def create_pdf(lang_code, checking_level, version): global download_dir, unfoldingWord_dir # Create PDF via ConTeXt try: print_ok('BEGINNING: ', 'PDF generation.') out_dir = os.path.join(download_dir, 'make_pdf') make_dir(out_dir) # generate a tex file print('Generating tex file...', end=' ') tex_file = os.path.join(out_dir, '{0}.tex'.format(lang_code)) # make sure it doesn't already exist if os.path.isfile(tex_file): os.remove(tex_file) with OBSTexExport(lang_code, tex_file, 0, '360px', checking_level) as tex: tex.run() print('finished.') # run context print_notice('Running context - this may take several minutes.') # noinspection PyTypeChecker trackers = ','.join([ 'afm.loading', 'fonts.missing', 'fonts.warnings', 'fonts.names', 'fonts.specifications', 'fonts.scaling', 'system.dump' ]) cmd = 'context --paranoid --batchmode --trackers={0} "{1}"'.format( trackers, tex_file) try: subprocess.check_call(cmd, shell=True, stderr=subprocess.STDOUT, cwd=out_dir) except subprocess.CalledProcessError as e: if e.message: raise e print('Finished running context.') print('Copying PDF to API...', end=' ') version = version.replace('.', '_') if version[0:1] != 'v': version = 'v' + version pdf_file = os.path.join(unfoldingWord_dir, lang_code, 'obs-{0}-{1}.pdf'.format(lang_code, version)) shutil.copyfile(os.path.join(out_dir, '{0}.pdf'.format(lang_code)), pdf_file) print('finished.') finally: print_ok('FINISHED:', 'generating PDF.')
def create_pdf(lang_code, checking_level, version): global download_dir, unfoldingWord_dir # Create PDF via ConTeXt try: print_ok('BEGINNING: ', 'PDF generation.') out_dir = os.path.join(download_dir, 'make_pdf') make_dir(out_dir) # generate a tex file print('Generating tex file...', end=' ') tex_file = os.path.join(out_dir, '{0}.tex'.format(lang_code)) # make sure it doesn't already exist if os.path.isfile(tex_file): os.remove(tex_file) with OBSTexExport(lang_code, tex_file, 0, '360px', checking_level) as tex: tex.run() print('finished.') # run context print_notice('Running context - this may take several minutes.') # noinspection PyTypeChecker trackers = ','.join(['afm.loading', 'fonts.missing', 'fonts.warnings', 'fonts.names', 'fonts.specifications', 'fonts.scaling', 'system.dump']) cmd = 'context --paranoid --batchmode --trackers={0} "{1}"'.format(trackers, tex_file) try: subprocess.check_call(cmd, shell=True, stderr=subprocess.STDOUT, cwd=out_dir) except subprocess.CalledProcessError as e: if e.message: raise e print('Finished running context.') print('Copying PDF to API...', end=' ') version = version.replace('.', '_') if version[0:1] != 'v': version = 'v' + version pdf_file = os.path.join(unfoldingWord_dir, lang_code, 'obs-{0}-{1}.pdf'.format(lang_code, version)) shutil.copyfile(os.path.join(out_dir, '{0}.pdf'.format(lang_code)), pdf_file) print('finished.') finally: print_ok('FINISHED:', 'generating PDF.')
