IRCloggy #git 2020-09-22

Logs Search ←Prev date Next date→ Channels Documentation

Provider of IRC logs since 2005.
WARNING: As Freenode became unjoinable and lost all warnings in topics, we cannot log channels on Freenode anymore.

2020-09-22

sybariten and actually, what should i correctly call this newest repo at bitbucket00:01
rewt sybariten: `git fetch` to fetch commits that the remote has that you don't00:02
sybariten uh... i'm not sure it did something.... i mean it didnt ask for PW , which i would have expected00:05
hm wait i'm stupid , this repo specificcally, is public... hm00:06
← tiin57 left00:12
← perrier-jouet left00:15
→ tiin57 joined00:15
→ perrier-jouet joined00:16
sybariten no rewt i can't really see it happening ... i'm checking against the bitbucket site, the files there are different than what i have in my local dir here00:16
rewt what does git status say?00:17
and are you looking at the same branch?00:17
← Gurkenglas left00:19
← yokoyamachan left00:21
← blahboybaz left00:22
→ nvmd joined00:22
→ fstd_ joined00:23
sybariten hm, as far as i know i only have one branch.... isn that a slightly more advanced feature of git? Let me try git status00:24
aaah okay, now git status says i'm behind with 8 commits... so i should do git pull?00:24
rewt `git pull` is the same as `git fetch` followed by `git merge`, and since you've already done fetch, you can do just merge00:25
sybariten kind of confusing to me that the actual command git fetch itself, never said "you're behind with 8 commits" ... but that i have to do git status. But this is probably logical with how git works.00:25
← lankanmon left00:25
← arcatech left00:26
← fstd left00:27
fstd_ → fstd00:27
sybariten ok yeah that worked fine, thank you rewt !00:27
→ th_ joined00:31
← mowcat left00:36
← Aki-lucky left00:37
→ arcatech joined00:39
← tiin57 left00:42
← flex14 left00:44
→ tiin57 joined00:45
→ cupstudio joined00:52
cupstudio Hi... this might be a dumb question, because I haven't been able to find an answer using google.00:52
When I merge my development branch into master, master gets a merge commit00:53
Then it looks like develop is behind master - should I update develop from master and bring the merge commit in? That seems odd...00:53
maybe I'm just expecting the two branches to be equal and they won't ever be00:54
← Betal left00:57
← andreas303 left00:59
→ Betal joined00:59
→ andreas303 joined01:03
→ Xenguy joined01:04
← mozgalom left01:04
→ inkbottle joined01:06
← kilobyte_ch left01:06
← maxzor left01:06
→ mozgalom joined01:07
← zebrag left01:08
← tiin57 left01:12
← dfee left01:14
→ tiin57 joined01:16
← POQDavid left01:18
→ POQDavid joined01:19
← mindCrime_ left01:19
→ kilobyte_ch joined01:20
← F0rTex left01:20
→ F0rTex joined01:22
← aw1 left01:24
← kerframil left01:31
← dionysus69 left01:36
→ macrover joined01:38
→ greggerz joined01:41
← tiin57 left01:42
← vicfred left01:42
→ flex14 joined01:45
→ tiin57 joined01:46
← inkbottle left01:47
← heyitsrama left01:47
← mozgalom left01:47
→ inkbottle joined01:47
← flex14 left01:47
Shentino_ → Shentino01:48
→ flex14 joined01:48
← flex14 left01:51
→ flex14 joined01:51
← andreas303 left01:52
→ kerframil joined01:52
→ yonder joined01:52
→ andreas303 joined01:54
← gast0n left01:55
← flex14 left01:55
→ flex14 joined01:55
← flex14 left01:59
→ flex14 joined01:59
→ lankanmon joined02:00
→ mns joined02:00
→ heyitsrama joined02:04
← mns left02:04
→ gschanuel9 joined02:08
← gschanuel left02:11
gschanuel9 → gschanuel02:11
← tiin57 left02:12
← cupstudio left02:13
→ tiin57 joined02:15
← heyitsrama left02:16
← macrover left02:18
→ dastan joined02:18
→ Muimi joined02:20
← tryte left02:21
← oprypin left02:21
→ tryte joined02:21
→ raven-au_ joined02:21
← raven-au left02:21
raven-au_ → raven-au02:21
→ Wetmelon joined02:22
← bdeferme_ left02:22
→ oprypin joined02:23
→ Fusl joined02:25
→ bdeferme_ joined02:25
→ Xenguy_ joined02:27
← Xenguy left02:27
← Kaisyu7 left02:30
← lagothrix left02:31
