mirrors/searxng
1
0
Fork 0
mirror of https://github.com/searxng/searxng.git synced 2024-11-23 03:11:00 +00:00
Code Issues Projects Releases Wiki Activity

[fix] engine: Anna's Archive - grep results from '.js-scroll-hidden' elements

Browse source 
The renderuing of the WEB page is very strange; except the firts position all
other positions of Anna's result page are enclosed in SGML comments.  These
cooments are *uncommented* by some JS code, see query of the class
'.js-scroll-hidden' in Anna's HTML template [1].

[1] https://annas-software.org/AnnaArchivist/annas-archive/-/blob/main/allthethings/templates/macros/md5_list.html

Signed-off-by: Markus Heiser <markus.heiser@darmarit.de>
This commit is contained in:
Markus Heiser 2023-06-28 09:16:49 +02:00
parent e2df6b77a3
commit 87e7926ae9
1 changed files with 22 additions and 20 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

42
searx/engines/annas_archive.py
View file

@ -97,14 +97,6 @@ aa_ext: str = ''
""" """
# xpath queries
xpath_results: str = '//main//a[starts-with(@href,"/md5")]'
xpath_url: str = ".//@href"
xpath_title: str = ".//h3/text()[1]"
xpath_authors: str = './/div[contains(@class, "italic")]'
xpath_publisher: str = './/div[contains(@class, "text-sm")]'
xpath_file_info: str = './/div[contains(@class, "text-xs")]'
def init(engine_settings=None): # pylint: disable=unused-argument def init(engine_settings=None): # pylint: disable=unused-argument
"""Check of engine's settings.""" """Check of engine's settings."""
@ -131,24 +123,34 @@ def response(resp) -> List[Dict[str, Optional[str]]]:
results: List[Dict[str, Optional[str]]] = [] results: List[Dict[str, Optional[str]]] = []
dom = html.fromstring(resp.text) dom = html.fromstring(resp.text)
for item in dom.xpath(xpath_results): for item in eval_xpath_list(dom, '//main//div[contains(@class, "h-[125]")]/a'):
result: Dict[str, Optional[str]] = {} results.append(_get_result(item))
result["url"] = base_url + item.xpath(xpath_url)[0] # The rendering of the WEB page is very strange; except the first position
# all other positions of Anna's result page are enclosed in SGML comments.
# These comments are *uncommented* by some JS code, see query of class
# '.js-scroll-hidden' in Anna's HTML template:
# https://annas-software.org/AnnaArchivist/annas-archive/-/blob/main/allthethings/templates/macros/md5_list.html
result["title"] = extract_text(eval_xpath(item, xpath_title)) for item in eval_xpath_list(dom, '//main//div[contains(@class, "js-scroll-hidden")]'):
item = html.fromstring(item.xpath('./comment()')[0].text)
result["content"] = "{publisher}. {authors}. {file_info}".format( results.append(_get_result(item))
authors=extract_text(eval_xpath(item, xpath_authors)),
publisher=extract_text(eval_xpath(item, xpath_publisher)),
file_info=extract_text(eval_xpath(item, xpath_file_info)),
)
results.append(result)
return results return results
def _get_result(item):
return {
'template': 'paper.html',
'url': base_url + item.xpath('./@href')[0],
'title': extract_text(eval_xpath(item, './/h3/text()[1]')),
'publisher': extract_text(eval_xpath(item, './/div[contains(@class, "text-sm")]')),
'authors': [extract_text(eval_xpath(item, './/div[contains(@class, "italic")]'))],
'content': extract_text(eval_xpath(item, './/div[contains(@class, "text-xs")]')),
'img_src': item.xpath('.//img/@src')[0],
}
def fetch_traits(engine_traits: EngineTraits): def fetch_traits(engine_traits: EngineTraits):
"""Fetch languages and other search arguments from Anna's search form.""" """Fetch languages and other search arguments from Anna's search form."""
# pylint: disable=import-outside-toplevel # pylint: disable=import-outside-toplevel