RESOLVED FIXED 167990
Remove most cases of #if ENABLE(CACHE_PARTITIONING) 
https://bugs.webkit.org/show_bug.cgi?id=167990
Summary Remove most cases of #if ENABLE(CACHE_PARTITIONING)
Antti Koivisto
Reported 2017-02-08 08:14:47 PST
This is ifdef is all over the place.
Attachments
patch (60.59 KB, patch)
2017-02-08 08:26 PST, Antti Koivisto 		sam: review+
DetailsFormatted DiffDiff
patch (58.46 KB, patch)
2017-02-09 01:21 PST, Antti Koivisto 		no flags
DetailsFormatted DiffDiff
patch (57.38 KB, patch)
2017-02-09 02:25 PST, Antti Koivisto 		no flags
DetailsFormatted DiffDiff
Show Obsolete (2) View All Add attachment proposed patch, testcase, etc.
Antti Koivisto
Comment 1 2017-02-08 08:26:17 PST
Created attachment 300901 [details] patch
Sam Weinig
Comment 2 2017-02-08 09:00:35 PST
Comment on attachment 300901 [details] patch View in context: https://bugs.webkit.org/attachment.cgi?id=300901&action=review I love this! rs=me when you get it compiling. > Source/WebCore/loader/FrameLoader.cpp:1204 > request.setDomainForCachePartition(m_frame.tree().top().document()->securityOrigin().domainForCachePartition()); Should this be using topOrigin? > Source/WebCore/loader/cache/MemoryCache.cpp:562 > static NeverDestroyed<String> httpString("http"); This can't really be necessary now that we have ASCIILiteral, right? > Source/WebCore/platform/network/ResourceRequestBase.cpp:665 > +#else > + return domain; > +#endif Is this the right decision? Maybe we should we just return emptyString() if there is no PUBLIC_SUFFIX_LIST support? Or maybe we should #error if you enable ENABLE(CACHE_PARTITIONING) without ENABLE(PUBLIC_SUFFIX_LIST)? > Source/WebKit2/NetworkProcess/cache/NetworkCacheSpeculativeLoadManager.cpp:97 > + > > revalidationRequest.makeUnconditional(); No need for this extra newline.
Antti Koivisto
Comment 3 2017-02-09 01:21:28 PST
Created attachment 301023 [details] patch
Antti Koivisto
Comment 4 2017-02-09 02:25:23 PST
Created attachment 301026 [details] patch
Antti Koivisto
Comment 5 2017-02-09 03:11:22 PST
https://trac.webkit.org/r211946
Note You need to log in before you can comment on or make changes to this bug.