wolfang/anki-hsk-creator
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

solve problerms with menaing trasnlations,

Browse Source 
mightneed to change trasnlation engine
This commit is contained in:
Wolfang Torres
2026-06-29 22:49:50 +08:00
parent 9299020738
commit 977bb422c0
6 changed files with 109 additions and 64 deletions
Download Patch File Download Diff File Expand all files Collapse all files

30
src/anki_hsk_creator/__main__.py
View File

@@ -124,39 +124,49 @@ def main():
break
elif option == "s":
input_file = cli_select_files()
if DICT_TYPE in input_file.input_file.suffixes:
if DICT_TYPE is input_file.file_type:
dict_selected = cli_select_dictionay_tsv()
if dict_selected:
language_id = cli_select_language()
print(
f"pre-processing file {input_file} with language {language_id}"
f"pre-processing {DICT_TYPE} file {input_file} with language {language_id}"
)
pre_process_a_dictionary_file(input_file, language_id)
else:
print(f"Processing file {input_file} with language {language_id}")
print(
f"Processing {DICT_TYPE} file {input_file} with language {language_id}"
)
language_id = cli_select_language(
input_file.available_dictionary_languages
)
process_a_dictionary_file(input_file, language_id)
elif PHRASES_TYPE in input_file.input_file.suffixes:
elif PHRASES_TYPE is input_file.file_type:
language_id = cli_select_language()
print(f"processing file {input_file} with language {language_id}")
print(
f"processing {PHRASES_TYPE} file {input_file} with language {language_id}"
)
process_a_phrases_file(input_file, language_id)
elif DICTATION_TYPE in input_file.input_file.suffixes:
elif DICTATION_TYPE is input_file.file_type:
language_id = cli_select_language()
print(f"processing file {input_file} with language {language_id}")
print(
f"processing {DICTATION_TYPE} file {input_file} with language {language_id}"
)
process_a_dictation_file(input_file, language_id)
elif COMPLETED_TYPE in input_file.input_file.suffixes:
elif COMPLETED_TYPE is input_file.file_type:
language_id = input_file.input_file.suffixes[1][1:]
print(f"processing file {input_file} with language {language_id}")
print(
f"processing {COMPLETED_TYPE} file {input_file} with language {language_id}"
)
proccess_a_completed_file(input_file, language_id)
else:
print(f"File {input_file} is not recognised")
elif option == "f":
language_id = cli_select_language()
langs = (
LANGUAGES.AvailableLanguages if language_id == "all" else [language_id]
)
in_folder = cli_select_folder()
for language_id in langs:
in_folder = cli_select_folder()
print(f"Selected: {in_folder} with language {language_id}")
for dirpath, dirnames, filenames in tuple(
in_folder.absolute_input_folder.walk()

24
src/anki_hsk_creator/anki_generation.py
View File

@@ -12,7 +12,7 @@ from genanki import Deck, Model, Note, Package
from pinyin_tone_converter.pinyin_tone_converter import PinyinToneConverter
# Local
from .constants import DICTATION_TYPE, DICT_TYPE, PHRASES_TYPE
from .constants import COMPLETED_TYPE, DICTATION_TYPE, DICT_TYPE, PHRASES_TYPE
from .utility import (
DictionaryResult,
ProcessFile,
@@ -377,8 +377,22 @@ def output_anki_package(
)
deck.add_note(note)
audios.append(result.audio_path)
elif process_file.file_type is COMPLETED_TYPE:
for result in results:
note = Note(
model=SIMPLE_HSK_MODEL,
fields=[
result.meaning,
PinyinToneConverter().convert_text(result.pinyin),
result.character,
f"[sound:{result.audio_path.name}]",
],
)
deck.add_note(note)
audios.append(result.audio_path)
decks.append(deck)
package = Package(decks)
package.media_files = audios
package.write_to_file(final_file)
return final_file
if decks:
package = Package(decks)
package.media_files = audios
package.write_to_file(final_file)
return final_file