→ lagothrix joined02:31
← greggerz left02:35
Xenguy_ → Xenguy02:37
→ yamashitatan joined02:38
→ macrover joined02:38
← cuppajoeman left02:39
→ Kaisyu7 joined02:39
← tiin57 left02:42
← pikapika left02:43
← macrover left02:43
← acetakwas left02:43
→ pikapika joined02:43
← POQDavid left02:43
→ acetakwas joined02:44
→ POQDavid joined02:44
→ dimi1947 joined02:45
→ tiin57 joined02:45
→ crose joined02:46
← Jookia left02:49
→ Jookia joined02:49
← acetakwas left02:51
← justanotheruser left02:51
→ acetakwas joined02:52
← Cabanoss- left02:53
→ Goplat joined02:55
← acetakwas left02:56
→ Cabanossi joined02:58
→ acetakwas joined02:58
→ snowkidimd joined02:59
← davispuh left03:03
← arcatech left03:05
← nikivi left03:06
→ nikivi joined03:06
← tiin57 left03:12
← wallace_mu left03:12
→ tiin57 joined03:15
← kerframil left03:18
← relipse left03:21
→ kerframil joined03:34
← AnAverageHuman left03:36
→ veegee joined03:37
→ blahboybaz joined03:42
← tiin57 left03:42
← yamashitatan left03:44
→ tiin57 joined03:45
→ vicfred joined03:46
igemnace you can expect them to be equal if a) you bring the merge commit back into develop (like you said), or b) if you forgo having the merge commit at all (e.g. a fast forward)03:47
some shops do b). they rebase all their branches onto master to make sure it's cleanly fast-forwardable every time. it depends on what your team values out of the VCS03:48
oh... he's not even here03:48
→ milkt joined03:52
← flex14 left03:54
→ macrover joined03:54
← iNs left04:03
→ iNs joined04:04
← thomasross left04:04
→ dfee joined04:05
← ninjin_ left04:06
→ ninjin_ joined04:06
← pac1_ left04:08
← donofrio left04:10
← tiin57 left04:12
→ heyitsrama joined04:12
→ tiin57 joined04:15
← veegee left04:22
← dfee left04:26
→ arcatech joined04:26
← jaziz left04:27
← fuzzmz left04:29
→ fuzzmz joined04:29
← rnsanchez left04:30
← vngx left04:30
← dastan left04:39
→ dastan joined04:39
← tiin57 left04:42
→ Newami joined04:43
→ tiin57 joined04:45
← corelax left04:45
← Newami left04:46
← vicfred left04:47
← heyitsrama left04:49
← macrover left04:54
→ corelax joined04:55
← dastan left04:56
← corelax left04:59
← FH_thecat left05:00
→ vicfred joined05:02
→ justanotheruser joined05:07
→ corelax joined05:10
← tiin57 left05:12
→ hexnewbie joined05:14
→ tiin57 joined05:14
← snowkidimd left05:14
← arcatech left05:22
← corelax left05:22
→ duderonomy joined05:22
→ corelax joined05:22
→ dfee joined05:23
← inkbottle left05:29
→ konrados joined05:37
← tiin57 left05:42
→ tiin57 joined05:45
← konrados left05:46
→ konrados joined05:49
→ theoceaniscool joined05:49
← blahboybaz left05:50
← kupi left06:02
← terabyte left06:03
→ mightyjens joined06:07
→ mirrorbird joined06:08
← tiin57 left06:12
→ qpdb joined06:13
→ tiin57 joined06:16
← kjartan left06:22
→ kjartan joined06:23
← lankanmon left06:25
→ tarkus joined06:26
← tryte left06:28
→ rfuentess joined06:28
→ tryte joined06:28
← nvmd left06:34
← tpaper left06:35
→ tpaper joined06:37
← independent left06:38
← HourGate left06:39
→ independent joined06:39
→ lankanmon joined06:41
← tiin57 left06:42
← kjartan left06:42
→ morfin joined06:44
morfin hello. How can i change old unpushed commit?06:45
→ tiin57 joined06:46
morfin seems like i need to rebase?06:47
← LordRyan left06:47
→ kjartan joined06:48
→ benfelin joined06:48
→ maxzor joined06:48
→ a-l-e joined06:49
selckin 'interactive rebase' is the magic keyword06:49
morfin sounds scary06:50
← mirrorbird left06:50
selckin as all things git are initially06:51
but its quite simple06:51
← Wetmelon left06:52
← thiago left07:01
← unluckyshrubbery left07:02
← kjartan left07:04
→ FH_thecat joined07:06
← sgn left07:11
← durham left07:11
← tiin57 left07:12
← Xenguy left07:13
→ jaziz joined07:13
→ tiin57 joined07:15
→ kjartan joined07:17
→ Anticom joined07:21
← Hello71 left07:22
← yonder left07:22
→ Hello71 joined07:22
→ chele joined07:24
→ sgn joined07:25
→ leeN joined07:25
← Goplat left07:28
← leeN left07:29
→ zillolo joined07:37
