Perl 6 - the future is here, just unevenly distributed

IRC log for #git, 2017-02-22

| Channels | #git index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
00:00 peterbecich joined #git
00:00 exitcode1 joined #git
00:02 shgysk8zer0 joined #git
00:04 humboldt joined #git
00:08 azerus joined #git
00:09 miczac joined #git
00:11 cdg joined #git
00:12 dminuoso joined #git
00:16 jstimm joined #git
00:18 impermanence joined #git
00:20 humboldt joined #git
00:20 ToBeCloud joined #git
00:21 perlpilot joined #git
00:22 hahuang65 joined #git
00:22 milki joined #git
00:23 NeverDie joined #git
00:24 d1z joined #git
00:27 roelmonnens joined #git
00:28 ash_workz joined #git
00:30 clemf joined #git
00:31 humboldt joined #git
00:34 ispn joined #git
00:35 shgysk8zer0_ joined #git
00:39 axisys joined #git
00:40 chipotle joined #git
00:40 pur3eval joined #git
00:41 Oem joined #git
00:44 Learath2 joined #git
00:47 Noldorin joined #git
00:52 Rodya_ joined #git
00:52 xaviergmail_ joined #git
00:53 sunrunner202 joined #git
00:53 styler2go joined #git
00:54 acetakwas joined #git
00:54 _main_ joined #git
00:55 invisbl_ joined #git
00:57 azerus joined #git
00:57 mikecmpbll joined #git
00:59 s2013 joined #git
00:59 Phanes joined #git
01:01 gopar joined #git
01:03 humboldt joined #git
01:05 nettoweb joined #git
01:05 paws_ joined #git
01:06 Murr joined #git
01:06 BlackFox joined #git
01:06 bjoe2k4 joined #git
01:07 PtxDK joined #git
01:07 cagomez joined #git
01:07 spm_draget joined #git
01:08 Blkt joined #git
01:08 KnightsOfNi joined #git
01:08 heftig joined #git
01:08 bittyx joined #git
01:09 manitu joined #git
01:09 explody joined #git
01:11 nettoweb1 joined #git
01:12 buckowski joined #git
01:13 Guest5972 joined #git
01:14 ecuanaso joined #git
01:15 Tony_ joined #git
01:15 humboldt joined #git
01:17 miczac joined #git
01:18 PtxDK joined #git
01:19 spm_draget joined #git
01:19 MonsterKiller joined #git
01:19 heftig joined #git
01:19 mischat joined #git
01:19 Blkt joined #git
01:20 bittyx joined #git
01:20 NumberSt- joined #git
01:20 madewokherd joined #git
01:21 buckowski joined #git
01:23 suy joined #git
01:24 NumberSt- joined #git
01:25 cqi joined #git
01:25 thiago joined #git
01:25 thiago joined #git
01:26 ujjain joined #git
01:26 ujjain joined #git
01:26 humboldt joined #git
01:29 xall joined #git
01:31 nettoweb joined #git
01:32 bvcosta joined #git
01:32 danimal2 joined #git
01:32 a_thakur joined #git
01:34 dave0x6d joined #git
01:35 pur3eval joined #git
01:36 SwiftMatt joined #git
01:38 gugah joined #git
01:39 matoro joined #git
01:40 humboldt joined #git
01:45 redhedded1 joined #git
01:45 humboldt joined #git
01:46 a_thakur joined #git
01:48 gopar joined #git
01:52 jimklo joined #git
01:52 chipotle joined #git
01:53 Rodya_ joined #git
01:53 prg3 joined #git
01:54 alexandre9099 joined #git
01:57 humboldt_ joined #git
01:59 mehola joined #git
02:00 cluelessperson joined #git
02:00 raedah joined #git
02:00 raedah how do I do a cherry-pick but not just with the last commit from a branch, but with the last 2 commits from a branch?
02:02 jameser joined #git
02:03 UniFreak joined #git
02:04 gopar joined #git
02:06 Sasazuka joined #git
02:07 cdg joined #git
02:08 d1z joined #git
02:08 duderonomy joined #git
02:09 danimal2 left #git
02:09 flippo joined #git
02:13 deltab raedah: pick one, then pick the other?
02:15 humboldt_ joined #git
02:15 roelmonnens joined #git
02:16 xall joined #git
02:16 gopar joined #git
02:21 mischat joined #git
02:21 Vampire0_ joined #git
02:24 def_jam joined #git
02:25 Vampire0 raedah, `git cherry-pick branch~..branch`
02:27 prsn joined #git
02:28 Wulf4 joined #git
02:29 finalbeta joined #git
02:31 humboldt_ joined #git
02:34 raedah git cherry-pick mybranch~1 mybranch~2    works
02:36 humboldt joined #git
02:37 glspi joined #git
02:37 thiago joined #git
02:38 atomi joined #git
02:40 mellernoia joined #git
02:46 s2013 joined #git
02:46 pur3eval joined #git
02:46 cqi joined #git
02:46 d^sh joined #git
02:49 pbrewczynski joined #git
02:50 nettoweb joined #git
02:50 humboldt joined #git
02:53 Rodya_ joined #git
02:54 eblip joined #git
02:59 orbyt_ joined #git
02:59 orbyt_ Is there an option for git log to show all global recent commits?
02:59 orbyt_ Eg. instead of the ones in the current repository
03:00 Murr so you mean, hire a botnet, probe the ssh ports of the entire internet, enumerate git repos, and log the last commits?
03:00 mehola joined #git
03:00 pks joined #git
03:00 orbyt_ No, global to the system.
03:00 kadoban orbyt_: Do you mean those of remotes you have in the repo? Otherwise, probably not.
03:01 kadoban Git doesn't have a system global registry of repos or anything like that.
03:01 Murr indeed
03:01 aavrug joined #git
03:01 Murr and the various hosting systems don't have a common api to enumerate repos, I would think
03:02 kadoban Each specific one might, but ... ya I highly doubt there's a shared one.
03:02 orbyt_ Sorry I mispoke, I meant instead of the ones in the current branch.
03:03 orbyt_ git log only logs the current branchs git commit history
03:03 kadoban Perhaps you mean --all ?  !lol is a common invocation. Or I like gitk --all personally.
03:03 gitinfo A nifty view of branches, tags, and other refs: git log --oneline --graph --decorate --all
03:03 ojdo joined #git
03:03 Bucciarati joined #git
03:03 raijin joined #git
03:04 orbyt_ kadoban what is git lol?
03:04 orbyt_ or !lol rather
03:04 gitinfo A nifty view of branches, tags, and other refs: git log --oneline --graph --decorate --all
03:04 kadoban What gitinfo said. It's a trigger for the bot to say a thing.
03:04 orbyt_ !lol
03:04 gitinfo A nifty view of branches, tags, and other refs: git log --oneline --graph --decorate --all
03:05 orbyt_ why is the trigger lol?
03:05 cyphase joined #git
03:05 kadoban Log Oneline aLl maybe? I dunno.
03:05 kadoban Or maybe Log OneLine
03:05 kadoban It's a fairly common alias for that command, not quite sure where it came from. It's memorable *shrug*
03:06 qqx joined #git
03:07 meZee joined #git
03:07 trelane joined #git
03:07 trelane joined #git
03:07 flippo I like --stat too
03:08 orbyt_ git log --all was really all I was looking for. I had checked the docs first but even they do not list an --all option: https://git-scm.com/book/en/v2/Git​-Basics-Viewing-the-Commit-History
03:09 kadoban The real docs are the manpage, they list *everything* unless it's a mistake. man git log
03:09 gitinfo the git-log manpage is available at http://jk.gs/git-log.html
03:09 flippo The effect of --all on log seems like a corner case
03:09 Peng_ joined #git
03:09 kadoban But honestly, unless you have something against GUIs, I'd look at gitk --all. I like it quite a bit better.
03:09 flippo Ick, GUIs.
03:10 roelmonnens joined #git
03:10 deltab gitg looks nicer, imho
03:10 fstd_ joined #git
03:10 flippo sudo apt install gitg
03:10 deltab or tig
03:10 orbyt_ thanks guys.
03:15 hexagoxel joined #git
03:17 Vortex34 joined #git
03:20 lowercaseman joined #git
03:21 prsn joined #git
03:22 mischat joined #git
03:22 mehola joined #git
03:24 MuffinMedic joined #git
03:26 hhee joined #git
03:30 Caelum joined #git
03:33 arescorpio joined #git
03:33 nevodka joined #git
03:35 cyphase joined #git
03:37 jstimm joined #git
03:38 jeffreylevesque joined #git
03:38 kyan joined #git
03:38 humboldt joined #git
03:38 Khisanth joined #git
03:38 NumberStation08 joined #git
03:42 hahuang65 joined #git
03:42 a_thakur joined #git
03:43 a_thakur joined #git
03:44 EV3RYDAYPR0GRESS joined #git
03:45 humboldt joined #git
03:46 govg joined #git
03:52 nowhereman joined #git
03:52 sunri5e joined #git
03:54 Rodya_ joined #git
03:54 chachasmooth joined #git
03:57 gugah joined #git
03:57 ozmage joined #git
03:59 Zialus joined #git
04:00 ozmage_ joined #git
04:00 dermoth joined #git
04:02 roelmonnens joined #git
04:05 pur3eval joined #git
04:09 rchavik joined #git
04:13 Tech joined #git
04:15 pks joined #git
04:16 prsn joined #git
04:17 peterbecich joined #git
04:19 raedah left #git
04:23 mischat joined #git
04:26 Goplat joined #git
04:28 pks joined #git
04:30 kurdzman joined #git
04:36 humboldt joined #git
04:37 wilornel Hey #git! I was wondering if any of you have setup any git continuous push/pull setups? I have two devices I log info on and always think "I'll sync them manuall later" and end up doing so ~once a quarter
04:37 wilornel I just found out about a tool that will let me track saves in a wip branch, which I think fits my preferences. I can also do a .git/hooks/post-commit which does `git push origin wip/dev; git push origin dev`
04:37 wilornel However, I now need to have something doing `git pull` on my other device before I start doing edits
04:38 ayogi joined #git
04:39 hexagoxel joined #git
04:41 humboldt joined #git
04:43 Cogitabundus joined #git
04:45 mehola joined #git
04:45 Cogitabundus joined #git
04:46 Cogitabundus joined #git
04:46 xall joined #git
04:50 rwp wilornel, Not to avoid git but you could always use the "unison" tool to sync your work areas. Might do it more than once a quarter then.
