Null dereference loading Blink layout test http/tests/misc/detach-during-notifyDone...
[WebKit-https.git] / LayoutTests / ChangeLog
1 2015-11-30  Jiewen Tan  <jiewen_tan@apple.com>
2
3         Null dereference loading Blink layout test http/tests/misc/detach-during-notifyDone.html
4         https://bugs.webkit.org/show_bug.cgi?id=149309
5         <rdar://problem/22748363>
6
7         Reviewed by Brent Fulgham.
8
9         The test case is from Blink r175601:
10         https://codereview.chromium.org/317513002
11         The test case will generate a set of weird ordering events that affects the documentLoader:
12         1. The subframe finishes loading, and since the frame’s testRunner is not set to wait until
13         done, WebKitTestRunner stops the load (by calling WKBundlePageStopLoading()).
14         2. This causes the in-progress XHR to be aborted, which causes its readyState to become DONE
15         (this bug doesn’t always reproduce because sometimes the XHR has already finished before the
16         frame finishes loading).
17         3. The onreadystatechange callback is executed, which sets innerHTML on the parent frame.
18         4. Setting innerHTML disconnects the subframe, nulling out its DocumentLoader.
19         5. We return to WebFrameLoaderClient::dispatchDidFinishLoad() from step #1, but now the
20         FrameLoader’s DocumentLoader is null. And WebKit crashes here.
21
22         Note that steps 2-4 happen synchronously inside WebFrameLoaderClient::dispatchDidFinishLoad().
23
24         * http/tests/misc/detach-during-notifyDone-expected.txt: Added.
25         * http/tests/misc/detach-during-notifyDone.html: Added.
26         * http/tests/misc/resources/detached-frame.html: Added.
27
28 2015-11-30  Commit Queue  <commit-queue@webkit.org>
29
30         Unreviewed, rolling out r192819.
31         https://bugs.webkit.org/show_bug.cgi?id=151681
32
33         This change broke existing layout tests on Windows (Requested
34         by ryanhaddad on #webkit).
35
36         Reverted changeset:
37
38         "Unify font-variant-* with font-variant shorthand"
39         https://bugs.webkit.org/show_bug.cgi?id=149773
40         http://trac.webkit.org/changeset/192819
41
42 2015-11-30  Brady Eidson  <beidson@apple.com>
43
44         Modern IDB: Set the correct source on the IDBRequest for cursor updates
45         https://bugs.webkit.org/show_bug.cgi?id=151665
46
47         Reviewed by Andy Estes.
48
49         * platform/mac-wk1/TestExpectations:
50
51 2015-11-30  Ryan Haddad  <ryanhaddad@apple.com>
52
53         Marking fast/forms/state-restore-per-form.html as a flaky timeout on mac-wk2
54         https://bugs.webkit.org/show_bug.cgi?id=150542
55
56         Unreviewed test gardening.
57
58         * platform/mac-wk2/TestExpectations:
59
60 2015-11-30  Brady Eidson  <beidson@apple.com>
61
62         Modern IDB: Correct handling of cursors finishing iteration.
63         https://bugs.webkit.org/show_bug.cgi?id=151664
64
65         Reviewed by Andy Estes.
66
67         * platform/mac-wk1/TestExpectations:
68
69 2015-11-30  Myles C. Maxfield  <mmaxfield@apple.com>
70
71         Unify font-variant-* with font-variant shorthand
72         https://bugs.webkit.org/show_bug.cgi?id=149773
73
74         Reviewed by Darin Adler.
75
76         Update tests. Also temporarily skip existing font-features tests until
77         https://bugs.webkit.org/show_bug.cgi?id=149774 is fixed.
78
79         * css3/font-variant-parsing-expected.txt:
80         * css3/font-variant-parsing.html:
81         * fast/css/font-property-priority-expected.txt:
82         * fast/css/font-shorthand-expected.txt:
83         * fast/css/parsing-font-variant-ligatures-expected.txt:
84         * fast/css/parsing-font-variant-ligatures.html:
85         * fast/inspector-support/style-expected.txt:
86         * fast/text/font-variant-shorthand-expected.txt: Added.
