Add compat to Crashpad

Add compat to Crashpad.

compat includes headers providing definitions normally provided by the
system, in cases where the system SDK does not always provide the
correct or up-to-date definitions, and cases where code on different
platforms needs to access definitions normally only available on one
platform.

To provide definitions on a single platform, where the system SDK may
not provide the definitions correctly, use subdirectories named for the
platform, such as “mac”.

To provide definitions normally available on only one platform to
others, use subdirectories that identify that they are to be used on
platforms other than the one that originated their definitions, such as
“non_win”.

In all cases, headers should be named as they are natively in their
respective SDKs, so that it’s possible to #include them according to
their usual names.

R=rsesek@chromium.org

Committed: https://chromium.googlesource.com/crashpad/crashpad/+/dfe81014f7a7

[Mark Mentovai, 2014-07-31 16:02:21]

Here are the first 4k lines or so. Most other pieces will be smaller than this,
but I don’t see much value in splitting this up.

[Robert Sesek, 2014-07-31 16:55:41]

https://codereview.chromium.org/432843002/diff/1/compat/compat.gyp
File compat/compat.gyp (right):

https://codereview.chromium.org/432843002/diff/1/compat/compat.gyp#newcode18
compat/compat.gyp:18: 'target_name': 'compat',
May want to re-upload with --no-find-copies.

https://codereview.chromium.org/432843002/diff/1/compat/mac/mach-o/loader.h
File compat/mac/mach-o/loader.h (right):

https://codereview.chromium.org/432843002/diff/1/compat/mac/mach-o/loader.h#n...
compat/mac/mach-o/loader.h:1: // Copyright 2014 The Crashpad Authors. All rights
reserved.
Same comment about --no-find-copies

https://codereview.chromium.org/432843002/diff/1/compat/non_win/dbghelp.h
File compat/non_win/dbghelp.h (right):

https://codereview.chromium.org/432843002/diff/1/compat/non_win/dbghelp.h#new...
compat/non_win/dbghelp.h:25: //! \file
Not sure you need a file-level comment here, if not, do you need this Doxygen
tag?

https://codereview.chromium.org/432843002/diff/1/compat/non_win/dbghelp.h#new...
compat/non_win/dbghelp.h:76: //! This structure must be present at the beginning
of a minidump file (at ::RVA
Does Doxygen do something special to ::RVA, or is it just your convention? I
believe there's a \see or \c directive that can hyperlink in the generated docs
if it doesn't -- something to consider.

https://codereview.chromium.org/432843002/diff/1/compat/non_win/dbghelp.h#new...
compat/non_win/dbghelp.h:120: //! \brief A variable-length UTF-16-encoded string
carried within a minidump
UTF-16 LE or BE, or is it the same as the minidump file itself?

https://codereview.chromium.org/432843002/diff/1/compat/non_win/dbghelp.h#new...
compat/non_win/dbghelp.h:619: //!  - MINIDUMP_MISC_INFO::ProcessCreateTime
Does Doxygen automatically turn these into bulleted lists?

https://codereview.chromium.org/432843002/diff/1/compat/non_win/dbghelp.h#new...
compat/non_win/dbghelp.h:748: //! This field is Windows-specific, and has no
meaning on other operating
Process UID maybe? Would only be interesting if < 500 probably.

https://codereview.chromium.org/432843002/diff/1/compat/non_win/dbghelp.h#new...
compat/non_win/dbghelp.h:763: //! This field is Windows-specific, and has no
meaning on other operating
Could be a sandbox check on Mac.

https://codereview.chromium.org/432843002/diff/1/crashpad.doxy
File crashpad.doxy (right):

https://codereview.chromium.org/432843002/diff/1/crashpad.doxy#newcode1
crashpad.doxy:1: # Doxyfile 1.8.7
I didn't review this file all that carefully.

[Mark Mentovai, 2014-07-31 17:06:55]

https://codereview.chromium.org/432843002/diff/1/compat/non_win/dbghelp.h
File compat/non_win/dbghelp.h (right):

https://codereview.chromium.org/432843002/diff/1/compat/non_win/dbghelp.h#new...
compat/non_win/dbghelp.h:25: //! \file
rsesek wrote:
> Not sure you need a file-level comment here, if not, do you need this Doxygen
> tag?