dest='gitrepo', default=False, required=True, help='Git repository where the source can be found.') parser.add_argument( '-t', '--tag', dest='tag', default='master', required=False, help='Branch or tag to use as the source. Default is master.') args = parser.parse_args(sys.argv[1:]) # prompt user to update meta.json print_notice( 'Check meta.yaml in the git repository and update the information if needed.' ) prompt('Press Enter to continue when ready...') try: print_ok('STARTING: ', 'publishing TA repository.') main(args.gitrepo, args.tag) print_ok('ALL FINISHED: ', 'publishing TA repository.') print_notice('Don\'t forget to notify the interested parties.') finally: # delete temp files if os.path.isdir(download_dir): shutil.rmtree(download_dir, ignore_errors=True)
print('finished.') return chapters if __name__ == '__main__': print() parser = argparse.ArgumentParser(description=__doc__, formatter_class=argparse.RawDescriptionHelpFormatter) parser.add_argument('-r', '--gitrepo', dest='gitrepo', default=False, required=True, help='Git repository where the source can be found.') parser.add_argument('-t', '--tag', dest='tag', default='master', required=False, help='Branch or tag to use as the source. Default is master.') parser.add_argument('-p', '--nopdf', dest='nopdf', action='store_true', help='Do not produce a PDF.') args = parser.parse_args(sys.argv[1:]) # prompt user to update status.json print_notice('Check status.json in the git repository and update the information if needed.') prompt('Press Enter to continue when ready...') try: print_ok('STARTING: ', 'publishing OBS repository.') main(args.gitrepo, args.tag, args.nopdf) print_ok('ALL FINISHED: ', 'publishing OBS repository.') print_notice('Don\'t forget to notify the interested parties.') finally: # delete temp files if os.path.isdir(download_dir): shutil.rmtree(download_dir, ignore_errors=True)
'--gitrepo', dest='gitrepo', default=False, required=True, help='Git repository where the source can be found.') parser.add_argument( '-t', '--tag', dest='tag', default='master', required=False, help='Branch or tag to use as the source. Default is master.') parser.add_argument('-p', '--nopdf', dest='nopdf', action='store_true', help='Do not produce a PDF.') args = parser.parse_args(sys.argv[1:]) try: print_ok('STARTING: ', 'publishing OBS repository.') main(args.gitrepo, args.tag, args.nopdf) print_ok('ALL FINISHED: ', 'publishing OBS repository.') print_notice('Don\'t forget to notify the interested parties.') finally: # delete temp files if os.path.isdir(download_dir): shutil.rmtree(download_dir, ignore_errors=True)
def main(git_repo, tag, domain): global download_dir, out_template # clean up the git repo url if git_repo[-4:] == '.git': git_repo = git_repo[:-4] if git_repo[-1:] == '/': git_repo = git_repo[:-1] # initialize some variables today = ''.join(str(datetime.date.today()).rsplit('-')[0:3]) download_dir = '/tmp/{0}'.format(git_repo.rpartition('/')[2]) make_dir(download_dir) downloaded_file = '{0}/{1}.zip'.format(download_dir, git_repo.rpartition('/')[2]) file_to_download = join_url_parts(git_repo, 'archive/' + tag + '.zip') books_published = {} metadata_obj = None usfm_dir = None # download the repository try: print('Downloading {0}...'.format(file_to_download), end=' ') if not os.path.isfile(downloaded_file): download_file(file_to_download, downloaded_file) finally: print('finished.') try: print('Unzipping...'.format(downloaded_file), end=' ') unzip(downloaded_file, download_dir) finally: print('finished.') # examine the repository for root, dirs, files in os.walk(download_dir): if 'meta.json' in files: # read the metadata try: print('Reading the metadata...', end=' ') metadata_obj = BibleMetaData(os.path.join(root, 'meta.json')) finally: print('finished.') if 'usfm' in dirs: usfm_dir = os.path.join(root, 'usfm') # if we have everything, exit the loop if usfm_dir and metadata_obj: break # check for valid repository structure if not metadata_obj: print_error('Did not find meta.json in {}'.format(git_repo)) sys.exit(1) if not usfm_dir: print_error('Did not find the usfm directory in {}'.format(git_repo)) sys.exit(1) # get the versification data vrs = Bible.get_versification(metadata_obj.versification) # type: list<Book> out_dir = out_template.format(domain, metadata_obj.slug, metadata_obj.lang) # walk through the usfm files usfm_files = glob(os.path.join(usfm_dir, '*.usfm')) errors_found = False for usfm_file in usfm_files: # read the file with codecs.open(usfm_file, 'r', 'utf-8') as in_file: book_text = in_file.read() # get the book id book_search = id_re.search(book_text) if not book_search: print_error('Book id not found in {}'.format(usfm_file)) sys.exit(1) book_id = book_search.group(1) print('Beginning {}...'