Last Comment Bug 778921 - "warning: invalid manifest line element "pref(javascript.options.xml.content,true)" running jstests.py
: "warning: invalid manifest line element "pref(javascript.options.xml.content,...
Status: RESOLVED FIXED
:
Product: Core
Classification: Components
Component: JavaScript Engine (show other bugs)
: Other Branch
: All All
: -- normal (vote)
: mozilla17
Assigned To: Jason Orendorff [:jorendorff]
: general
: Jason Orendorff [:jorendorff]
Mentors:
Depends on:
Blocks:
  Show dependency treegraph
 
Reported: 2012-07-30 13:54 PDT by Jason Orendorff [:jorendorff]
Modified: 2012-07-31 19:17 PDT (History)
1 user (show)
See Also:
Crash Signature:
(edit)
QA Whiteboard:
Iteration: ---
Points: ---
Has Regression Range: ---
Has STR: ---


Attachments
v1 (10.34 KB, patch)
2012-07-30 15:36 PDT, Jason Orendorff [:jorendorff]
luke: review+
Details | Diff | Splinter Review
Rebased on top of terrence's patch in bug 778383 (7.94 KB, patch)
2012-07-31 09:36 PDT, Jason Orendorff [:jorendorff]
jorendorff: review+
Details | Diff | Splinter Review

Description Jason Orendorff [:jorendorff] 2012-07-30 13:54:35 PDT

    
Comment 1 Jason Orendorff [:jorendorff] 2012-07-30 15:36:36 PDT
Created attachment 647326 [details] [diff] [review]
v1

E4X is disabled by default in the shell; you have to turn it on using
  options('allow_xml');

Up to now, the root shell.js script contained such an options() call, so E4X was effectively enabled for all JS tests, but only when they run in the shell.

With this patch, allow_xml is off by default in tests, and you can turn it on by using the magic comment

// |reftest| pref(javascript.options.xml.content,true)

The magic comment already worked in the jsreftest test harness, so it is already there for all E4X-using tests that run in the browser.

Up to now the shell test harness didn't know about this pref() thingy and it just spat out a warning. Now the shell test harness also honors this setting. I had to add the magic comment to a few tests that don't run in the browser.
Comment 2 Jason Orendorff [:jorendorff] 2012-07-31 09:36:17 PDT
Created attachment 647572 [details] [diff] [review]
Rebased on top of terrence's patch in bug 778383

Carrying forward review.
Comment 3 Jason Orendorff [:jorendorff] 2012-07-31 12:06:35 PDT
Embarrassing:

https://hg.mozilla.org/integration/mozilla-inbound/rev/9fa54a1fc055

I think this is the first time Python's mutating += has bitten me. Pythons don't usually bite (they crush their prey).

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