Doxygen won’t show macro documentation unless the file has \file in it, which
kind of sucks.

http://www.doxygen.nl/manual/docblocks.html, search for “Let's repeat that”.

https://codereview.chromium.org/432843002/diff/1/compat/non_win/dbghelp.h#new...
compat/non_win/dbghelp.h:76: //! This structure must be present at the beginning
of a minidump file (at ::RVA
rsesek wrote:
> Does Doxygen do something special to ::RVA, or is it just your convention? I
> believe there's a \see or \c directive that can hyperlink in the generated
docs
> if it doesn't -- something to consider.

It doesn’t linkify RVA unless the colons are present. The colons are the most
succinct way to get it to do the right thing.

https://codereview.chromium.org/432843002/diff/1/compat/non_win/dbghelp.h#new...
compat/non_win/dbghelp.h:619: //!  - MINIDUMP_MISC_INFO::ProcessCreateTime
rsesek wrote:
> Does Doxygen automatically turn these into bulleted lists?

Yes, it has markdown support (http://doxygen.nl/manual/markdown.html). I can
stick the generated docs somewhere if you’d like to see them.

https://codereview.chromium.org/432843002/diff/1/compat/non_win/dbghelp.h#new...
compat/non_win/dbghelp.h:748: //! This field is Windows-specific, and has no
meaning on other operating
rsesek wrote:
> Process UID maybe? Would only be interesting if < 500 probably.

I think we’ll wind up defining better fields to hold this…

https://codereview.chromium.org/432843002/diff/1/compat/non_win/dbghelp.h#new...
compat/non_win/dbghelp.h:763: //! This field is Windows-specific, and has no
meaning on other operating
rsesek wrote:
> Could be a sandbox check on Mac.

…and this. I don’t know how much more recycling of existing fields we’ll want to
do. There’s a bunch more data I’d ultimately like to capture.

https://codereview.chromium.org/432843002/diff/1/crashpad.doxy
File crashpad.doxy (right):

https://codereview.chromium.org/432843002/diff/1/crashpad.doxy#newcode1
crashpad.doxy:1: # Doxyfile 1.8.7
rsesek wrote:
> I didn't review this file all that carefully.

Yeah, I thought about checking in a virgin template first and then letting you
review the diffs in this change. Would you like me to do that?

[Robert Sesek, 2014-07-31 17:25:36]

LGTM

https://codereview.chromium.org/432843002/diff/1/compat/non_win/dbghelp.h
File compat/non_win/dbghelp.h (right):

https://codereview.chromium.org/432843002/diff/1/compat/non_win/dbghelp.h#new...
compat/non_win/dbghelp.h:76: //! This structure must be present at the beginning
of a minidump file (at ::RVA
On 2014/07/31 17:06:54, Mark Mentovai wrote:
> rsesek wrote:
> > Does Doxygen do something special to ::RVA, or is it just your convention? I
> > believe there's a \see or \c directive that can hyperlink in the generated
> docs
> > if it doesn't -- something to consider.
> 
> It doesn’t linkify RVA unless the colons are present. The colons are the most
> succinct way to get it to do the right thing.

Got it, I thought that might be the case.

https://codereview.chromium.org/432843002/diff/1/crashpad.doxy
File crashpad.doxy (right):

https://codereview.chromium.org/432843002/diff/1/crashpad.doxy#newcode1
crashpad.doxy:1: # Doxyfile 1.8.7
On 2014/07/31 17:06:55, Mark Mentovai wrote:
> rsesek wrote:
> > I didn't review this file all that carefully.
> 
> Yeah, I thought about checking in a virgin template first and then letting you
> review the diffs in this change. Would you like me to do that?

No, I think it's fine, since this isn't really code.

[Mark Mentovai, 2014-07-31 17:45:55]

Committed patchset #3 manually as rdfe81014f7a7 (presubmit successful).