04:50 deltab you could change your prompt to tell you when you haven't sync'd recently
04:51 mloy what about a login script that cd's to the dir and runs git pull?
04:53 humboldt joined #git
04:54 humboldt_ joined #git
04:56 Cabanoss- joined #git
04:56 qqx joined #git
04:56 salamanderrake joined #git
04:58 Cogitabundus joined #git
05:00 dermoth joined #git
05:01 MattMaker joined #git
05:01 thebope joined #git
05:02 qqx joined #git
05:04 chipotle joined #git
05:04 brent__ joined #git
05:06 qqx joined #git
05:06 ruxu joined #git
05:08 salamanderrake joined #git
05:09 cdown joined #git
05:09 xall joined #git
05:11 venmx joined #git
05:13 JanC_ joined #git
05:13 kasual joined #git
05:13 eroux joined #git
05:13 ash_workz joined #git
05:14 recola joined #git
05:15 Sample joined #git
05:16 svm_invictvs joined #git
05:18 wilbert joined #git
05:24 mischat joined #git
05:27 humboldt joined #git
05:31 s2013 joined #git
05:37 explody joined #git
05:40 fuchstronaut joined #git
05:41 prsn joined #git
05:44 ckruczek left #git
05:44 nowhereman joined #git
05:44 ckruczek joined #git
05:48 humboldt joined #git
05:49 humboldt_ joined #git
05:52 jfelchner joined #git
05:52 xall joined #git
05:53 dreiss joined #git
05:54 jfelchner Hey all.  I'm trying to setup a git config for a remote such that I fetch all branches _except_ those with prefix (eg origin/foo/*).  If there a place you all can point me to with the proper syntax for that?
05:55 howdoi joined #git
05:55 kyan joined #git
05:56 Rodya_ joined #git
05:57 salamanderrake joined #git
05:58 pigeon joined #git
05:58 _ikke_ jfelchner: Hmm, not sure there is an exclusion syntax
05:59 a_thakur joined #git
05:59 pur3eval joined #git
06:01 jfelchner Ah
06:01 jfelchner _ikke_ thank you for replying.
06:05 SwiftMatt joined #git
06:06 kurdzman joined #git
06:06 rscata joined #git
06:07 navidr joined #git
06:10 a_thakur joined #git
06:12 ecuanaso joined #git
06:12 MattMaker joined #git
06:14 MineCoins joined #git
06:14 zefferno joined #git
06:16 midgetsp_ left #git
06:16 dec0n joined #git
06:16 a_thakur joined #git
06:20 explody joined #git
06:24 HurricaneHarry joined #git
06:25 mischat joined #git
06:31 zeroed joined #git
06:31 zeroed joined #git
06:32 jwmann joined #git
06:35 User458764 joined #git
06:35 rawl79 joined #git
06:37 rawl79 left #git
06:37 a_thakur joined #git
06:41 zulutango joined #git
06:43 a_thakur joined #git
06:44 Murii joined #git
06:45 ljc joined #git
06:46 svm_invictvs joined #git
06:47 zeroed joined #git
06:49 kurdzman joined #git
06:49 nowhereman joined #git
06:55 Raging_Hog joined #git
06:55 solenodic joined #git
06:55 ThomasLocke joined #git
06:56 nowhereman joined #git
06:56 fatalhalt joined #git
06:56 User458764 joined #git
06:58 bocaneri joined #git
07:00 sunrunner202 left #git
07:00 a_thakur joined #git
07:01 afuentes joined #git
07:01 sarri joined #git
07:01 sarri joined #git
07:11 User458764 joined #git
07:11 venmx joined #git
07:13 a_thakur joined #git
07:13 MattMaker joined #git
07:14 Cthalupa joined #git
07:16 JeroenT joined #git
07:20 dec0n joined #git
07:20 ISmithers joined #git
07:22 shgysk8zer0_ joined #git
07:25 mischat joined #git
07:26 roelmonnens joined #git
07:26 srandon1213 joined #git
07:27 srandon1213 hello all, can somebody explain to me commonly how collaboration is done in teams ? for example, should we create a new branch for new feature and work on a branch... and if more people is working on a branch.. we should work on separate part of the branch right  ?
07:27 srandon1213 is this correct  ?
07:27 Taikibansei joined #git
07:27 _ikke_ !workflow
07:27 gitinfo Finding the right workflow for you is critical for the success of any SCM project.  Git is very flexible with respect to workflow.  See http://sethrobertson.github.c​om/GitBestPractices/#workflow for a list of references about choosing branching and distributed workflows.
07:28 xaviergmail_ joined #git
07:29 jast srandon1213: yeah, that's one way of working that seems to avoid a fair bit of pain
07:31 nidr0x joined #git
07:31 jnavila joined #git
07:32 pks joined #git
07:33 srandon1213 jast: ok there are other workflows  ?
07:34 mischat joined #git
07:35 a_thakur joined #git
07:36 waveclaw joined #git
07:37 dwmw2_gone joined #git
07:37 JeroenT_ joined #git
07:38 srandon1213 jast: i mean commonly used workflows ?
07:38 zulutango joined #git
07:40 lordjancso joined #git
07:42 _ikke_ !gitflow !github
07:42 gitinfo The description of the gitflow branch workflow model is at http://nvie.com/posts/a-succ​essful-git-branching-model/ while a tool to help implement this workflow is at https://github.com/petervanderdoes/gitflow  See http://sethrobertson.github.c​om/GitBestPractices/#workflow for other workflow suggestions/references
07:42 gitinfo Note that git != github. Feel free to ask us about Github-specific features (Forks, Pull Requests, Wikis, etc), but there are no guarantees. There is a #github channel, which might help too (again, no guarantees)
07:42 _ikke_ !github_flow
07:42 gitinfo This is the workflow followed by github: http://scottchacon.com/2011/08/31/github-flow.html
07:44 kurdzman_ joined #git
07:46 Alienpruts joined #git
07:50 a_thakur joined #git
07:53 a_thakur joined #git
07:53 Oatmeal joined #git
07:54 jast srandon1213: it depends a lot on the size of the project. on small project it can be enough to just have everyone go wild on the main branch. with mature projects you may want/have to maintain multiple stable versions, so then more branches are involved. in Git's development and the Linux kernel, proposed patches go through things like mailing lists for review; some projects even use dedicated code review tools integrated with Git.
07:54 jast if you want to do code reviews, I very much recommend using feature branches a lot or having patches go through some other process before applying them to the main branch
07:55 jast and, while we're on the subject, I very much recommend code reviews by developers who are fairly experienced with the code base :}
07:57 Rodya_ joined #git
08:00 diogenese joined #git
08:00 dreiss joined #git
08:00 Tobbi joined #git
08:01 MattMaker joined #git
08:02 solenodic joined #git
08:05 elect joined #git
08:06 MattMaker joined #git
08:09 tvw joined #git
08:11 xall_ joined #git
08:14 qt-x joined #git
08:14 King_Hual joined #git
08:14 dikiaap_ joined #git
08:16 djb-irc joined #git
08:17 dikiaap_ left #git
08:17 Tobbi joined #git
08:17 pks joined #git
08:17 kexmex joined #git
08:17 iqiqwi joined #git
08:20 TomyWork joined #git
08:20 dwmw2_gone joined #git
08:22 waveclaw joined #git
08:23 stliu joined #git
08:24 Dumblez0r joined #git
08:29 MattMaker joined #git
08:34 zeroed joined #git
08:34 zeroed joined #git
08:36 djb-irc joined #git
08:37 senaps_ joined #git
08:37 sarbs joined #git
08:37 waveclaw joined #git
08:37 a_thakur joined #git
08:39 armyriad joined #git
08:40 Oatmeal joined #git
08:40 troulouliou_div2 joined #git
08:44 Balliad joined #git
08:44 TomyWork joined #git
08:44 Tobbi joined #git
08:49 overlord_tm joined #git
08:49 Anja_ joined #git
08:51 kurkale6ka joined #git
08:52 miczac joined #git
08:52 SwiftMatt joined #git
08:55 a_thakur joined #git
08:58 flaviodesousa joined #git
08:58 Rodya_ joined #git
08:59 diogenese joined #git
09:00 MineCoins joined #git
09:00 Tobbi_ joined #git
09:03 p4trick joined #git
09:04 bongjovi joined #git
09:04 p4trick joined #git
09:04 mikecmpbll joined #git
09:05 waveclaw joined #git
09:06 nikomo joined #git
09:06 solenodic joined #git
09:07 alcohol joined #git
09:08 p4trix joined #git
09:09 p4trick joined #git
09:12 venmx joined #git
09:14 nettoweb joined #git
09:16 User458764 joined #git
09:16 ozmage joined #git
09:16 albel727 joined #git
09:16 JeroenT joined #git
09:20 ThomasTang joined #git
09:20 marcogmonteiro joined #git
09:20 clmsy joined #git
09:21 achadwick joined #git
09:22 nikomo Is it possible to create a tag that just keeps pointing at the latest commit? Got a project right now where rc builds are pointing at the tag for the last stable release of the repo that contains data, but there's been months of work happening in the data repo - would be nice to have a v4.0.6 tag in the data repo that tracks HEAD automatically, and the CI system for rc builds just packages off the v4.0.6 tag
09:22 nikomo the tag would then, on release, just be updated to point to whatever commit they've stopped at, in the data repo. so the 4.0.6 release build keeps pointing at the same tag
09:22 _ikke_ nikomo: No, a tag is assumed to never change
09:22 nikomo Dang
09:23 _ikke_ nikomo: A branch on the other hand would expected to change
09:23 hackvana joined #git
09:23 nikomo So, basically, the repo maintainers are using the wrong tool for this?
09:23 hackvana left #git
09:23 grawity a tag that tracks HEAD automatically sounds like the utterly wrong tool
09:24 grawity just build from the stable/v4.0 branch or whatever
09:25 nikomo Looking at they, they seem to keep rewriting the tag when they want to move it
09:25 _ikke_ nikomo: You would create new tags to point to new RCs
09:25 a_thakur_ joined #git
09:25 _ikke_ Yeah, that's not good
09:25 nikomo The rc1 is right now tag 4.0.6, I'm guessing they're going to rewrite the tag when rc2 comes around etc. sigh https://github.com/KiCad/kicad-source-mirror/c​ommit/3c6959b7de5c7a4e0d527b2952a60e7ede79d5c5
09:26 marenz joined #git
09:27 redeemed joined #git
09:28 Tobbi_ joined #git
09:28 jast yeah, there's really no way around that. tags can't be moved without rewriting them.