.format(book_id), end=' ') # get book versification info book = next((b for b in vrs if b.book_id == book_id), None) if not book: print_error('Book versification data was not found for "{}"'.format(book_id)) sys.exit(1) # remove \s5 lines book_text = s5_re.sub('', book_text) # get the usfm for the book book.set_usfm(book_text) # do basic checks book.verify_usfm_tags() book.verify_chapters_and_verses(True) if book.validation_errors: errors_found = True # get chunks for this book Bible.chunk_book(metadata_obj.versification, book) book.apply_chunks() # produces something like '01-GEN.usfm' book_file_name = '{0}-{1}.usfm'.format(str(book.number).zfill(2), book.book_id) print('Writing ' + book_file_name + '...', end=' ') write_file('{0}/{1}'.format(out_dir, book_file_name), book.usfm) meta = ['Bible: OT'] if book.number > 39: meta = ['Bible: NT'] books_published[book.book_id.lower()] = {'name': book.name, 'meta': meta, 'sort': str(book.number).zfill(2), 'desc': '' } print('finished.') # stop if errors were found if errors_found: print_error('These USFM errors must be corrected before publishing can continue.') sys.exit(1) print('Writing status.json...', end=' ') status = {"slug": '{0}'.format(metadata_obj.slug.lower()), "name": metadata_obj.name, "lang": metadata_obj.lang, "date_modified": today, "books_published": books_published, "status": {"checking_entity": metadata_obj.checking_entity, "checking_level": metadata_obj.checking_level, "comments": metadata_obj.comments, "contributors": metadata_obj.contributors, "publish_date": today, "source_text": metadata_obj.source_text, "source_text_version": metadata_obj.source_text_version, "version": metadata_obj.version } } write_file('{0}/status.json'.format(out_dir), status, indent=2) print('finished.') print() print('Publishing to the API...') with api_publish(out_dir) as api: api.run() print('Finished publishing to the API.') # update the catalog print() print('Updating the catalogs...', end=' ') update_catalog() print('finished.') print_notice('Check {0} and do a git push'.format(out_dir))
def main(git_repo, tag, domain): global download_dir, out_template # clean up the git repo url if git_repo[-4:] == '.git': git_repo = git_repo[:-4] if git_repo[-1:] == '/': git_repo = git_repo[:-1] # initialize some variables today = ''.join(str(datetime.date.today()).rsplit('-')[0:3]) # str(datetime.date.today()) download_dir = '/tmp/{0}'.format(git_repo.rpartition('/')[2]) make_dir(download_dir) downloaded_file = '{0}/{1}.zip'.format(download_dir, git_repo.rpartition('/')[2]) file_to_download = join_url_parts(git_repo, 'archive/' + tag + '.zip') manifest = None metadata_obj = None content_dir = '' usfm_file = None # download the repository try: print('Downloading {0}...'.format(file_to_download), end=' ') if not os.path.isfile(downloaded_file): download_file(file_to_download, downloaded_file) finally: print('finished.') try: print('Unzipping...'.format(downloaded_file), end=' ') unzip(downloaded_file, download_dir) finally: print('finished.') # examine the repository for root, dirs, files in os.walk(download_dir): if 'manifest.json' in files: # read the manifest try: print('Reading the manifest...', end=' ') manifest = load_json_object(os.path.join(root, 'manifest.json')) content_dir = root # look for the usfm file for the whole book found_usfm = glob(os.path.join(content_dir, '*.usfm')) if len(found_usfm) == 1: usfm_file = os.path.join(content_dir, found_usfm[0]) finally: print('finished.') if 'meta.json' in files: # read the metadata try: print('Reading the metadata...', end=' ') metadata_obj = BibleMetaData(os.path.join(root, 'meta.json')) finally: print('finished.') # if we have everything, exit the loop if manifest and metadata_obj: break # check for valid repository structure if not manifest: print_error('Did not find manifest.json in {}'.format(git_repo)) sys.exit(1) if not metadata_obj: print_error('Did not find meta.json in {}'.format(git_repo)) sys.exit(1) # get the versification data print('Getting versification info...', end=' ') vrs = Bible.get_versification(metadata_obj.versification) # type: list<Book> # get the book object for this repository book = next((b for b in vrs if b.book_id.lower() == manifest['project']['id']), None) # type: Book if