#22636: Add Travis configs so GitHub forks get CI coverage -------------------------------------------------+------------------------- Reporter: catalyst | Owner: | patrickod Type: defect | Status: | needs_review Priority: High | Milestone: Tor: | 0.3.2.x-final Component: Core Tor/Tor | Version: Severity: Normal | Resolution: Keywords: continuous-integration ci testing | Actual Points: 1 best-practice unit-testing new-developers | travis review-group-21 | Parent ID: | Points: .5 Reviewer: nickm | Sponsor: -------------------------------------------------+-------------------------
Comment (by nickm): Replying to [comment:24 isis]: > Replying to [comment:21 nickm]: > > Hi! I have some questions: > > - If we merge this, will it start spamming the #tor-bots IRC? > > It will spam #tor-bots when someone makes a commit which breaks things when the commit's parent was previously working, and it will also spam for a commit which fixes things when the commit's parent was previously broken. Do we want different behaviour? Maybe we only want that for some people? I don't think we care if some person we've never heard of starts breaking their github tor branches. > > - Do we want to consider adding --enable-fragile-hardening? > > We can do that, if that is what we expect to pass. (I think that's a reasonable expectation. Also, side note, we should make sure to document our expectations in doc/HACKING better!) Added in [https://gitweb.torproject.org/user/isis/tor.git/commit/?h=bug22636&id=c91a57ccf90308c6728184b43519f96b61acb95d commit] `c91a57ccf9`. Cool > > - Should we install the optional dependencies (zstd, lzma2, scrypt, ...) > > Sure, done in [https://gitweb.torproject.org/user/isis/tor.git/commit/?h=bug22636&id=1bb00fb812c0df7a574ed62e9f53b0e8192c7d04 commit] `1bb00fb812`. (Except, for some reason, Ubuntu Trusty doesn't have a libzstd package available.) Well, at least we tried :) > > - In the homebrew stanza, should all of those lines have "brew outdated openssl" or only the openssl one? > > Oh oops, good catch. Fixed in [https://gitweb.torproject.org/user/isis/tor.git/commit/?h=bug22636&id=8f8689f70235dc19cbc5092ea148af5772a9cdc3 commit] `8f8689f70235`. > > > - Should we skip rust installation if we won't be building with rust, in order to save time? > > Yes, this is a great idea. Done in [https://gitweb.torproject.org/user/isis/tor.git/commit/?h=bug22636&id=e5dd07a4c64fd2a4132ab1f6dec64640a95da35e commit] `e5dd07a4c64`. It doesn't seem to save any time though, probably if I had to guess because there's some transparent proxy setup with a cache. > > I've updated `bug22636_0.2.4_squashed` and `bug22636_0.3.1_squashed` and tests [https://travis-ci.org/isislovecruft/tor/builds/257127758 are] [https://travis-ci.org/isislovecruft/tor/builds/257126606 passing] respectively. > > Replying to [comment:23 nickm]: > > Oh, one more thing: > > > > --enable-fatal-warnings was called --enable-gcc-warnings before 0.2.9.x > > Oh right. Thanks! There's now a `bug22636_0.2.9` [https://gitweb.torproject.org/user/isis/tor.git/log/?h=bug22636_0.2.9 branch] for 0.2.9 and 0.3.0 which contains everything from the `bug22636_0.2.4_squashed` branch, except it uses --enable-fatal-warnings instead. `bug22636_0.2.4_squashed` has now been changed to use --enable- gcc-warnings. Sounds good! I'll give it all another look over, and merge. -- Ticket URL: <https://trac.torproject.org/projects/tor/ticket/22636#comment:25> Tor Bug Tracker & Wiki <https://trac.torproject.org/> The Tor Project: anonymity online _______________________________________________ tor-bugs mailing list tor-bugs@lists.torproject.org https://lists.torproject.org/cgi-bin/mailman/listinfo/tor-bugs