09:28 drodger joined #git
09:28 nikomo I'll let them know they're stupid^H^H^H^H^H^H using tools in a non-standard way
09:28 Andrew_K joined #git
09:29 Anticom joined #git
09:30 xanadu joined #git
09:30 MattMaker joined #git
09:31 troulouliou_div2 joined #git
09:32 Murii joined #git
09:34 notebox joined #git
09:35 ifra201 joined #git
09:35 Murii joined #git
09:36 Murii joined #git
09:37 notebox joined #git
09:37 chll_ joined #git
09:37 seni13 joined #git
09:38 Murii joined #git
09:38 durre joined #git
09:39 finalbeta1 joined #git
09:39 Murii joined #git
09:40 seni1 joined #git
09:41 Murii joined #git
09:42 Puffball joined #git
09:42 mischat joined #git
09:42 nd_ joined #git
09:44 finalbeta joined #git
09:44 clickjack joined #git
09:45 hhee joined #git
09:46 anohigisavay joined #git
09:47 theoceaniscool joined #git
09:47 dimi1947 joined #git
09:47 anohigisavay hi. i have come to a situation where i fail to checkout another branch
09:48 anohigisavay i added a .gitattributes and set text=auto to avoid messed line endings across the team
09:49 anohigisavay then i accidentally touched a file with CRLF line endings
09:49 anohigisavay and apparently git has automatically normalized it with LF
09:50 anohigisavay and now that file is marked as 'modified' forever
09:50 anohigisavay and i can't checkout a legacy branch without .gitattributes
09:53 t0by joined #git
09:53 t0by joined #git
09:54 hhee joined #git
09:54 kexmex joined #git
09:56 anohigisavay i fixed it by manually setting .git/HEAD to point to the target branch and issuing `git reset --hard`
09:57 anohigisavay looks bad idea though
09:58 ferr1 joined #git
09:59 jameser joined #git
10:00 ljc joined #git
10:02 xaviergmail_ joined #git
10:02 osse anohigisavay: in that case you could do   git reset --hard targetbranch   probably
10:03 anohigisavay osse, i don't think so. this will point my current branch to targetbranch
10:04 ismithers joined #git
10:04 anohigisavay i mean, say i'm at develop. if i do `git reset --hard master` then develop and master will point to the same commit
10:04 zeroed joined #git
10:05 netj joined #git
10:06 osse oh, i thought that was what you wanted
10:06 osse nvm me
10:07 anohigisavay :D np. i just wanted to leave the branch untouched, discard CRLF changes and go to another branch
10:08 peterppp joined #git
10:08 anohigisavay when i did `git checkout` it complained my CRLF changes would be overwritten and refused to checkout
10:10 osse anohigisavay: i think you could have done  git checkout HEAD -- .  to clear the unwanted "changes"
10:11 osse it will reply attributes and whatnot
10:11 Tobbi_ joined #git
10:12 Wulf4 joined #git
10:12 anohigisavay osse, i tried that and it still automatically converted all my line endings and left the file at modified state
10:12 anohigisavay that's just how text=auto works, i suppose
10:13 osse if the committed files are CRLF then they need to be converted before text=auto works properly
10:14 osse I did this once to our repo. It was a mess but definitely worth it
10:14 mischat joined #git
10:15 anohigisavay i was following this guide: https://help.github.com/articl​es/dealing-with-line-endings/
10:15 vuoto joined #git
10:16 anohigisavay in which i have a chance to convert all badly formed files in one shot
10:17 anohigisavay but then i noticed we were having a huge refactoring going on
10:17 anohigisavay so i decided to just keep the .gitattributes file in a branch and get to it after we merged the refactor branch
10:19 causasui joined #git
10:20 anohigisavay (i fear to encounter a heap of merge conflicts with only line ending changes and actual changes)
10:24 noidea joined #git
10:24 noidea joined #git
10:26 thebope joined #git
10:28 ixpict joined #git
10:30 bannakaffalatta joined #git
10:31 masuberu joined #git
10:31 MattMaker joined #git
10:32 ruxu joined #git
10:33 sostenuto joined #git
10:35 ecuanaso joined #git
10:35 hyperair joined #git
10:36 ahr3n joined #git
10:38 s1scha joined #git
10:41 nettoweb joined #git
10:43 mehola joined #git
10:44 leeN joined #git
10:44 marcogmonteiro joined #git
10:44 dersand joined #git
10:46 Tobbi_ joined #git
10:47 User458764 joined #git
10:49 bruce_lee joined #git
10:51 causasui joined #git
10:55 aard_ joined #git
10:57 d^sh joined #git
10:57 Keksike joined #git
10:57 watersoul joined #git
10:59 PCatinean joined #git
10:59 ozmage joined #git
10:59 Keksike Hey! How should I go about doing this: I have made a new branch off a branch, and made some changes to the new one. I then made some changes to the old one too, and now I want to take the changes from the old one to the new one. What should I do?
11:00 Rodya_ joined #git
11:00 Keksike If I simply pull the old branch into the new one, it will ofcourse overrun the changes I made to the new branch.
11:01 holgersson left #git
11:01 holgersson joined #git
11:01 solenodic joined #git
11:01 holgersson left #git
11:02 Snugglebash joined #git
11:04 intellix joined #git
11:06 a_thakur joined #git
11:09 bvcosta joined #git
11:10 Tobbi_ joined #git
11:10 nowhereman joined #git
11:11 osse Keksike: git merge oldbranch
11:11 sarbs joined #git
11:12 venmx joined #git
11:15 iauu joined #git
11:17 Keksike osse: but that will remove the changes I have done to the newbranch
11:18 osse Keksike: no
11:18 osse Keksike: it's *merge*, not overwrite
11:20 rafalcpp can I have a commit signed by separate GPG keys?
11:20 KekSi and you'll obviously have to merge manually in case there's collisions
11:20 grawity rafalcpp: separate from what?
11:20 rafalcpp e.g. devel A writes a commit, and in description adds "Reviewed-by: B". A -S signs it.  And then devel B also -S sign it
11:20 grawity no, that's not supported
11:20 rafalcpp grawity: by two keys
11:21 rafalcpp hmm too bad that would be nice. Other next idea to do so?
11:21 rafalcpp can B do an empty commit that doesn't change any files but just adds commit message (or e.g. copies old one)?
11:22 a3Dman joined #git
11:22 grawity that's possible, although I'm not sure if I see the point of all that
11:23 rafalcpp grawity: how else do you prove that both: A commited some code, and B says he reviewed it?
11:24 rafalcpp (without need to trust anyone besides these devels)
11:24 grawity if B is willing to PGP-sign stuff, then sign the email message saying such
11:24 rafalcpp grawity: and put it where?
11:24 rafalcpp users that will want to check this, do trust the developers (but not some git server), and want to download stuff from some (untrusted) git server
11:25 grawity well
11:25 grawity then they really *have to* trust someone who can commit directly to the repo
11:25 rafalcpp grawity: no, why?
11:26 rafalcpp code can be physicially pushed by someone yet another
11:26 mischat_ joined #git
11:26 rafalcpp either way, how to commit a commit that does no change to files?
11:28 nevodka rafalcpp why
11:28 grawity --allow-empty-commit
11:28 Keksike ok, thanks osse and KekSi
11:28 nevodka its --allow-empty
11:29 nevodka not -commit
11:29 Keksike offtopic: KekSi: finally I meet the person who has stolen my nick on freenode! damn you!
11:29 nevodka also rafalcpp next time just man git commit
11:29 gitinfo the git-commit manpage is available at http://jk.gs/git-commit.html
11:29 rafalcpp thanks;  nevodka why do this? as written above: so that users that trust only pubkey A and pubkey B can verify on their own that given code was commited by A and that B said he reviewed it
11:30 Macaveli joined #git
11:30 grawity have B merge A's branch and sign the merge
11:30 grawity though I guess that doesn't scale
11:30 rafalcpp grawity: yeah that is actually even better. Actually reviews would rather happen at merge point
11:32 MattMaker joined #git
11:34 spac3h joined #git
11:34 humboldt_ joined #git
11:35 UniFreak joined #git
11:36 kaldoran joined #git
11:37 humboldt joined #git
11:39 clickjack joined #git
11:40 hhee joined #git
11:43 M-ou-se sometimes when i push to my server, i get 'remote: fatal: git cat-file <hash>: bad file'. sometimes a few of those lines.
11:43 M-ou-se but afterwards everything seems to work fine. second push doesn't give such an error. pulling works fine, etc.
11:44 humboldt joined #git
11:45 osse M-ou-se: That comes from a serverside hook
11:45 M-ou-se ah!
11:46 osse Probably post-receive since it doesn't affect the outcome of the push
11:46 M-ou-se osse: didn't think of that at all. thanks. i was scared my files were getting corrupted. but yes, i have a hacky hook installed.
