Allow empty header value in the blink HTTP header parser.

Allow empty header value in the blink HTTP header parser.

The HTTP header parser in Blink forbids empty header values which should
be allowed. This CL fix the issue.

BUG=380075
R=tyoshino@chromium.org

Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=175532

[tyoshino (SeeGerritForStatus), 2014-06-04 08:24:12]

https://codereview.chromium.org/318613002/diff/1/Source/platform/network/HTTP...
File Source/platform/network/HTTPParsers.cpp (right):

https://codereview.chromium.org/318613002/diff/1/Source/platform/network/HTTP...
Source/platform/network/HTTPParsers.cpp:636: for (i = start; i < size && s[i] ==
' '; ++i) {
remove "i = start"?

https://codereview.chromium.org/318613002/diff/1/Source/platform/network/HTTP...
Source/platform/network/HTTPParsers.cpp:652: failureReason = "CR doesn't follow
LF after value at " + trimInputSample(&s[i + 1], size - i - 1);
not your mistake but this should be
"LF doesn't follow CR"

https://codereview.chromium.org/318613002/diff/1/Source/platform/network/HTTP...
Source/platform/network/HTTPParsers.cpp:656: *position = i + 1;
is this correct?

https://codereview.chromium.org/318613002/diff/1/Source/platform/network/HTTP...
Source/platform/network/HTTPParsers.cpp:672: failureReason = "CR doesn't follow
LF at " + trimInputSample(0, size);
ditto

https://codereview.chromium.org/318613002/diff/1/Source/platform/network/HTTP...
Source/platform/network/HTTPParsers.cpp:680: failureReason = "Invalid UTF-8
sequence in header name";
It's better to say that that header name was missing. WDYT?

https://codereview.chromium.org/318613002/diff/1/Source/platform/network/HTTP...
File Source/platform/network/HTTPParsersTest.cpp (right):

https://codereview.chromium.org/318613002/diff/1/Source/platform/network/HTTP...
Source/platform/network/HTTPParsersTest.cpp:192: 
add a test that takes input of two lines and check the return value.

[yhirano, 2014-06-04 09:14:03]

https://codereview.chromium.org/318613002/diff/1/Source/platform/network/HTTP...
File Source/platform/network/HTTPParsers.cpp (right):

https://codereview.chromium.org/318613002/diff/1/Source/platform/network/HTTP...
Source/platform/network/HTTPParsers.cpp:636: for (i = start; i < size && s[i] ==
' '; ++i) {
On 2014/06/04 08:24:12, tyoshino wrote:
> remove "i = start"?

Done.

https://codereview.chromium.org/318613002/diff/1/Source/platform/network/HTTP...
Source/platform/network/HTTPParsers.cpp:652: failureReason = "CR doesn't follow
LF after value at " + trimInputSample(&s[i + 1], size - i - 1);
On 2014/06/04 08:24:12, tyoshino wrote:
> not your mistake but this should be
> "LF doesn't follow CR"

Done.

https://codereview.chromium.org/318613002/diff/1/Source/platform/network/HTTP...
Source/platform/network/HTTPParsers.cpp:656: *position = i + 1;
On 2014/06/04 08:24:12, tyoshino wrote:
> is this correct?

Currently parseHTTPHeader returns (the number of consumed bytes - 1) and this CL
doesn't change the behavior.
I don't like the current behavior, but I would like to fix it in another CL.

https://codereview.chromium.org/318613002/diff/1/Source/platform/network/HTTP...
Source/platform/network/HTTPParsers.cpp:672: failureReason = "CR doesn't follow
LF at " + trimInputSample(0, size);
On 2014/06/04 08:24:12, tyoshino wrote:
> ditto

Done.

https://codereview.chromium.org/318613002/diff/1/Source/platform/network/HTTP...
Source/platform/network/HTTPParsers.cpp:680: failureReason = "Invalid UTF-8
sequence in header name";
On 2014/06/04 08:24:12, tyoshino wrote:
> It's better to say that that header name was missing. WDYT?
I added some messages:
 - "Header name is missing" if the name is empty.
 - "Invalid UTF-8 sequence in header name" if the name is invalid.
 - "Invalid UTF-8 sequence in header value" if the value is invalid.

https://codereview.chromium.org/318613002/diff/1/Source/platform/network/HTTP...
File Source/platform/network/HTTPParsersTest.cpp (right):

https://codereview.chromium.org/318613002/diff/1/Source/platform/network/HTTP...
Source/platform/network/HTTPParsersTest.cpp:192: 
On 2014/06/04 08:24:12, tyoshino wrote:
> add a test that takes input of two lines and check the return value.

Done.

[tyoshino (SeeGerritForStatus), 2014-06-05 02:07:32]

lgtm

https://codereview.chromium.org/318613002/diff/20001/Source/platform/network/...
File Source/platform/network/HTTPParsers.cpp (right):

https://codereview.chromium.org/318613002/diff/20001/Source/platform/network/...
Source/platform/network/HTTPParsers.cpp:668: return false;
nice!

[yhirano, 2014-06-05 03:48:47]

+tkent for OWNER review.

[tkent, 2014-06-05 03:58:23]

rslgtm

[yhirano, 2014-06-05 04:02:59]

The CQ bit was checked by yhirano@chromium.org

[commit-bot: I haz the power, 2014-06-05 04:04:12]

CQ is trying da patch. Follow status at
 https://chromium-status.appspot.com/cq/yhirano@chromium.org/318613002/20001

[commit-bot: I haz the power, 2014-06-05 05:16:05]

Change committed as 175532