Rebaseline block-test* tests for Mac....

Rebaseline block-test* tests for Mac.

On Mac, when trying to connect to 255.255.255.255, we get an EACCES error.  I
don't think it is correct for me to map that to ERR_ADDRESS_INVALID or
ERR_ADDRESS_UNREACHABLE, so I'm going to map it to ERR_ACCESS_DENIED.  This
means that I need to rebaseline the chromium-mac results for these tests.

I think it is okay to have platform specific behavior here.  I kind of prefer
that the error codes we show are more closely in sync with the operating
system.  That way it may be easier for a sophisticated user to make sense of
our error codes.

R=wtc
BUG=18665
TEST=none


Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=23176

[wtc, 2009-08-07 02:13:20]

LGTM, but let's map EACCES to ERR_ACCESS_DENIED.  ERR_FAILED
is bad.

Here is the connect(2) man page for Mac OS X:
http://developer.apple.com/documentation/Darwin/Reference/Manpages/man2/conne...

It says:
  [EACCES]    The destination address is a broadcast
              address and the socket option SO_BROADCAST
              is not set.

Is 255.255.255.255 a broadcast address?  Is there a
non-broadcast address that's guaranteed to be unreachable?
We may be able to get a consistent error code across
platforms with that kind of address.

[Peter Kasting, 2009-08-07 02:16:25]

On Thu, Aug 6, 2009 at 7:13 PM, <wtc@chromium.org> wrote:

> Is 255.255.255.255 a broadcast address?


IIRC there's a block of addresses with the first octet being the top end of
the range (248 - 255 or something similar) that are all broadcast.

PK


> http://codereview.chromium.org/164114
>

[wtc, 2009-08-07 02:25:25]

Another option:

In http://codereview.chromium.org/159904, we're
adding our first context-specific error mapping
function, MapConnectError.

So you could map EACCES to ERR_ADDRESS_INVALID or
ERR_ADDRESS_UNREACHABLE in MapConnectError in
tcp_client_socket_libevent.cc (assuming you think they
are better than ERR_ACCESS_DENIED).

[wtc, 2009-08-07 02:28:10]

I forgot to say -- map EACCES to ERR_ACCESS_DENIED in
MapPosixError, and optionally (if you like) map EACCES
to ERR_ADDRESS_INVALID or ERR_ADDRESS_UNREACHABLE in
MapConnectError.  ERR_FAILED is bad.

[darin (slow to review), 2009-08-07 06:03:50]

This sounds like a good solution to me.  I like making the error reporting
to the URLRequest consumer the same across platforms.-Darin

On Thu, Aug 6, 2009 at 7:28 PM, <wtc@chromium.org> wrote:

> I forgot to say -- map EACCES to ERR_ACCESS_DENIED in
> MapPosixError, and optionally (if you like) map EACCES
> to ERR_ADDRESS_INVALID or ERR_ADDRESS_UNREACHABLE in
> MapConnectError.  ERR_FAILED is bad.
>
>
> http://codereview.chromium.org/164114
>

[darin (slow to review), 2009-08-11 23:35:40]

I decided to just go with mapping EACCES to ERR_ACCESS_DENIED.  I think it is
better for our error codes to be consistent with the OS than to be self
consistent across operating systems.

[wtc, 2009-08-12 00:08:14]

LGTM.