not book: print_error('Book versification data was not found for "{}"'.format(manifest['project']['id'])) sys.exit(1) print('finished') if usfm_file: read_unified_file(book, usfm_file) else: read_chunked_files(book, content_dir, metadata_obj) # do basic checks print('Running USFM checks...', end=' ') book.verify_chapters_and_verses(True) if book.validation_errors: print_error('These USFM errors must be corrected before publishing can continue.') sys.exit(1) else: print('finished.') # insert paragraph markers print('Inserting paragraph markers...', end=' ') Bible.insert_paragraph_markers(book) print('finished.') # get chunks for this book print('Chunking the text...', end=' ') Bible.chunk_book(metadata_obj.versification, book) book.apply_chunks() print('finished.') # save the output out_dir = out_template.format(domain, metadata_obj.slug) # produces something like '01-GEN.usfm' book_file_name = '{0}-{1}.usfm'.format(str(book.number).zfill(2), book.book_id) print('Writing ' + book_file_name + '...', end=' ') write_file('{0}/{1}'.format(out_dir, book_file_name), book.usfm) print('finished.') # look for an existing status.json file print('Updating the status for {0}...'.format(metadata_obj.lang), end=' ') status_file = '{0}/status.json'.format(out_dir) if os.path.isfile(status_file): status = BibleStatus(status_file) else: status = BibleStatus() status.update_from_meta_data(metadata_obj) # add this book to the list of "books_published" status.add_book_published(book) # update the "date_modified" status.date_modified = today print('finished.') # save the status.json file print('Writing status.json...', end=' ') status_json = json.dumps(status, sort_keys=True, indent=2, cls=BibleEncoder) write_file(status_file, status_json) print('finished') # let the API know it is there print('Publishing to the API...') with api_publish(out_dir) as api: api.run() print('Finished publishing to the API.') # update the catalog print() print('Updating the catalogs...', end=' ') update_catalog() print('finished.') print_notice('Check {0} and do a git push'.format(out_dir))
obs_chapter.frames.sort(key=lambda f: f['id']) # add this chapter to the OBS object chapters.append(obs_chapter) print('finished.') return chapters if __name__ == '__main__': print('') parser = argparse.ArgumentParser(description=__doc__, formatter_class=argparse.RawDescriptionHelpFormatter) parser.add_argument('-r', '--gitrepo', dest='gitrepo', default=False, required=True, help='Git repository where the source can be found.') parser.add_argument('-t', '--tag', dest='tag', default='master', required=False, help='Branch or tag to use as the source. Default is master.') parser.add_argument('-p', '--nopdf', dest='nopdf', action='store_true', help='Do not produce a PDF.') args = parser.parse_args(sys.argv[1:]) try: print_ok('STARTING: ', 'publishing OBS repository.') main(args.gitrepo, args.tag, args.nopdf) print_ok('ALL FINISHED: ', 'publishing OBS repository.') print_notice('Don\'t forget to notify the interested parties.') finally: # delete temp files if os.path.isdir(download_dir): shutil.rmtree(download_dir, ignore_errors=True)
this_lang = next(l for l in langs if l['lc'] == args.lang) if not this_lang: print_error('Information for language "{0}" was not found.'.format( args.lang)) sys.exit(1) if 'github' not in args.gitrepo and 'git.door43.org' not in args.gitrepo: print_warning( 'Currently only github and git.door43.org repositories are supported.' ) sys.exit(0) # prompt user to update status.txt print_notice( 'Check status-{0}.json in the git repository and update the information if needed.' .format(args.lang)) prompt('Press Enter to continue when ready...') print() # do the import import_obs(this_lang, args.gitrepo, args.url, args.nopdf) # update the catalog print_ok('STARTING: ', 'updating the catalogs.') update_catalog() print_ok('FINISHED: ', 'updating the catalogs.') # update OBS in-progress print_ok('STARTING: ', 'updating OBS in progress.') ObsInProgress.run()