72
src/anki_hsk_creator/api.py
View File

@@ -266,38 +266,54 @@ def proccess_a_completed_file(process_file: ProcessFile, language_id: str) -> Pa
def folder_proccess(process_folder: ProcessFolder, language_id: str):
process_folder.language_id = language_id
print(f"Proccesing folder {process_folder} with language {language_id}")
final_file = process_folder.output_name.with_suffix(".apkg")
if final_file.is_file():
print("File already proccessed")
return final_file
TTS.create_tts()
TRANS.create_translator(LANGUAGES.CN, language_id)
CCCEDICT.create_cedict(language_id)
results = {}
for process_file in process_folder.input_files:
print(f"Proccessing {process_file}")
if process_file.file_type is DICT_TYPE:
with process_file.absolute_input_file.open(
"r", encoding="utf8", newline="\n"
) as file:
words_list = [word.strip() for word in file.readlines() if word]
dictionary_bulk_process(words_list, process_file)
results[process_file] = dictionary_process(process_file)
elif process_file.file_type is DICTATION_TYPE:
with process_file.absolute_input_file.open(
"r", encoding="utf8", newline="\n"
) as file:
text_lines = [
line.strip() for line in file.read().split("。") if line.strip()
]
results[process_file] = dictation_process(text_lines, process_file)
elif process_file.file_type is PHRASES_TYPE:
with process_file.absolute_input_file.open(
"r", encoding="utf8", newline="\n"
) as file:
text_lines = [line.strip() for line in file.readlines() if line.strip()]
results[process_file] = translator_process(text_lines, process_file)
elif process_file.file_type is COMPLETED_TYPE:
process_file.language_id = process_file.input_file.suffixes[1][1:]
with process_file.absolute_input_file.open(
"r", encoding="utf8", newline="\n"
) as file:
text_lines = [line.strip() for line in file.readlines() if line.strip()]
results[process_file] = completed_process(text_lines, process_file)
try:
if process_file.file_type is DICT_TYPE:
with process_file.absolute_input_file.open(
"r", encoding="utf8", newline="\n"
) as file:
words_list = [word.strip() for word in file.readlines() if word]
dictionary_bulk_process(words_list, process_file)
results[process_file] = dictionary_process(process_file)
elif process_file.file_type is DICTATION_TYPE:
with process_file.absolute_input_file.open(
"r", encoding="utf8", newline="\n"
) as file:
text_lines = [
line.strip() for line in file.read().split("。") if line.strip()
]
results[process_file] = dictation_process(text_lines, process_file)
elif process_file.file_type is PHRASES_TYPE:
with process_file.absolute_input_file.open(
"r", encoding="utf8", newline="\n"
) as file:
text_lines = [
line.strip() for line in file.readlines() if line.strip()
]
results[process_file] = translator_process(text_lines, process_file)
elif process_file.file_type is COMPLETED_TYPE:
completed_language_id = process_file.input_file.suffixes[1][1:]
if completed_language_id != language_id:
continue
process_file.language_id = completed_language_id
with process_file.absolute_input_file.open(
"r", encoding="utf8", newline="\n"
) as file:
text_lines = [
line.strip() for line in file.readlines() if line.strip()
]
results[process_file] = completed_process(text_lines, process_file)
except AttributeError, AssertionError:
print(f"Error procesing {process_file} with language {language_id}")
continue
return output_anki_package(process_folder, results)

12
src/anki_hsk_creator/constants.py
View File

@@ -35,8 +35,11 @@ class LANGUAGES:
RU = "ru"
TR = "tr"
TH = "th"
JP = "jp"
AvailableLanguages = (ES, EN, FR, RU, TR, TH, JP)
JA = "ja"
KO = "ko"
PB = "pb"
VI = "vi"
AvailableLanguages = (ES, EN, FR, RU, TR, TH, JA, KO, PB, VI)
LanguageNames = {
EN: "English",
ES: "Spanish",
@@ -44,5 +47,8 @@ class LANGUAGES:
RU: "Russian",
TR: "Turkish",
TH: "Thai",
JP: "Japanese",
JA: "Japanese",
KO: "Korean",
PB: "Portuguese (Brazil)",
VI: "Vietnamese",
}

29
src/anki_hsk_creator/proccessor.py
View File

@@ -106,24 +106,24 @@ def dictionary_bulk_process(words_list: list[str], process_file: ProcessFile):
number = 1
for words in words_list:
word = words.split()[0]
hint = " ".join(words.split()[1:]) if len(words.split()) > 1 else None
hint = words.split(maxsplit=1)[1] if len(words.split()) > 1 else None
entries_en = dictionary_en.get(word)
entries = dictionary.get(word)
if entries:
pos_meanings = (
[
meaning
for entry, entry_en in zip(entries, entries_en)
for meaning, meaning_en in zip(
entry.meanings, entry_en.meanings
)
if hint in meaning_en or hint in meaning
]
if hint
else [meaning for entry in entries for meaning in entry.meanings]
)
all_meanings = [meaning for entry in entries for meaning in entry.meanings]
pos_meanings = [
meaning
for entry, entry_en in zip(entries, entries_en)
for meaning, meaning_en in zip(
entry.meanings, entry_en.meanings
)
if hint and (hint in meaning_en or hint in meaning)
]
pos_meanings = pos_meanings or all_meanings
if not pos_meanings:
raise ValueError(f"Not menaing found for hint {hint}, {all_meanings}")
meanings_text = "\n".join(
f"{n}: {meaning}" for n, meaning in enumerate(pos_meanings)
f"{n+1}: {meaning}" for n, meaning in enumerate(pos_meanings)
)
entry = entries[0]
tsv_writer.writerow(
@@ -212,7 +212,6 @@ def dictionary_process(process_file: ProcessFile) -> list[DictionaryResult]:
if not audio_path.exists():
wavs, sr = TTS.generate(f"{line['simplified']}。")
sf.write(audio_path, wavs[0], sr)
print(line)
result = DictionaryResult(
**line, audio_path=audio_path, language_id=process_file.language_id
)

6
src/anki_hsk_creator/utility.py
View File

@@ -54,7 +54,6 @@ class TRANS:
TRANS.PACKAGES,
)
)
print(f"available packages {packages[:5]}")
packages_to_install = []
ready = False
for in_package in packages:
@@ -121,11 +120,10 @@ class TranslatedEntry:
if not self._translated_meanings:
for meaning in self.entry.meanings:
if self.language_id != LANGUAGES.EN:
print(f"translating from {LANGUAGES.EN} to {self.language_id}")
print(f"-> {meaning}")
trans_meaning = argostranslate.translate.translate(
meaning, LANGUAGES.EN, self.language_id
)
print(f"{meaning}-> {trans_meaning}")
else:
trans_meaning = meaning
self._translated_meanings.append(trans_meaning)
@@ -318,6 +316,8 @@ class ProcessFile:
"""Posible name for the output file, still missing the filetype"""
if self.language_id is None:
raise ValueError("Not a valid language selected")
if self.file_type is COMPLETED_TYPE:
return self.out_folder / f"{self.input_file.stem}.temp"
return self.out_folder / f"{self.input_file.stem}.{self.language_id}.temp"
@property