11:50 glebihan_ joined #git
11:51 glebihan joined #git
11:59 humboldt joined #git
12:00 ankit01ojha joined #git
12:00 durre joined #git
12:00 Tobbi_ joined #git
12:03 roelmonnens joined #git
12:05 zulutango joined #git
12:06 mischat joined #git
12:06 Anticom joined #git
12:06 peterppp joined #git
12:07 chipotle joined #git
12:07 jameser joined #git
12:08 humboldt joined #git
12:08 anry joined #git
12:11 robattila256 joined #git
12:11 venmx joined #git
12:12 a_thakur joined #git
12:13 a_thakur joined #git
12:13 b3kNi joined #git
12:14 thebope joined #git
12:15 jameser joined #git
12:16 shtrb joined #git
12:16 bvcosta joined #git
12:16 humboldt joined #git
12:17 jameser joined #git
12:17 jorgevidaurre joined #git
12:18 xaviergmail_ joined #git
12:19 stamina joined #git
12:20 robotrolll joined #git
12:21 notebox joined #git
12:23 YuGiOhJCJ joined #git
12:25 peterppp joined #git
12:25 souther joined #git
12:26 thebinary joined #git
12:26 humboldt joined #git
12:27 govg joined #git
12:28 thebird joined #git
12:29 humboldt_ joined #git
12:30 explody joined #git
12:31 souther joined #git
12:33 MattMaker joined #git
12:33 bvtton joined #git
12:33 diego1 joined #git
12:36 raijin joined #git
12:36 vuoto joined #git
12:37 jleclanche_ joined #git
12:37 kutenai_ joined #git
12:37 souther joined #git
12:37 KG4PEQ_ joined #git
12:37 glebihan joined #git
12:37 Snugglebash joined #git
12:37 humboldt_ joined #git
12:38 deepy joined #git
12:38 rxo_ joined #git
12:39 dtype joined #git
12:39 TooLmaN joined #git
12:39 ohcibi joined #git
12:39 rfv joined #git
12:39 dustinm joined #git
12:39 maxorator joined #git
12:40 XDS2010 joined #git
12:40 timwis joined #git
12:40 icedp joined #git
12:40 uranellus joined #git
12:41 uranellus joined #git
12:41 gsingh93 joined #git
12:42 souther joined #git
12:42 Olipro joined #git
12:43 Tobbi_ joined #git
12:43 PaulePanter joined #git
12:43 d0nn1e joined #git
12:45 cqi joined #git
12:45 kristofferR joined #git
12:46 nicolai86 joined #git
12:46 MattMaker joined #git
12:48 souther joined #git
12:49 jameser joined #git
12:49 Anticom joined #git
12:50 rorro joined #git
12:55 treehug88 joined #git
12:55 bvcosta joined #git
12:55 treia joined #git
12:56 miczac joined #git
13:01 mischat joined #git
13:01 Rodya_ joined #git
13:01 scoobertron joined #git
13:02 chamunks- joined #git
13:03 peterppp joined #git
13:03 yena_ joined #git
13:04 bcardarella_ joined #git
13:04 addshore_ joined #git
13:05 Chinorro_ joined #git
13:05 Pie_Napple joined #git
13:05 Guest55219 joined #git
13:05 yaiyan joined #git
13:06 addshore joined #git
13:06 fahadash joined #git
13:06 AlexanderS joined #git
13:06 dsdeiz joined #git
13:06 Nicmavr joined #git
13:07 synthroid joined #git
13:08 thebope joined #git
13:08 dvaske joined #git
13:10 b3kNik joined #git
13:11 Tobbi_ joined #git
13:12 c0c0 joined #git
13:12 frobnic joined #git
13:13 AciD` joined #git
13:13 AciD` joined #git
13:13 acetakwas joined #git
13:13 dedicated joined #git
13:13 MuffinMedic joined #git
13:13 humboldt joined #git
13:13 whaley joined #git
13:14 howdoi joined #git
13:14 GGMethos joined #git
13:15 VampiricPadraig joined #git
13:15 VampiricPadraig joined #git
13:15 markovh joined #git
13:15 rasto_ joined #git
13:17 jameser joined #git
13:17 pwndave joined #git
13:19 GGMethos joined #git
13:19 cdesai joined #git
13:23 miczac joined #git
13:27 a_thakur joined #git
13:27 marcogmonteiro joined #git
13:28 rvgate joined #git
13:28 peacememories joined #git
13:28 jeffreylevesque joined #git
13:30 a_thakur_ joined #git
13:30 aspiers joined #git
13:31 jameser joined #git
13:31 humboldt joined #git
13:32 p4trix joined #git
13:32 Raed joined #git
13:34 explody joined #git
13:34 ghoti joined #git
13:35 tgy left #git
13:35 overlord_tm joined #git
13:35 miczac_ joined #git
13:36 navidr joined #git
13:36 toydestroyer joined #git
13:36 Kestrel-029 joined #git
13:36 Caelum joined #git
13:36 b3kNi joined #git
13:37 m4sk1n joined #git
13:38 ome joined #git
13:38 icedp joined #git
13:38 kurdzman joined #git
13:39 Tobbi_ joined #git
13:44 jameser joined #git
13:46 hhee joined #git
13:47 Masber_080 joined #git
13:48 bongjovi joined #git
13:48 Masber_080 hi
13:48 gitinfo Masber_080: hi! I'd like to automatically welcome you to #git, a place full of helpful gits. Got a question? Just ask it — chances are someone will answer fairly soon. The topic has links with more information about git and this channel. NB. it can't hurt to do a backup (type !backup for help) before trying things out, especially if they involve dangerous keywords such as --hard, clean, --force/-f, rm and so on.
13:50 nettoweb joined #git
13:51 AKPWD joined #git
13:53 re1 joined #git
13:53 Kartagis left #git
13:53 ash_workz joined #git
13:54 nettoweb1 joined #git
13:54 BlackPhoenix joined #git
13:55 Tech joined #git
13:56 Darcidride joined #git
13:57 Anticom__ joined #git
13:57 miczac joined #git
13:58 Anticom__ Sorry, flaky internet connection
13:58 gugah joined #git
14:00 govg joined #git
14:01 mischat joined #git
14:02 Rodya_ joined #git
14:02 Tobbi_ joined #git
14:02 thebope joined #git
14:02 oaao joined #git
14:03 raijin joined #git
14:05 jameser joined #git
14:05 aspiers joined #git
14:06 jimi_ joined #git
14:07 cqi joined #git
14:09 mikecmpbll joined #git
14:13 Snugglebash joined #git
14:15 rnsanchez joined #git
14:16 humboldt joined #git
14:16 bannakaffalatta joined #git
14:16 jameser joined #git
14:17 courrier joined #git
14:17 theoceaniscool joined #git
14:19 humboldt joined #git
14:21 Anticom joined #git
14:21 notebox joined #git
14:21 lowercas_ joined #git
14:22 ShekharReddy joined #git
14:23 chipotle joined #git
14:26 oaaao joined #git
14:26 _UniFreak joined #git
14:26 lektrik joined #git
14:27 zzz joined #git
14:27 lektrik joined #git
14:27 svara1 joined #git
14:27 UTAN_dev joined #git
14:28 jameser joined #git
14:28 griffindy joined #git
14:31 dreiss joined #git
14:32 marc_v92 joined #git
14:33 Achylles joined #git
14:34 Achylles joined #git
14:34 texinwien_ joined #git
14:34 e14 joined #git
14:34 Tobbi_ joined #git
14:35 mischat joined #git
14:35 XenophonF joined #git
14:36 qqx joined #git
14:37 r_rios joined #git
14:37 crose joined #git
14:39 qqx joined #git
14:39 jost joined #git
14:39 cdown joined #git
14:40 govg joined #git
14:42 modin joined #git
14:42 marcogmonteiro joined #git
14:45 nopacien13 left #git
14:45 xaviergmail_ joined #git
14:45 irqq joined #git
14:46 ResidentBiscuit joined #git
14:47 cdg joined #git
14:48 GodGinrai joined #git
14:49 Es0teric joined #git
14:51 AaronMT joined #git
14:52 skered joined #git
14:52 mischat_ joined #git
14:52 CtrlC joined #git
14:53 mischat_ joined #git
14:53 Tobbi_ joined #git
14:54 nickabbey joined #git
14:54 Infamous left #git
14:56 thebope joined #git
14:58 Raging_Hog joined #git
14:58 re1 joined #git
14:58 Achylles joined #git
15:00 Dougie187 joined #git
15:02 Awesomecase joined #git
15:02 mischat joined #git
15:02 shtrb left #git
15:04 MineCoins joined #git
15:06 kexmex joined #git
15:07 seveneve1 joined #git
15:07 jameser joined #git
15:09 LeBlaaanc joined #git
15:09 pur3eval joined #git
15:09 denisMone joined #git
15:11 Tobbi_ joined #git
15:11 raspado joined #git
15:12 jagob joined #git
15:12 le_melomane joined #git
15:13 cdown_ joined #git
15:13 intellix joined #git
15:15 govg joined #git
15:16 re1 joined #git
15:16 kpease_ joined #git
15:17 humboldt joined #git
15:19 Celelibi Is there a way to setup a global hook for all my projects?
15:20 _ikke_ I don't think so. The only thing I can think of is adjusting the repository template, but that only affects new repositories
15:20 jast you can set a different hooks dir to use by default
15:20 _ikke_ Ah, that might work too
15:20 kpease joined #git
15:20 jast but that's not a hassle only if all of your repos will be the same
15:21 _ikke_ right
15:21 _ikke_ It doesn't allow you to have separate hooks anymore
15:21 jast the config option is core.hooksPath (man git-config)
15:21 gitinfo the git-config manpage is available at http://jk.gs/git-config.html
15:21 kpease_ joined #git
15:22 Celelibi I think changing the template dir would be the right solution if I can put symlinks in it.
15:23 prsn joined #git
15:24 Celelibi I think it's a design limitation to have a single script that run as a hook instead of a whole dir. So I would have pre-commit that runs everything in pre-commit.d
15:24 _ikke_ on my system it's /usr/lib/git-core/templates
15:25 jast I'm fairly sure symlinks work just fine
15:25 Celelibi But at the same time I would like to have my hook that prevent pushing a commit whose message start with "nopush!".
15:25 scoobertron joined #git
15:25 jast and if you want to use hook subdirs, put standard hooks in place that iterate over a subdir in a location of your choice and run all the scripts within
15:25 qqx Yes, symlinks work fine for the templates.
15:25 Celelibi And the actual script should be in a single place so it can be updated.
15:26 Achylles joined #git
15:26 Celelibi So yeah, I might put this in place at some point.
15:27 blackwind_123 joined #git
15:28 marc_v92 joined #git
15:28 shinnya joined #git
15:28 Tobbi joined #git
15:29 _ikke_ Celelibi: https://public-inbox.org/git/7vd4f3z​8xu.fsf@gitster.siamese.dyndns.org/
15:30 _ikke_ Why multiple hooks weren't implemented
15:31 aspiers joined #git
15:33 publio joined #git
15:35 Celelibi Ok. Fair enough.
15:35 TomyLobo joined #git
15:35 clemf joined #git
15:35 Es0teric joined #git
15:37 iamthad joined #git
15:37 le_melomane joined #git
15:38 Levex joined #git
15:38 nettoweb joined #git
15:39 MattMaker joined #git
15:41 Nilesh_ joined #git
15:42 mischat_ joined #git
15:46 Es0teric joined #git
15:47 mischat joined #git
15:48 satifant joined #git
15:50 thebope joined #git
15:54 shgysk8zer0_ joined #git
15:56 Es0teric joined #git
15:56 aavrug joined #git
15:57 tomasm- joined #git
15:58 tomasm- hey I had to fix some code on a github project, I created my own fork, made the changes there, now how do I submit the changes (notify) back to the original codebase repo/owner?
