420858 - Browser only Regressions in ecma_3/Operators/11.13.1-001.js, js1_5/Regress/regress-252892.js, js1_5/Regress/regress-321757.js

Status: VERIFIED FIXED

(Core :: XPConnect, defect, P1)

Description

[Bob Clary [:bc] (inactive)]

range <http://bonsai.mozilla.org/cvsquery.cgi?module=PhoenixTinderbox&date=explicit&mindate=1204592820&maxdate=1204595699>

jst, this looks like you.

Scroll to the bottom of <http://tinderbox.mozilla.org/showlog.cgi?log=MozillaTest/1204617600.1204630302.31138.gz&fulltext=1> to see the possible regressions.

1. <http://test.bclary.com/tests/mozilla.org/js/js-test-driver-standards.html?test=ecma_3/Operators/11.13.1-001.js;language=type;text/javascript>

2. <http://test.bclary.com/tests/mozilla.org/js/js-test-driver-standards.html?test=js1_5/Regress/regress-252892.js;language=type;text/javascript>

3. <http://test.bclary.com/tests/mozilla.org/js/js-test-driver-standards.html?test=js1_5/Regress/regress-321757.js;language=type;text/javascript>

Comment 1

[Johnny Stenback (:jst)]

Attachment: Don't resolve when binding names.

Comment 2

[Brendan Eich [:brendan]]

Comment on attachment 307366 [details] [diff] [review]
Don't resolve when binding names.

>+  if ((flags & (JSRESOLVE_ASSIGNING)) && (JSOp)*cx->fp->pc != JSOP_BINDNAME &&

It occurs that JSOP_BINDNAME should not have JOF_SET, aka JOF_ASSIGNING (which induces JSRESOLVE_ASSIGNING). But it should not be non-assigning either, an r-value context, since there is no "get" instead of "set" -- instead, there's a "bind" operation. Suggests JOF_BINDING, but that is for the future. For now, an explicit single-op special case is good enough.

/be

Comment 3

[Johnny Stenback (:jst)]

Fix checked in. Bob, please reopen or file a separate bug if there's still tests failing.

Comment 4

[Bob Clary [:bc] (inactive)]

v