87         * fast/text/font-variant-shorthand.html: Added.
88         * platform/mac/TestExpectations:
89         * platform/mac/fast/writing-mode/broken-ideograph-small-caps-expected.txt:
90
91 2015-11-30  Brady Eidson  <beidson@apple.com>
92
93         Modern IDB: After versionchange transactions abort, fire onerror on the original IDBOpenDBRequest.
94         https://bugs.webkit.org/show_bug.cgi?id=151648
95
96         Reviewed by Andy Estes.
97
98         * platform/mac-wk1/TestExpectations:
99         * storage/indexeddb/modern/abort-requests-cancelled-expected.txt:
100         * storage/indexeddb/modern/abort-requests-cancelled.html:
101         * storage/indexeddb/modern/aborted-put-expected.txt:
102         * storage/indexeddb/modern/aborted-put.html:
103         * storage/indexeddb/modern/createobjectstore-basic-expected.txt:
104         * storage/indexeddb/modern/createobjectstore-basic.html:
105         * storage/indexeddb/modern/deleteindex-2-expected.txt:
106         * storage/indexeddb/modern/deleteindex-2.html:
107         * storage/indexeddb/modern/deleteobjectstore-1-expected.txt:
108         * storage/indexeddb/modern/deleteobjectstore-1.html:
109         * storage/indexeddb/modern/idbdatabase-transaction-failures-expected.txt:
110         * storage/indexeddb/modern/versionchange-abort-then-reopen-expected.txt:
111         * storage/indexeddb/modern/versionchange-abort-then-reopen.html:
112
113 2015-11-30  Chris Dumez  <cdumez@apple.com>
114
115         location.origin is undefined in a web worker
116         https://bugs.webkit.org/show_bug.cgi?id=151614
117
118         Reviewed by Darin Adler.
119
120         Update existing layout test to confirm the existence of location.origin when in a
121         WorkerGlobalScope.
122
123         * fast/workers/resources/worker-location.js:
124         * fast/workers/worker-location-expected.txt:
125
126 2015-11-30  Brady Eidson  <beidson@apple.com>
127
128         Modern IDB: Support updating cursor values when the object store uses inline keys.
129         https://bugs.webkit.org/show_bug.cgi?id=151647
130
131         Reviewed by Andy Estes.
132
133         * platform/mac-wk1/TestExpectations:
134
135 2015-11-30  Brady Eidson  <beidson@apple.com>
136
137         Modern IDB: openCursor() fix resulting in at least 4 more passing tests.
138         https://bugs.webkit.org/show_bug.cgi?id=151630
139
140         Reviewed by Andy Estes.
141
142         * platform/mac-wk1/TestExpectations:
143         * storage/indexeddb/modern/cursor-2-expected.txt:
144         * storage/indexeddb/modern/cursor-2.html:
145         * storage/indexeddb/modern/cursor-3-expected.txt:
146         * storage/indexeddb/modern/cursor-3.html:
147         * storage/indexeddb/modern/cursor-4-expected.txt:
148         * storage/indexeddb/modern/cursor-4.html:
149         * storage/indexeddb/modern/deleteindex-1-expected.txt:
150         * storage/indexeddb/modern/deleteindex-1.html:
151         * storage/indexeddb/modern/deleteindex-2-expected.txt:
152         * storage/indexeddb/modern/deleteindex-2.html:
153         * storage/indexeddb/modern/index-4-expected.txt:
154         * storage/indexeddb/modern/index-4.html:
155         * storage/indexeddb/modern/index-cursor-1-expected.txt:
156         * storage/indexeddb/modern/index-cursor-1.html:
157         * storage/indexeddb/modern/index-cursor-2-expected.txt:
158         * storage/indexeddb/modern/index-cursor-2.html:
159         * storage/indexeddb/modern/index-cursor-3-expected.txt:
160         * storage/indexeddb/modern/index-cursor-3.html:
161
162 2015-11-30  Ryan Haddad  <ryanhaddad@apple.com>
163
164         Marking http/tests/cache/disk-cache/disk-cache-request-max-stale.html as flaky on Yosemite WK2
165         https://bugs.webkit.org/show_bug.cgi?id=143159
166
167         Unreviewed test gardening.
168
169         * platform/mac-wk2/TestExpectations:
170
171 2015-11-30  Brady Eidson  <beidson@apple.com>
172
173         Modern IDB: Support keyPath injection into object store records.
174         https://bugs.webkit.org/show_bug.cgi?id=151640
175
176         Reviewed by Andy Estes.
177
178         * platform/mac-wk1/TestExpectations:
179
180 2015-11-29  Antoine Quint  <graouts@apple.com>
181
182         Browser does not fall back to SVG attribute value when CSS style value is invalid or not supported
183         https://bugs.webkit.org/show_bug.cgi?id=147932
184
185         Reviewed by Dean Jackson.
186
187         Testing that we correctly fall back to the presentation attribute for SVGPaint and SVGColor values
188         specified with an invalid keyword in a `style` attribute. We also update the expected output for
189         svg/css/svg-attribute-parser-mode.html which is now in line with values returned by Firefox and
190         Chrome, where we correctly use the default value instead of null objects, which was definitely
191         an error.
192
193         * svg/css/invalid-color-cascade-expected.svg: Added.
194         * svg/css/invalid-color-cascade.svg: Added.
195         * svg/css/invalid-paint-cascade-expected.svg: Added.
196         * svg/css/invalid-paint-cascade.svg: Added.
197         * svg/css/script-tests/svg-attribute-parser-mode.js:
198         * svg/css/svg-attribute-parser-mode-expected.txt:
199
200 2015-11-18  Andy Estes  <aestes@apple.com>
201
202         [Content Filtering] Crash in DocumentLoader::notifyFinished() when allowing a media document to load
203         https://bugs.webkit.org/show_bug.cgi?id=151433
204         rdar://problem/23506594
205
206         Reviewed by Alexey Proskuryakov.
207
208         * contentfiltering/allow-media-document-expected.txt: Added.
209         * contentfiltering/allow-media-document.html: Added.
210         * contentfiltering/resources/test.mp4: Added.
211
212 2015-11-25  Pranjal Jumde  <pjumde@apple.com>
213
214         Checks for buffer-overflows when reading characters from textRun
215         https://bugs.webkit.org/show_bug.cgi?id=151055
216         <rdar://problem/23251789>
217
218         Reviewed by Myles C. Maxfield.
219
220         * dom/html/level1/core/151055_asan.html:
221         * dom/html/level1/core/151055_asan-expected.txt:
222
223 2015-11-24  Commit Queue  <commit-queue@webkit.org>
224
225         Unreviewed, rolling out r192536, r192722, and r192743.
226         https://bugs.webkit.org/show_bug.cgi?id=151593
227
228         Still causing trouble. (Requested by kling on #webkit).
229
230         Reverted changesets:
231
232         "[JSC] JSPropertyNameEnumerator could be destructorless."
233         https://bugs.webkit.org/show_bug.cgi?id=151242
234         http://trac.webkit.org/changeset/192536
235
236         "REGRESSION(r192536): Null pointer dereference in
237         JSPropertyNameEnumerator::visitChildren()."
238         https://bugs.webkit.org/show_bug.cgi?id=151495
239         http://trac.webkit.org/changeset/192722
240
241         "REGRESSION(r192536): Null pointer dereference in
242         JSPropertyNameEnumerator::visitChildren()."
243         https://bugs.webkit.org/show_bug.cgi?id=151495
244         http://trac.webkit.org/changeset/192743
245
246 2015-11-24  Xabier Rodriguez Calvar  <calvaris@igalia.com>
247
248         [Streams API] Implement pipeTo method in readable Stream
249         https://bugs.webkit.org/show_bug.cgi?id=151588
250
251         Reviewed by Darin Adler.
252
253         Test "Piping to a writable stream that does not consume the writes fast enough exerts backpressure on the
254         source" was moved to its own file because it causes timing issues.