15:58 moritz tomasm-: you create a pull reuest
15:59 shgysk8zer0_ joined #git
15:59 tomasm- moritz, yeah it just says "Choose different branches or forks above to discuss and review changes" and is disabled..... both branches say "develop"... i dont think either of them are my copy
16:00 moritz tomasm-: did you already commit and push your changes?
16:00 tomasm- I did the changes using the web browser
16:01 moritz sorry, never did that; try asking in #github, maybe?
16:01 roelmonnens joined #git
16:02 LosPup joined #git
16:02 jost_ joined #git
16:02 tomasm- ok well my branch is 2 commits ahead of the original
16:03 Rodya_ joined #git
16:03 WayToDoor joined #git
16:09 Timvde joined #git
16:09 clemf joined #git
16:10 crayon joined #git
16:11 mischat_ joined #git
16:12 mischat_ joined #git
16:13 jagob joined #git
16:15 mischat joined #git
16:15 Tacoder joined #git
16:15 nickabbey joined #git
16:16 re1 joined #git
16:18 dedicated_ joined #git
16:18 jnavila joined #git
16:19 navidr joined #git
16:20 cdown joined #git
16:22 User458764 joined #git
16:23 User458764 joined #git
16:24 ExoUNX joined #git
16:25 WayToDoor joined #git
16:30 intellix joined #git
16:31 boombatower joined #git
16:33 ams__ joined #git
16:33 mischat joined #git
16:35 marc_v92 joined #git
16:36 Gustavo6046 joined #git
16:39 marc_v92 joined #git
16:40 thebird joined #git
16:40 clemf joined #git
16:40 PioneerAxon joined #git
16:40 MattMaker joined #git
16:40 rivarun joined #git
16:41 Cavallari joined #git
16:44 thebope joined #git
16:45 lss8 joined #git
16:45 lss8 joined #git
16:45 brent__ joined #git
16:46 bvcosta joined #git
16:47 finalbeta joined #git
16:48 mehola joined #git
16:48 hue joined #git
16:48 telephone joined #git
16:48 telephone joined #git
16:51 nettoweb joined #git
16:55 madewokherd joined #git
16:57 miczac joined #git
16:58 brent__ joined #git
16:58 lucido joined #git
17:00 lucido hi, I did git checkout somehash, then change a few things around while playing with the code. How can I get back to another brach and discard the changes I made to the working copy at that hash?
17:01 lucido git status says HEAD detached at c4ed43b
17:01 xaviergmail_ joined #git
17:01 skered joined #git
17:01 moritz git checkout .; git checkout otherbranch
17:02 lucido moritz, what does the dot mean?
17:02 moritz lucido: "current directory"
17:02 xaviergmail_ joined #git
17:02 nickabbey joined #git
17:03 lucido moritz, how can git checkout a a directory
17:03 lucido I cant interpret that
17:03 zzz joined #git
17:04 Rodya_ joined #git
17:05 nickabbe_ joined #git
17:05 pks joined #git
17:06 aavrug_ joined #git
17:06 qqx lucido: When <paths> or --patch are given, git checkout does not switch branches. It updates the named paths in the working tree from the index file or from a named <tree-ish>
17:07 qqx That's a quote from the git-checkout man page.
17:07 gugah joined #git
17:07 kexmex joined #git
17:07 qqx The . would satisfy the <paths> condition.
17:08 qqx Although that assumes that the only modified files are in or below your current directory. If that's not the case you could use `:/` instead of `.`
17:08 lucido qqx, thanks, I think I need to read up on how git works
17:09 GT4066 joined #git
17:09 perlpilot lucido: Read !progit
17:09 gitinfo lucido: [!book] There are several good books available about git; 'Pro Git' is probably the best: http://git-scm.com/book but also look at !bottomup !cs !gcs !designers !gitt !vcbe and !parable
17:09 lucido thanks
17:11 multi_io joined #git
17:12 jstimm joined #git
17:13 ExoUNX joined #git
17:15 nettoweb joined #git
17:16 hasc joined #git
17:17 finalbeta joined #git
17:18 elsevero joined #git
17:18 solarfly joined #git
17:19 Levex joined #git
17:22 orbyt_ joined #git
17:23 pur3eval joined #git
17:23 tang^ joined #git
17:24 raynold joined #git
17:25 CheckDavid joined #git
17:26 a_thakur joined #git
17:28 Ryanar joined #git
17:28 Peetz0r joined #git
17:29 j7k6 joined #git
17:30 math-alpha joined #git
17:30 rgrinberg joined #git
17:37 thebird joined #git
17:38 Murii_ joined #git
17:38 nettoweb joined #git
17:38 thebope joined #git
17:40 PioneerAxon joined #git
17:41 MattMaker joined #git
17:42 NimeshNeema joined #git
17:43 JamesBaxter joined #git
17:45 jnavila joined #git
17:45 Dougie187 left #git
17:46 a_thakur_ joined #git
17:48 guampa joined #git
17:51 Jonuz joined #git
17:52 ghost00 joined #git
17:52 aspiers joined #git
17:56 glspi joined #git
17:56 svm_invictvs joined #git
17:57 User458764 joined #git
17:57 cebor_ joined #git
17:58 nickabbey joined #git
17:59 chardan joined #git
17:59 ij left #git
18:00 Ryanar joined #git
18:00 ToBeCloud joined #git
18:02 le_melomane joined #git
18:03 aspiers joined #git
18:05 chardan joined #git
18:05 Rodya_ joined #git
18:07 matsaman joined #git
18:12 Hozy joined #git
18:13 TomiNewbie joined #git
18:14 User458764 joined #git
18:15 MineCoins joined #git
18:15 moei joined #git
18:16 Es0teric joined #git
18:17 Tobbi joined #git
18:17 nod joined #git
18:17 dedicated joined #git
18:18 rwb joined #git
18:19 TomiNewbie I have 2 different servers I want to add to version control. And they share some code.
18:19 TomiNewbie Should I create one folder for each server, and a third folder for the shared code. And add the library as a submodule to the servers?
18:19 TomiNewbie Or should I create those 3 folders, but package the library as a composer package?
18:21 dreiss joined #git
18:21 TomiNewbie (And add the package to servers' composer.json as a dependency.)
18:22 davidfetter_ge joined #git
18:22 davidfetter_ge hi
18:22 maurer So, I don't know about your particular use case, but general rules of thumb:
18:22 kurdzman joined #git
18:22 maurer 1.) Only use submodule if you think it is rare that you'd want to edit one package while editing the other, e.g. if you could think of it as referencing someone else's project
18:22 davidfetter_ge i have an issue where build artifacts are accidentally getting checked in. is a pre-commit hook the appropriate place to run the cleaner script that makes this not happen?
18:23 maurer 2.) If you're making local changes to a subproject, but want to be able to push out only changes to that subproject for publishing purposes, or want to take upstream updates to it, check out git subtree
18:23 osse davidfetter_ge: that's a last resort thing. the appropriate thing to do: 1) have a proper .gitignore 2) run git status 3) don't use commit -a or add . blindly
18:24 pur3eval joined #git
18:24 maurer 3.) If neither of these apply, think carefully about why you want to put the files in separate repos - they probably don't belong in them
18:24 maurer (my list there is to TomiNewbie )
18:24 davidfetter_ge osse, sadly, the artifact is a file that's been modified by the build, but could conceivably be modified on purpose in the future. is this just a case of Don't Do That™?
18:24 maurer (If you don't feel like that helps you, feel free to explain in more detail what you mean by "two difrferent servers", and who writes/maintains the code)
18:25 maurer davidfetter_ge: My reccomendation here would be to make a pre-commit hook with a warning
18:25 osse davidfetter_ge: why may it be checked in in the future? why does it get modified=
18:25 maurer e.g. "Warning, you are about to commit an update to THE_BIG_ASSET, did you mean to do this? (y/n)"
18:26 davidfetter_ge maurer, the warning is complex to construct. the cleaner is already working.
18:26 davidfetter_ge osse, it's called manifest.yml, and the build injects a shortened SHA
18:26 davidfetter_ge i guess i could write up a warning for that case using y2j, now that i think of it
18:26 maurer davidfetter_ge: I'm mainly advising against the cleaner because git commit is normally assumed not to change the state of the repository. I'm imagining a script deleting files messing with e.g. git rebase
18:27 osse solution: don't put the sha1 there. it will never be correct
18:27 osse if you mean the commit sha
18:27 davidfetter_ge it's never correct to commit it, but it's required as part of the build
18:27 maurer davidfetter_ge: If that's the issue, you can add it to .gitignore
18:27 maurer You can have files listed in .gitignore present in the repository
18:27 maurer .gitignore just prevents them from accidentally being added to the index
18:27 osse i'd change the build procedure
18:28 davidfetter_ge maurer, i did not know that
18:28 qqx maurer: You can do that but it's useless.
18:28 guampa joined #git
18:28 osse maurer: that only works if it's not part of the repo initially
18:28 maurer Oh, derp
18:28 maurer my bad
18:28 le_melomane joined #git
18:29 re1 joined #git
18:29 TomiNewbie That is very helpful. It looks like I want the library in just 1 repository.
18:30 synthroi_ joined #git
18:30 TomiNewbie But I still want to have the production branch of my servers in version control, and they use that library. Currently each server has its own copy of the library.
18:30 duderonomy joined #git
18:31 TomiNewbie Is composer an appropriate solution or am I missing a simpler git solution?
18:31 Tech joined #git
18:32 synthro__ joined #git
18:32 thebope joined #git
18:32 thiago joined #git
18:33 Es0teric joined #git
18:34 chipotle joined #git
18:34 whaley joined #git
18:35 le_melomane joined #git
18:36 theoceaniscool joined #git
18:36 joshszep joined #git
18:36 davidfetter_ge so it turns out that i can get a true or false out of a pretty simple command.  how do i put that into a pre-commit hook that warns and fails?
18:37 osse davidfetter_ge: as output?
18:37 osse it prints true or false?
18:37 kurdzman joined #git
18:37 davidfetter_ge osse, yep
18:37 osse is the hook a shell script ?
