i FSrSSKJrJr SSKJrJr SSKr"SS\5r g) zGeneric information extractor) ExtractorMessage)configtextNc\rSrSrSrSrSrSrSr\ R"SS5(a\S - r\S - rS r S r S r SrSrSrg)GenericExtractorz+Extractor for images in a generic web page.generic)z {category}z {subcategory}z{path}z {imageurl}z(?i)(?Pg(?:eneric)?:)) extractorr enabled?zj(?Phttps?://)?(?P[-\w\.]+)(?P/[^?#]*)?(?:\?(?P[^#]*))?(?:\#(?P.*))?z$generic:https://www.nongnu.org/lzip/cUSUl[R"X5 US(aUSRS5SUlO%UR R S5 USUlUS(a USUlO7SUl[R"URUR5UlUS Ul URUS-Ul g) Ndomainr r:rz.Falling back on generic information extractor.schemezhttps://path) subcategoryr__init__ partitionurlloginforrensure_http_schemerroot)selfmatchs I/app/mltbenv/lib/python3.13/site-packages/gallery_dl/extractor/generic.pyrGenericExtractor.__init__&s ?4'  Qx))#.q1DH HHMMJ KQxDH ?/DK$DK..txxEDH&M KK%/1 c## URUR5RnURU5nUR U5n[ U5US'[US5n[RSU4v UHjuUS'upVU(a.URU5 SU;a[R"XR5 O[R"XR5 [RXR4v Ml g![ a Nf=f7f)ztGet page, extract metadata & images, yield them in suitable messages Adapted from common.GalleryExtractor.items() countrnum extensionN) requestrrmetadataimageslen TypeError enumerater Directoryupdatenameext_from_urlUrl)rpagedataimgsr)rimgdatas ritemsGenericExtractor.items>s ||DHH%**}}T"{{4  IDM4#T))+1 'DK# G$g-))#4%%c0++s( (,2    s+AC; C+BC;+ C85C;7C88C;c[R"USS5[R"USS5[R"USS5[R"USS5[R"USS5[R"USS5[R"US S5[R"US S5[R"US S5[R"US S5S . nUR5VVs0sH%up4U(dMU[R"U5_M' nnnURR SS5US'UR US'U$s snnf)z8Extract generic webpage metadata, return them in a dict.zzz""ii7>"ii7>"ii8#?"ii:CA"ii?F"ii;SB"iiA3H' 015 B #4==## Byy((b1V ((Y Cs = EEcSnSn[R"U5RU5n[R"U5RU5nXE-n[R"S5RU5nU(aUSR S5UlOoUR RS5(a!UR R S5UlO.[RRUR 5Ul/nUHn U RS5(aURU 5 M,U RS5(a/URURU RS5-5 MqU RS5(a URURU -5 MURUR S-U -5 M [ R#U5nUV s/sHoSU 04PM n n U $s sn f) aExtract image urls, return a list of (image url, metadata) tuples. The extractor aims at finding as many _likely_ image urls as possible, using two strategies (see below); since these often overlap, any duplicate urls will be removed at the end of the process. Note: since we are using re.findall() (see below), it's essential that the following patterns contain 0 or at most 1 capturing group, so that re.findall() return a list of urls (instead of a list of tuples of matching groups). All other groups used in the pattern should be non-capturing (?:...). 1: Look in src/srcset attributes of img/video/source elements See: https://www.w3schools.com/tags/att_src.asp https://www.w3schools.com/tags/att_source_srcset.asp We allow both absolute and relative urls here. Note that srcset attributes often contain multiple space separated image urls; this pattern matches only the first url; remaining urls will be matched by the "imageurl_pattern_ext" pattern below. zE(?i)<(?:img|video|source)\s[^>]*src(?:set)?=[\"']?(?P[^\"'\s>]+)zY(?i)(?:[^?&#\"'>\s]+)\.(?:jpe?g|jpe|png|gif|web[mp]|mp4|mkv|og[gmv]|opus)(?:[^\"'<>\s]*)?z/(?i)(?:[^\"' >]+)rrChttpz//imageurl)rrefindallsearchrstripbaseurlrendswithosrdirname startswithappendrlstriprdictfromkeys) rr1imageurl_pattern_srcimageurl_pattern_ext imageurls_src imageurls_ext imageurls basematch absimageurlsur)s rr)GenericExtractor.imagesys6 #     45==dC  45==dC !1 GG >@@Ft   $U+2237DLxx  %%#xxs3 !wwtxx8  A||F####A&d####DKK!((3-$?@c""##DIIM2##DLL3$6$:;}}\2 1== 1z1o& = >s2H)rRrrrrrN)__name__ __module__ __qualname____firstlineno____doc__category directory_fmt archive_fmtpatternrgetexamplerr5r(r)__static_attributes__r!rr r sd5H=MK 0G zz*I664  #G5G20)6@ar!r ) rhcommonrrr$rros.pathrTr rpr!rrss!$&LyLr!