Don't use AddSubstring for external natives

Don't use AddSubstring for external natives

The input string is not necessarily zero terminated

BUG=chromium:443230
R=adamk@chromium.org
LOG=n

Committed: https://crrev.com/a09168be68e09c8374829895ac335d4df7c65991
Cr-Commit-Position: refs/heads/master@{#25879}

[adamk, 2014-12-17 19:36:22]

It looks to me like the caller that's triggering this DCHECK,
NativesStore::NameFromId, is abusing the SimpleStringBuilder API. All other
callers of SimpleStringBuilder are using it with NUL-terminated strings, which
is what Finalize() returns and what AddSubstring is documented to deal with
("The input string must have enough characters"). NameFromId really just wants a
Vector.

It seems like it should be easy enough for NameForId to simply shove the correct
bytes into the Vector it's returning.

[jochen (gone - plz use gerrit), 2014-12-18 09:32:26]

jochen@chromium.org changed reviewers:
+ svenpanne@chromium.org

[jochen (gone - plz use gerrit), 2014-12-18 09:32:27]

fair enough

+svenpanne because timzeon

[Sven Panne, 2014-12-18 10:32:32]

lgtm

https://codereview.chromium.org/795383003/diff/20001/src/natives-external.cc
File src/natives-external.cc (right):

https://codereview.chromium.org/795383003/diff/20001/src/natives-external.cc#...
src/natives-external.cc:85: const char native[] = "native ";
Probably not a big deal here, but note that GCC actually copies characters here.
Clang is smarter here. :-/

[jochen (gone - plz use gerrit), 2014-12-18 10:33:57]

The CQ bit was checked by jochen@chromium.org

[commit-bot: I haz the power, 2014-12-18 10:34:09]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/795383003/20001

[commit-bot: I haz the power, 2014-12-18 11:01:49]

Committed patchset #2 (id:20001)

[commit-bot: I haz the power, 2014-12-18 18:05:21]

Patchset 2 (id:??) landed as
https://crrev.com/a09168be68e09c8374829895ac335d4df7c65991
Cr-Commit-Position: refs/heads/master@{#25879}