← tiin57 left07:42
→ oxymoron93 joined07:43
→ tiin57 joined07:45
→ wrobinson joined07:50
← wrobinso1 left07:52
← Mattiaslndstrm left07:56
→ PJBoy joined07:57
→ dermoth_ joined08:03
← dermoth left08:03
dermoth_ → dermoth08:03
← cd left08:04
← zumba_addict left08:08
← cyphase left08:08
← tiin57 left08:12
← mir100 left08:13
→ tiin57 joined08:15
← benfelin left08:19
← dimi1947 left08:20
← duderonomy left08:27
→ Tawonga joined08:29
→ unluckyshrubbery joined08:30
→ cyphase joined08:31
← hussam left08:34
→ duderonomy joined08:36
← lankanmon left08:38
← tiin57 left08:42
→ lankanmon joined08:42
→ tiin57 joined08:45
← Furai left08:46
→ Noti joined08:47
→ LordRyan joined08:58
→ mozgalom joined08:58
→ Furai joined08:59
← cerna left09:00
→ calcul0n joined09:07
→ OnkelBlade joined09:09
→ neverwas joined09:09
← tiin57 left09:11
← cdown left09:13
→ Lucas_Gray joined09:14
→ tiin57 joined09:15
→ cdown joined09:15
→ cdown_ joined09:17
← Jookia left09:18
← tubaman left09:18
→ Jookia joined09:19
← cdown left09:20
cdown_ → cdown09:20
→ mobidrop joined09:25
→ dionysus69 joined09:27
← DiffieHellman left09:31
→ cosimone joined09:32
← opal left09:35
← Betal left09:36
→ mirrorbird joined09:37
→ opal joined09:37
→ feriman joined09:39
← tiin57 left09:42
← EvanCarroll left09:44
← dhollinger left09:44
← Guest20 left09:44
→ tiin57 joined09:45
→ EvanCarroll joined09:45
→ dhollinger joined09:45
→ Guest20 joined09:48
→ benfelin joined09:52
← maxzor left09:57
→ maxzor joined09:58
← howdoi left10:00
← flowinho left10:01
→ flowinho joined10:01
← sunri5e left10:04
→ sunri5e joined10:05
→ Mattiaslndstrm joined10:05
→ flex14 joined10:08
→ calcul0n_ joined10:08
← sunri5e left10:10
← nickmass left10:10
→ sunri5e joined10:11
← calcul0n left10:11
← tiin57 left10:11
→ nickmass joined10:12
→ tiin57 joined10:15
→ Tobbi_1 joined10:16
selckin any way to make git stash show, list new files?10:19
← Tobbi_ left10:19
Tobbi_1 → Tobbi_10:19
→ jast joined10:19
← BtbN left10:19
→ BtbN joined10:20
osse selckin: what do you mean, exactly? make them stand out from the rest?10:21
you can use --diff-filter=A to filter the rest out10:21
selckin doesn't show them either, maybe i changed some default, talking about, echo foo > bar; git stash -u; git stash show # always empty no mather the options i can find to pass10:23
← acetakwas left10:25
osse huh10:25
← POQDavid left10:25
osse weird implementation10:25
← pikapika left10:25
osse the untracked files are stored in a separate object10:25
→ POQDavid joined10:25
→ pikapika joined10:25
osse selckin: try this: git log --graph HEAD..stash10:26
→ acetakwas joined10:26
← Noti left10:26
→ benyoo joined10:27
osse git ls-tree stash^310:27
selckin aliasing the log graph then, thanks!10:28
osse better: git log --graph HEAD~..stash10:28
one commit was missing10:28
ikke It makes sense untracked objects are stored seperately10:28
← Tobbi_ left10:28
ikke You don't want them end up in the index10:29
→ Gurkenglas joined10:29
→ vngx joined10:29
osse agreed, but it would also make sense for git stash show to show them10:29
→ ehostunreach joined10:30
osse git stash show should show what was stashed, regardless of impl details, imo10:30
← la_croix left10:30
→ Tobbi_ joined10:31
→ p-core joined10:31
→ la_croix joined10:33
← tiin57 left10:42
→ tgunr joined10:42
→ minhaj-sixbyte joined10:43
← minhaj-sixbyte left10:43
← flex14 left10:43
→ Guest20_ joined10:44
→ minhaj-sixbyte joined10:44
← minhaj-sixbyte left10:44
→ tiin57 joined10:45
→ minhaj-sixbyte joined10:45
← Guest20 left10:45
← minhaj-sixbyte left10:45
→ cdown_ joined10:46
selckin agree!10:47
← impermanence left10:48
osse selckin: git diff --diff-filter=A --name-only stash~ stash^310:48
BAM10:48
← cdown left10:48
cdown_ → cdown10:48
selckin what does stash^3 mean here10:50
oh the other object10:50
← oatmealraisin left10:51
osse third parent of the stash commit10:52
it only exists if there are untracked files10:52
← kjartan left10:54
← chele left11:04
