Chromium Code Reviews
chromiumcodereview-hr@appspot.gserviceaccount.com (chromiumcodereview-hr) | Please choose your nickname with Settings | Help | Chromium Project | Gerrit Changes | Sign out
(410)

Issue 2101014: Eliminate the establishing_tunnel_ internal state and move to explicit... (Closed)

Created:
10 years, 7 months ago by Mike Belshe
Modified:
9 years, 5 months ago
CC:
chromium-reviews, cbentzel+watch_chromium.org, darin-cc_chromium.org
Visibility:
Public.

Description

Eliminate the establishing_tunnel_ internal state and move to explicit states in the state machine for proxy tunnel establishment. Original work from svandebo. BUG=none TEST=existing Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=48173

Patch Set 1 #

Patch Set 2 : '' #

Patch Set 3 : '' #

Patch Set 4 : '' #

Patch Set 5 : '' #

Patch Set 6 : '' #

Total comments: 4

Patch Set 7 : '' #

Total comments: 5
Unified diffs Side-by-side diffs Delta from patch set Stats (+227 lines, -133 lines) Patch
M chrome/app/generated_resources.grd View 1 chunk +3 lines, -0 lines 0 comments Download
M chrome/browser/tab_contents/tab_contents.cc View 1 chunk +2 lines, -0 lines 0 comments Download
M net/base/load_states.h View 1 2 3 4 1 chunk +4 lines, -0 lines 0 comments Download
M net/base/net_log_event_type_list.h View 3 4 2 chunks +20 lines, -14 lines 0 comments Download
M net/http/http_network_transaction.h View 1 2 3 4 5 chunks +12 lines, -8 lines 0 comments Download
M net/http/http_network_transaction.cc View 1 2 3 4 5 6 17 chunks +186 lines, -111 lines 5 comments Download

Messages

Total messages: 5 (0 generated)
Mike Belshe
10 years, 7 months ago (2010-05-24 22:07:40 UTC) #1
vandebo (ex-Chrome)
LGTM. The changes are just mostly just rebasing changes - you could have just asked ...
10 years, 7 months ago (2010-05-24 23:46:54 UTC) #2
vandebo (ex-Chrome)
http://codereview.chromium.org/2101014/diff/29001/30005 File net/http/http_network_transaction.cc (right): http://codereview.chromium.org/2101014/diff/29001/30005#newcode965 net/http/http_network_transaction.cc:965: // ------------------------------------------------------------------------- Don't need these delimiters in the committed ...
10 years, 7 months ago (2010-05-24 23:48:38 UTC) #3
Mike Belshe
http://codereview.chromium.org/2101014/diff/29001/30005 File net/http/http_network_transaction.cc (right): http://codereview.chromium.org/2101014/diff/29001/30005#newcode965 net/http/http_network_transaction.cc:965: // ------------------------------------------------------------------------- On 2010/05/24 23:48:39, vandebo wrote: > Don't ...
10 years, 7 months ago (2010-05-24 23:54:15 UTC) #4
eroman
10 years, 7 months ago (2010-05-26 03:21:26 UTC) #5
http://codereview.chromium.org/2101014/diff/38001/15006
File net/http/http_network_transaction.cc (right):

http://codereview.chromium.org/2101014/diff/38001/15006#newcode483
net/http/http_network_transaction.cc:483: ssl_connect_start_time_.is_null())
Using the time variable feels a bit hockey.

http://codereview.chromium.org/2101014/diff/38001/15006#newcode521
net/http/http_network_transaction.cc:521: if (headers->response_code() == 407) {
Before it was only failing in the case where the URL was using SSL (i.e.
establishing tunnel), whereas now it always fails to read on 407.

At the very least the error code is wrong now (since this may not apply to the
tunnel case). Even if the error code were correct, I don't think we want this
new behavior, but should continue to show the error page for failed 407 when on
non-https.

http://codereview.chromium.org/2101014/diff/38001/15006#newcode632
net/http/http_network_transaction.cc:632: NULL);
nit: line these up with the paren of the previous line.

http://codereview.chromium.org/2101014/diff/38001/15006#newcode1012
net/http/http_network_transaction.cc:1012: return result;
No call to ClearTunnelState() in this case?

http://codereview.chromium.org/2101014/diff/38001/15006#newcode1028
net/http/http_network_transaction.cc:1028: ClearTunnelState();
The repeated calls to ClearTunnelState() seems pretty error prone. Especially if
someone else modifies this function in the future.

Powered by Google App Engine
This is Rietveld 408576698