net: Move the logic reading upload elements to UploadData::Element.

net: Move the logic reading upload elements to UploadData::Element.

This is yet another clean-up patch to simplify UploadDataStream
in preparation for making UploadDataStream::Read asynchronous.

Managing the offset and the file stream inside UploadData::Element
is cleaner anyway.

BUG=72001
TEST=net_unittests


Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=123677

[willchan no longer on Chromium, 2012-02-24 21:54:40]

LGTM

https://chromiumcodereview.appspot.com/9452040/diff/1/net/base/upload_data.cc
File net/base/upload_data.cc (right):

https://chromiumcodereview.appspot.com/9452040/diff/1/net/base/upload_data.cc...
net/base/upload_data.cc:141: int UploadData::Element::ReadFromMemorySync(char*
buf, int buf_len) {
Can you DCHECK that buf_len is positive (it shouldn't be 0, right?)?

https://chromiumcodereview.appspot.com/9452040/diff/1/net/base/upload_data.cc...
net/base/upload_data.cc:158: int UploadData::Element::ReadFromFileSync(char*
buf, int buf_len) {
Ditto on DCHECK'ing |buf_len|'s values.

[satorux1, 2012-02-24 22:35:22]

Turned out URLRequestTestHTTP.Redirect301Tests failed, because UploadData was
reused for another UploadDataStream, and offset_ in each upload element already
pointed to the end. The fix was as easy as resetting the offset when
UploadDataStream is created. now rerunning try bots.

http://codereview.chromium.org/9452040/diff/1/net/base/upload_data.cc
File net/base/upload_data.cc (right):

http://codereview.chromium.org/9452040/diff/1/net/base/upload_data.cc#newcode141
net/base/upload_data.cc:141: int UploadData::Element::ReadFromMemorySync(char*
buf, int buf_len) {
On 2012/02/24 21:54:40, willchan wrote:
> Can you DCHECK that buf_len is positive (it shouldn't be 0, right?)?

Done. Updated the comment as well.

http://codereview.chromium.org/9452040/diff/1/net/base/upload_data.cc#newcode158
net/base/upload_data.cc:158: int UploadData::Element::ReadFromFileSync(char*
buf, int buf_len) {
On 2012/02/24 21:54:40, willchan wrote:
> Ditto on DCHECK'ing |buf_len|'s values.

Done.

[commit-bot: I haz the power, 2012-02-25 21:46:06]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/satorux@chromium.org/9452040/6003

[commit-bot: I haz the power, 2012-02-25 22:59:22]

Change committed as 123677