→ chele joined11:04
→ kjartan joined11:05
→ mat001_ joined11:08
→ Noti joined11:08
← mat001 left11:09
← tiin57 left11:12
← mirrorbird left11:14
→ tiin57 joined11:15
← Muimi left11:17
← kerframil left11:18
← mgedmin left11:22
→ mgedmin joined11:25
→ cdown_ joined11:26
← p-core left11:26
→ p-core joined11:27
← splud left11:27
← Guest20_ left11:28
← cdown left11:29
cdown_ → cdown11:29
→ SteffanW joined11:32
→ splud joined11:33
← Noti left11:33
→ cuppajoeman joined11:38
← magic_ninja left11:41
← tiin57 left11:42
→ magic_ninja joined11:43
← benyoo left11:45
→ benyoo joined11:45
← p-core left11:45
→ tiin57 joined11:46
→ oatmealraisin joined11:50
← kjartan left11:51
← benyoo left11:51
← mijndert left11:52
← la_croix left11:53
→ mijndert joined11:53
← theoceaniscool left11:54
→ la_croix joined11:54
→ kjartan joined11:55
← flowinho left11:56
→ flowinho joined11:57
← ehostunreach left11:57
← kjartan left12:00
→ kjartan joined12:07
← tiin57 left12:11
→ tiin57 joined12:15
← tarkus left12:17
← igemnace left12:18
→ igemnace joined12:20
← Kaivo left12:26
← dqx_ left12:27
← cheater left12:34
← milkt left12:35
→ milkt joined12:35
→ flex14 joined12:36
← la_croix left12:36
← jaziz left12:37
→ la_croix joined12:39
← nikivi left12:40
← acetakwas left12:40
← POQDavid left12:40
← pikapika left12:41
→ POQDavid joined12:41
→ acetakwas joined12:41
→ pikapika joined12:41
→ nikivi joined12:41
← tiin57 left12:41
← acetakwas left12:42
← pikapika left12:43
→ acetakwas joined12:43
→ pikapika joined12:45
→ tiin57 joined12:45
→ Maxattax joined12:47
→ impermanence joined12:48
← POQDavid left12:49
← pikapika left12:49
→ POQDavid joined12:49
→ pikapika joined12:50
← nikivi left12:51
→ nikivi joined12:51
← kjartan left12:52
← nikivi left12:52
← acetakwas left12:53
→ nikivi joined12:53
→ acetakwas joined12:53
→ kjartan joined12:53
← pikapika left12:55
→ pikapika joined12:56
→ Wetmelon joined12:58
← DarthRadar left13:06
→ ttl joined13:07
→ mowcat joined13:09
→ Xenguy joined13:10
← Xenguy left13:10
← tiin57 left13:12
→ tiin57 joined13:15
→ nvmd joined13:16
→ CodeSlingerPaul joined13:18
→ Muimi joined13:25
← ghost43 left13:28
→ donofrio joined13:28
← opal left13:31
→ ghost43 joined13:31
→ opal joined13:32
→ sudoforge joined13:36
← dhollinger left13:37
→ dhollinger joined13:41
← tgamblin left13:41
← tiin57 left13:42
← Lucas_Gray left13:43
→ theoceaniscool joined13:44
→ LionsMane joined13:45
→ tiin57 joined13:45
→ thijso joined13:48
← OnkelBlade left13:49
→ OnkelBlade joined13:50
→ mns joined13:51
→ p-core joined13:52
→ macrover joined13:54
← epoll left13:54
→ epoll joined13:55
→ thiago joined13:55
← impermanence left13:57
→ haymawk joined13:57
→ rafasc joined14:06
← tiin57 left14:12
← kjartan left14:14
→ tiin57 joined14:15
→ kjartan joined14:18
← macrover left14:22
← p-core left14:22
→ Aki-lucky joined14:23
← mns left14:23
→ hussam joined14:24
→ gast0n joined14:26
→ rak joined14:28
rak Is there an environment variable or equivalent that I can rely on in git hooks that points to GIT_DIR? AFAICT, GIT_DIR isn't set when hooks run14:28
I just want to know what the root/base directory for the git repo is.14:29
← OnkelBlade left14:29
→ irc_user joined14:30
→ OnkelBlade joined14:30
osse rak: if the repo is bare then GIT_DIR is simply the current directory14:31
rafasc rak: git rev-parse has a couple of --options, like14:31
→ Lucas_Gray joined14:31
osse otherwise, yeah that thing14:31
→ impermanence joined14:31
rafasc git rev-parse --git-dir, for example.14:31
→ relipse joined14:34
→ AnAverageHuman joined14:35
→ tgamblin joined14:38
rak rev-parse does the trick. Thanks all!14:38
→ mns joined14:38
← tiin57 left14:41
← benfelin left14:42
← mowcat left14:43
→ greggerz joined14:43
← konrados left14:43
→ tiin57 joined14:45
→ iffraff joined14:47
iffraff Hi, when I do a git log, I see all the commits sorted by date. However for commits that were merged in, the date seems to be ... well I think it's the date the commit got merged into the current branch or into master. Not the date that it was created. So I kind of loose the real time line14:49
