mini_installer: Change to use _countof rather than base's arraysize macro.

Replace mini_installer's use of arraysize macro with range-based for loops
and/or _countof.

* Remove the duplication in the decompress.h

https://msdn.microsoft.com/en-us/library/ms175773.aspx

This issue arised when doing https://codereview.chromium.org/1115993004/.

BUG=None
TEST=trybots
R=grt@chromium.org

Committed: https://crrev.com/016d71e2e048f0578e2c53c3f741755173ef8cb2
Cr-Commit-Position: refs/heads/master@{#328623}

[tfarina, 2015-05-06 15:54:55]

PTAL.

[grt (UTC plus 2), 2015-05-06 16:46:34]

Thanks for the cleanup.

https://codereview.chromium.org/1126163003/diff/1/chrome/installer/mini_insta...
File chrome/installer/mini_installer/configuration_test.cc (right):

https://codereview.chromium.org/1126163003/diff/1/chrome/installer/mini_insta...
chrome/installer/mini_installer/configuration_test.cc:63: for (size_t i = 0; i <
_countof(kCommandLines); ++i) {
use new C++ here:
  for (const wchar_t* command_line : kCommandLines) {

https://codereview.chromium.org/1126163003/diff/1/chrome/installer/mini_insta...
File chrome/installer/mini_installer/decompress.cc (right):

https://codereview.chromium.org/1126163003/diff/1/chrome/installer/mini_insta...
chrome/installer/mini_installer/decompress.cc:194: wchar_t path[MAX_PATH] = {0};
change this to:
    static const DWORD kBufferSize = MAX_PATH;
    wchar_t path[kBufferSize];

and then use kBufferSize in place of _countof(path) below.

https://codereview.chromium.org/1126163003/diff/1/chrome/installer/mini_insta...
chrome/installer/mini_installer/decompress.cc:195: for (int i = 0; i <
_countof(candidate_paths); ++i) {
this also looks like a job for:
    for (const wchar_t* candidate_path : candidate_paths) {

https://codereview.chromium.org/1126163003/diff/1/chrome/installer/mini_insta...
File chrome/installer/mini_installer/mini_installer.cc (right):

https://codereview.chromium.org/1126163003/diff/1/chrome/installer/mini_insta...
chrome/installer/mini_installer/mini_installer.cc:742: for (int i = 0; i <
_countof(kDirectoryPrefixes); ++i) {
  for (const wchar_t* directory_prefix : kDirectoryPrefixes) {

[tfarina, 2015-05-06 17:21:56]

https://codereview.chromium.org/1126163003/diff/1/chrome/installer/mini_insta...
File chrome/installer/mini_installer/configuration_test.cc (right):

https://codereview.chromium.org/1126163003/diff/1/chrome/installer/mini_insta...
chrome/installer/mini_installer/configuration_test.cc:63: for (size_t i = 0; i <
_countof(kCommandLines); ++i) {
On 2015/05/06 16:46:34, grt wrote:
> use new C++ here:
>   for (const wchar_t* command_line : kCommandLines) {

Done.

https://codereview.chromium.org/1126163003/diff/1/chrome/installer/mini_insta...
File chrome/installer/mini_installer/decompress.cc (right):

https://codereview.chromium.org/1126163003/diff/1/chrome/installer/mini_insta...
chrome/installer/mini_installer/decompress.cc:194: wchar_t path[MAX_PATH] = {0};
On 2015/05/06 16:46:34, grt wrote:
> change this to:
>     static const DWORD kBufferSize = MAX_PATH;
>     wchar_t path[kBufferSize];
> 
> and then use kBufferSize in place of _countof(path) below.

