Add ProcessSnapshotMinidump, the beginning of the minidump variant of the Snapshot family

Add ProcessSnapshotMinidump, the beginning of the minidump variant of
the Snapshot family.

For the time being, only ProcessSnapshotMinidump::AnnotationsSimpleMap()
is implemented. In order to complete the initial uploader for Crashpad,
ModuleSnapshotMinidump::AnnotationsSimpleMap() is also needed, to be
accessed by ProcessSnapshotMinidump::Modules().

TEST=snapshot_test ProcessSnapshotMinidump.*
R=rsesek@chromium.org

Committed: https://chromium.googlesource.com/crashpad/crashpad/+/40b931bd8eabbe1f2d71ef6f885076cb74632ed6

[Mark Mentovai, 2015-02-18 22:56:33]

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

[Mark Mentovai, 2015-02-18 22:56:34]

This depends on https://codereview.chromium.org/913223008/.

[Robert Sesek, 2015-02-19 15:13:20]

https://codereview.chromium.org/932153003/diff/1/snapshot/minidump/minidump_s...
File snapshot/minidump/minidump_simple_string_dictionary_reader.h (right):

https://codereview.chromium.org/932153003/diff/1/snapshot/minidump/minidump_s...
snapshot/minidump/minidump_simple_string_dictionary_reader.h:37:
FileReaderInterface* file_reader,
Is it worth a note that the file_reader offset will remain unchanged? I'm not
clear why it's necessary to restore the SEEK_CUR though.

https://codereview.chromium.org/932153003/diff/1/snapshot/minidump/minidump_s...
snapshot/minidump/minidump_simple_string_dictionary_reader.h:39:
std::map<std::string, std::string>* dictionary);
No requirement for empty on entry? (Maybe not because this will be used to
aggregate all annotations from the process and the modules, but then how will we
deal with duplicate keys?).

https://codereview.chromium.org/932153003/diff/1/snapshot/minidump/process_sn...
File snapshot/minidump/process_snapshot_minidump.cc (right):

https://codereview.chromium.org/932153003/diff/1/snapshot/minidump/process_sn...
snapshot/minidump/process_snapshot_minidump.cc:164: const auto it =
stream_map_.find(kMinidumpStreamTypeCrashpadInfo);
Make this a reference, too?

https://codereview.chromium.org/932153003/diff/1/snapshot/minidump/process_sn...
File snapshot/minidump/process_snapshot_minidump_test.cc (right):

https://codereview.chromium.org/932153003/diff/1/snapshot/minidump/process_sn...
snapshot/minidump/process_snapshot_minidump_test.cc:88: const char kValue1[] =
"THE FIRST VALUE EVER!";
SECOND VALUE EVER?

[Mark Mentovai, 2015-02-26 18:42:19]

Patchset #5 (id:80001) has been deleted

[Mark Mentovai, 2015-02-26 18:43:03]

Updated

https://codereview.chromium.org/932153003/diff/1/snapshot/minidump/minidump_s...
File snapshot/minidump/minidump_simple_string_dictionary_reader.h (right):

https://codereview.chromium.org/932153003/diff/1/snapshot/minidump/minidump_s...
snapshot/minidump/minidump_simple_string_dictionary_reader.h:37:
FileReaderInterface* file_reader,
Robert Sesek wrote:
> Is it worth a note that the file_reader offset will remain unchanged? I'm not
> clear why it's necessary to restore the SEEK_CUR though.

I’m getting rid of all of the restoration stuff. I thought it might be useful,
but it’s just becoming pervasive, so I’m leaving it out until it proves itself
necessary, and even then it’l probably be better handled via a scoper.

[Robert Sesek, 2015-02-26 19:17:19]

LGTM

[Mark Mentovai, 2015-02-26 19:43:31]

Committed patchset #5 (id:100001) manually as
40b931bd8eabbe1f2d71ef6f885076cb74632ed6 (presubmit successful).