Status: Untriaged Owner: dpra...@chromium.org Labels: Type-Bug Pri-2 OS-All Area-Misc LayoutTests
New issue 24203 by dpra...@chromium.org: [GKURL] setQuery() does not work on non-standard URLs http://code.google.com/p/chromium/issues/detail?id=24203 ( Migrating from http://buganizer/issue?id=1089231 ). ericroman on March 26 2008 17:34 (New) Summary [GKURL] setQuery() does not work on non-standard URLs Component Chrome > Deprecated > Compatibility Reporter ericroman CC brettw, chrome-bugs Type Feature request Priority P2 Severity S2 In prod false Notes When posting forms, WebCore uses KURL.setQuery() to append the GET parameters to the URL. In GKURL this has no effect on non-standard URLs. The end result is no GET parameters are appended to the URL. WebKit's layout tests include some of the following non-standard form action urls: about:blank mailto:f...@foo.com data:text/html,<script>if (window.layoutTestController) layoutTestController.notifyDone();</script>PASS<br> ***note on data:urls** By far the weirdest one is data:url, since blindly appending ?var1=val1&var2=val2... really doesn't have any meaning. However, it looks like both Safari and Firefox do this. In fact, in what looks like a bug, both Safari and Firefox will do wonky stuff if you have a question mark already in the data:url (i.e. it is probably just scanning for the first question mark, and considering that the start of the "query") This difference accounts for the following layout test failures: V8 | KJS # LayoutTests\http\tests\navigation\onload-navigation-iframe-timeout.html = FAIL V8 | KJS # LayoutTests\http\tests\navigation\onload-navigation-iframe.html = FAIL V8 | KJS # LayoutTests\fast\events\stopPropagation-submit.html = FAIL V8 | KJS # LayoutTests\fast\forms\mailto\advanced-get.html = FAIL V8 | KJS # LayoutTests\fast\forms\mailto\advanced-put.html = FAIL V8 | KJS # LayoutTests\fast\forms\mailto\get-multiple-items-text-plain.html = FAIL V8 | KJS # LayoutTests\fast\forms\mailto\get-multiple-items-x-www-form-urlencoded.html = FAIL V8 | KJS # LayoutTests\fast\forms\mailto\get-multiple-items.html = FAIL V8 | KJS # LayoutTests\fast\forms\mailto\get-non-ascii-text-plain.html = FAIL V8 | KJS # LayoutTests\fast\forms\mailto\get-non-ascii.html = FAIL V8 | KJS # LayoutTests\fast\forms\mailto\get-overwrite-query.html = FAIL V8 | KJS # LayoutTests\fast\forms\mailto\post-append-query.html = FAIL V8 | KJS # LayoutTests\fast\forms\mailto\post-multiple-items-multipart-form-data.html = FAIL V8 | KJS # LayoutTests\fast\forms\mailto\post-multiple-items-text-plain.html = FAIL V8 | KJS # LayoutTests\fast\forms\mailto\post-multiple-items-x-www-form-urlencoded.html = FAIL V8 | KJS # LayoutTests\fast\forms\mailto\post-multiple-items.html = FAIL ericroman on March 26 2008 17:37 (New) CC brettw, chrome-bugs, ojan Notes 1027218 has been marked a duplicate of this issue mal on March 27 2008 16:14 (New) Component Chrome > Deprecated > WebKit Hotlist Cr__Rel_1.0 mal on March 27 2008 16:14 (New) Depends on 771509 brettw on April 01 2008 10:36 (New) Notes What does IE do for setQuery on data URLs? This behavior sounds very wrong to me, and I would be resistant to adding it if there is not requirement for web compatibility. We will need to do this for mailto URLs, though. ojan on April 01 2008 11:43 (New) CC brettw, chrome-bugs, ianh, ojan Notes Does IE support data URLs? I thought not. It does just seem totally broken to have query params for data URLs. I tried finding documentation about data URLs in HTML5 and couldn't. Hixie, is this spec'ed anywhere you know of? I saw HTML5 mentions data URLs all over the place, but doesn't actually define them anywhere. That said, the data urls are jsut two of the failing tests. The mailto bug we can go ahead and fix. brettw on April 01 2008 11:49 (New) Notes javascript URLs should be handled the same as data URLs would be in this case. ojan on April 01 2008 13:30 (New) Notes So, according to http://tools.ietf.org/html/rfc2397, for data: urls, if there are parameters, they should be placed as part of the mediatype (unless I'm reading it incorrectly). I think we should the data: and javascript: bugs upstream to webkit and let them hash it out. mal on April 10 2008 15:40 (Assigned) Assignee ericroman Type Bug Priority P1 Notes Looks like this would fix a bunch of layout tests, so we should fix it (at least mailto:) sooner (P1). I don't think Brett will get to this, given his plate so I'm assigning to Eric. pamg on April 11 2008 17:59 (Assigned) Notes For LayoutTests\fast\forms\mailto\get-overwrite-query.html also note that the query should replace an existing value, and the @ in the query parameters should be escaped. chromedashboard-buganizer on April 21 2008 02:15 (Assigned) URLs affected http://tools.ietf.org/html/rfc2397 eseidel on April 28 2008 17:18 (Assigned) Notes If you're not hacking on this one, I'll see if I can't fix it tomorrow and send you a patch. Assign it to me if that's the case. ericroman on April 28 2008 20:18 (Accepted) URLs affected <none> Notes eseidel: i'll give this one a stab tonight. sorry for being delinquent on it. ericroman on May 06 2008 11:25 (Fixed / Fixed) Notes specifically addressed the mailto: case in googleurl 76. ericroman on May 07 2008 13:55 (Fixed / Fixed) Notes filed: https://bugs.webkit.org/show_bug.cgi?id=18929 since I don't think it makes sense to support this for non-standard urls other than mailto. i.e. doing this is non-nonsensical for data:urls. laforge on October 28 2008 12:38 (Verified / Fixed) Verifier laforge Notes Cleanup operation, batch updating old 'FIXED' issues to 'VERIFIED' that have not been modified since 9/2/08. Status Verified Resolution Fixed Reporter ericroman Assignee ericroman Verifier laforge CC brettw, chrome-bugs, ianh, ojan Type Bug Priority P1 Severity S2 Depends on 771509, 1155751 Blocking Changelists Found in Targeted to Verified in In prod false Hotlists Cr__Rel_1.0 Email: buganizer-system+20506+1089...@google.com -- You received this message because you are listed in the owner or CC fields of this issue, or because you starred this issue. You may adjust your issue notification preferences at: http://code.google.com/hosting/settings --~--~---------~--~----~------------~-------~--~----~ Automated mail from issue updates at http://crbug.com/ Subscription options: http://groups.google.com/group/chromium-bugs -~----------~----~----~----~------~----~------~--~---