a) is this true :) b) if so is there a way to see the created date stamp?14:50
rafasc git log --author-date-order14:50
← sweatsuit left14:50
← AnAverageHuman left14:51
rafasc it's also a good idea to use something like !lol14:51
gitinfo A nifty view of branches, tags, and other refs: git log --oneline --graph --decorate --all14:51
rafasc So you can see the topological relation between commits.14:51
→ Ironman3000 joined14:52
Ironman3000 Hi, does anyone here use BitBucket and knows how to create a pull request on the linux CLI?14:52
iffraff sweeet! thanks folks!14:54
→ arcatech joined14:56
rafasc Ironman3000: there's probably something out there, but it's only possible if the 3rd party provides a api for it.14:57
https://community.atlassian.com/t5/Bitbucket-questions/Re-How-do-I-create-a-pull-request-from-the-command-line/qaq-p/585738/comment-id/16408#M1640814:57
this seems the closest I could find googling "bitbucket merge request cli"14:57
Ironman3000 thanks14:58
rafasc https://stackoverflow.com/questions/8721730/bitbucket-send-a-pull-request-via-command-line/15046966#1504696614:58
← iffraff left14:59
← flex14 left14:59
→ sweatsuit joined15:02
← tgamblin left15:03
← SteffanW left15:11
← tiin57 left15:12
← OnkelBlade left15:12
osse !hub15:13
gitinfo Hub is a wrapper around git which provides specific github features to the git cli. See https://github.com/github/hub for details.15:13
osse hmm15:13
!spindle15:13
gitinfo [!git-spindle] https://github.com/seveas/git-spindle is a handy tool to use the features of GitHub (and GH Enterprise), GitLab, BitBucket and possibly more from the command line.15:13
osse Ironman3000: see above15:14
→ tiin57 joined15:15
shailangsa anybody know why "error.GitError: None: remote omnirom does not have refs/heads/android-8.1" appears when trying to sync the repo using the 8.1 branch for the minimal twrp manifest using "repo init --depth=1 -u git://github.com/minimal-manifest-twrp/platform_manifest_twrp_omni.git -b twrp-8.1" , https://pastebin.com/qg1a4wr6 ?15:16
← oxymoron93 left15:17
osse Where is Seveas anyway15:17
← a-l-e left15:20
ikke Last message was end of Augus15:20
← mightyjens left15:21
← zillolo left15:21
← chele left15:21
→ AnAverageHuman joined15:21
← Lucas_Gray left15:23
← mobidrop left15:23
→ Lucas_Gray joined15:26
→ mir100 joined15:28
← mns left15:31
→ hiro joined15:31
hiro ok hello everybody. huge channel i see :D15:32
this problem keeps on coming up for me: https://stackoverflow.com/questions/6155359/git-diff-ignore-renaming-of-variables15:32
and i waste a lot of time doing either 1) sed -i 's/newname/oldname/' * */* */*/*; git checkout -b oldnames; commit -a; git diff mainbranch oldnames15:33
or 2) the answer from stackoverflow (git diff | egrep -v 'oldName|newName' | egrep '^[+-]')15:33
→ tgamblin joined15:34
hiro 2) is a bit limited bec. it drops also commits where on one line the variable name got changed in *addition* to another change that i actually would like to see15:34
← greggerz left15:35
hiro i hope after nearly 10 years of everybody finding this stackoverflow on google, there might be more robust solutions now?15:35
in my case sed sometimes works better, but only if the variable name is very unique15:36
otherwise it would be better to use more cues from the diff instead to prefilter everything, but that's a lot of steps and it happens way too often for me15:37
osse hiro: Not what you're looking for, but I suggest you stage the changes. That way git diff won't display them15:37
hiro the motivation here is to try and quickly understand a commit log that has lots of commiters that have no idea about basic git hygiene :)15:38
to quickly see all the relevant changes that actually functionally effect anything at all15:39
there is git diff -w which helps a lot for commits where people mix in whitespace changes15:39
