#!/usr/bin/env python3
"""ocr2tex.py PDFPAGE PRINTED [--head 'Title'] — turn tesseract scaffold into a draft
page file (paragraphs joined, LaTeX specials escaped, quotes normalised).
The draft MUST then be proofread against the page image and corrected."""
import sys, re, argparse
ap=argparse.ArgumentParser(); ap.add_argument("page",type=int); ap.add_argument("printed",type=int)
ap.add_argument("--head",default=None); ap.add_argument("--bookmark",default=None)
a=ap.parse_args()
txt=open(f"work/ocr/p{a.page:04d}.txt").read()
lines=txt.splitlines()
if a.head and lines and lines[0].strip().lower()==a.head.lower(): lines=lines[1:]
body="\n".join(lines).strip()
paras=[re.sub(r"\s*\n\s*"," ",p).strip() for p in re.split(r"\n\s*\n",body) if p.strip()]
def esc(s):
    s=s.replace("\\","\\textbackslash{}")
    for c in "&%$#_{}": s=s.replace(c,"\\"+c)
    s=re.sub(r"(\d)-(\d)",r"\1-\2",s)
    s=s.replace("‘","`").replace("’","'").replace("“","``").replace("”","''")
    s=re.sub(r"\b(\w+)-\s+(\w)",r"\1\2",s)   # de-hyphenate line breaks (check!)
    return s
out=[f"% PDF page {a.page} — printed page {a.printed}",f"\\origpage{{{a.printed}}}"]
if a.bookmark: out.append(f"\\pdfbookmark[1]{{{a.bookmark}}}{{bm{a.page}}}")
if a.head: out.append(f"\\chaphead{{{a.head}}}")
out.append("\n\n".join(esc(p) for p in paras))
p=f"src/pages/p{a.page:04d}.tex"; open(p,"w").write("\n".join(out)+"\n"); print(p, len(paras),"paras")