def import_obs(lang_data, git_repo, door43_url, no_pdf): global download_dir, root, api_dir lang_code = lang_data['lc'] # pre-flight checklist link_source = '/var/www/vhosts/api.unfoldingword.org/httpdocs/obs/jpg/1/en' if not os.path.isdir(link_source): print_error( 'Image source directory not found: {0}.'.format(link_source)) sys.exit(1) if git_repo[-1:] != '/': git_repo += '/' if no_pdf: tools_dir = None else: tools_dir = '/var/www/vhosts/door43.org/tools' if not os.path.isdir(tools_dir): tools_dir = os.path.expanduser('~/Projects/tools') # prompt if tools not found if not os.path.isdir(tools_dir): tools_dir = None print_notice( 'The tools directory was not found. The PDF cannot be generated.' ) resp = prompt( 'Do you want to continue without generating a PDF? [Y|n]: ') if resp != '' and resp != 'Y' and resp != 'y': sys.exit(0) if git_repo: if git_repo[-1:] == '/': git_repo = git_repo[:-1] download_dir = '/tmp/{0}'.format(git_repo.rpartition('/')[2]) make_dir(download_dir) # make sure OBS is initialized on Dokuwiki test_dir = root.format(lang_code, '') if not os.path.isdir(test_dir): print_warning( 'It seems OBS has not been initialized on Door43.org for {0}'. format(lang_code)) sys.exit(1) elif door43_url: print_error('URL not yet implemented.') return else: print_error('Source not provided.') return # get the source files from the git repository if 'github' in git_repo: # https://github.com/unfoldingWord/obs-ru # https://raw.githubusercontent.com/unfoldingWord/obs-ru/master/obs-ru.json raw_url = git_repo.replace('github.com', 'raw.githubusercontent.com') elif 'git.door43.org' in git_repo: raw_url = join_url_parts(git_repo, 'raw') else: # this is to keep IntelliJ happy, is should have been caught in sub main return # download needed files from the repository file_suffix = '-{0}.json'.format(lang_code.lower()) files_to_download = [ join_url_parts(raw_url, 'master/obs' + file_suffix), join_url_parts(raw_url, 'master/status' + file_suffix) ] for file_to_download in files_to_download: downloaded_file = os.path.join(download_dir, file_to_download.rpartition('/')[2]) try: print('Downloading {0}...'.format(file_to_download), end=' ') download_file(file_to_download, downloaded_file) finally: print('finished.') # read the files from the git repository file_suffix = '-{0}.json'.format(lang_code.lower()) obs_obj = None status_obj = None # front_matter_found = False # back_matter_found = False try: print('Examining the files...', end=' ') for root_path, dirs, files in os.walk(download_dir): if not len(files): continue for git_file in files: if git_file == 'obs' + file_suffix: obs_obj = OBS(os.path.join(root_path, git_file)) elif git_file == 'status' + file_suffix: status_obj = OBSStatus(os.path.join(root_path, git_file)) # elif 'front-matter' in git_file: # front_matter_found = True # elif 'back-matter' in git_file: # back_matter_found = True finally: print('finished.') # check data integrity if not obs_obj.verify_all(): sys.exit(1) if not status_obj: print_error( 'The file "status{0}" was not found in the git repository.'.format( file_suffix)) sys.exit(1) # create Dokuwiki pages print_ok('Begin: ', 'creating Dokuwiki pages.') for chapter in obs_obj.chapters: chapter_title = '====== {0} ======'.format(chapter['title']) chapter_ref = '//{0}//'.format(chapter['ref']) chapter_body = '' chapter_num = chapter['number'].zfill(2) for frame in chapter['frames']: chapter_body += '{{{{{0}?direct&}}}}\n\n{1}\n\n'.format( frame['img'], frame['text']) file_name = root.format(lang_code, chapter_num + '.txt') print(' Writing {0}'.format(file_name)) with codecs.open(file_name, 'w', 'utf-8-sig') as out_file: out_file.write('{0}\n\n{1}{2}\n'.format(chapter_title, chapter_body, chapter_ref)) print_ok('Finished: ', 'creating Dokuwiki pages.') # Create image symlinks on api.unfoldingword.org try: print('Creating symlink to images directory...', end=' ') link_name = '/var/www/vhosts/api.unfoldingword.org/httpdocs/obs/jpg/1/{0}'.format( lang_code.lower()) if not os.path.isfile(link_name) and not os.path.isdir( link_name) and not os.path.islink(link_name): os.symlink(link_source, link_name) finally: print('finished.') # Create PDF via ConTeXt if not no_pdf and tools_dir and os.path.isdir(tools_dir): try: print_ok('Beginning: ', 'PDF generation.') script_file = os.path.join(tools_dir, 'obs/book/pdf_export.sh') out_dir = api_dir.format(lang_code) make_dir(out_dir) process = subprocess.Popen([ script_file, '-l', lang_code, '-c', status_obj.checking_level, '-v', status_obj.version, '-o', out_dir ], shell=True, stdout=subprocess.PIPE, stderr=subprocess.PIPE) # wait for the process to terminate out, err = process.communicate() exit_code = process.returncode out = out.strip().decode('utf-8') err = err.strip().decode('utf-8') # the error message may be in stdout if exit_code != 0: if not err: err = out out = None if err: print_error(err, 2) if out: print(' ' + out) print(' PDF subprocess finished with exit code {0}'.format( exit_code)) finally: print_ok('Finished:', 'generating PDF.')