18:38 davidfetter_ge the hook is TBD, so if a shell script works, that's what it'll be
18:38 osse davidfetter_ge: it can be anything. git treats it just like an executable. you can write them in sh, bash, python, ruby, php, assembly, ...
18:38 osse your choice
18:38 davidfetter_ge the command is a jq query that i can massage
18:39 hasc joined #git
18:39 davidfetter_ge right. so basically, a warning to STDERR and a return code of non-zero will do the trick?
18:39 osse yes
18:39 osse #!/bin/sh
18:39 davidfetter_ge great :)
18:39 osse [ "$(command here)" != true ]
18:39 osse done
18:40 djb-irc joined #git
18:40 * davidfetter_ge tries to figure out whether it's worth figuring out how to get jq to send things to stderr
18:40 m0j0 joined #git
18:40 osse i think you should figure out how to avoid needing to change that file \o/
18:40 eneasvva joined #git
18:41 eneasvva Hi guys
18:41 eneasvva I'm in a very big problem
18:41 osse davidfetter_ge: even simpler:  rename it to manifest.yml.in, make the thing that puts the sha1 in there write to manifest.yml. done.
18:41 eneasvva I renamed the branches in local and delete remote branch in order to push new names branches
18:41 davidfetter_ge osse, i assure you that work's already been done to that end, but mistakes still happen
18:42 eneasvva but surprise, my git client can't do fetch or pull at now
18:42 MattMaker joined #git
18:42 eneasvva it tells my that all is already update but is false
18:43 osse davidfetter_ge: mistakes are fine.  but by editing a committed file as part of an automatic build you're asking for it
18:43 bannakaf_ joined #git
18:43 davidfetter_ge osse, if i were alone on this team, i would definitely do it this way. i have to choose battles carefully
18:44 osse if you really are between a rock and a hard place on this one then I wouldn't even bother with the hook. let them deal with it.
18:44 osse if you write a pre-commit hook you would need to tell them to use it anyway
18:45 osse maybe i'm spoiled
18:45 davidfetter_ge it's not that they're unwilling, just under a lot of time pressure
18:46 davidfetter_ge changing the build system in this way needs to be a very simple and bullet-proof change. i've noticed that changes that are bullet-proof take a lot of development effort
18:46 osse Depending on how the automatic editing of the file happens I'd say my solution takes somewhere between 1 and 3 minutes to implement :p
18:46 osse and they don't have to do anything, probably. they just notice that  manifest.yml  no longer exists and manifest.yml.in does and go "oh, ok then"
18:47 davidfetter_ge ok, i think i have enough takeaways to form the basis of a short and productive discussion with the team
18:47 s1scha joined #git
18:48 osse davidfetter_ge: i'm not shitting on your pre-commit hook btw. there's nothing wrong with having that. but sharing such a hook among a team is a lot of work in its own. they aren't automatically used when a new person clones or pulls, etc.
18:48 davidfetter_ge osse, thanks for clarifying that :)
18:48 intellix joined #git
18:48 e14 joined #git
18:48 davidfetter_ge that is indeed a task that's easy to get wrong
18:49 * davidfetter_ge wonders whether there's a way to have git inits that do this
18:49 guampa joined #git
18:49 davidfetter_ge like if we're bought into the hook system, is there some straightforward way to stamp out our template?
18:50 osse there is no way with a vanilla git. but you can commit the hooks as script files in the repo and add a script to install them, or make custom target in a build system, or whatever is appropriate
18:50 osse davidfetter_ge: what do you mean by that?
18:50 davidfetter_ge osse, so let's imagine we want everybody's git repos to have certain hooks in .git/hooks
18:51 xall_ joined #git
18:51 davidfetter_ge surely somebody's figured out ways to make this so. is a git wrapper a usual way?
18:51 osse i don't know what's usual
18:52 Eugene davidfetter_ge - the way I understand your original question, you don't want compiler output to be tracked? What's wrong with gitignore?
18:52 qqx Git does use a template directory to populate the contents of new repos as part of init, but in most cases that just moves the problem.
18:52 osse Eugene: the compiler's output shares name and path with an input
18:52 osse sed -i style
18:52 Eugene So he's crazy, got it
18:53 Eugene You can do a pre-commit hook and put it inside the repo(under hooks/). Add a note to the README explaining that users need to symlink it into .git/hooks/
18:53 jstimm joined #git
18:54 kasual joined #git
18:54 Eugene You can also do a pre-receive hook(on your git server) to explicitly deny any such accidental commit contents from being pushed. You should explain in the error message how to remove/rewrite these commits out
18:54 Eugene But wrapping `git` for this is silly
18:54 causasui joined #git
18:55 osse 'tis better to wrap git than to git wrap
18:56 osse -- Olivia Wilde
18:56 Eugene git-rap --> !jokes
18:56 gitinfo A complete listing of the git-jokes used in the /topic can be found at https://madeitwor.se/git-jokes (patches welcome)
18:58 elsevero joined #git
18:58 solarfly joined #git
18:58 nidr0x joined #git
19:00 tureba joined #git
19:03 cdesai joined #git
19:03 Tobbi joined #git
19:03 doener joined #git
19:05 Rodya_ joined #git
19:06 kexmex joined #git
19:07 durre joined #git
19:07 tureba joined #git
19:10 Levex joined #git
19:13 jas4711 joined #git
19:14 matoro joined #git
19:14 cdesai joined #git
19:16 Murii_ joined #git
19:19 Sasazuka joined #git
19:21 ShapeShifter499 joined #git
19:23 Dougie187 joined #git
19:25 mischat joined #git
19:26 thebope joined #git
19:27 Kurvivor joined #git
19:28 Kurvivor hello! need help with restoring state of my repository
19:28 elsevero_ joined #git
19:28 Kurvivor due to shenengians its .git folder was lost
19:28 modin joined #git
19:28 Kurvivor it currently contains local changes
19:28 jnavila joined #git
19:28 fatalhalt joined #git
19:28 _ikke_ Kurvivor: Then you need to clone it again
19:28 Kurvivor i have reinitialised repository, added remote and fetched from it
19:28 _ikke_ right
19:29 dreiss joined #git
19:29 goTAN joined #git
19:29 Kurvivor is there a way to somehow tell git that "current working folder is the tip of branch 'master' with some changes"?
19:29 jimsio joined #git
19:29 Kurvivor it seems that reset --soft might do it, but for that branches must already exist
19:30 gp5st joined #git
19:30 elsevero joined #git
19:30 Kurvivor how can i create local branch from remote one without changing working directory?
19:30 gp5st does `git fetch` and `git push` just run `ssh` with a command?
19:32 _ikke_ Kurvivor: update-ref master origin/master
19:32 gp5st s/does/do/
19:32 qqx gp5st: For ssh remotes, yes. At least for the communication part.
19:32 gp5st qqx yes, I should have specified :)
19:32 devslash joined #git
19:32 Eugene Yup. man git-remote-helper
19:32 Eugene man git-remote-helpers ?
19:32 gitinfo the git-remote-helpers manpage is available at http://jk.gs/git-remote-helpers.html
19:32 Eugene Figures
19:32 Balliad joined #git
19:33 kurdzman joined #git
19:34 thiago joined #git
19:34 devslash im learning git and want to understand something
19:35 gp5st devslash: that's always a good first step
19:35 JamesBaxter joined #git
19:36 devslash if i check out a repo, make a change to a file and commit my changes,what happens when 2 people have committed changes to the same file
19:38 kurdzman joined #git
19:38 elsevero joined #git
19:38 kadoban devslash: When you want to combine them, usually you "merge". Often git can figure out automatically how to combine your changes. Otherwise it tells you there's a conflict and you have to resolve it (which just means, changing the file to be such that it includes both sets of changes, whatever that means)
19:39 devslash ok so you have to resolve the merge conflict
19:39 BlackMaria joined #git
19:39 kadoban Potentially, yes.
19:39 kadoban Not all changes to the same file are conflicts though. And even if they are it's usually pretty easy to resolve once you get used to it.
19:39 qqx And a merge needs to be done even if the changes weren't to the same file. Although in that case it's pretty much guaranteed that git can automatically resolve things.
19:39 devslash and how would you specify one merge over the other
19:40 devslash err commit
19:40 kadoban devslash: Not sure what that means. !eek goes through some of the details.
19:40 gitinfo devslash: [!eekaconflict] Merge conflicts are a natural part of collaboration. When facing one, *don't panic*. Read "How to resolve conflicts" in man git-merge and http://git-scm.com/book/ch3-​2.html#Basic-Merge-Conflicts then carefully go through the conflicts. Picking one side verbatim is not always the right choice! A nice video explaining merge conflicts: https://www.youtube.com/watch?v=zz7NuSCH6II
19:43 MattMaker joined #git
19:47 gp5st also, as kadoban alluded to, manual merges don't happen as frequently as they do in svn or other source control systems. git's pretty good about only making you do it when there is really nothing it can do automatically
19:47 synthroid joined #git
19:48 Dandelock joined #git
19:49 devslash so the order is something like git clone for an existing repo or git init for a new repo, create file(s), git add --all, git commit -m "comment", git merge right
19:49 xall joined #git
19:50 kadoban Usually something more like you push the commit(s) somewhere and tell upstream to merge them for you. But sure, close enough.
19:50 kadoban It depends a lot on workflow, which git is flexible with.
