Small tool to extract labels and merge json files for robisim74/angular-l10n

extract_translation_labels.py

import re
import json
import glob
import os
import pathlib

# translation.translate('ABC')
PATTERN_TS_1 = re.compile(r"""translation.translate\(['"]([^'"]+)['"]\)""")

# {{ 'ABC' | translate:lang }}
PATTERN_TEMPLATE_1 = re.compile(r"""\{\{\s*['"]([^'"]+)['"]\s*\|\s*translate:lang\s*\}\}""")

NEEDTRANSLATIONSLABEL = "TRANSLATE:{}"

languages = set()

def loadTranslations():
    fnames = glob.glob('./src/assets/locale-??.json')
    # print("fnames: {}".format(fnames))
    
    translations = {}

    for fname in fnames:
        lang = re.findall(r"locale-(\w+).json", fname)[0]
        languages.add(lang)
        print("Loading language {} from {}".format(lang, fname))
        with open(fname) as f:
            jsonContent = json.loads(f.read())
        
        for label, translated in jsonContent.items():
            # print("label {} translated as {}".format(label, translated))
            t = translations.get(label, {})
            t[lang] = translated
            translations[label] = t
    
    return translations


def saveTranslations(translations):
    for lang in languages:
        fname = "./src/assets/locale-{}.json".format(lang)

        outdict = { label: tvalues.get(lang, NEEDTRANSLATIONSLABEL.format(label)) for label, tvalues in translations.items() }

        with open(fname, 'w') as f:
            f.write(json.dumps(outdict, indent=2, sort_keys=True))


def mergeTranslations(translations, extractedLabels):
    for label in extractedLabels:
        if label not in translations:
            translations[label] = { lang: NEEDTRANSLATIONSLABEL.format(label) for lang in languages }
    
    return translations


def scanTSFile(fname):
    with open(fname) as f:
        contents = f.read()
    
    extractedLabels = {}
    results = PATTERN_TS_1.findall(contents)
    if results:
        for label in results:
            print("Found label {} in file {}".format(repr(label), fname))
            extractedLabels[label] = NEEDTRANSLATIONSLABEL.format(label)
    
    return extractedLabels


def scanTemplateFile(fname):
    with open(fname) as f:
        contents = f.read()
    
    extractedLabels = {}
    results = PATTERN_TEMPLATE_1.findall(contents)
    if results:
        for label in results:
            print("Found label {} in file {}".format(repr(label), fname))
            extractedLabels[label] = NEEDTRANSLATIONSLABEL.format(label)
    
    return extractedLabels


def scanPath(srcPath, translations):
    path_iter = list(pathlib.Path('src/').glob('**/*.ts')) + \
                list(pathlib.Path('src/').glob('**/*.html'))

    for fname in path_iter:
        fext = os.path.splitext(fname)[1]
        # print(fext)
        extractedLabels = {}
        if fext == '.ts':
            extractedLabels = scanTSFile(fname)
        elif fext == '.html':
            extractedLabels = scanTemplateFile(fname)

        translations = mergeTranslations(translations, extractedLabels)
    
    return translations


if __name__ == "__main__":
    print("Extract and Merge translations")
    
    print("Loading translations")
    translations = loadTranslations()
    print("Scanning...")
    translations = scanPath('src/', translations)
    print("Saving translations")
    saveTranslations(translations)