#!/usr/bin/env python3
"""Apply enriched bodies from drafts/out/ back into the live source files.

Tense guides → Conjuga/Conjuga/conjuga_data.json (tenseGuides[].body)
Grammar notes → Conjuga/Conjuga/Models/GrammarNote.swift (body: \"\"\"...\"\"\")

Filename conventions in drafts/out/:
    tense__<tenseId>.md       — body to drop into the matching tenseGuide
    note__<noteId>.md          — body to drop into the matching GrammarNote(...) declaration

Run from anywhere; uses absolute paths anchored at the repo root.
"""

from __future__ import annotations
import json
import re
import sys
from pathlib import Path

REPO = Path('/Users/m4mini/Desktop/code/Spanish')
OUT_DIR = REPO / 'Conjuga/Scripts/guide-enrichment/out'
TENSE_JSON = REPO / 'Conjuga/Conjuga/conjuga_data.json'
NOTES_SWIFT = REPO / 'Conjuga/Conjuga/Models/GrammarNote.swift'


def read_draft(path: Path) -> str:
    """Drafts may start with comment blocks like `# Title: ...`, `# Category:
    ...`, and `# ENRICHED BODY` separated by blank lines. Strip every leading
    line that is blank or starts with `#` until we reach the actual body."""
    raw = path.read_text(encoding='utf-8')
    lines = raw.splitlines()
    start = 0
    for i, line in enumerate(lines):
        stripped = line.strip()
        if stripped == '' or stripped.startswith('#'):
            start = i + 1
            continue
        # Hit the first real content line — keep everything from here.
        start = i
        break
    body = '\n'.join(lines[start:]).strip()
    if not body:
        raise ValueError(f'Empty body after stripping header in {path}')
    return body


def apply_tense_guides() -> int:
    data = json.loads(TENSE_JSON.read_text(encoding='utf-8'))
    drafts = sorted(OUT_DIR.glob('tense__*.md'))
    by_id = {g['tenseId']: g for g in data['tenseGuides']}
    applied = 0
    for path in drafts:
        tense_id = path.stem.removeprefix('tense__')
        if tense_id not in by_id:
            print(f'  SKIP {tense_id}: not in tenseGuides', file=sys.stderr)
            continue
        body = read_draft(path)
        by_id[tense_id]['body'] = body
        applied += 1
        print(f'  applied tense: {tense_id}  ({len(body)} chars)')
    TENSE_JSON.write_text(
        json.dumps(data, ensure_ascii=False, separators=(',', ':')),
        encoding='utf-8'
    )
    return applied


# Match each GrammarNote(...) declaration. Body uses """...""" — may contain
# anything except a triple-quote.
NOTE_PATTERN = re.compile(
    r'(GrammarNote\(\s*id:\s*"([^"]+)",\s*'
    r'title:\s*"(?:[^"\\]|\\.)*",\s*'
    r'category:\s*"[^"]+",\s*'
    r'body:\s*""")(.*?)(""")',
    re.DOTALL
)


def apply_grammar_notes() -> int:
    src = NOTES_SWIFT.read_text(encoding='utf-8')
    drafts = sorted(OUT_DIR.glob('note__*.md'))
    by_id = {p.stem.removeprefix('note__'): p for p in drafts}
    applied = [0]

    def replace_match(m):
        prefix, note_id, _, suffix = m.group(1), m.group(2), m.group(3), m.group(4) if m.lastindex and m.lastindex >= 4 else m.group(3)
        return m.group(0)  # placeholder, see real callback below

    def real_replace(m):
        prefix = m.group(1)
        note_id = m.group(2)
        suffix = m.group(4)
        if note_id not in by_id:
            return m.group(0)
        body = read_draft(by_id[note_id])
        if '"""' in body:
            raise ValueError(f'Body for {note_id} contains triple-quote — would break Swift parser')
        # Re-indent to match the existing Swift block. The existing format uses
        # 8 spaces of leading indent inside body lines. We don't enforce that —
        # the Swift compiler handles multiline string indentation by stripping
        # the leading whitespace common to all lines based on the closing """.
        # Just write the body verbatim.
        applied[0] += 1
        print(f'  applied note: {note_id}  ({len(body)} chars)')
        return f'{prefix}\n{body}\n{suffix}'

    new_src = NOTE_PATTERN.sub(real_replace, src)
    NOTES_SWIFT.write_text(new_src, encoding='utf-8')
    return applied[0]


def main():
    if not OUT_DIR.exists():
        print(f'No drafts/out directory at {OUT_DIR}', file=sys.stderr)
        sys.exit(1)

    print(f'=== Tense guides ===')
    tense_count = apply_tense_guides()
    print(f'\n=== Grammar notes ===')
    note_count = apply_grammar_notes()

    print(f'\nTotal applied: {tense_count} tense guides + {note_count} grammar notes')
    if tense_count == 0 and note_count == 0:
        print('Nothing applied — drafts/out/ was empty.', file=sys.stderr)
        sys.exit(2)


if __name__ == '__main__':
    main()