19:51 synthroi_ joined #git
19:51 kurdzman joined #git
19:51 devslash I might be getting a job where all work is saved using git to push code changes
19:51 devslash I think I understand enough about git as far as checking out code, making changes and pushing the changes
19:52 aidalgol joined #git
19:53 Gsham joined #git
19:53 l4v2 joined #git
19:53 irqq joined #git
19:53 synthro__ joined #git
19:54 pur3eval joined #git
19:56 nettoweb joined #git
19:56 dave0x6d joined #git
19:56 cliluw joined #git
19:57 xaviergmail_ joined #git
19:58 explody joined #git
20:02 kurdzman joined #git
20:03 Levex joined #git
20:05 Balliad joined #git
20:05 hhee joined #git
20:05 a_thakur joined #git
20:06 Rodya_ joined #git
20:07 kurdzman joined #git
20:07 nowhereman joined #git
20:08 gopar joined #git
20:08 guampa joined #git
20:09 blackwind_123 joined #git
20:09 aspiers joined #git
20:09 acidjnk joined #git
20:09 acidjnk hello
20:10 feodoran joined #git
20:11 bannakaffalatta joined #git
20:12 humboldt joined #git
20:12 devslash how do you uncommit a change that youve already made
20:12 osse devslash: git reset HEAD~1
20:13 gp5st devslash: short version is you don't, you make a commit that undoes it
20:13 devslash oh
20:13 devslash oh i see ok
20:13 devslash why HEAD~1 and not HEAD
20:13 ISmithers joined #git
20:13 Es0teric joined #git
20:13 gp5st devslash: longer version is the reset or re-base IF AND ONLY IF you havn't pushed your commit (then someone else could be using it)
20:14 _ikke_ devslash: HEAD is the current commit
20:14 gp5st devslash: HEAD is the current commit...the one you want to get rid of
20:14 devslash i know. you said HEAD~1
20:14 _ikke_ git reset requires a commit to go to
20:14 devslash wouldnt you do git reset HEAD
20:14 _ikke_ not the commit you want to undo
20:14 gp5st ~1 means back 1 parent
20:14 devslash ahh
20:14 _ikke_ devslash: THe only thing that would do is unstage everything
20:14 gp5st ^ means back one parent too
20:14 devslash thanks
20:14 gp5st HEAD^
20:15 gp5st but again, if you've pushed, please don't force push
20:15 devslash what do you mean force push
20:16 whaley_ joined #git
20:16 gp5st devslash: you could always `git revert HEAD` as well, which makes an exactly opposite commit as the most recent commit, allowing you to explain why you wanted to get rid of it
20:16 humboldt joined #git
20:16 gp5st devslash: if you can't fast-forward when you push, you can't push and need to "force" it
20:17 devslash but if i made a commit and push the changes, i could also just revert the changes in the file then commit and push those changes again right
20:17 Murii_ joined #git
20:17 devslash i mean, would that be considered better practice
20:18 gp5st that would be a forced push if you get rid of the commit; you mean create a new commit undoing the changes (git revert), then that would be a normal push
20:18 gp5st if you mean*
20:18 devslash yes
20:18 ahr3n joined #git
20:18 devslash i mean a normal push
20:18 m0viefreak joined #git
20:18 devslash so if I want to undo a push made previously, I could always undo the change and repush the new file
20:19 gp5st yes, that is best practice for a commit that's been pushed
20:19 devslash but once Ive pushed it, I'd use HEAD~1  to check out the file before the push that I want to undo is that right ?
20:20 gp5st no, `git reset HEAD~1` will make HEAD it's parent, disregarding the old head. if you pushed that it wouldn't work and you'd need to force it
20:20 gp5st it's a big tree
20:20 gp5st HEAD is just a pointer to where you currently are in the tree
20:20 thebope joined #git
20:21 gp5st if you `git reset HEAD~1` it moves the pointer up the tree, and then if you commit again, it creates a second child and points head at that. the other child, where you started, is now orphaned
20:21 raspado joined #git
20:22 devslash ok so please help me understand
20:22 nitesh joined #git
20:22 devslash HEAD points to latest push that contains a file i want to revert to the previous push
20:22 gp5st no
20:23 gp5st HEAD points to the current state of the working tree
20:23 jstimm joined #git
20:23 blueyed joined #git
20:23 Sample joined #git
20:23 devslash oh I see
20:23 gp5st it need not be the "latest" of anything
20:24 disi joined #git
20:24 gp5st if you commit, HEAD is made to be the latest commit
20:24 devslash ok
20:24 sofly joined #git
20:24 devslash after the commit comes push right ?
20:25 redhedded1 joined #git
20:25 gp5st it doesn't have to
20:25 gp5st you can commit many times before you push
20:25 sq-one joined #git
20:25 gp5st in fact, there isn't a need to push if there isn't a copy of the repo
20:25 moritz https://www.youtube.com/watch?v=Ed2TIqWc_bs push! push! push!
20:26 gp5st I sometimes just create a repo for something I'm working on, and if I don't like it, it never gets pushed anywhere
20:26 moritz SCNR
20:26 neunon joined #git
20:27 feodoran "git clone ssh://user@host:path local/path" fails with "fatal: No path specified. See 'man git-pull' for valid url syntax" why is that?
20:27 gitinfo the git-pull manpage is available at http://jk.gs/git-pull.html
20:27 grawity because that's not a valid URL syntax
20:28 sofly joined #git
20:28 sofly joined #git
20:28 grawity it's either user@host:path (SCP/RCP syntax) or ssh://user@host/path (URL syntax, like http)
20:28 explody joined #git
20:29 gp5st feodaron: when you have the protocol in the url, the : is for a port specifier. scp/rcp syntax is a special snowflake
20:30 feodoran ah, thank you
20:31 feodoran the path after after host: is it relative to home or absolute?
20:31 __rgn joined #git
20:31 grawity for SCP syntax it's relative
20:31 shgysk8zer0_ joined #git
20:31 grawity in URLs, I think it's absolute
20:31 grawity unless you use ssh://host/~/path of course
20:32 hahuang65 joined #git
20:32 grawity hmm, ~ not accepted there anymore?
20:32 grawity I was sure it used to work
20:32 clemf_ joined #git
20:32 feodoran absolute works, thx
20:32 ToxicFrog I think it's up to the server to handle the ~; it's not guaranteed to work.
20:33 ToxicFrog I've only seen it used commonly in http:// and https:// URLs.
20:34 Kurvivor i have this problem: warning: refname 'master' is ambiguous.
20:34 robattila256 joined #git
20:34 rwp For ~ in ssh it always resolves to the $HOME of the user. For ~ in web urls https/http it depends upon the configuration of the web server. Can work. Works for me when I set it up on my servers. But won't if the web host has not configured it.
20:34 kadoban Kurvivor: Did you make a tag called master or something?
20:35 Kurvivor i am pretty sure i did not, but i will check
20:35 Kurvivor no, i have not
20:35 Kurvivor however, i have run update-ref on master
20:36 gp5st ToxicFrog: that's an apache thing that drops you into the user following the ~'s $HOME/public folder
20:37 thebird joined #git
20:37 kadoban What did you run exactly? Was this just before the problem started? I'm not very familiar with update-ref, probably can't help.
20:38 kadoban You could check "find .git/refs/ -type f" maybe, see if there's two things called master, I dunno.
20:38 LiftLeft joined #git
20:38 kadoban and/or maybe  find .git/ -name 'master' I dunno
20:39 Kurvivor it seems that just switching to some other branchand discarding master (git branch -d master) got rid of ambiguouity
20:39 Kurvivor thanks for ansering me in any case
20:39 Balliad joined #git
20:39 raspado joined #git
20:39 kadoban Odd
20:40 kadoban Hope it deleted the right one I guess ...
20:41 diplomat270159 joined #git
20:42 diplomat270159 just a test message to try out HexChat
20:43 kadoban diplomat270159: I'm sure you can find a channel with fewer than 1200 people to test our your client, if you have further tests.
20:43 Atm0spher1c joined #git
20:43 MattMaker joined #git
20:43 diplomat270159 np ttyl sorry
20:43 kevr left #git
20:47 durham joined #git
20:47 le_melomane joined #git
20:47 pbrewczynski joined #git
20:47 texinwien_ joined #git
20:48 buckowski joined #git
20:53 e joined #git
20:53 durham joined #git
20:55 durham joined #git
20:55 hahuang65 joined #git
20:58 Balliad joined #git
21:00 r_rios joined #git
21:01 r_rios Is there a way I can remove all files from my current branch that are not present in another branch?
21:01 Es0teric joined #git
21:01 r_rios I know I can copy all files that are not in my current branch by using git checkout branch -- .
21:02 r_rios But that doesn't remove any files
21:03 osse git diff --name-only -z --diff-filter=A HEAD anotherbranch | xargs -0 git rm
21:04 cdg joined #git
21:04 osse r_rios:
21:05 r_rios Thanks :)
21:06 bannakaffalatta joined #git
21:07 Rodya_ joined #git
21:12 j7k6 joined #git
21:13 davimore joined #git
21:13 Rodya_ joined #git
21:13 askb joined #git
21:15 thebope joined #git
21:15 Rodya__ joined #git
21:15 crose joined #git
21:16 Masber_080 joined #git
21:17 Anja joined #git
21:17 rudi_s joined #git
21:18 le_melomane joined #git
21:18 VA3VNA joined #git
21:18 cdg_ joined #git
21:19 finalbeta joined #git
21:19 GodGinrai joined #git
21:19 VA3VNA joined #git
21:23 Levex joined #git
21:29 KostasDi joined #git
21:30 KostasDi hi there
21:30 gitinfo KostasDi: hi! I'd like to automatically welcome you to #git, a place full of helpful gits. Got a question? Just ask it — chances are someone will answer fairly soon. The topic has links with more information about git and this channel. NB. it can't hurt to do a backup (type !backup for help) before trying things out, especially if they involve dangerous keywords such as --hard, clean, --force/-f, rm and so on.
21:30 mgver joined #git
21:31 KostasDi When there are identical lines at the beginning and end of a hunk, is there a way to instruct git on which one to use?
21:31 aspiers joined #git
21:31 chardan joined #git
21:33 kfoonamalik joined #git
21:34 F0rTh3J3st joined #git
21:34 KostasDi for example, lets say I'm adding "def\none" in line 3 in this example: https://www.diffchecker.com/jL2VYSof but it thinks one\ndef was added in the end
21:35 MarioBranco joined #git
21:35 Es0teric joined #git
21:35 porsche944 joined #git
21:36 Levex joined #git
21:36 anbz joined #git
21:36 goTAN joined #git
21:37 KostasDi any ideas?
21:38 Levex joined #git
21:39 cdg joined #git
21:39 feodoran left #git
21:43 griffindy joined #git
21:43 pur3eval joined #git
21:44 MattMaker joined #git
21:46 Es0teric joined #git
21:47 osse KostasDi: there are ways to influence the output of git diff
21:47 osse KostasDi: but in the end it doesn't matter
21:47 osse KostasDi: git doesn't store the diff. it stores the two version of the file
21:47 KostasDi true, just being nitpicky
21:48 APic
21:48 KostasDi the patience algorithm is still giving me the same result
21:49 ISmithers joined #git
21:49 nowhereman joined #git
21:49 KostasDi short of modifying the actual commit, is there a nicer way to do it?
