pixiv: fix illust id parsing (fix #4043).

* Tighten up illust id parsing to avoid misparsing ids from
  non-illust urls (sketch urls and novel urls).

* Move id parsing tests from post_test.rb to sources/pixiv_test.rb.

* Drop support for touch.pixiv.net urls. These urls are no longer used
  by Pixiv and aren't present as the source of any posts on Danbooru.
This commit is contained in:
evazion
2019-01-13 14:20:30 -06:00
parent 3f95b5163c
commit 8cadef2dd7
3 changed files with 134 additions and 112 deletions

View File

@@ -1,4 +1,47 @@
require 'csv'
# Pixiv
#
# * https://i.pximg.net/img-original/img/2014/10/03/18/10/20/46324488_p0.png
#
# * https://i.pximg.net/c/250x250_80_a2/img-master/img/2014/10/29/09/27/19/46785915_p0_square1200.jpg
# * https://i.pximg.net/img-master/img/2014/10/03/18/10/20/46324488_p0_master1200.jpg
#
# * https://www.pixiv.net/member_illust.php?mode=medium&illust_id=46324488
# * https://www.pixiv.net/member_illust.php?mode=manga&illust_id=46324488
# * https://www.pixiv.net/member_illust.php?mode=manga_big&illust_id=46324488&page=0
#
# * https://www.pixiv.net/member.php?id=339253
# * https://www.pixiv.net/member_illust.php?id=339253&type=illust
# * https://www.pixiv.net/u/9202877
# * https://www.pixiv.net/stacc/noizave
# * http://www.pixiv.me/noizave
#
# Fanbox
#
# * https://fanbox.pixiv.net/images/post/39714/JvjJal8v1yLgc5DPyEI05YpT.png
# * https://pixiv.pximg.net/fanbox/public/images/creator/1566167/profile/Ix6bnJmTaOAFZhXHLbWyIY1e.jpeg
#
# * https://pixiv.pximg.net/c/400x400_90_a2_g5/fanbox/public/images/creator/1566167/profile/Ix6bnJmTaOAFZhXHLbWyIY1e.jpeg
# * https://pixiv.pximg.net/c/1200x630_90_a2_g5/fanbox/public/images/post/186919/cover/VCI1Mcs2rbmWPg0mmiTisovn.jpeg
#
# * https://www.pixiv.net/fanbox/creator/1566167/post/39714
# * https://www.pixiv.net/fanbox/creator/1566167
#
# Novels
#
# * https://i.pximg.net/novel-cover-original/img/2019/01/14/01/15/05/10617324_d84daae89092d96bbe66efafec136e42.jpg
# * https://i.pximg.net/c/600x600/novel-cover-master/img/2019/01/14/01/15/05/10617324_d84daae89092d96bbe66efafec136e42_master1200.jpg
# * https://img-novel.pximg.net/img-novel/work_main/XtFbt7gsymsvyaG45lZ8/1554.jpg?20190107110435
#
# * https://www.pixiv.net/novel/show.php?id=10617324
# * https://novel.pixiv.net/works/1554
#
# Sketch
#
# * https://img-sketch.pixiv.net/uploads/medium/file/4463372/8906921629213362989.jpg
# * https://img-sketch.pximg.net/c!/w=540,f=webp:jpeg/uploads/medium/file/4463372/8906921629213362989.jpg
# * https://sketch.pixiv.net/items/1588346448904706151
# * https://sketch.pixiv.net/@0125840
#
module Sources
module Strategies
@@ -218,37 +261,45 @@ module Sources
# even though it makes sense to reference page_url here, it will only look
# at (url, referer_url).
def illust_id