link_source = '/var/www/vhosts/api.unfoldingword.org/httpdocs/obs/jpg/1/en' if not os.path.isdir(link_source): print_error('Image source directory not found: {0}.'.format(link_source)) sys.exit(1) if no_pdf: tools_dir = None else: tools_dir = '/var/www/vhosts/door43.org/tools' if not os.path.isdir(tools_dir): tools_dir = os.path.expanduser('~/Projects/tools') # prompt if tools not found if not os.path.isdir(tools_dir): tools_dir = None print_notice('The tools directory was not found. The PDF cannot be generated.') resp = prompt('Do you want to continue without generating a PDF? [Y|n]: ') if resp != '' and resp != 'Y' and resp != 'y': sys.exit(0) today = ''.join(str(datetime.date.today()).rsplit('-')[0:3]) print('Loading languages...', end=' ') lang_dict = OBS.load_lang_strings() print('finished.') print('Loading the catalog...', end=' ') uw_cat_path = os.path.join(unfoldingWord_dir, 'obs-catalog.json') uw_catalog = load_json_object(uw_cat_path, []) uw_cat_langs = [x['language'] for x in uw_catalog] cat_path = os.path.join(export_dir, 'obs-catalog.json')
def rechunk_this_one(api_directory): global id_re, s5_re print_notice('Processing {}'.format(api_directory)) # read the status.json file with codecs.open(os.path.join(api_directory, 'status.json'), 'r', 'utf-8-sig') as in_file: status = json.loads(in_file.read()) # determine versification if status['lang'] == 'ru': versification = 'rsc' elif status['lang'] == 'hi' or status['lang'] == 'sr-Latn' or status[ 'lang'] == 'hu' or status['lang'] == 'ta': versification = 'ufw-odx' elif status['lang'] == 'bn': versification = 'ufw-bn' elif status['lang'] == 'ar': versification = 'avd' elif status['lang'] == 'kn': versification = 'ufw-rev' else: versification = 'ufw' versification_data = Bible.get_versification( versification) # type: list<Book> # remove all .sig files for f in os.listdir(api_directory): if f.endswith('.sig'): os.remove(os.path.join(api_directory, f)) # rechunk files in this directory usfm_files = glob(os.path.join(api_directory, '*.usfm')) errors_found = False for usfm_file in usfm_files: if usfm_file.endswith('LICENSE.usfm'): continue # read the file with codecs.open(usfm_file, 'r', 'utf-8') as in_file: book_text = in_file.read() # get the book id book_search = id_re.search(book_text) if not book_search: print_error('Book id not found in {}'.format(usfm_file)) sys.exit(1) book_id = book_search.group(1) print('Beginning {}...'.format(book_id), end=' ') # get book versification info book = next((b for b in versification_data if b.book_id == book_id), None) if not book: print_error( 'Book versification data was not found for "{}"'.format( book_id)) sys.exit(1) # remove \s5 lines book_text = s5_re.sub('', book_text) # get the usfm for the book book.set_usfm(book_text) # do basic checks book.verify_chapters_and_verses(True) if book.validation_errors: errors_found = True # get chunks for this book Bible.chunk_book(versification, book) book.apply_chunks() # produces something like '01-GEN.usfm' book_file_name = '{0}-{1}.usfm'.format( str(book.number).zfill(2), book.book_id) print('Writing ' + book_file_name + '...', end=' ') write_file(usfm_file, book.usfm) print('finished.') if errors_found: print_error( 'These USFM errors must be corrected before publishing can continue.' ) sys.exit(1) # rebuild source for tS print() print('Publishing to the API...') with api_publish(api_directory) as api: api.run() print('Finished publishing to the API.') # update the catalog print() print('Updating the catalogs...', end=' ') update_catalog() print('finished.')
