Cutlery/calibre
1
0
Fork 0
mirror of https://github.com/kovidgoyal/calibre.git synced 2025-07-09 03:04:10 -04:00
Code Issues Packages Projects Releases Wiki Activity

Update scientific_american.recipe

Browse Source
This commit is contained in:
unkn0w7n 2024-03-02 09:37:54 +05:30
parent eb05330943
commit c0f1f6670d
1 changed files with 9 additions and 7 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

16
recipes/scientific_american.recipe
View File

@ -67,22 +67,24 @@ class ScientificAmerican(BasicNewsRecipe):
if not curr_issue_link: if not curr_issue_link:
self.abort_recipe_processing("Unable to find issue link") self.abort_recipe_processing("Unable to find issue link")
issue_url = 'https://www.scientificamerican.com' + curr_issue_link.a["href"] issue_url = 'https://www.scientificamerican.com' + curr_issue_link.a["href"]
# for past editions https://www.scientificamerican.com/archive/issues/
# issue_url = 'https://www.scientificamerican.com/issue/sa/2024/01-01/'
soup = self.index_to_soup(issue_url) soup = self.index_to_soup(issue_url)
script = soup.find("script", id="__NEXT_DATA__") script = soup.find("script", id="__DATA__")
if not script: if not script:
self.abort_recipe_processing("Unable to find script") self.abort_recipe_processing("Unable to find script")
JSON = script.contents[0].split('JSON.parse(`')[1].replace("\\\\", "\\")
data = json.JSONDecoder().raw_decode(JSON)[0]
issue_info = ( issue_info = (
json.loads(script.contents[0]) data
.get("props", {}) .get("initialData", {})
.get("pageProps", {}) .get("issueData", {})
.get("issue", {})
) )
if not issue_info: if not issue_info:
self.abort_recipe_processing("Unable to find issue info") self.abort_recipe_processing("Unable to find issue info")
image_id, ext = splitext(issue_info["image"]) self.cover_url = issue_info["image_url"] + "?w=800"
self.cover_url = f"https://static.scientificamerican.com/sciam/cache/file/{image_id}_source{ext}?w=800"
edition_date = datetime.strptime(issue_info["issue_date"], "%Y-%m-%d") edition_date = datetime.strptime(issue_info["issue_date"], "%Y-%m-%d")
self.timefmt = f" [{edition_date:%B %Y}]" self.timefmt = f" [{edition_date:%B %Y}]"