Make importer.cc compile on Posix.

Make importer.cc compile on Posix.

Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=10528

[Evan Martin, 2009-02-26 22:58:34]

http://codereview.chromium.org/27242/diff/1/3
File chrome/browser/importer/importer.cc (right):

http://codereview.chromium.org/27242/diff/1/3#newcode579
Line 579: #if defined(OS_WIN)
Should we just not even have the MS_IE option in the ProfileType enum?

[James Hawkins, 2009-02-26 23:08:29]

http://codereview.chromium.org/27242/diff/1/3
File chrome/browser/importer/importer.cc (right):

http://codereview.chromium.org/27242/diff/1/3#newcode579
Line 579: #if defined(OS_WIN)
On 2009/02/26 22:58:34, Evan Martin wrote:
> Should we just not even have the MS_IE option in the ProfileType enum?

I can ifdef the MS_IE case out, but I'd have to add a default case that returns
NULL.  It seems like we don't want to have a catch-all like that because of the
NOTREACHED below.

[Evan Martin, 2009-02-26 23:11:48]

http://codereview.chromium.org/27242/diff/1/3
File chrome/browser/importer/importer.cc (right):

http://codereview.chromium.org/27242/diff/1/3#newcode579
Line 579: #if defined(OS_WIN)
On 2009/02/26 23:08:29, James Hawkins wrote:
> On 2009/02/26 22:58:34, Evan Martin wrote:
> > Should we just not even have the MS_IE option in the ProfileType enum?
> 
> I can ifdef the MS_IE case out, but I'd have to add a default case that
returns
> NULL.  It seems like we don't want to have a catch-all like that because of
the
> NOTREACHED below.

I meant, could you ifdef out the MS_IE case from the ProfileType enum?  (I don't
know if this would be ugly, it's just a thought that would expose other places
where MS_IE is assumed...)

[Evan Martin, 2009-02-26 23:31:40]

LG

[Evan Martin, 2009-02-26 23:31:54]

http://codereview.chromium.org/27242/diff/1007/1012
File chrome/browser/importer/importer.h (right):

http://codereview.chromium.org/27242/diff/1007/1012#newcode33
Line 33: FIREFOX2 = 1,
Is that necessary?  I think it's better to remove all explicit assignment of
these enum values (including the IE one).

[Mark Mentovai, 2009-02-27 01:25:43]

LGYPTM