mirror of
https://github.com/harvard-edge/cs249r_book.git
synced 2026-07-20 04:45:11 -05:00
96 lines
3.3 KiB
Python
96 lines
3.3 KiB
Python
#!/usr/bin/env python3
|
|
"""Apply edited alt-text from copy editor JSON extraction.
|
|
|
|
Reads the alt_text_edits.json and updates fig-alt= attributes in QMD files.
|
|
|
|
Usage:
|
|
python3 fix_alt_text.py --edits data/alt_text_edits.json --dry-run book/quarto/contents/vol1/
|
|
python3 fix_alt_text.py --edits data/alt_text_edits.json book/quarto/contents/vol1/
|
|
"""
|
|
import argparse
|
|
import json
|
|
import re
|
|
from pathlib import Path
|
|
|
|
|
|
def load_alt_text_edits(filepath: Path) -> dict[str, str]:
|
|
"""Load the alt-text edits, returning {label: alt_text}."""
|
|
with open(filepath) as f:
|
|
data = json.load(f)
|
|
mapping = {}
|
|
skipped = 0
|
|
for entry in data:
|
|
label = entry.get("label", "").strip()
|
|
alt_text = entry.get("alt_text", "").strip()
|
|
if label and alt_text:
|
|
# Clean up whitespace issues from docx extraction
|
|
alt_text = re.sub(r'\s+', ' ', alt_text).strip()
|
|
|
|
# Skip entries with OCR corruption (single letter + space + word)
|
|
# e.g. "s witch t ransformer" instead of "Switch Transformer"
|
|
if re.search(r'\b[a-z] [a-z]{2,}', alt_text):
|
|
skipped += 1
|
|
continue
|
|
|
|
# Enforce 250 char limit
|
|
if len(alt_text) > 250:
|
|
alt_text = alt_text[:247] + "..."
|
|
mapping[label] = alt_text
|
|
if skipped:
|
|
print(f"Skipped {skipped} entries with OCR corruption from docx extraction")
|
|
return mapping
|
|
|
|
|
|
def process_file(filepath: Path, edits: dict[str, str], dry_run: bool) -> int:
|
|
"""Update fig-alt attributes in a single QMD file."""
|
|
text = filepath.read_text(encoding="utf-8")
|
|
count = 0
|
|
|
|
for label, new_alt in edits.items():
|
|
# Pattern: fig-alt="..." in a div that has #fig-LABEL or #label
|
|
# The div looks like: ::: {#fig-xxx ... fig-alt="CURRENT ALT TEXT"}
|
|
pattern = rf'({{[^}}]*#{re.escape(label)}\b[^}}]*fig-alt=")([^"]*?)(")'
|
|
|
|
def replacer(m):
|
|
nonlocal count
|
|
old_alt = m.group(2)
|
|
if old_alt.strip() != new_alt:
|
|
count += 1
|
|
if dry_run:
|
|
short_old = old_alt[:60] + "..." if len(old_alt) > 60 else old_alt
|
|
short_new = new_alt[:60] + "..." if len(new_alt) > 60 else new_alt
|
|
print(f" {label}: \"{short_old}\" -> \"{short_new}\"")
|
|
return m.group(1) + new_alt + m.group(3)
|
|
return m.group(0)
|
|
|
|
text = re.sub(pattern, replacer, text)
|
|
|
|
if count > 0 and not dry_run:
|
|
filepath.write_text(text, encoding="utf-8")
|
|
|
|
return count
|
|
|
|
|
|
def main():
|
|
parser = argparse.ArgumentParser(description="Apply edited alt-text")
|
|
parser.add_argument("path", type=Path, help="Directory to process")
|
|
parser.add_argument("--dry-run", action="store_true")
|
|
parser.add_argument("--edits", type=Path, required=True,
|
|
help="Path to alt_text_edits.json")
|
|
args = parser.parse_args()
|
|
|
|
edits = load_alt_text_edits(args.edits)
|
|
print(f"Loaded {len(edits)} alt-text edits")
|
|
|
|
files = sorted(args.path.rglob("*.qmd"))
|
|
total = 0
|
|
for f in files:
|
|
total += process_file(f, edits, dry_run=args.dry_run)
|
|
|
|
mode = "DRY RUN" if args.dry_run else "APPLIED"
|
|
print(f"\n{mode}: {total} alt-text updates across {len(files)} files")
|
|
|
|
|
|
if __name__ == "__main__":
|
|
main()
|