github/bilingual_book_maker
1
1
Fork 0
mirror of https://github.com/yihong0618/bilingual_book_maker.git synced 2025-06-05 19:15:34 +00:00
Code Issues Packages Projects Releases Wiki Activity

support config tags to translate (#107)

Browse Source
This commit is contained in:
hleft 2023-03-08 22:11:43 +08:00 committed by GitHub
parent 3472f3e673
commit b25c4ca873
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
3 changed files with 13 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

8
book_maker/cli.py
View File

@ -82,6 +82,13 @@ def main():
type=str, type=str,
help="specify base url other than the OpenAI's official API address", help="specify base url other than the OpenAI's official API address",
) )
parser.add_argument(
"--translate-tags",
dest="translate_tags",
type=str,
default="p",
help="example --translate-tags p,blockquote",
)
options = parser.parse_args() options = parser.parse_args()
PROXY = options.proxy PROXY = options.proxy
@ -121,6 +128,7 @@ def main():
model_api_base=model_api_base, model_api_base=model_api_base,
is_test=options.test, is_test=options.test,
test_num=options.test_num, test_num=options.test_num,
translate_tags=options.translate_tags,
) )
e.make_bilingual_book() e.make_bilingual_book()

7
book_maker/loader/epub_loader.py
View File

@ -23,12 +23,14 @@ class EPUBBookLoader(BaseBookLoader):
model_api_base=None, model_api_base=None,
is_test=False, is_test=False,
test_num=5, test_num=5,
translate_tags="p",
): ):
self.epub_name = epub_name self.epub_name = epub_name
self.new_epub = epub.EpubBook() self.new_epub = epub.EpubBook()
self.translate_model = model(key, language, model_api_base) self.translate_model = model(key, language, model_api_base)
self.is_test = is_test self.is_test = is_test
self.test_num = test_num self.test_num = test_num
self.translate_tags = translate_tags
try: try:
self.origin_book = epub.read_epub(self.epub_name) self.origin_book = epub.read_epub(self.epub_name)
@ -68,10 +70,11 @@ class EPUBBookLoader(BaseBookLoader):
def make_bilingual_book(self): def make_bilingual_book(self):
new_book = self._make_new_book(self.origin_book) new_book = self._make_new_book(self.origin_book)
all_items = list(self.origin_book.get_items()) all_items = list(self.origin_book.get_items())
trans_taglist = self.translate_tags.split(",")
all_p_length = sum( all_p_length = sum(
0 0
if i.get_type() != ITEM_DOCUMENT if i.get_type() != ITEM_DOCUMENT
else len(bs(i.content, "html.parser").findAll("p")) else len(bs(i.content, "html.parser").findAll(trans_taglist))
for i in all_items for i in all_items
) )
pbar = tqdm(total=self.test_num) if self.is_test else tqdm(total=all_p_length) pbar = tqdm(total=self.test_num) if self.is_test else tqdm(total=all_p_length)
@ -81,7 +84,7 @@ class EPUBBookLoader(BaseBookLoader):
for item in self.origin_book.get_items(): for item in self.origin_book.get_items():
if item.get_type() == ITEM_DOCUMENT: if item.get_type() == ITEM_DOCUMENT:
soup = bs(item.content, "html.parser") soup = bs(item.content, "html.parser")
p_list = soup.findAll("p") p_list = soup.findAll(trans_taglist)
is_test_done = self.is_test and index > self.test_num is_test_done = self.is_test and index > self.test_num
for p in p_list: for p in p_list:
if is_test_done or not p.text or self._is_special_text(p.text): if is_test_done or not p.text or self._is_special_text(p.text):

BIN
test_books/Liber_Esther.epub Normal file
View File

Binary file not shown.