github/gallery-dl
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

[imgth] use text.extract_all

Browse Source
This commit is contained in:
Mike Fährmann
2015-11-03 00:11:08 +01:00
parent 353ac1e00b
commit 1352187758
1 changed files with 9 additions and 10 deletions
Download Patch File Download Diff File Expand all files Collapse all files

19
gallery_dl/extractor/imgth.py
View File

@@ -58,16 +58,15 @@ class ImgthExtractor(Extractor):
def get_job_metadata(self, page): def get_job_metadata(self, page):
"""Collect metadata for extractor-job""" """Collect metadata for extractor-job"""
title, pos = text.extract(page, '<h1>', '</h1>') data = {
count, pos = text.extract(page, 'total of images in this gallery: ', ' ', pos)
date , pos = text.extract(page, 'created on ', ' by <', pos)
_ , pos = text.extract(page, 'href="/users/', '', pos)
user , pos = text.extract(page, '>', '<', pos)
return {
"category": info["category"], "category": info["category"],
"gallery-id": self.gid, "gallery-id": self.gid,
"title": title,
"user": user,
"date": date,
"count": count,
} }
data, _ = text.extract_all(page, (
("title", '<h1>', '</h1>'),
("count", 'total of images in this gallery: ', ' '),
("date" , 'created on ', ' by <'),
(None , 'href="/users/', ''),
("user" , '>', '<'),
), values=data)
return data