21:51 hahuang61 joined #git
21:51 safe joined #git
21:52 Klumben joined #git
21:52 osse probably not in in this case
21:52 phy1729 The commit doesn't store the diff
21:53 osse maybe --indent-heuristic
21:57 njbair joined #git
21:57 dsdeiz joined #git
22:00 ToBeCloud joined #git
22:00 thiago joined #git
22:03 gopar joined #git
22:03 DarkPsydeLord joined #git
22:04 gopar joined #git
22:05 dysfigured joined #git
22:05 swords joined #git
22:05 dysfigured joined #git
22:09 ResidentBiscuit joined #git
22:09 thebope joined #git
22:09 Rodya_ joined #git
22:10 aspiers joined #git
22:13 kyan Hi, where can I find a hook or option or something to make git yell at me when I try to git add a file that's bigger than like 1mb or something? Thanks!
22:13 zartu joined #git
22:15 cjohnson pre-commit hook
22:15 cjohnson https://git-scm.com/book/gr/​v2/Customizing-Git-Git-Hooks
22:15 hahuang61 joined #git
22:15 cjohnson this would obvioulsy only happen for those who install the given hook
22:16 kyan cool, thanks!
22:16 kyan (it's just for me)
22:17 blackwind_123 joined #git
22:17 Gsham joined #git
22:17 mikecmpbll joined #git
22:18 Chrona joined #git
22:19 xaviergmail_ joined #git
22:20 averell joined #git
22:20 jeffreylevesque joined #git
22:20 jimi_ joined #git
22:21 clemf joined #git
22:22 chardan I'm having a rebase problem I just can't get my head around. "git log" shows two commits. I want to squish them into one and delete all history (to push to github). "git rebase -i HEAD~1" proceeds, offering only one commit... so, "git log" still shows both commits. How can I see what's happening?
22:22 chardan Frankly, I'm just feeling like deleting all the git stuff and checking the files into a new project, but I know that's not The Right Way(TM).
22:23 cdg joined #git
22:24 rwp chardan, Get back one more commit before that commit. git rebase -i HEAD^^
22:24 ISmithers joined #git
22:24 Dougie187 joined #git
22:24 s1scha joined #git
22:25 rwp Or even one or three more behind it. "git rebase -i HEAD^^^^^" No harm in having a few more in the interactive rebase list.
22:26 chardan rwp: Hm... "fatal: Needed a single revision" / "invalid upstream HEAD^^^^^"
22:26 rwp Sorry. You didn't have enough commits in the list then.
22:26 alip joined #git
22:27 rwp Do you *only* have the two and you want to squeeze from two into one?  Hmm...  Let me try that.
22:27 chardan rwp: Yep, that's the case.
22:27 rwp That hits somewhat of a special case.
22:27 chardan rwp: And I want any of the history kept-- essentially "make head the only thing".
22:29 chardan Hmmm... maybe "rebase -i --root"...
22:29 * chardan reads a bit.
22:30 rqo joined #git
22:30 rwp The --root seems to work.
22:30 chardan here goes...
22:31 rwp Worked okay for me in my test case that I ran for you.
22:31 chardan I am wondering if something else is messed up: in the rebase editor, it shows the commits in the "wrong" order...
22:32 rwp It always shows them upside down for my taste.  That is normal.
22:32 Awesomecase joined #git
22:32 rwp It is *applying* them in top to down order. Which is why that is the order presented.
22:32 netj joined #git
22:32 rwp But it is the opposite of log order.
22:32 chardan Well... the commit I want is at the bottom. And if I say 's', it refuses.
22:32 rwp If the log showed them in application order top to bottom then the log would be in the reversed direction too.
22:33 chardan They are the opposite.
22:33 rwp Hmm...  Worked for me in my test case.  Try building a simple test case in another directory and try it there.  Just to prove that is the general way to do things.
22:33 chardan Sure, I'll try it.
22:33 chardan Thanks.
22:34 thiago chardan: the rebase editor shows them in the opposite order of git log
22:34 rwp I simply did an init, followed by 'date > file1' adding and committing, followed by 'date > file2' adding and committing. Then rebase -i --root, and squished the second commit into the first. Worked for me.
22:35 rwp After being in the editor for the first pass to select the commits and their disposition then it starts a second editor pass to allow you to combine the commit messages together.  Fix the commit message as desired, save, exit, and it squishes the commits together.
22:36 rwb joined #git
22:36 chardan rwp: Ok, I have that same situation, sec... aaan... okay, I'm not sure I entirely understand what I did differently, but it worked... thanks very much!
22:36 chardan (Now to try to understand what I actually did. ;>)
22:37 rwp chardan, Glad to help and good luck!
22:37 chardan rwp: Much appreciated, thanks folks!
22:37 MarioBranco joined #git
22:38 GavinMagnus joined #git
22:39 DarkPsydeLord joined #git
22:39 Darren_ joined #git
22:40 __rgn joined #git
22:40 Dougie187 joined #git
22:40 roelmonnens joined #git
22:42 dsdeiz joined #git
22:42 dsdeiz joined #git
22:43 skered joined #git
22:45 MattMaker joined #git
22:46 Sleepy63 joined #git
22:46 roelmonnens joined #git
22:47 joshszep joined #git
22:49 ozmage joined #git
22:49 joshszep joined #git
22:49 DarkPsydeLord joined #git
22:50 shgysk8zer0_ joined #git
22:50 DarkPsydeLord joined #git
22:51 pur3eval joined #git
22:51 DarkPsydeLord joined #git
22:53 mehola joined #git
23:00 theoceaniscool joined #git
23:00 exitcode1 joined #git
23:01 fmeerkoetter joined #git
23:01 matoro joined #git
23:01 sfix joined #git
23:03 thebope joined #git
23:03 goTAN joined #git
23:04 GavinMagnus left #git
23:05 j7k6 joined #git
23:13 Kulrak joined #git
23:13 Kulrak joined #git
23:16 Ntemis joined #git
23:17 Ntemis sup guys
23:17 elsevero joined #git
23:17 lo0open joined #git
23:17 SwiftMatt joined #git
23:17 Ntemis i would very much appreciate some help with updating my branch
23:18 lo0open Hey. I want to skip all the files under the directory bin/. How can this be achieved?
23:18 lo0open What do I need to put i .gitignore? :)
23:18 lo0open *in
23:18 Ntemis i have a branch i want to update with git pull remotename branchname but i want to pull from a specific commit and after
23:18 Ntemis doable?
23:19 Ntemis i must not use previous commits as i divert a lot and wont be clean
23:19 Ntemis but from one point and after all commits are a ok
23:19 Ntemis can anyone help me out?
23:20 goTAN joined #git
23:28 MasterNayru joined #git
23:32 recola joined #git
23:33 recola left #git
23:34 msonntag joined #git
23:34 Ntemis I have a branch i want to update with git pull remotename branchname but i want to pull from a specific commit and after i must not use previous commits as i divert a lot and wont be clean but from one point and after all commits are ok to merge can anyone help me out?
23:34 qqx Ntemis: No, that's not possible to do with pull.
23:35 Ntemis oh so how i do it? possible?
23:35 Ntemis i dont want to cherry-pick one by one
23:35 qqx The closest you can to is to fetch the remote branch and then cherry-pick the commits that you want.
23:35 Ntemis too many
23:35 Ntemis 1 month work
23:36 qqx You can pass firstId..lastId as a range to cherry-pick.
23:36 Ntemis it stops at merge commits
23:37 qqx There are merge commits? I'm out.
23:37 Ntemis :)
23:37 Ntemis ty btw
23:38 Ntemis pull eats everything :)
23:38 Ntemis but as you said i cant use it
23:38 Ntemis shame
23:39 Ntemis i wish i could do pull commit1...commitfinish
23:39 qqx Pull by default does a merge, which by design includes everything.
23:39 Ntemis if i use cherry-pick i have to stop at every merge commit
23:40 Ntemis not counter productive
23:40 qqx I know if there's going to be a good way to handle that with merge commits.
23:41 Ntemis must be a way
23:41 Ntemis we just dont know it yet
23:41 Ntemis :)
23:41 Ntemis just thought of one
23:42 jimi_ joined #git
23:42 Ntemis fork the repo and merge all the commits i want into one then cherry-pick it from the other
23:42 Ntemis one commit to rule them all
23:43 Ntemis but i wont keep authorship this way :(
23:44 qqx If that's an acceptable solution there's no need to fork. Could just do `git merge --squash`.
23:44 Ntemis branch i want to fetch is remote
23:45 Ntemis cant squash something is not local
23:45 qqx So you'd have to fetch first.
23:45 kyan joined #git
23:45 Ntemis ah i see
23:45 qqx That's going to be a requirement for any solution.
23:45 Ntemis yes you are right
23:45 qqx It might be possible with `git filter-branch`.
23:45 Ntemis hmm
23:45 Ntemis will try
23:46 Ntemis will keep authorships?
23:46 ESphynx joined #git
23:46 MattMaker joined #git
23:46 qqx Yes, that should preserve author/committer id/time
23:46 ESphynx guys, I’m going insane... Why would ‘git stash show -p stash@{0}’ shows me something completely different than ‘git stash list -p stash@{0}’ for the first stash diff it shows?
23:47 Ntemis ty qqx
23:48 gugah joined #git
23:49 ertes joined #git
23:49 ESphynx http://stackoverflow.com/questions/2​1738579/difference-between-git-stash​-show-p-stashn-and-git-show-stashn -- major annoyance had me pulling my hair...
23:50 madewokherd joined #git
23:50 svm_invictvs joined #git
23:50 Rodya_ joined #git
23:52 mischat joined #git
23:56 StrongBit joined #git
23:56 StrongBit hi
23:56 gitinfo StrongBit: hi! I'd like to automatically welcome you to #git, a place full of helpful gits. Got a question? Just ask it — chances are someone will answer fairly soon. The topic has links with more information about git and this channel. NB. it can't hurt to do a backup (type !backup for help) before trying things out, especially if they involve dangerous keywords such as --hard, clean, --force/-f, rm and so on.
23:56 StrongBit I am using git extensions
23:56 StrongBit and I have a gogs server using a self signed certificate
23:57 StrongBit I want to be able to connect to it
23:57 StrongBit and clone a repo
23:57 StrongBit what should I do?
23:57 thebope joined #git
23:57 foist_ joined #git
23:58 DolphinDream joined #git
23:58 foist_ left #git
23:59 Rodya_ joined #git

| Channels | #git index | Today | | Search | Google Search | Plain-Text | summary