Remove CGW_ArrayTemplate and its O(n^2 log n) sort.

Remove CGW_ArrayTemplate and its O(n^2 log n) sort.

R=thestig@chromium.org

Committed: https://pdfium.googlesource.com/pdfium/+/0841b0f37678ba4962247f5636e9390718fc027e

[Tom Sepez, 2016-02-01 18:40:08]

tsepez@chromium.org changed reviewers:
+ thestig@chromium.org

[Tom Sepez, 2016-02-01 18:40:08]

Lei, for review.  Now that we have tests for the two uses of this, we can remove
it.

[Lei Zhang, 2016-02-01 19:12:55]

https://codereview.chromium.org/1652613002/diff/40001/fpdfsdk/src/fsdk_basefo...
File fpdfsdk/src/fsdk_baseform.cpp (right):

https://codereview.chromium.org/1652613002/diff/40001/fpdfsdk/src/fsdk_basefo...
fpdfsdk/src/fsdk_baseform.cpp:2108: auto iter = std::find(m_Annots.begin(),
m_Annots.end(), pAnnot);
IYWU paranoia: #include <algorithm>

https://codereview.chromium.org/1652613002/diff/40001/fpdfsdk/src/fsdk_basefo...
fpdfsdk/src/fsdk_baseform.cpp:2154: if (sa.empty())
We can probably just omit this?

https://codereview.chromium.org/1652613002/diff/40001/fpdfsdk/src/fsdk_basefo...
fpdfsdk/src/fsdk_baseform.cpp:2161: for (int i = sa.size() - 1; i >= 0; i--) {
for (auto it = sa.rbegin(); it != sa.rend(); ++it) here, and nLeftTopIndex = it
- sa.begin() below, or make |nLeftTopIndex| an iterator as well.

https://codereview.chromium.org/1652613002/diff/40001/fpdfsdk/src/fsdk_basefo...
fpdfsdk/src/fsdk_baseform.cpp:2175: for (int i = 0; i < sa.size(); ++i) {
As is, shadow variable |i| here and in the for loops below.

https://codereview.chromium.org/1652613002/diff/40001/fpdfsdk/src/javascript/...
File fpdfsdk/src/javascript/Field.cpp (right):

https://codereview.chromium.org/1652613002/diff/40001/fpdfsdk/src/javascript/...
fpdfsdk/src/javascript/Field.cpp:3094:
std::vector<std::unique_ptr<CFX_WideString>> swSort;
#include <vector>, <memory> and <algorithm>

[Tom Sepez, 2016-02-01 20:14:23]

https://codereview.chromium.org/1652613002/diff/40001/fpdfsdk/src/fsdk_basefo...
File fpdfsdk/src/fsdk_baseform.cpp (right):

https://codereview.chromium.org/1652613002/diff/40001/fpdfsdk/src/fsdk_basefo...
fpdfsdk/src/fsdk_baseform.cpp:2108: auto iter = std::find(m_Annots.begin(),
m_Annots.end(), pAnnot);
On 2016/02/01 19:12:55, Lei Zhang wrote:
> IYWU paranoia: #include <algorithm>

Done.

https://codereview.chromium.org/1652613002/diff/40001/fpdfsdk/src/fsdk_basefo...
fpdfsdk/src/fsdk_baseform.cpp:2154: if (sa.empty())
On 2016/02/01 19:12:55, Lei Zhang wrote:
> We can probably just omit this?

Done.

https://codereview.chromium.org/1652613002/diff/40001/fpdfsdk/src/fsdk_basefo...
fpdfsdk/src/fsdk_baseform.cpp:2161: for (int i = sa.size() - 1; i >= 0; i--) {
On 2016/02/01 19:12:55, Lei Zhang wrote:
> for (auto it = sa.rbegin(); it != sa.rend(); ++it) here, and nLeftTopIndex =
it
> - sa.begin() below, or make |nLeftTopIndex| an iterator as well.

Meh, I think both of those are more complicated to understand.

https://codereview.chromium.org/1652613002/diff/40001/fpdfsdk/src/fsdk_basefo...
fpdfsdk/src/fsdk_baseform.cpp:2175: for (int i = 0; i < sa.size(); ++i) {
On 2016/02/01 19:12:55, Lei Zhang wrote:
> As is, shadow variable |i| here and in the for loops below.

