def transfer_math_equations_newline(article: Article,
doc: MarkdownDoc) -> MarkdownDoc:
doc.transfer_math_block_by_line(
test=lambda line: line.endswith(r'\\'),
transfer=lambda line: line[:-2] + r'\newline',
)
return doc
def add_lcn_footer(article: Article, doc: MarkdownDoc) -> MarkdownDoc:
print('add lcn footer')
script_path = os.path.realpath(__file__)
project_path = Path(script_path).parent.parent
file = project_path / 'snippet' / 'footer' / 'lcn.md'
footer = MarkdownDoc.from_file(file)
doc.footer = footer.body
return doc
def extract_meta(doc: MarkdownDoc) -> MetaInfo:
name = None
doc_name = None
title_en = None
title_cn = None
original_url = None
translator_name = None
translator_page = None
p = doc.find_one(lambda p: isinstance(p, Quote))
assert p is not None
for line in p.line_strings():
if '原文地址' in line:
original_address = extract_link(line)
title_en = original_address.text
original_url = original_address.url
elif '本文永久链接' in line:
permalink = extract_link(line)
url = permalink.url
path = Path(urlparse(url).path)
name = path.stem
doc_name = path.name
elif '译者' in line:
translator = extract_link(line)
translator_name = translator.text
translator_page = translator.url
heading1: Optional[Heading] = doc.find_one(
lambda p: isinstance(p, Heading) and p.level == 1)
if heading1 is not None:
title_cn = heading1.text
return MetaInfo(
base=BaseInfo(
name=name,
docName=doc_name,
titleEn=title_en,
titleCn=title_cn,
tags=['翻译'],
),
translation=TranslationInfo(
originalUrl=original_url,
translatorName=translator_name,
translatorPage=translator_page,
goldMiner=GoldMinerTranslationInfo(postUrl='', ),
),
)
def group_code_blocks(article: Article, doc: MarkdownDoc) -> MarkdownDoc:
code_block_groups = doc.find_adjacent(lambda p: isinstance(p, CodeBlock))
for group in code_block_groups:
for p in group:
code_block: CodeBlock = p
code_block.language = code_block.language + ' []'
return doc
def save_body_to_temp_zhihu(article: Article, doc: MarkdownDoc) -> None:
body_string = doc.full_body_string()
# workaround: replace math equations as zhihu special format
body_string = replace_math_equations_zhihu(body_string)
filename = article.meta.base.docName
file = Path.home() / 'Desktop' / filename
with file.open('w') as f:
print(body_string, file=f)
print(f'document body exported to file {file}')
def transfer_image_uri_as_public(article: Article,
doc: MarkdownDoc) -> MarkdownDoc:
def transfer(uri: str) -> str:
base_url_path = settings.image.oss.baseDir
article_name = article.meta.base.name
file_name = Path(uri).name
url_path = Path(base_url_path).joinpath(article_name, file_name)
url = ParseResult(
scheme='http',
netloc=settings.image.oss.publicHost,
path=str(url_path),
params='',
query='',
fragment='',
).geturl()
return url
doc.transfer_image_uri(
test=lambda image: image.is_local(),
transfer=transfer,
)
return doc
def save_doc(self, doc: MarkdownDoc) -> None:
self._mkdir()
doc.save(self.doc_path())
def read_doc(self) -> MarkdownDoc:
doc_file = self.path_to(self.meta.base.docName)
if not doc_file.exists():
raise RuntimeError(f'doc file not exists: {doc_file}')
return MarkdownDoc.from_file(doc_file)
def copy_body(article: Article, doc: MarkdownDoc) -> None:
pyperclip.copy(doc.full_body_string())
print('document body copied to clipboard')
def add_zhihu_footer(article: Article, doc: MarkdownDoc) -> MarkdownDoc:
print('add zhihu footer')
file = get_project_path() / 'snippet' / 'footer' / 'zhihu.md'
footer = MarkdownDoc.from_file(file)
doc.footer = footer.body
return doc
def extract_title(article: Article, doc: MarkdownDoc) -> MarkdownDoc:
heading1 = doc.remove_start(
lambda p: isinstance(p, Heading) and p.level == 1)
if heading1 is not None:
doc.title = heading1.text
return doc
def remove_footer(article: Article, doc: MarkdownDoc) -> MarkdownDoc:
doc.remove_end_while(
lambda p: isinstance(p, Quote) or isinstance(p, HorizontalRule))
return doc
def extract_gold_miner_header(article: Article,
doc: MarkdownDoc) -> MarkdownDoc:
doc.header = doc.remove_start_while(lambda p: isinstance(p, Quote))
return doc