# http://img18.pixiv.net/img/evazion/14901720.png
#
# http://i2.pixiv.net/img18/img/evazion/14901720.png
# http://i2.pixiv.net/img18/img/evazion/14901720_m.png
# http://i2.pixiv.net/img18/img/evazion/14901720_s.png
# http://i1.pixiv.net/img07/img/pasirism/18557054_p1.png
# http://i1.pixiv.net/img07/img/pasirism/18557054_big_p1.png
#
# http://i1.pixiv.net/img-inf/img/2011/05/01/23/28/04/18557054_64x64.jpg
# http://i1.pixiv.net/img-inf/img/2011/05/01/23/28/04/18557054_s.png
#
# http://i1.pixiv.net/c/600x600/img-master/img/2014/10/02/13/51/23/46304396_p0_master1200.jpg
# http://i1.pixiv.net/img-original/img/2014/10/02/13/51/23/46304396_p0.png
#
# http://i1.pixiv.net/img-zip-ugoira/img/2014/10/03/17/29/16/46323924_ugoira1920x1080.zip
if url =~ %r!/(\d+)(?:_\w+)?\.(?:jpg|jpeg|png|gif|zip)!i
return $1
end
return nil if novel_id.present? || fanbox_id.present?
[url, referer_url].each do |x|
parsed_urls.each do |url|
# http://www.pixiv.net/member_illust.php?mode=medium&illust_id=18557054
# http://www.pixiv.net/member_illust.php?mode=big&illust_id=18557054
# http://www.pixiv.net/member_illust.php?mode=manga&illust_id=18557054
# http://www.pixiv.net/member_illust.php?mode=manga_big&illust_id=18557054&page=1
if x =~ /illust_id=(\d+)/i
return $1
end
if url.host == "www.pixiv.net" && url.path == "/member_illust.php" && url.query_values["illust_id"].present?
return url.query_values["illust_id"].to_i
# http://www.pixiv.net/i/18557054
if x =~ %r!pixiv\.net/i/(\d+)!i
return $1
elsif url.host == "www.pixiv.net" && url.path =~ %r!\A/i/(?<illust_id>\d+)\z!i
return $~[:illust_id].to_i
# http://img18.pixiv.net/img/evazion/14901720.png
# http://i2.pixiv.net/img18/img/evazion/14901720.png
# http://i2.pixiv.net/img18/img/evazion/14901720_m.png
# http://i2.pixiv.net/img18/img/evazion/14901720_s.png
# http://i1.pixiv.net/img07/img/pasirism/18557054_p1.png
# http://i1.pixiv.net/img07/img/pasirism/18557054_big_p1.png
elsif url.host =~ %r!\A(?:i\d+|img\d+)\.pixiv\.net\z!i &&
url.path =~ %r!\A(?:/img\d+)?/img/\w+/(?<illust_id>\d+)(?:_\w+)?\.(?:jpg|jpeg|png|gif|zip)!i
return $~[:illust_id].to_i
# http://i1.pixiv.net/img-inf/img/2011/05/01/23/28/04/18557054_64x64.jpg
# http://i1.pixiv.net/img-inf/img/2011/05/01/23/28/04/18557054_s.png
# http://i1.pixiv.net/c/600x600/img-master/img/2014/10/02/13/51/23/46304396_p0_master1200.jpg
# http://i1.pixiv.net/img-original/img/2014/10/02/13/51/23/46304396_p0.png
# http://i1.pixiv.net/img-zip-ugoira/img/2014/10/03/17/29/16/46323924_ugoira1920x1080.zip
# https://i.pximg.net/img-original/img/2014/10/03/18/10/20/46324488_p0.png
# https://i.pximg.net/img-master/img/2014/10/03/18/10/20/46324488_p0_master1200.jpg
#
# but not:
#
# https://i.pximg.net/novel-cover-original/img/2019/01/14/01/15/05/10617324_d84daae89092d96bbe66efafec136e42.jpg
# https://img-sketch.pixiv.net/uploads/medium/file/4463372/8906921629213362989.jpg
elsif url.host =~ %r!\A(?:i\.pximg\.net|i\d+\.pixiv\.net)\z!i &&
url.path =~ %r!\A(/c/\w+)?/img-[a-z-]+/img/\d{4}(?:/\d{2}){5}/(?<illust_id>\d+)(?:_\w+)?\.(?:jpg|jpeg|png|gif|zip)!i
return $~[:illust_id].to_i
end
end