width: 1200px;
            margin: 0 auto;
            background: white;
            padding: 30px;
            border-radius: 8px;
            box-shadow: 0 2px 10px rgba(0,0,0,0.1);
        }

        h1 {
            color: #333;
            margin-bottom: 10px;
        }

        .description {
            color: #666;
            margin-bottom: 25px;
            font-size: 14px;
        }

        .section {
            margin-bottom: 25px;
        }

        label {
            display: block;
            font-weight: 600;
            margin-bottom: 8px;
            color: #444;
        }

        textarea {
            width: 100%;
            min-height: 250px;
            padding: 15px;
            border: 2px solid #ddd;
            border-radius: 6px;
            font-family: 'Courier New', monospace;
            font-size: 14px;
            resize: vertical;
            transition: border-color 0.3s;
        }

        textarea:focus {
            outline: none;
            border-color: #4CAF50;
        }

        .button-group {
            display: flex;
            gap: 10px;
            margin-bottom: 25px;
        }

        button {
            padding: 12px 24px;
            font-size: 16px;
            font-weight: 600;
            border: none;
            border-radius: 6px;
            cursor: pointer;
            transition: all 0.3s;
        }

        .process-btn {
            background-color: #4CAF50;
            color: white;
        }

        .process-btn:hover {
            background-color: #45a049;
        }

        .copy-btn {
            background-color: #2196F3;
            color: white;
        }

        .copy-btn:hover {
            background-color: #0b7dda;
        }

        .clear-btn {
            background-color: #f44336;
            color: white;
        }

        .clear-btn:hover {
            background-color: #da190b;
        }

        .stats {
            background-color: #f9f9f9;
            padding: 15px;
            border-radius: 6px;
            margin-bottom: 25px;
            display: flex;
            gap: 30px;
            font-size: 14px;
        }

        .stat-item {
            color: #666;
        }

        .stat-item strong {
            color: #333;
        }

        .success-message {
            display: none;
            background-color: #4CAF50;
            color: white;
            padding: 12px;
            border-radius: 6px;
            margin-bottom: 20px;
            text-align: center;
        }

        .success-message.show {
            display: block;
        }
    </style>
</head>
<body>
    <div class="container">
        <h1>Blank Line & Whitespace Remover</h1>
        <p class="description">Paste your text below to remove all blank lines and trailing whitespace from each line.</p>

        <div id="successMessage" class="success-message">Copied to clipboard!</div>

        <div class="section">
            <label for="inputText">Input Text:</label>
            <textarea id="inputText" placeholder="Paste your text here..."></textarea>
        </div>

        <div class="button-group">
            <button class="process-btn" onclick="processText()">Process Text</button>
            <button class="copy-btn" onclick="copyOutput()">Copy Output</button>
            <button class="clear-btn" onclick="clearAll()">Clear All</button>
        </div>

        <div class="stats" id="stats" style="display: none;">
            <div class="stat-item">
                Original length: <strong id="originalLength">0</strong> characters
            </div>
            <div class="stat-item">
                Processed length: <strong id="processedLength">0</strong> characters
            </div>
            <div class="stat-item">
                Removed: <strong id="removedChars">0</strong> characters
            </div>
        </div>

        <div class="section">
            <label for="outputText">Output Text:</label>
            <textarea id="outputText" placeholder="Processed text will appear here..." readonly></textarea>
        </div>
    </div>

    <script>
        function processText() {
            const input = document.getElementById('inputText').value;

            if (!input.trim()) {
                alert('Please paste some text first!');
                return;
            }

            // Split by newlines, trim trailing whitespace from each line,
            // filter out blank lines, then join with newlines
            const processed = input
                .split('\n')
                .map(line => line.trimEnd())
                .filter(line => line.length > 0)
                .join('\n');

            document.getElementById('outputText').value = processed;

            // Update statistics
            const originalLength = input.length;
            const processedLength = processed.length;
            const removedChars = originalLength - processedLength;

            document.getElementById('originalLength').textContent = originalLength;
            document.getElementById('processedLength').textContent = processedLength;
            document.getElementById('removedChars').textContent = removedChars;
            document.getElementById('stats').style.display = 'flex';
        }

        function copyOutput() {
            const output = document.getElementById('outputText');

            if (!output.value) {
                alert('No output to copy! Process some text first.');
                return;
            }

            output.select();
            document.execCommand('copy');