print_error( 'Image source directory not found: {0}.'.format(link_source)) sys.exit(1) if no_pdf: tools_dir = None else: tools_dir = '/var/www/vhosts/door43.org/tools' if not os.path.isdir(tools_dir): tools_dir = os.path.expanduser('~/Projects/tools') # prompt if tools not found if not os.path.isdir(tools_dir): tools_dir = None print_notice( 'The tools directory was not found. The PDF cannot be generated.' ) resp = prompt( 'Do you want to continue without generating a PDF? [Y|n]: ') if resp != '' and resp != 'Y' and resp != 'y': sys.exit(0) today = ''.join(str(datetime.date.today()).rsplit('-')[0:3]) print('Loading languages...', end=' ') lang_dict = OBS.load_lang_strings() print('finished.') print('Loading the catalog...', end=' ') uw_cat_path = os.path.join(unfoldingWord_dir, 'obs-catalog.json') uw_catalog = load_json_object(uw_cat_path, [])
def rechunk_this_one(api_directory): global id_re, s5_re print_notice('Processing {}'.format(api_directory)) # read the status.json file with codecs.open(os.path.join(api_directory, 'status.json'), 'r', 'utf-8-sig') as in_file: status = json.loads(in_file.read()) # determine versification if status['lang'] == 'ru': versification = 'rsc' elif status['lang'] == 'hi' or status['lang'] == 'sr-Latn' or status['lang'] == 'hu' or status['lang'] == 'ta': versification = 'ufw-odx' elif status['lang'] == 'bn': versification = 'ufw-bn' elif status['lang'] == 'ar': versification = 'avd' elif status['lang'] == 'kn': versification = 'ufw-rev' else: versification = 'ufw' versification_data = Bible.get_versification(versification) # type: list<Book> # remove all .sig files for f in os.listdir(api_directory): if f.endswith('.sig'): os.remove(os.path.join(api_directory, f)) # rechunk files in this directory usfm_files = glob(os.path.join(api_directory, '*.usfm')) errors_found = False for usfm_file in usfm_files: if usfm_file.endswith('LICENSE.usfm'): continue # read the file with codecs.open(usfm_file, 'r', 'utf-8') as in_file: book_text = in_file.read() # get the book id book_search = id_re.search(book_text) if not book_search: print_error('Book id not found in {}'.format(usfm_file)) sys.exit(1) book_id = book_search.group(1) print('Beginning {}...'.format(book_id), end=' ') # get book versification info book = next((b for b in versification_data if b.book_id == book_id), None) if not book: print_error('Book versification data was not found for "{}"'.format(book_id)) sys.exit(1) # remove \s5 lines book_text = s5_re.sub('', book_text) # get the usfm for the book book.set_usfm(book_text) # do basic checks book.verify_chapters_and_verses(True) if book.validation_errors: errors_found = True # get chunks for this book Bible.chunk_book(versification, book) book.apply_chunks() # produces something like '01-GEN.usfm' book_file_name = '{0}-{1}.usfm'.format(str(book.number).zfill(2), book.book_id) print('Writing ' + book_file_name + '...', end=' ') write_file(usfm_file, book.usfm) print('finished.') if errors_found: print_error('These USFM errors must be corrected before publishing can continue.') sys.exit(1) # rebuild source for tS print() print('Publishing to the API...') with api_publish(api_directory) as api: api.run() print('Finished publishing to the API.') # update the catalog print() print('Updating the catalogs...', end=' ') update_catalog() print('finished.')
