File size: 4,120 Bytes
734569e | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 | """
Convert grpo_vertex_v3.md → grpo_vertex_v3.ipynb
Parses markdown with ```python fenced code blocks into Jupyter notebook cells.
- ```python blocks → code cells
- Everything else → markdown cells
- Consecutive markdown sections are merged into single cells
- Empty cells are skipped
"""
import json
import re
import sys
from pathlib import Path
def md_to_notebook(md_text: str) -> dict:
"""Parse markdown into notebook cells."""
cells = []
# Split on ```python and ``` boundaries
# We need to track whether we're inside a code block
lines = md_text.split('\n')
current_type = 'markdown' # or 'code'
current_lines = []
for line in lines:
if line.strip() == '```python':
# Flush current markdown
if current_lines and current_type == 'markdown':
text = '\n'.join(current_lines).strip()
if text:
cells.append(make_markdown_cell(text))
current_lines = []
current_type = 'code'
elif line.strip() == '```' and current_type == 'code':
# Flush current code
if current_lines:
code = '\n'.join(current_lines)
# Remove trailing whitespace but keep structure
code = code.rstrip()
if code:
cells.append(make_code_cell(code))
current_lines = []
current_type = 'markdown'
else:
current_lines.append(line)
# Flush remaining
if current_lines:
text = '\n'.join(current_lines).strip()
if text:
if current_type == 'code':
cells.append(make_code_cell(text))
else:
cells.append(make_markdown_cell(text))
# Build notebook
notebook = {
"nbformat": 4,
"nbformat_minor": 5,
"metadata": {
"kernelspec": {
"display_name": "Python 3 (ipykernel)",
"language": "python",
"name": "python3"
},
"language_info": {
"name": "python",
"version": "3.10.0",
"mimetype": "text/x-python",
"file_extension": ".py"
}
},
"cells": cells
}
return notebook
def make_code_cell(source: str) -> dict:
"""Create a code cell."""
return {
"cell_type": "code",
"execution_count": None,
"metadata": {},
"outputs": [],
"source": source.split('\n') # Will be joined with \n later
}
def make_markdown_cell(source: str) -> dict:
"""Create a markdown cell."""
return {
"cell_type": "markdown",
"metadata": {},
"source": source.split('\n')
}
def format_notebook(notebook: dict) -> str:
"""
Format notebook JSON with proper source line handling.
Each line in source needs a trailing \n except the last.
"""
for cell in notebook["cells"]:
lines = cell["source"]
if lines:
# Add \n to all lines except the last
cell["source"] = [line + '\n' for line in lines[:-1]] + [lines[-1]]
return json.dumps(notebook, indent=1, ensure_ascii=False)
def main():
input_path = Path(sys.argv[1]) if len(sys.argv) > 1 else Path("grpo_vertex_v3.md")
output_path = input_path.with_suffix('.ipynb')
if not input_path.exists():
print(f"Error: {input_path} not found")
sys.exit(1)
md_text = input_path.read_text(encoding='utf-8')
notebook = md_to_notebook(md_text)
# Stats
code_cells = sum(1 for c in notebook["cells"] if c["cell_type"] == "code")
md_cells = sum(1 for c in notebook["cells"] if c["cell_type"] == "markdown")
output_path.write_text(format_notebook(notebook), encoding='utf-8')
print(f"✓ Converted {input_path} → {output_path}")
print(f" {code_cells} code cells, {md_cells} markdown cells")
print(f" Size: {output_path.stat().st_size / 1024:.1f} KB")
if __name__ == "__main__":
main()
|