            // Show success message
            const successMsg = document.getElementById('successMessage');
            successMsg.classList.add('show');
            setTimeout(() => {
                successMsg.classList.remove('show');
            }, 2000);
        }

        function clearAll() {
            document.getElementById('inputText').value = '';
            document.getElementById('outputText').value = '';
            document.getElementById('stats').style.display = 'none';
        }

        // Auto-focus on input textarea when page loads
        window.onload = function() {
            document.getElementById('inputText').focus();
        };
    </script>
</body>
</html>
	import sys


	def is_separator(line):
	"""Check if a line is a horizontal separator (e.g. ├───┼───┤, ╞═══╪═══╡, etc.)"""
	stripped = line.strip()
	# Separator lines are made up of box-drawing border characters and have no letter/digit content
	border_chars = set('─━═┼╪╫┬┴├┤┌┐└┘╔╗╚╝╠╣╦╩╬╞╡╥╨╤╧╟╢+\|-=')
	return bool(stripped) and all(c in border_chars or c.isspace() for c in stripped)


	def parse_columns(line):
	"""Split a │-delimited line into cell values."""
	parts = line.split('│')
	if len(parts) >= 3:
	return [part.strip() for part in parts[1:-1]]
	return None


	def main():
	if len(sys.argv) != 2:
	print("Usage: python script.py <input_file>")
	sys.exit(1)

	filename = sys.argv[1]
	with open(filename, 'r') as f:
	lines = f.read().splitlines()

	# Group lines between separators into logical rows
	logical_rows = []
	current_group = []

	for line in lines:
	if is_separator(line):
	if current_group:
	logical_rows.append(current_group)
	current_group = []
	else:
	stripped = line.strip()
	if stripped.startswith('│') and '│' in stripped[1:]:
	current_group.append(line)

	# Don't forget the last group if the table doesn't end with a separator
	if current_group:
	logical_rows.append(current_group)

	if not logical_rows:
	print("No table found in the input file.")
	sys.exit(1)

	# Merge multi-line groups into single rows, joining cell content with spaces
	rows = []
	for group in logical_rows:
	merged = None
	for line in group:
	cols = parse_columns(line)
	if cols is None:
	continue
	if merged is None:
	merged = cols
	else:
	# Pad to same length if needed
	while len(merged) < len(cols):
	merged.append('')
	while len(cols) < len(merged):
	cols.append('')
	# Append each cell's continuation text
	for i in range(len(merged)):
	if cols[i]:
	merged[i] = (merged[i] + ' ' + cols[i]).strip()
	if merged and any(merged):
	rows.append(merged)

	if not rows:
	print("No table found in the input file.")
	sys.exit(1)

	# First row is headers, rest are data
	headers = rows[0]
	data_rows = rows[1:]

	# Escape any pipes inside cell content so they don't break Markdown
	def escape_cell(cell):
	return cell.replace('\|', '\\\|')

	md_table = '\| ' + ' \| '.join(escape_cell(h) for h in headers) + ' \|\n'
	md_table += '\| ' + ' \| '.join(['---'] * len(headers)) + ' \|\n'
	for row in data_rows:
	# Pad short rows
	while len(row) < len(headers):
	row.append('')
	md_table += '\| ' + ' \| '.join(escape_cell(c) for c in row) + ' \|\n'

	print(md_table)


	if __name__ == "__main__":
	main()
	<!DOCTYPE html>
	<html lang="en">
	<head>
	<meta charset="UTF-8">
	<meta name="viewport" content="width=device-width, initial-scale=1.0">
	<title>Blank Line & Whitespace Remover</title>
	<style>
	* {
	margin: 0;
	padding: 0;
	box-sizing: border-box;
	}

	body {
	font-family: -apple-system, BlinkMacSystemFont, 'Segoe UI', Roboto, Oxygen, Ubuntu, Cantarell, sans-serif;
	padding: 20px;
	background-color: #f5f5f5;
	line-height: 1.6;
	}