def import_obs(lang_data, git_repo, door43_url, no_pdf): global download_dir, root, api_dir lang_code = lang_data['lc'] # pre-flight checklist link_source = '/var/www/vhosts/api.unfoldingword.org/httpdocs/obs/jpg/1/en' if not os.path.isdir(link_source): print_error('Image source directory not found: {0}.'.format(link_source)) sys.exit(1) if git_repo[-1:] != '/': git_repo += '/' if no_pdf: tools_dir = None else: tools_dir = '/var/www/vhosts/door43.org/tools' if not os.path.isdir(tools_dir): tools_dir = os.path.expanduser('~/Projects/tools') # prompt if tools not found if not os.path.isdir(tools_dir): tools_dir = None print_notice('The tools directory was not found. The PDF cannot be generated.') resp = prompt('Do you want to continue without generating a PDF? [Y|n]: ') if resp != '' and resp != 'Y' and resp != 'y': sys.exit(0) if git_repo: if git_repo[-1:] == '/': git_repo = git_repo[:-1] download_dir = '/tmp/{0}'.format(git_repo.rpartition('/')[2]) make_dir(download_dir) # make sure OBS is initialized on Dokuwiki test_dir = root.format(lang_code, '') if not os.path.isdir(test_dir): print_warning('It seems OBS has not been initialized on Door43.org for {0}'.format(lang_code)) sys.exit(1) elif door43_url: print_error('URL not yet implemented.') return else: print_error('Source not provided.') return # get the source files from the git repository if 'github' in git_repo: # https://github.com/unfoldingWord/obs-ru # https://raw.githubusercontent.com/unfoldingWord/obs-ru/master/obs-ru.json raw_url = git_repo.replace('github.com', 'raw.githubusercontent.com') elif 'git.door43.org' in git_repo: raw_url = join_url_parts(git_repo, 'raw') else: # this is to keep IntelliJ happy, is should have been caught in sub main return # download needed files from the repository file_suffix = '-{0}.json'.format(lang_code.lower()) files_to_download = [ join_url_parts(raw_url, 'master/obs' + file_suffix), join_url_parts(raw_url, 'master/status' + file_suffix) ] for file_to_download in files_to_download: downloaded_file = os.path.join(download_dir, file_to_download.rpartition('/')[2]) try: print('Downloading {0}...'.format(file_to_download), end=' ') download_file(file_to_download, downloaded_file) finally: print('finished.') # read the files from the git repository file_suffix = '-{0}.json'.format(lang_code.lower()) obs_obj = None status_obj = None # front_matter_found = False # back_matter_found = False try: print('Examining the files...', end=' ') for root_path, dirs, files in os.walk(download_dir): if not len(files): continue for git_file in files: if git_file == 'obs' + file_suffix: obs_obj = OBS(os.path.join(root_path, git_file)) elif git_file == 'status' + file_suffix: status_obj = OBSStatus(os.path.join(root_path, git_file)) # elif 'front-matter' in git_file: # front_matter_found = True # elif 'back-matter' in git_file: # back_matter_found = True finally: print('finished.') # check data integrity if not obs_obj.verify_all(): sys.exit(1) if not status_obj: print_error('The file "status{0}" was not found in the git repository.'.format(file_suffix)) sys.exit(1) # create Dokuwiki pages print_ok('Begin: ', 'creating Dokuwiki pages.') for chapter in obs_obj.chapters: chapter_title = '====== {0} ======'.format(chapter['title']) chapter_ref = '//{0}//'.format(chapter['ref']) chapter_body = '' chapter_num = chapter['number'].zfill(2) for frame in chapter['frames']: chapter_body += '{{{{{0}?direct&}}}}\n\n{1}\n\n'.format(frame['img'], frame['text']) file_name = root.format(lang_code, chapter_num + '.txt') print(' Writing {0}'.format(file_name)) with codecs.open(file_name, 'w', 'utf-8-sig') as out_file: out_file.write('{0}\n\n{1}{2}\n'.format(chapter_title, chapter_body, chapter_ref)) print_ok('Finished: ', 'creating Dokuwiki pages.') # Create image symlinks on api.unfoldingword.org try: print('Creating symlink to images directory...', end=' ') link_name = '/var/www/vhosts/api.unfoldingword.org/httpdocs/obs/jpg/1/{0}'.format(lang_code.lower()) if not os.path.isfile(link_name) and not os.path.isdir(link_name) and not os.path.islink(link_name): os.symlink(link_source, link_name) finally: print('finished.') # Create PDF via ConTeXt if not no_pdf and tools_dir and os.path.isdir(tools_dir): try: print_ok('Beginning: ', 'PDF generation.') script_file = os.path.join(tools_dir, 'obs/book/pdf_export.sh') out_dir = api_dir.format(lang_code) make_dir(out_dir) process = subprocess.Popen([script_file, '-l', lang_code, '-c', status_obj.checking_level, '-v', status_obj.version, '-o', out_dir], shell=True, stdout=subprocess.PIPE, stderr=subprocess.PIPE) # wait for the process to terminate out, err = process.communicate() exit_code = process.returncode out = out.strip().decode('utf-8') err = err.strip().decode('utf-8') # the error message may be in stdout if exit_code != 0: if not err: err = out out = None if err: print_error(err, 2) if out: print(' ' + out) print(' PDF subprocess finished with exit code {0}'.format(exit_code)) finally: print_ok('Finished:', 'generating PDF.')