Done.

https://codereview.chromium.org/1126163003/diff/1/chrome/installer/mini_insta...
chrome/installer/mini_installer/decompress.cc:195: for (int i = 0; i <
_countof(candidate_paths); ++i) {
On 2015/05/06 16:46:34, grt wrote:
> this also looks like a job for:
>     for (const wchar_t* candidate_path : candidate_paths) {

Done.

https://codereview.chromium.org/1126163003/diff/1/chrome/installer/mini_insta...
File chrome/installer/mini_installer/mini_installer.cc (right):

https://codereview.chromium.org/1126163003/diff/1/chrome/installer/mini_insta...
chrome/installer/mini_installer/mini_installer.cc:742: for (int i = 0; i <
_countof(kDirectoryPrefixes); ++i) {
On 2015/05/06 16:46:34, grt wrote:
>   for (const wchar_t* directory_prefix : kDirectoryPrefixes) {

Done.

[grt (UTC plus 2), 2015-05-06 18:10:15]

please change the description to something like:

Replace mini_installer's use of arraysize macro with range-based for loops
and/or _countof.

https://codereview.chromium.org/1126163003/diff/20001/chrome/installer/mini_i...
File chrome/installer/mini_installer/configuration_test.cc (right):

https://codereview.chromium.org/1126163003/diff/20001/chrome/installer/mini_i...
chrome/installer/mini_installer/configuration_test.cc:7: #include <stdlib.h>
is this needed now?

https://codereview.chromium.org/1126163003/diff/20001/chrome/installer/mini_i...
File chrome/installer/mini_installer/decompress.cc (right):

https://codereview.chromium.org/1126163003/diff/20001/chrome/installer/mini_i...
chrome/installer/mini_installer/decompress.cc:8: #include <stdlib.h>
is this still needed?

[tfarina, 2015-05-06 18:16:08]

CL description updated.

https://codereview.chromium.org/1126163003/diff/20001/chrome/installer/mini_i...
File chrome/installer/mini_installer/configuration_test.cc (right):

https://codereview.chromium.org/1126163003/diff/20001/chrome/installer/mini_i...
chrome/installer/mini_installer/configuration_test.cc:7: #include <stdlib.h>
On 2015/05/06 18:10:15, grt wrote:
> is this needed now?

Done.

https://codereview.chromium.org/1126163003/diff/20001/chrome/installer/mini_i...
File chrome/installer/mini_installer/decompress.cc (right):

https://codereview.chromium.org/1126163003/diff/20001/chrome/installer/mini_i...
chrome/installer/mini_installer/decompress.cc:8: #include <stdlib.h>
On 2015/05/06 18:10:15, grt wrote:
> is this still needed?

Done.

[grt (UTC plus 2), 2015-05-06 18:19:56]

lgtm

[tfarina, 2015-05-06 21:57:20]

The CQ bit was checked by tfarina@chromium.org

[commit-bot: I haz the power, 2015-05-06 21:58:55]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1126163003/40001

[commit-bot: I haz the power, 2015-05-06 22:05:27]

Committed patchset #3 (id:40001)

[commit-bot: I haz the power, 2015-05-06 22:06:21]

Patchset 3 (id:??) landed as
https://crrev.com/016d71e2e048f0578e2c53c3f741755173ef8cb2
Cr-Commit-Position: refs/heads/master@{#328623}

[grt (UTC plus 2), 2015-05-07 13:28:13]

A revert of this CL (patchset #3 id:40001) has been created in
https://codereview.chromium.org/1133613002/ by grt@chromium.org.

The reason for reverting is: Broke signing build..

[tfarina, 2015-05-07 13:43:44]

Greg, what was the error? Could you send me a link to it?

How do you suggest we fix it when relanding?

[grt (UTC plus 2), 2015-05-07 13:52:39]

On 2015/05/07 13:43:44, tfarina wrote:
> Greg, what was the error? Could you send me a link to it?
> 
> How do you suggest we fix it when relanding?

It turns out that we still can't use C++-11 features in
chrome/installer/mini_installer, much to my chagrin. You could use _countof
everywhere instead of range-based for loops.