def validated_string_language(self, title, restrict=None):
try:
restrict_codes = {iso639.to_iso639_2(l) for l in restrict}
detected = detect(title)
threealpha = iso639.to_iso639_2(detected)
ok = True if restrict is None else (threealpha in restrict_codes)
if ok:
language = self.language_object_from_code(threealpha)
if language is not None:
# If there was a restricted set of languages passed to this function,
# we have confidence that we've matched the language of the title
# because it is one of the restricted languages.
# (If no restriction was supplied, then we use whatever language
# was detected in the title.)
return language
else:
# The detected language of the title was not a member of the restricted
# set, so we lack confidence to proceed.
return None
except iso639.NonExistentLanguageError as e:
warnings.warn('*** Unrecognized language code detected: %s' % (e,))
except KeyError as e:
warnings.warn(
'*** LANGUAGE: detected but unrecognized title language %r: %s' % (e.args[0], title)
)
except Exception as e:
print('*** detect_title_language error: %r' % (e,))
return None
def test_language_coverage():
"""Check if WikiPron covers languages with a sufficient amount of data.
If any warnings are raised, they should be suppressed by expanding
the LANGUAGE_CODES dict to handle the relevant languages.
"""
categories = _get_language_categories()
sizes = _get_language_sizes(categories)
for language, size in sizes.items():
if size < _MIN_LANGUAGE_SIZE:
continue
if language in ("Mon", "Translingual"):
# "mon" is the ISO 639 code for Mongolian, but there is also
# the Mon language (ISO 639 code: "mnw").
continue
try:
language_code = iso639.to_iso639_2(language)
except iso639.NonExistentLanguageError:
# Check if WikiPron can handle `language` directly.
language_code = language
try:
language_inferred = wikipron.Config(key=language_code).language
except iso639.NonExistentLanguageError:
warnings.warn(f'WikiPron cannot handle "{language}".')
continue
if language_inferred != language:
warnings.warn(
f'WikiPron resolves the key "{language_code}" to '
f'"{language_inferred}", '
f'which is not "{language}" on Wiktionary.'
)
def to_3_letter_lang(lang):
if len(lang) == 2:
if iso639.is_valid639_1(lang):
return iso639.to_iso639_2(lang)
if len(lang) == 3:
if iso639.is_valid639_2(lang):
return lang
return False
def index_ebooks_from_library(site,
_uuid="",
library="",
start=0,
stop=0,
dir=".",
num=1000,
force_refresh=False):
offset = 0 if not start else start - 1
num = min(1000, num)
server = site.rstrip('/')
api = server + '/ajax/'
lib = library
library = '/' + library if library else library
timeout = 15
print(f"\nIndexing library: {lib} from server: {server} ")
url = api + 'search' + library + '?num=0'
print(f"\nGetting ebooks count of library: {lib} from server:{server} ")
# print(url)
try:
r = requests.get(url, verify=False, timeout=(timeout, 30))
r.raise_for_status()
except requests.RequestException as e:
print("Unable to open site:", url)
return
# pass
except Exception as e:
print("Other issue:", e)
return
# pass
except:
print("Wazza !!!!")
sys.exit(1)
total_num = int(r.json()["total_num"])
total_num = total_num if not stop else stop
print()
print(f"Total count={total_num} from {server}")
# library=r.json()["base_url"].split('/')[-1]
# base_url=r.json()["base_url"]
db = init_site_db(site, _uuid=_uuid, dir=dir)
print()
range = offset + 1
while offset < total_num:
remaining_num = min(num, total_num - offset)
# print()
# print("Downloading ids: offset="+str(offset), "num="+str(remaining_num))
print('\r {:180.180}'.format(
f'Downloading ids: offset={str(offset)} count={str(remaining_num)} from {server}'
),
end='')
# url=server+base_url+'?num='+str(remaining_num)+'&offset='+str(offset)+'&sort=timestamp&sort_order=desc'
url = api + 'search' + library + '?num=' + str(
remaining_num) + '&offset=' + str(
offset) + '&sort=timestamp&sort_order=desc'
# print("->", url)
try:
r = requests.get(url, verify=False, timeout=(timeout, 30))
r.raise_for_status()
except requests.RequestException as e:
print("Connection issue:", e)
return
# pass
except Exception as e:
print("Other issue:", e)
return
# pass
except:
print("Wazza !!!!")
return
# print("Ids received from:"+str(offset), "to:"+str(offset+remaining_num-1))
# print()
# print("Downloading metadata from", str(offset+1), "to", str(offset+remaining_num))
print('\r {:180.180}'.format(
f'Downloading metadata from {str(offset+1)} to {str(offset+remaining_num)}/{total_num} from {server}'
),
end='')
books_s = ",".join(str(i) for i in r.json()['book_ids'])
url = api + 'books' + library + '?ids=' + books_s
# url=server+base_url+'/books?ids='+books_s
# print("->", url)
# print ('\r{:190.190}'.format(f'url= {url} ...'), end='')
try:
r = requests.get(url, verify=False, timeout=(60, 60))
r.raise_for_status()
except requests.RequestException as e:
print("Connection issue:", e)
return
# pass
except Exception as e:
print("Other issue:", e)
return
# pass
except:
print("Wazza !!!!")
