#!/usr/bin/env python3
"""Translation check (run by build.sh):
 1. every L.t("...") / L.f("...") key in the code has a Galician entry in L.java,
 2. English and Galician use the same placeholders in the same order,
 3. no English text reaches the screen without going through L (setText/setTitle/... literals),
 4. no unused entries.
usage: check_i18n.py <project dir>"""
import os
import re
import sys

root = sys.argv[1] if len(sys.argv) > 1 else "."
LIT = r'"((?:[^"\\]|\\.)*)"'


def unescape(s):
    out, i = [], 0
    while i < len(s):
        c = s[i]
        if c == "\\" and i + 1 < len(s):
            n = s[i + 1]
            if n == "u":
                out.append(chr(int(s[i + 2:i + 6], 16)))
                i += 6
                continue
            out.append({"n": "\n", "t": "\t", "r": "\r", '"': '"', "'": "'", "\\": "\\"}.get(n, n))
            i += 2
            continue
        out.append(c)
        i += 1
    return "".join(out)


def placeholders(s):
    return re.findall(r"%[sdn%]", s)


lsrc = open(os.path.join(root, "src/org/gingercal/L.java"), encoding="utf-8").read()
table = {}
for m in re.finditer(r"put\(\s*" + LIT + r"\s*,\s*" + LIT + r"\s*\);", lsrc, re.S):
    en, gal = unescape(m.group(1)), unescape(m.group(2))
    if en in table:
        print("DUPLICATE key:", en[:60])
    table[en] = gal

problems, used = [], set()
UI_CALL = re.compile(r"\b(setText|setTitle|setMessage|setHint|setPositiveButton|setNegativeButton|setNeutralButton|"
                     r"toast\([^,]+,|ProgressDialog\.show\([^,]+,[^,]+,|Rows\.add|rows\.add|Rows\.add|"
                     r"new DavException\([^,]+,|new IOException\(|label\(|field\([^,]+,)\s*\(?\s*" + LIT)
for d in ("src", "android"):
    for dirpath, _, files in os.walk(os.path.join(root, d)):
        for f in files:
            if not f.endswith(".java") or f == "L.java":
                continue
            path = os.path.join(dirpath, f)
            text = open(path, encoding="utf-8").read()
            for m in re.finditer(r"\bL\.[tf]\(\s*" + LIT, text):
                key = unescape(m.group(1))
                used.add(key)
                if key not in table:
                    problems.append("%s: no Galician for %r" % (f, key[:70]))
                elif placeholders(key) != placeholders(table[key]):
                    problems.append("%s: placeholders differ for %r" % (f, key[:70]))
            for line_no, line in enumerate(text.splitlines(), 1):
                for m in UI_CALL.finditer(line):
                    s = unescape(m.group(2))
                    if re.search(r"[A-Za-z]{2,}", s) and s not in ("Galego", "English", "Lingua / Language"):
                        problems.append("%s:%d: untranslated UI text %r" % (f, line_no, s[:60]))
                for m in re.finditer(r"\.add\(\s*" + LIT, line):
                    s = unescape(m.group(1))
                    if " " in s and re.search(r"[A-Za-z]{3,} [A-Za-z]{3,}", s):
                        problems.append("%s:%d: untranslated list text %r" % (f, line_no, s[:60]))

for key in sorted(set(table) - used):
    problems.append("unused entry in L.java: %r" % key[:70])
for p in problems:
    print("I18N:", p)
print("translations: %d entries, %d keys used, %s" % (len(table), len(used), "OK" if not problems else "PROBLEMS"))
sys.exit(1 if problems else 0)