osse git diff -S/-G might help. Would be nice with a way to negate them. That would be perfect15:39
hiro but basically i'd need the same for loads of other bad habits that people tend to have15:39
osse: yeah esp. if i can do a "strict" match where it only works if there is no other change but the matched regex on the line15:40
← tiin57 left15:42
→ tiin57 joined15:45
← AnAverageHuman left15:45
→ bamdad joined15:45
hiro osse: to negate them i think it might be possible to make a branch from before the diversion, apply this patch, rebase upon it and then revert :D15:48
this patch == git log -S oldname -S newname -w -p > patch15:48
→ zebrag joined15:50
→ Kaivo joined15:50
→ Muimi_ joined15:51
→ thomasross joined15:52
← Muimi left15:53
Muimi_ → Muimi15:53
← tryte left15:55
→ tryte joined15:55
← cebor left15:55
→ Aquazi joined15:56
→ cd joined15:56
→ cebor joined15:56
← cebor left15:58
→ cebor joined15:58
→ init_ joined16:00
← tilpner left16:03
→ tilpner joined16:04
← tilpner left16:05
→ tilpner joined16:06
← Lunatrius left16:07
← Kaisyu7 left16:07
→ Lunatrius joined16:07
→ Kaisyu7 joined16:08
← roflin left16:08
← warthog9 left16:08
→ warthog9 joined16:08
→ roflin joined16:10
← tiin57 left16:11
→ laertus joined16:15
laertus once i do a "git fetch", how do i diff the changes?16:15
→ mowcat joined16:15
→ tiin57 joined16:15
laertus "git diff" by itself doesn't show anything, but "git fetch" did show that some changes were there16:16
ikke laertus: diff against the remote tracking branch16:16
for example: git diff HEAD @{u}16:16
laertus what does the @{u} stand for?16:17
ikke short for HEAD@{upstream}, the remote tracking branch for the currently checked out branch16:18
laertus hmmm... i'm tracking origin/master, and i had first tried "git diff origin/master" but seems to give me different results than "git diff HEAD @{u}"16:19
← mikkel left16:19
← zebrag left16:19
ikke The command you ran would be equal to git diff origin/master HEAD, the order is switched16:20
laertus ah16:20
ikke (it should show the inverse)16:20
laertus ok, so would "git diff HEAD origin/master" be the same as "git diff HEAD @{u}" ?16:20
ikke If you are on master, then yes16:20
laertus i see16:21
thank you16:21
ikke you're welcome16:21
laertus actually, i have another question...16:22
ikke You have 2 more questions before your trial ends :P16:22
laertus :)16:22
some time back i did a "git fetch" on another repo, and after reviewing the changes did a "git apply" (from within magit), and it seems i got a merge conflict16:22
in #emacs i was advised not to use "git apply" and instead do a fast-forward merge16:23
ikke apply in emacs is something different from git apply16:23
laertus is that good advice, and if so why would a fast-forward merge work when an apply failed?16:23
ikke You would use apply only in certain circumstances16:24
A fast-forward merge is certainly the best thing to do when grabbing upstream changes16:25
→ benfelin joined16:26
laertus ok, thank you16:27
and how would i actually do a fast-forward merge directly from the command line?16:27
ikke git merge --ff-only @{u}16:27
← rfuentess left16:27
laertus great, thanks again16:27
i think i've used up all my questions :)16:28
ikke git rebase @{u} would also do a fast-forward when possible (and rebase otherwise)16:28
→ mikkel joined16:32
→ a-l-e joined16:35
← PacKetSlayer left16:35
← mikkel left16:36
→ PacKetSlayer joined16:37
→ shtrb joined16:37
shtrb Is there a way I could extract a GPG public key from commits (or from github) ?16:38
ikke Not from commits, it would only contain the signatures16:39
← tgunr left16:39
shtrb My only option is to request each person to send their public keys correct ?16:39
ikke or use some kind of gpg key server16:40
shtrb yes (ofc). thanks16:40
ikke github might publish users gpg public keys as well16:41
shtrb Oh , thanks , how ?16:41
if you have a clue .16:41
← tiin57 left16:41
ikke https://github.com/<username>.keys perhaps, but not sure if that's only ssh keys, or also gpg keys16:42