	.container {
	max-width: 1200px;
	margin: 0 auto;
	background: white;
	padding: 30px;
	border-radius: 8px;
	box-shadow: 0 2px 10px rgba(0,0,0,0.1);
	}

	h1 {
	color: #333;
	margin-bottom: 10px;
	}

	.description {
	color: #666;
	margin-bottom: 25px;
	font-size: 14px;
	}

	.section {
	margin-bottom: 25px;
	}

	label {
	display: block;
	font-weight: 600;
	margin-bottom: 8px;
	color: #444;
	}

	textarea {
	width: 100%;
	min-height: 250px;
	padding: 15px;
	border: 2px solid #ddd;
	border-radius: 6px;
	font-family: 'Courier New', monospace;
	font-size: 14px;
	resize: vertical;
	transition: border-color 0.3s;
	}

	textarea:focus {
	outline: none;
	border-color: #4CAF50;
	}

	.button-group {
	display: flex;
	gap: 10px;
	margin-bottom: 25px;
	}

	button {
	padding: 12px 24px;
	font-size: 16px;
	font-weight: 600;
	border: none;
	border-radius: 6px;
	cursor: pointer;
	transition: all 0.3s;
	}

	.process-btn {
	background-color: #4CAF50;
	color: white;
	}

	.process-btn:hover {
	background-color: #45a049;
	}

	.copy-btn {
	background-color: #2196F3;
	color: white;
	}

	.copy-btn:hover {
	background-color: #0b7dda;
	}

	.clear-btn {
	background-color: #f44336;
	color: white;
	}

	.clear-btn:hover {
	background-color: #da190b;
	}

	.stats {
	background-color: #f9f9f9;
	padding: 15px;
	border-radius: 6px;
	margin-bottom: 25px;
	display: flex;
	gap: 30px;
	font-size: 14px;
	}

	.stat-item {
	color: #666;
	}

	.stat-item strong {
	color: #333;
	}

	.success-message {
	display: none;
	background-color: #4CAF50;
	color: white;
	padding: 12px;
	border-radius: 6px;
	margin-bottom: 20px;
	text-align: center;
	}

	.success-message.show {
	display: block;
	}
	</style>
	</head>
	<body>
	<div class="container">
	<h1>Blank Line & Whitespace Remover</h1>
	<p class="description">Paste your text below to remove all blank lines and trailing whitespace from each line.</p>

	<div id="successMessage" class="success-message">Copied to clipboard!</div>

	<div class="section">
	<label for="inputText">Input Text:</label>
	<textarea id="inputText" placeholder="Paste your text here..."></textarea>
	</div>

	<div class="button-group">
	<button class="process-btn" onclick="processText()">Process Text</button>
	<button class="copy-btn" onclick="copyOutput()">Copy Output</button>
	<button class="clear-btn" onclick="clearAll()">Clear All</button>
	</div>

	<div class="stats" id="stats" style="display: none;">
	<div class="stat-item">
	Original length: <strong id="originalLength">0</strong> characters
	</div>
	<div class="stat-item">
	Processed length: <strong id="processedLength">0</strong> characters
	</div>
	<div class="stat-item">
	Removed: <strong id="removedChars">0</strong> characters
	</div>
	</div>

	<div class="section">
	<label for="outputText">Output Text:</label>
	<textarea id="outputText" placeholder="Processed text will appear here..." readonly></textarea>
	</div>
	</div>

	<script>
	function processText() {
	const input = document.getElementById('inputText').value;

	if (!input.trim()) {
	alert('Please paste some text first!');
	return;
	}

	// Split by newlines, trim trailing whitespace from each line,
	// filter out blank lines, then join with newlines
	const processed = input
	.split('\n')
	.map(line => line.trimEnd())
	.filter(line => line.length > 0)
	.join('\n');

	document.getElementById('outputText').value = processed;

	// Update statistics
	const originalLength = input.length;
	const processedLength = processed.length;
	const removedChars = originalLength - processedLength;

	document.getElementById('originalLength').textContent = originalLength;
	document.getElementById('processedLength').textContent = processedLength;
	document.getElementById('removedChars').textContent = removedChars;
	document.getElementById('stats').style.display = 'flex';
	}

	function copyOutput() {
	const output = document.getElementById('outputText');

	if (!output.value) {
	alert('No output to copy! Process some text first.');
	return;
	}

	output.select();
	document.execCommand('copy');

	// Show success message
	const successMsg = document.getElementById('successMessage');
	successMsg.classList.add('show');
	setTimeout(() => {
	successMsg.classList.remove('show');
	}, 2000);
	}

	function clearAll() {
	document.getElementById('inputText').value = '';
	document.getElementById('outputText').value = '';
	document.getElementById('stats').style.display = 'none';
	}

	// Auto-focus on input textarea when page loads
	window.onload = function() {
	document.getElementById('inputText').focus();
	};
	</script>
	</body>
	</html>