Backport SQLite memcmp patch.

Backport SQLite memcmp patch.

http://www.sqlite.org/src/info/d73435587b

Verified that the amalgamation came out with all the right patches by
comparing the amalgamation diff to the appropriately-ordered diffs of
the original files.

BUG=178677
Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=199345

[Scott Hess - ex-Googler, 2013-05-08 22:22:38]

I think you were the last to ask about this, so ...

[wtc, 2013-05-08 22:39:27]

Patch set 1 LGTM. Thanks.

By the way, when I told you a memcmp() implementation could
compare the bytes backward, I was wrong. I forgot that
memcmp() needs to report an ordering rather than just
equality. The ordering is defined as the ordering of the
first bytes that differ. So we need to compare bytes from
the beginning.

https://codereview.chromium.org/15070002/diff/1/third_party/sqlite/src/src/bu...
File third_party/sqlite/src/src/build.c (right):

https://codereview.chromium.org/15070002/diff/1/third_party/sqlite/src/src/bu...
third_party/sqlite/src/src/build.c:2483: &&
sqlite3StrNICmp(&pTab->zName[7],"altertab_",9)!=0 ){

Did you notice that he changed some string comparisons, such
as this one, to be case-insensitive?

https://codereview.chromium.org/15070002/diff/1/third_party/sqlite/src/src/ex...
File third_party/sqlite/src/src/expr.c (right):

https://codereview.chromium.org/15070002/diff/1/third_party/sqlite/src/src/ex...
third_party/sqlite/src/src/expr.c:582: n = sqlite3Strlen30(z);

The variable |n| can be removed. It is assigned a value here
but not used.

[Scott Hess - ex-Googler, 2013-05-08 22:56:51]

Removed the variable.  It still compiles, so what could go wrong?  :-).

https://codereview.chromium.org/15070002/diff/1/third_party/sqlite/src/src/bu...
File third_party/sqlite/src/src/build.c (right):

https://codereview.chromium.org/15070002/diff/1/third_party/sqlite/src/src/bu...
third_party/sqlite/src/src/build.c:2483: &&
sqlite3StrNICmp(&pTab->zName[7],"altertab_",9)!=0 ){
On 2013/05/08 22:39:27, wtc wrote:
> Did you notice that he changed some string comparisons, such
> as this one, to be case-insensitive?

Yeah.  I don't think it will matter to any of our code, fortunately.

https://codereview.chromium.org/15070002/diff/1/third_party/sqlite/src/src/ex...
File third_party/sqlite/src/src/expr.c (right):

https://codereview.chromium.org/15070002/diff/1/third_party/sqlite/src/src/ex...
third_party/sqlite/src/src/expr.c:582: n = sqlite3Strlen30(z);
On 2013/05/08 22:39:27, wtc wrote:
> The variable |n| can be removed. It is assigned a value here
> but not used.

This was one of the spots where I had to manually patch.  I'll remove with my
fingers crossed :-).

[commit-bot: I haz the power, 2013-05-09 21:43:40]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/shess@chromium.org/15070002/8001

[commit-bot: I haz the power, 2013-05-10 02:52:45]

Change committed as 199345