[GitHub] trafficserver pull request: TS-4388: Fix TSHttpTxnParentProxySet

2016-05-02 Thread jpeach
Github user jpeach closed the pull request at: https://github.com/apache/trafficserver/pull/606 --- If your project is set up for it, you can reply to this email and have your reply appear on GitHub as well. If your project does not have this feature enabled and wishes so, or if the feature

[GitHub] trafficserver pull request: TS-4388: Fix TSHttpTxnParentProxySet

2016-05-02 Thread jpeach
Github user jpeach commented on the pull request: https://github.com/apache/trafficserver/pull/606#issuecomment-216290646 Merged. --- If your project is set up for it, you can reply to this email and have your reply appear on GitHub as well. If your project does not have this feature

[GitHub] trafficserver pull request: TS-4388: Fix TSHttpTxnParentProxySet

2016-04-30 Thread zwoop
Github user zwoop commented on the pull request: https://github.com/apache/trafficserver/pull/606#issuecomment-215984159 +1 on the refactoring / commits. --- If your project is set up for it, you can reply to this email and have your reply appear on GitHub as well. If your project doe

[GitHub] trafficserver pull request: TS-4388: Fix TSHttpTxnParentProxySet

2016-04-29 Thread jpeach
Github user jpeach commented on the pull request: https://github.com/apache/trafficserver/pull/606#issuecomment-215932527 OK I broke out some refactoring into 2 additional commits. --- If your project is set up for it, you can reply to this email and have your reply appear on GitHub a

[GitHub] trafficserver pull request: TS-4388: Fix TSHttpTxnParentProxySet

2016-04-29 Thread zwoop
Github user zwoop commented on the pull request: https://github.com/apache/trafficserver/pull/606#issuecomment-215751946 I agree with Phil, since the refactoring (variable name changes) is so big, it would be much nicer to have this as two separate PRs. --- If your project is set up

[GitHub] trafficserver pull request: TS-4388: Fix TSHttpTxnParentProxySet

2016-04-28 Thread jpeach
Github user jpeach commented on the pull request: https://github.com/apache/trafficserver/pull/606#issuecomment-215625974 I looked at the diff again, and though there are a few changes here, they are pretty obvious in a side-by-side diff. Basically every place that follows ``parent_re

[GitHub] trafficserver pull request: TS-4388: Fix TSHttpTxnParentProxySet

2016-04-28 Thread PSUdaemon
Github user PSUdaemon commented on the pull request: https://github.com/apache/trafficserver/pull/606#issuecomment-215622970 I like the robustness, it's just hard to tell that this commit is anything more than a refactor. I'd just separate the small fix out and leave the rest as one p

[GitHub] trafficserver pull request: TS-4388: Fix TSHttpTxnParentProxySet

2016-04-28 Thread jpeach
Github user jpeach commented on the pull request: https://github.com/apache/trafficserver/pull/606#issuecomment-215619876 The actual fix is to avoid dereferencing the ``rec`` pointer when it has the API sentinel value. I could have just make a small fix to the code path that I happene

[GitHub] trafficserver pull request: TS-4388: Fix TSHttpTxnParentProxySet

2016-04-28 Thread PSUdaemon
Github user PSUdaemon commented on the pull request: https://github.com/apache/trafficserver/pull/606#issuecomment-215619566 The 4th commit is so much refactoring with no change in behavior that it's hard to follow what the actual fix is. I agree that `r` is a poor variable name, and

[GitHub] trafficserver pull request: TS-4388: Fix TSHttpTxnParentProxySet

2016-04-28 Thread PSUdaemon
Github user PSUdaemon commented on a diff in the pull request: https://github.com/apache/trafficserver/pull/606#discussion_r61531390 --- Diff: proxy/http/HttpConfig.cc --- @@ -888,8 +888,6 @@ HttpConfig::startup() HttpEstablishStaticConfigLongLong(c.origin_min_keep_alive_conn

[GitHub] trafficserver pull request: TS-4388: Fix TSHttpTxnParentProxySet

2016-04-28 Thread JamesPeach
Github user JamesPeach commented on the pull request: https://github.com/apache/trafficserver/pull/606#issuecomment-215581577 The 99% case in the code is ``parent_result.result`` which didn't seem too bad to me. I hated, hated, hated ``parent_result.r`` ;) --- If your project is set

[GitHub] trafficserver pull request: TS-4388: Fix TSHttpTxnParentProxySet

2016-04-28 Thread zwoop
Github user zwoop commented on the pull request: https://github.com/apache/trafficserver/pull/606#issuecomment-215579807 I'm not sure I'm in love with result->result That seems pretty redundant. If you are going to bike shed this, maybe result->rc, or result->resu

[GitHub] trafficserver pull request: TS-4388: Fix TSHttpTxnParentProxySet

2016-04-28 Thread jpeach
Github user jpeach commented on the pull request: https://github.com/apache/trafficserver/pull/606#issuecomment-215578712 Ping @zwoop @PSUdaemon @jrushford --- If your project is set up for it, you can reply to this email and have your reply appear on GitHub as well. If your project

[GitHub] trafficserver pull request: TS-4388: Fix TSHttpTxnParentProxySet

2016-04-28 Thread jpeach
GitHub user jpeach opened a pull request: https://github.com/apache/trafficserver/pull/606 TS-4388: Fix TSHttpTxnParentProxySet This PR fixes ``TSHttpTxnParentProxySet``. The first two commits are minor cleanups of cruft. The third one fixes the API tests so that they don't