https://developer.github.com/v3/users/gpg_keys/#list-gpg-keys-for-a-user16:42
also an API endpoint apparently16:42
shtrb Thank you !16:42
ikke https://github.com/<user>.gpg_keys16:43
→ orbyt_v7 joined16:43
shtrb WOW ! thank you16:44
how did you find that urls ?16:44
ikke deduced it from the api documetation16:44
← POQDavid left16:45
→ tiin57 joined16:45
→ POQDavid joined16:45
shtrb facepalm16:46
← pikapika left16:52
→ pikapika joined16:52
← Lucas_Gray left17:00
→ archwizard56 joined17:01
← Ironman3000 left17:01
→ howdoi joined17:03
← archwizard left17:04
→ tgunr joined17:05
← mozgalom left17:05
← tiin57 left17:11
→ OnkelBlade joined17:13
→ Human_G33k joined17:14
→ tiin57 joined17:15
← HumanGeek left17:17
→ Guest20 joined17:19
← mir100 left17:20
← thiago left17:25
← mowcat left17:26
→ Muimi_ joined17:27
→ Betal joined17:28
← Muimi left17:30
Muimi_ → Muimi17:30
→ wootehfoot joined17:31
← lightstalker left17:32
→ lightstalker joined17:34
→ moo joined17:34
← wootehfoot left17:36
→ AnAverageHuman joined17:37
→ thiago joined17:38
← crose left17:38
← Muimi left17:40
→ davispuh joined17:41
← tiin57 left17:41
← Guest20 left17:43
→ tiin57 joined17:45
← arcatech left17:46
← moo left17:47
→ wootehfoot joined17:47
→ saveNexit joined17:51
← Anticom left17:52
→ strobelight joined17:54
→ mirrorbird joined17:54
→ arcatech joined17:55
→ mozgalom joined17:56
→ dqx_ joined17:58
← duderonomy left18:00
← saveNexit left18:00
← maxzor left18:03
← wad left18:04
← tgunr left18:07
→ stats4647 joined18:08
→ tgunr joined18:09
← AnAverageHuman left18:10
← tiin57 left18:12
→ tiin57 joined18:15
← tejr left18:17
→ tejr joined18:17
← Sauvin left18:24
→ UlrichH joined18:24
→ duderonomy joined18:25
← tgunr left18:27
→ tgunr joined18:30
hiro hahahahaha after many many hours i randomly realize a much more obvious and useulf solution to undo this. so simple. just do the git log -S oldname -S newname -w -p| sed 's/^-/+/; s/^+/-/' > patch !!!! then look at diff. done.18:33
noooo18:34
i'm wrong18:34
ikke hiro: diff -R18:34
hiro recursive diff?18:35
← flowinho left18:35
ikke reverse18:35
hiro yeah, but i made an error :(18:35
i'd need to add one more step18:35
though i can't accept that i'm the only one trying to come up with this in a slightly more stable way...18:35
← laertus left18:36
→ flowinho joined18:36
← orbyt_v7 left18:39
← tiin57 left18:41
← flowinho left18:42
→ flowinho joined18:43
← OnkelBlade left18:44
→ zebrag joined18:44
→ ericl joined18:45
ericl Does anyone know how MingW relates to Git?18:45
It shows up on my terminal18:45
→ tiin57 joined18:45
← mozgalom left18:45
hiro so for a trivial patch file like18:46
-test(bla)18:46
+test(blub)18:46
cat diff | sed -n '/^+/{s/^+/+/; p}; /^-/{s/^\(-.*\)bla\(.*\)$/\1blub\2/; p}'18:46
-test(blub)18:46
+test(blub)18:46
this will ignore all diffs that are just changing bla to blub18:46
→ mozgalom joined18:46
hiro but it can only happen once18:46
you can basically apply that patch and look at the diff afterwards18:46
it's a lot of back and forth18:46
but it can be automated fully if you just run parts of it multiple time until it all checks out18:47
but i *bet* somebody found an even better way18:47
this is very naive approach18:47
→ gehn joined18:47
ikke ericl: it's a third party system that git for windows uses18:48
gehn how might I take a branch and "turn it into" a set of uncommitted changes? I googled for this, but all I get is results for the opposite of what I want18:48
→ yonder joined18:48
ericl Oh I see what u mean18:48
thanks sm for the reply @ikke18:48
gehn like basically the equivalent of `git stash apply` but from a branch not a stash18:49
ericl This is an important distinction — instead of getting just a working copy, Git receives a full copy of nearly all data that the server has.18:50
Can someone explain this sentence?18:50
Is the difference for git that it copies the commit history too?18:50
