| Summary: | Clean up some ENABLE(PICTURE_SIZES) stuff, don't change function signatures inside #ifdefs | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| Product: | WebKit | Reporter: | Simon Fraser (smfr) <simon.fraser> | ||||||
| Component: | New Bugs | Assignee: | Simon Fraser (smfr) <simon.fraser> | ||||||
| Status: | RESOLVED FIXED | ||||||||
| Severity: | Normal | CC: | commit-queue, esprehn+autocc, gyuyoung.kim, koivisto, simon.fraser, yoav | ||||||
| Priority: | P2 | ||||||||
| Version: | 528+ (Nightly build) | ||||||||
| Hardware: | Unspecified | ||||||||
| OS: | Unspecified | ||||||||
| Bug Depends on: | 136942 | ||||||||
| Bug Blocks: | |||||||||
| Attachments: |
|
||||||||
|
Description
Simon Fraser (smfr)
2014-09-08 18:27:04 PDT
Created attachment 237828 [details]
Patch
Comment on attachment 237828 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=237828&action=review > Source/WebCore/html/parser/HTMLDocumentParser.cpp:329 > + m_preloadScanner->scan(m_preloader.get(), document()->baseElementURL(), document()->renderView(), document()->frame()); Seems bizarre to pass both a RenderView* and a Frame*. Why not just pass a Document& instead? It seems to me that SourceSizeList::parseSizesAttribute should also just take a Document&. And the same for every level in between. Yes, I thought the same thing. I'll do that. Created attachment 237862 [details]
Patch
Comment on attachment 237862 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=237862&action=review > Source/WebCore/html/parser/HTMLPreloadScanner.cpp:102 > for (HTMLToken::AttributeList::const_iterator iter = attributes.begin(); iter != attributes.end(); ++iter) { Opportunity for new-style for loop. > Source/WebCore/html/parser/HTMLPreloadScanner.cpp:112 > + sourceSize = SourceSizeList::parseSizesAttribute(m_sizesAttribute, document.renderView(), document.frame()); I suggest pushing down Document& even further, and changing the interface to SourceSizeList to take Document&. > Source/WebCore/html/parser/HTMLPreloadScanner.h:50 > + void scan(const HTMLToken&, PreloadRequestStream& requests, Document&); I don’t think the argument name “requests” adds anything here. > Source/WebCore/html/parser/HTMLPreloadScanner.h:130 > + void scan(HTMLResourcePreloader*, Document&); Should be HTMLResourcePreloader&. Comment on attachment 237862 [details] Patch Clearing flags on attachment: 237862 Committed r173464: <http://trac.webkit.org/changeset/173464> All reviewed patches have been landed. Closing bug. |