Bug 29356

Summary:

LayoutTests/platform/win/accessibility/scroll-to-anchor.html doesn't render as text if there's no accessibilityController

Product:

WebKit

Reporter:

Dirk Pranke <dpranke>

Component:

Tools / Tests

Assignee:

Dirk Pranke <dpranke>

Status:

RESOLVED INVALID

Severity:

Normal

CC:

fishd

Priority:

Version:

528+ (Nightly build)

Hardware:

OS:

OS X 10.5

Attachments:

Description	Flags
patch to handle accessibilityController not being present more gracefully	eric: review-
patch w/ revised indent	fishd: review-

Dirk Pranke

Reported 2009-09-17 20:09:28 PDT

Chromium does not yet implement accessibilityController() in test_shell (chromium's version of dumpRenderTree). The test script bails out without calling dumpAsText(), which causes Chromium's test driver (run_webkit_tests) to expect there to be an image baseline, and there isn't one.

Attachments
patch to handle accessibilityController not being present more gracefully (1.58 KB, patch) 2009-09-17 20:21 PDT, Dirk Pranke	eric: review-	Details Formatted Diff Diff
patch w/ revised indent (1.59 KB, patch) 2009-09-17 23:33 PDT, Dirk Pranke	fishd: review-	Details Formatted Diff Diff
Show Obsolete (1) View All Add attachment proposed patch, testcase, etc.

Dirk Pranke

Comment 1 2009-09-17 20:21:46 PDT

Created attachment 39746 [details] patch to handle accessibilityController not being present more gracefully

Eric Seidel (no email)

Comment 2 2009-09-17 22:31:08 PDT

Comment on attachment 39746 [details] patch to handle accessibilityController not being present more gracefully Indent is wrong. This makes no sense? Why not just skip this test? What purpose does it serve if accessibilityController is not there?

Dirk Pranke

Comment 3 2009-09-17 23:33:30 PDT

Created attachment 39749 [details] patch w/ revised indent

Dirk Pranke

Comment 4 2009-09-17 23:34:53 PDT

(In reply to comment #2) > (From update of attachment 39746 [details]) > Indent is wrong. Fixed! Thanks. > This makes no sense? Why not just skip this test? What purpose does it serve > if accessibilityController is not there? If we skip the test, then we're not running it at all. If we don't skip the test, we are at least insuring that we don't crash on this test. Another way of saying this is that I don't like skipping tests :)

Darin Fisher (:fishd, Google)

Comment 5 2009-09-18 10:19:32 PDT

Comment on attachment 39749 [details] patch w/ revised indent I don't think this is the right change to make. We should leave the test as is. Back in chromium land, we should just implement a no-op accessibilityController if we want to avoid skipping the test.

Dirk Pranke

Comment 6 2009-09-18 12:13:33 PDT

okey doke.

Note You need to log in before you can comment on or make changes to this bug.