#!/usr/bin/env python3
# -*- coding: utf-8 -*-
"""Build gdpr-content-blocker-en_US.po + .mo.

Robust and order-independent: every German source string (extracted into
hilfsdaten/strings.txt) is looked up in hilfsdaten/translations-en.json.
Adding a new __() string to the plugin only requires adding one key/value to
that JSON file — order does not matter. The build FAILS (and lists them) if any
extracted string has no translation.

Usage:  python hilfsdaten/extract-strings.py gdpr-content-blocker hilfsdaten/strings.txt
        python hilfsdaten/build-en-mo.py .
"""
import json
import os
import struct
import sys

base = sys.argv[1] if len(sys.argv) > 1 else '.'
here = os.path.dirname(os.path.abspath(__file__))

with open(os.path.join(here, 'translations-en.json'), encoding='utf-8') as f:
    TRANSLATIONS = json.load(f)

strings_file = os.path.join(base, 'hilfsdaten', 'strings.txt')
de = []
with open(strings_file, encoding='utf-8') as f:
    for line in f:
        if line.startswith('MSGID\t'):
            de.append(line[len('MSGID\t'):].rstrip('\n'))

missing = [s for s in de if s not in TRANSLATIONS]
if missing:
    print('MISSING %d translation(s) — add them to hilfsdaten/translations-en.json:' % len(missing))
    for m in missing:
        print('  ' + json.dumps(m, ensure_ascii=False))
    sys.exit(1)

header = (
    'Project-Id-Version: GDPR Content Blocker\n'
    'MIME-Version: 1.0\n'
    'Content-Type: text/plain; charset=UTF-8\n'
    'Content-Transfer-Encoding: 8bit\n'
    'Language: en_US\n'
)

entries = {'': header}
for s in de:
    entries[s] = TRANSLATIONS[s]


def po_escape(s):
    return s.replace('\\', '\\\\').replace('"', '\\"').replace('\n', '\\n')


po_path = os.path.join(base, 'gdpr-content-blocker', 'languages', 'gdpr-content-blocker-en_US.po')
with open(po_path, 'w', encoding='utf-8') as f:
    f.write('msgid ""\nmsgstr ""\n')
    for line in header.rstrip('\n').split('\n'):
        f.write('"%s\\n"\n' % po_escape(line))
    f.write('\n')
    for s in de:
        f.write('msgid "%s"\n' % po_escape(s))
        f.write('msgstr "%s"\n\n' % po_escape(TRANSLATIONS[s]))


def write_mo(entries, out):
    keys = sorted(entries.keys())
    ids = b''
    strs = b''
    offsets = []
    for k in keys:
        kb = k.encode('utf-8')
        vb = entries[k].encode('utf-8')
        offsets.append((len(ids), len(kb), len(strs), len(vb)))
        ids += kb + b'\x00'
        strs += vb + b'\x00'
    keystart = 7 * 4 + 16 * len(keys)
    valuestart = keystart + len(ids)
    ko = []
    vo = []
    for o1, l1, o2, l2 in offsets:
        ko += [l1, o1 + keystart]
        vo += [l2, o2 + valuestart]
    out_b = struct.pack('Iiiiiii', 0x950412de, 0, len(keys), 7 * 4, 7 * 4 + len(keys) * 8, 0, 0)
    out_b += struct.pack('i' * len(ko), *ko) + struct.pack('i' * len(vo), *vo) + ids + strs
    with open(out, 'wb') as f:
        f.write(out_b)


mo_path = os.path.join(base, 'gdpr-content-blocker', 'languages', 'gdpr-content-blocker-en_US.mo')
write_mo(entries, mo_path)
print('OK: %d strings -> %s + .mo' % (len(de), os.path.basename(po_path)))