Done.

https://codereview.chromium.org/1652613002/diff/40001/fpdfsdk/src/fsdk_basefo...
fpdfsdk/src/fsdk_baseform.cpp:2175: for (int i = 0; i < sa.size(); ++i) {
On 2016/02/01 19:12:55, Lei Zhang wrote:
> As is, shadow variable |i| here and in the for loops below.

Huh?  Aren't yhe other |i|s are already out of scope when we introduce |i| here?

https://codereview.chromium.org/1652613002/diff/40001/fpdfsdk/src/javascript/...
File fpdfsdk/src/javascript/Field.cpp (right):

https://codereview.chromium.org/1652613002/diff/40001/fpdfsdk/src/javascript/...
fpdfsdk/src/javascript/Field.cpp:3094:
std::vector<std::unique_ptr<CFX_WideString>> swSort;
On 2016/02/01 19:12:55, Lei Zhang wrote:
> #include <vector>, <memory> and <algorithm>

Done.

[Lei Zhang, 2016-02-01 21:18:15]

lgtm

https://codereview.chromium.org/1652613002/diff/40001/fpdfsdk/src/fsdk_basefo...
File fpdfsdk/src/fsdk_baseform.cpp (right):

https://codereview.chromium.org/1652613002/diff/40001/fpdfsdk/src/fsdk_basefo...
fpdfsdk/src/fsdk_baseform.cpp:2161: for (int i = sa.size() - 1; i >= 0; i--) {
On 2016/02/01 20:14:22, Tom Sepez wrote:
> On 2016/02/01 19:12:55, Lei Zhang wrote:
> > for (auto it = sa.rbegin(); it != sa.rend(); ++it) here, and nLeftTopIndex =
> it
> > - sa.begin() below, or make |nLeftTopIndex| an iterator as well.
> 
> Meh, I think both of those are more complicated to understand.

I was thinking of avoiding int vs size_t, but it looks like we don't do
-Wsign-conversion in Chromium either.

https://codereview.chromium.org/1652613002/diff/40001/fpdfsdk/src/fsdk_basefo...
fpdfsdk/src/fsdk_baseform.cpp:2175: for (int i = 0; i < sa.size(); ++i) {
On 2016/02/01 20:14:22, Tom Sepez wrote:
> On 2016/02/01 19:12:55, Lei Zhang wrote:
> > As is, shadow variable |i| here and in the for loops below.
> 
> Huh?  Aren't yhe other |i|s are already out of scope when we introduce |i|
here?

n/m, I read the diff wrong somewhere.

https://codereview.chromium.org/1652613002/diff/60001/fpdfsdk/src/fsdk_basefo...
File fpdfsdk/src/fsdk_baseform.cpp (right):

https://codereview.chromium.org/1652613002/diff/60001/fpdfsdk/src/fsdk_basefo...
fpdfsdk/src/fsdk_baseform.cpp:2195: if (sa.empty())
Here too.

[Tom Sepez, 2016-02-01 21:37:41]

Description was changed from

==========
Remove CGW_ArrayTemplate and its O(n^2 log n) sort.
==========

to

==========
Remove CGW_ArrayTemplate and its O(n^2 log n) sort.

R=thestig@chromium.org

Committed:
https://pdfium.googlesource.com/pdfium/+/0841b0f37678ba4962247f5636e9390718fc...
==========

[Tom Sepez, 2016-02-01 21:37:42]

Committed patchset #5 (id:80001) manually as
0841b0f37678ba4962247f5636e9390718fc027e (presubmit successful).

[Tom Sepez, 2016-02-01 21:40:33]

https://codereview.chromium.org/1652613002/diff/60001/fpdfsdk/src/fsdk_basefo...
File fpdfsdk/src/fsdk_baseform.cpp (right):

https://codereview.chromium.org/1652613002/diff/60001/fpdfsdk/src/fsdk_basefo...
fpdfsdk/src/fsdk_baseform.cpp:2195: if (sa.empty())
On 2016/02/01 21:18:15, Lei Zhang wrote:
> Here too.

Done.