This isn't strictly necessary at this point. The basic issue is that there are
a few new classes (iterator stuff, intl stuff, and typed objects) that use this
kind of constructor bootstrap scheme, and each of them will need something like
this to move to ClassSpecs. But that doesn't actually need to happen until we
convert them. I wrote this patch before I realized this, and I believe it's
correct, so I figured I'd get it landed along with everything. But I'm also ok
to skip it for now.
Bug 924839 - Remove a patch already part of ICU 52.1. See http://bugs.icu-project.org/trac/ticket/10283 but also note the relevant code was removed completely upstream. r=glandium
* * *
Bug 924839 - Remove another patch already part of ICU 52.1. See http://bugs.icu-project.org/trac/ticket/10290 for that. r=gaston
* * *
Bug 924839 - Remove another patch already in ICU 52.1. See http://bugs.icu-project.org/trac/ticket/10045 for more. r=Norbert
* * *
Bug 924839 - Remove another patch already applied upstream. See http://bugs.icu-project.org/trac/changeset/32937 for more. r=gaston
* * *
Bug 924839 - Update the ICU update script to update to 52.1, *without* applying any of our local patches. r=glandium
* * *
Bug 924839 - Make the ICU update script only do updating within intl/icu/source and nowhere else. r=glandium
* * *
Bug 924839 - Implement the changes that would be made by |cd intl/; ./update-icu.sh http://source.icu-project.org/repos/icu/icu/tags/release-52-1/;|, run with the prior changesets' changes made (thus not applying any of our local patches). These changes don't actually work without subsequent adjustments, but this provides a codebase upon which those adjustments can be made, for the purpose of generating local patches to be kept in intl/icu-patches/. rs=the-usual-suspects
* * *
Bug 924839 - Update the bug 899722 local patch to make runConfigureICU not override CC/CXX on BSD systems. r=gaston
* * *
Bug 924839 - Update the bug 724533 patch that makes ICU builds with MozillaBuild on Windows. r=glandium
* * *
Bug 924839 - Import an upstream patch fixing the genrb tool to properly handle the -R (--omitCollationRules) option. See http://bugs.icu-project.org/trac/ticket/10043 for the original bug report and a link to the ultimate upstream landing. r=Norbert
* * *
Bug 924839 - Import the upstream fix for http://bugs.icu-project.org/trac/ticket/10486 so that ICU with -DU_USING_ICU_NAMESPACE=0 will compile on Windows. r=Norbert
* * *
Bug 924839 - Adjust the update script to update ICU, then to apply all local patches (rather than skipping the second step). Thus if the update script is properly run, now, the final result should be no changes at all to the tree. NOT REVIEWED YET
* * *
Bug 924839 - Update jstests that depend on CLDR locale data to match CLDR 24. r=Norbert
Bug 924839 - Remove a patch already part of ICU 52.1. See http://bugs.icu-project.org/trac/ticket/10283 but also note the relevant code was removed completely upstream. r=glandium
* * *
Bug 924839 - Remove another patch already part of ICU 52.1. See http://bugs.icu-project.org/trac/ticket/10290 for that. r=gaston
* * *
Bug 924839 - Remove another patch already in ICU 52.1. See http://bugs.icu-project.org/trac/ticket/10045 for more. r=Norbert
* * *
Bug 924839 - Remove another patch already applied upstream. See http://bugs.icu-project.org/trac/changeset/32937 for more. r=gaston
* * *
Bug 924839 - Update the ICU update script to update to 52.1, *without* applying any of our local patches. r=glandium
* * *
Bug 924839 - Make the ICU update script only do updating within intl/icu/source and nowhere else. r=glandium
* * *
Bug 924839 - Implement the changes that would be made by |cd intl/; ./update-icu.sh http://source.icu-project.org/repos/icu/icu/tags/release-52-1/;|, run with the prior changesets' changes made (thus not applying any of our local patches). These changes don't actually work without subsequent adjustments, but this provides a codebase upon which those adjustments can be made, for the purpose of generating local patches to be kept in intl/icu-patches/. rs=the-usual-suspects
* * *
Bug 924839 - Update the bug 899722 local patch to make runConfigureICU not override CC/CXX on BSD systems. r=gaston
* * *
Bug 924839 - Update the bug 724533 patch that makes ICU builds with MozillaBuild on Windows. r=glandium
* * *
Bug 924839 - Import an upstream patch fixing the genrb tool to properly handle the -R (--omitCollationRules) option. See http://bugs.icu-project.org/trac/ticket/10043 for the original bug report and a link to the ultimate upstream landing. r=Norbert
* * *
Bug 924839 - Import the upstream fix for http://bugs.icu-project.org/trac/ticket/10486 so that ICU with -DU_USING_ICU_NAMESPACE=0 will compile on Windows. r=Norbert
* * *
Bug 924839 - Adjust the update script to update ICU, then to apply all local patches (rather than skipping the second step). Thus if the update script is properly run, now, the final result should be no changes at all to the tree. NOT REVIEWED YET
* * *
Bug 924839 - Update jstests that depend on CLDR locale data to match CLDR 24. r=Norbert
--HG--
extra : rebase_source : e993b6e4a507f92d1b1a24949185d00d1f6201f2
This lets js_ReportUncaughtException get the report directly from the underlying
Error object, rather than trying to duck-type it (which fails for security
wrappers).
These iteration counts were selected to be close to the minimum number
necessary to expose bugs while working on this patch (see comments in
Bug 961821 for more details here). Note that they are all 500
iterations (not 5000); please be wary of reducing the iteration count
further without first ensuring that the jit has time to compile this
code.
As part of above, alpha-renamed IonBuilder::setElemTryTyped to
IonBuilder::setElemTryTypedArray to keep clear the current distinction
between TypedObject and TypedArray.
Drive-by fix: Reference to Bug 894105 in comment had off-by-one typo.
Bug 961821: jit-support for writes into TypedObject arrays (r=nmatsakis).
As part of above, alpha-renamed IonBuilder::setElemTryTyped to
IonBuilder::setElemTryTypedArray to keep clear the current distinction
between TypedObject and TypedArray.
Drive-by fix: Reference to Bug 894105 in comment had off-by-one typo.
The former triggers all of the appropriate initialization, whereas the latter
is 'imaginary' in jsprototypes.h, which means that the is no initialization
function associated with it.
