Add MachOImageSegmentReader

Add MachOImageSegmentReader.

This is part of a family of classes whose chief entry point is
MachOImageReader. MachOImageReader depends on this class to read
segments, so this one is landing first. The bulk of this class will be
tested as part of MachOImageReader.

TEST=util_test MachOImageSegmentReader.*
R=rsesek@chromium.org

Committed: https://chromium.googlesource.com/crashpad/crashpad/+/1460c47d6ad6c93b49f9a934630e0f48fd687ce0

[Mark Mentovai, 2014-08-28 18:42:36]

mark@chromium.org changed reviewers:
+ rsesek@chromium.org

[Robert Sesek, 2014-09-02 16:40:09]

No test?

https://codereview.chromium.org/516983003/diff/1/util/mac/mach_o_image_segmen...
File util/mac/mach_o_image_segment_reader.cc (right):

https://codereview.chromium.org/516983003/diff/1/util/mac/mach_o_image_segmen...
util/mac/mach_o_image_segment_reader.cc:193: CHECK(index < sections_.size());
CHECK_LT?

https://codereview.chromium.org/516983003/diff/1/util/mac/mach_o_image_segmen...
File util/mac/mach_o_image_segment_reader.h (right):

https://codereview.chromium.org/516983003/diff/1/util/mac/mach_o_image_segmen...
util/mac/mach_o_image_segment_reader.h:59: mach_vm_address_t
load_command_address,
It's not clear to me where I'm supposed to get this address.

[Mark Mentovai, 2014-09-02 16:49:13]

rsesek wrote:
> No test?

As the CL description says, it’ll be tested as part of MachOImageReader.

I didn’t really think it was worthwhile to test it alone. MachOImageReader
(which will probably be the next review) depends on this class and makes pretty
heavy use of it.

[Robert Sesek, 2014-09-02 16:50:17]

On 2014/09/02 16:49:13, Mark Mentovai wrote:
> rsesek wrote:
> > No test?
> 
> As the CL description says, it’ll be tested as part of MachOImageReader.

What about the pieces that format section names? Will those be tested there,
too?

> I didn’t really think it was worthwhile to test it alone. MachOImageReader
> (which will probably be the next review) depends on this class and makes
pretty
> heavy use of it.

[Mark Mentovai, 2014-09-02 17:44:15]

Good point, the tests for that can be here.

[Robert Sesek, 2014-09-02 18:00:13]

LGTM

https://codereview.chromium.org/516983003/diff/60001/util/mac/mach_o_image_se...
File util/mac/mach_o_image_segment_reader_test.cc (right):

https://codereview.chromium.org/516983003/diff/60001/util/mac/mach_o_image_se...
util/mac/mach_o_image_segment_reader_test.cc:179: << base::StringPrintf("index
%zu, segment %s, section %s",
Not SCOPED_TRACE?

[Mark Mentovai, 2014-09-02 18:01:05]

https://codereview.chromium.org/516983003/diff/60001/util/mac/mach_o_image_se...
File util/mac/mach_o_image_segment_reader_test.cc (right):

https://codereview.chromium.org/516983003/diff/60001/util/mac/mach_o_image_se...
util/mac/mach_o_image_segment_reader_test.cc:179: << base::StringPrintf("index
%zu, segment %s, section %s",
rsesek wrote:
> Not SCOPED_TRACE?

Doesn’t really matter one way or the other when there’s only one gtesting thing
in the scope.

[Mark Mentovai, 2014-09-02 19:50:14]

Committed patchset #4 (id:60001) manually as 1460c47 (presubmit successful).