255
256         * streams/reference-implementation/pipe-to.html: Moved "Piping to a writable stream that does not consume the
257         writes fast enough exerts backpressure on the source" test to its own file.
258         * streams/pipe-to.html: Added with "Piping to a writable stream that does not consume the writes fast enough
259         exerts backpressure on the source" test.
260         * streams/reference-implementation/brand-checks.html: Fixed issue with the creation of a ReadableStreamReader.
261         * streams/pipe-to-expected.txt:
262         * streams/reference-implementation/brand-checks-expected.txt:
263         * streams/reference-implementation/pipe-through-expected.txt:
264         * streams/reference-implementation/pipe-to-expected.txt:
265         * streams/reference-implementation/pipe-to-options-expected.txt:
266         * streams/reference-implementation/readable-stream-templated-expected.txt: Expectations.
267         * platform/mac/TestExpectations:
268         * platform/win/TestExpectations: Flagged pipe-to test because of webkit.org/b/147933.
269
270 2015-11-24  Antti Koivisto  <antti@apple.com>
271
272         REGRESSION (r190983): Non-element, non-text nodes should not be distributed to slots
273         https://bugs.webkit.org/show_bug.cgi?id=151566
274         rdar://problem/23430177
275
276         Reviewed by Zalan Bujtas.
277
278         * fast/html/details-comment-crash-expected.html: Added.
279         * fast/html/details-comment-crash.html: Added.
280
281 2015-11-23  Brady Eidson  <beidson@apple.com>
282
283         Modern IDB: Unskip all indexeddb/mozilla tests that pass.
284         https://bugs.webkit.org/show_bug.cgi?id=151568
285
286         Reviewed by Alex Christensen.
287
288         * platform/mac-wk1/TestExpectations:
289
290 2015-11-23  Brady Eidson  <beidson@apple.com>
291
292         Modern IDB: When a transaction is aborted, call onerror handlers for all in-progress requests.
293         https://bugs.webkit.org/show_bug.cgi?id=151550
294
295         Reviewed by Alex Christensen.
296
297         * storage/indexeddb/modern/abort-requests-cancelled-expected.txt: Added.
298         * storage/indexeddb/modern/abort-requests-cancelled.html: Added.
299         * storage/indexeddb/modern/idbtransaction-objectstore-failures-expected.txt:
300         * storage/indexeddb/modern/idbtransaction-objectstore-failures.html:
301         * storage/indexeddb/modern/index-5-expected.txt:
302         * storage/indexeddb/modern/index-5.html:
303
304 2015-11-23  Brady Eidson  <beidson@apple.com>
305
306         Modern IDB: Unskip storage/indexeddb/mozilla/global-data.html.
307         https://bugs.webkit.org/show_bug.cgi?id=151557
308
309         Reviewed by Alex Christensen.
310
311         * platform/mac-wk1/TestExpectations:
312
313 2015-11-17  Sergio Villar Senin  <svillar@igalia.com>
314
315         ASSERTION FAILED: freeSpace >= 0 in WebCore::RenderGrid::computeTrackSizesForDirection
316         https://bugs.webkit.org/show_bug.cgi?id=151254
317
318         Reviewed by Darin Adler.
319
320         * fast/css-grid-layout/grid-quirks-mode-huge-margin-crash-expected.txt: Added.
321         * fast/css-grid-layout/grid-quirks-mode-huge-margin-crash.html: Added.
322
323 2015-11-22  Myles C. Maxfield  <mmaxfield@apple.com>
324
325         Add a test to make sure that font-variant: inherit works correctly.
326
327         Unreviewed.
328
329         * fast/text/font-variant-inherit-expected.html: Added.
330         * fast/text/font-variant-inherit.html: Added.
331
332 2015-11-22  Myles C. Maxfield  <mmaxfield@apple.com>
333
334         Font selection should not consult font-variant property
335         https://bugs.webkit.org/show_bug.cgi?id=151537
336
337         Reviewed by Simon Fraser.
338
339         * fast/text/font-selection-font-variant-expected.html: Added.
340         * fast/text/font-selection-font-variant.html: Added.
341
342 == Rolled over to ChangeLog-2015-11-21 ==