return
# print(len(r.json()), "received")
print('\r {:180.180}'.format(f'{len(r.json())} received'), end='')
books = []
for id, r_book in r.json().items():
uuid = r_book['uuid']
if not uuid:
print("No uuid for ebook: ignored")
continue
if r_book['authors']:
desc = f"({r_book['title']} / {r_book['authors'][0]})"
else:
desc = f"({r_book['title']})"
# print (f'\r--> {range}/{total_num} - {desc}', end='')
# print (f'\r{server}--> {range}/{total_num} - {desc}', end='')
print('\r {:180.180} '.format(
f'{range}/{total_num} ({server} : {uuid} --> {desc}'),
end='')
if not force_refresh:
# print("Checking local metadata:", uuid)
try:
book = load_metadata(dir, uuid)
except:
print("Unable to get metadata from:", uuid)
range += 1
continue
if book:
print("Metadata already present for:", uuid)
range += 1
continue
if not r_book['formats']:
# print("No format found for {}".format(r_book['uuid']))
range += 1
continue
book = {}
book['uuid'] = r_book['uuid']
book['id'] = id
book['library'] = lib
book['title'] = r_book['title']
book['authors'] = r_book['authors']
# book['desc']=""
book['desc'] = r_book['comments']
book['series'] = r_book['series']
s_i = r_book['series_index']
if (s_i):
book['series_index'] = int(s_i)
# book['edition']=0
book['identifiers'] = r_book['identifiers']
book['tags'] = r_book['tags']
book['publisher'] = r_book['publisher']
book['pubdate'] = r_book['pubdate']
if not r_book['languages']:
# if True:
text = r_book['title'] + ". "
if r_book['comments']:
text = r_book['comments']
s_language, prob = identifier.classify(text)
if prob >= 0.85:
language = iso639.to_iso639_2(s_language)
book['language'] = language
else:
book['language'] = ''
else:
book['language'] = iso639.to_iso639_2(r_book['languages'][0])
if r_book['cover']:
book['cover'] = True
else:
book['cover'] = False
book['last_modified'] = r_book['last_modified']
book['timestamp'] = r_book['timestamp']
book['formats'] = []
formats = r_book['formats']
for f in formats:
if 'size' in r_book['format_metadata'][f]:
size = int(r_book['format_metadata'][f]['size'])
else:
# print()
# print(f"Size not found for format '{f}' uuid={uuid}: skipped")
pass
#TODO query the size when the function to rebuild the full url is ready
#
# print("Trying to get size online: {}".format('url'))
# try:
# size=get_file_size(s['url'])
# except:
# print("Unable to access size for format '{}' : {} skipped".format(f, uuid))
# continue
book[f] = (size)
book['formats'].append(f)
if not book['formats']:
# if not c_format:
# print()
# print(f"No format found for {book['uuid']} id={book['id']} : skipped")
range += 1
# continue
books.append(book)
range += 1
# print()
# print("Saving metadata")
print('\r {:180.180}'.format(f'Saving metadata from {server}'), end='')
save_books_metadata_from_site(db, books)
print('\r {:180.180}'.format(
f'--> Saved {range-1}/{total_num} ebooks from {server}'),
end='')
print()
print()
# try:
# save_metadata(db, books)
# except:
# print("Unable to save book metadata")
offset = offset + num
def findSubtitlesNoneIso639(scanfolder, isoMode, disablelangdetect):
"""
Detect subtitles that do not comply with ISO-639.
TODO: Add more subtitle extensions (and read/parse them correctly for
language detection)
TODO: Seperate language detection better in different functions
TODO: Add percentage of certainty and possible other languages when
low certainty
TODO: Handle unicode better to detect languages like German and Dutch
better
TODO: Use table
"""
subtitleExts = ['.srt', '.sub', '.ass']
total = 0
incorrect = 0
detectedlang = 0
for subdir, dirnames, filenames in os.walk(scanfolder):
for filename in filenames:
incorrectSubtitle = False
extension = os.path.splitext(filename)[1].lower()
# subdirName = os.path.basename(os.path.normpath(subdir))
if extension in subtitleExts:
total = total + 1
langcodeFromFilename = getIsoLanguageCodeFromFilename(filename)
detectedLanguage = ""
detectedIsoMode = False
if is_valid639_1(langcodeFromFilename):
detectedIsoMode = "1"
detectedLanguage = iso639_to_name(langcodeFromFilename)
if is_valid639_2(langcodeFromFilename):
detectedIsoMode = "2"
detectedLanguage = iso639_to_name(langcodeFromFilename)
if detectedIsoMode is not isoMode:
isoShouldBe = ""
if isoMode == "1" and detectedIsoMode == "2":
isoShouldBe = to_iso639_1(langcodeFromFilename)
if isoMode == "2" and detectedIsoMode == "1":
isoShouldBe = to_iso639_2(langcodeFromFilename)
filepath = subdir + os.sep + filename
incorrectSubtitle = True
incorrect = incorrect + 1
warning = "Incorrectly named subtitle found at "
warning += bold(filepath)
printNotificationWarning(warning)
if detectedIsoMode is not False:
info = "\t\tLang code " + bold(langcodeFromFilename)
info += " (ISO 639-" + str(detectedIsoMode) + ") "
info += "detected. The ISO 639-" + isoMode + " code"
info += " for " + detectedLanguage + " is "
info += bold(isoShouldBe) + "."
printNotificationInfo(info)
if incorrectSubtitle and not disablelangdetect:
filepath = subdir + os.sep + filename
try:
with io.open(filepath, "r", encoding="utf-8") as mfile:
my_unicode_string = mfile.read()
possibleLanguage = "\tDetected language is likely to "
possibleLanguage += "be \"" + detect(my_unicode_string)
possibleLanguage += "\"\n"
detectedlang = detectedlang + 1
except Exception:
possibleLanguage = "\tLanguage detection failed\n"
info = "Found subtitle files " + bold(str(total)) + " of which "
info += bold(str(incorrect)) + " are incorrectly named!"
printNotificationInfo(info)