or is there anything else going on behind the scenes18:50
thanks18:50
← tgunr left18:51
→ tgunr joined18:52
ikke Yes, you get the entire history of the repository18:53
→ ash_worksi joined18:53
ash_worksi what is the command to pull a branch to just have it to checkout out but not to pull remote files to your actual file system (in otherwords don't overwrite the current files)18:54
fetch!18:54
ikke quick18:57
quack18:57
← ttl left18:57
Lewix I git fetch and I have the following:18:59
error: refs/heads/584-twilio-webhook-for-stop-lastsent does not point to a valid object!18:59
→ orbyt_v7 joined19:00
← arcatech left19:01
← haymawk left19:04
→ Phylock joined19:04
ikke did you have a computer crash or something like that?19:05
← gschanuel left19:05
Lewix ikke: nope, my coworker pushed, then I git fetch and it occured19:05
→ arcatech joined19:06
→ maxzor joined19:08
← gehn left19:10
→ gehn joined19:10
← tiin57 left19:11
← duderonomy left19:12
→ tiin57 joined19:15
← orbyt_v7 left19:15
← impermanence left19:17
→ mark4 joined19:19
mark4 im having extreme git issues here - im on branch A i want to be on branch B i say git checkout B and now im in detached head state19:20
what the hell is going on?19:20
i dont understand i need to be on branch B why does it not simply switch to branch B when i tell it to19:20
← wootehfoot left19:23
ikke mark4: seems like B is not a branch19:24
mark4 it is but i was specifying origin/feature/B not just feature/b19:25
sorry figured it out :)19:25
← vicfred left19:27
→ KnoP joined19:27
ikke :)19:27
Lewix: Just fetching is very unlikely to cause these issues19:28
→ gschanuel joined19:28
← cosimone left19:29
→ duderonomy joined19:30
→ cosimone joined19:30
→ Anticom joined19:30
← UlrichH left19:30
← a-l-e left19:31
← tgunr left19:31
→ leeN joined19:36
→ tgunr joined19:37
← dqx_ left19:37
← mark4 left19:38
ericl how can i see permissions on a git repo19:39
like which branches can i push too19:39
ikke ericl: depends on the remote system19:40
that's not something git itself offers (there is no built-in acl)19:40
ericl i'm trying to modify a github repo19:40
i got in on my computer using git clone19:40
and there's like a ton of branches19:40
i'm not sure if i can just edit the master branch directly on my local computer19:40
and tehn push it to remote19:40
or if i have to do pull request19:41
and i'm not even sure why i should fork19:41
this is so confusing19:41
→ dqx_ joined19:41
ikke Are you a contributor (and marked as such) in github for that project?19:41
← tiin57 left19:41
ericl yes19:42
i got added19:42
← Phylock left19:42
← tgamblin left19:44
← theoceaniscool left19:44
→ tiin57 joined19:45
← random_yanek left19:46
→ random_yanek joined19:47
← ericl left19:47
← st-gourichon-fid left19:51
← duderonomy left19:53
→ duderonomy joined19:58
← dionysus69 left20:03
→ matelot joined20:06
← shtrb left20:08
→ vicfred joined20:09
← Maxattax left20:10
gehn `git checkout master; git merge --no-commit my-branch` is close to what i want, except that I don't want to be considered in the middle of a merge, and also I want changes to be listed as not staged20:11
← tiin57 left20:11
← mozgalom left20:12
→ st-gourichon-fid joined20:13
→ mozgalom joined20:14
← ash_worksi left20:14
→ tiin57 joined20:15
← dqx_ left20:16
← random_yanek left20:16
← duderonomy left20:17
← strobelight left20:18
→ strobelight joined20:18
→ random_yanek joined20:20
→ cdown_ joined20:20
→ tgamblin joined20:21
← cdown left20:22
cdown_ → cdown20:22
→ dqx_ joined20:23
rafasc gehn: use git merge --squash my-branch instead.20:24
→ sqwishy joined20:26
rafasc after that, unstage the changes with: git reset;20:27
→ wootehfoot joined20:28
sqwishy I have `[credentials] helper` set in a git --local config & global. And running `git config --get-all credentials.helper` shows both. I think it's using the global helper first. Is there a way to use the local one first/only?20:28
← st-gourichon-fid left20:29
mozgalom → bomb-on20:29
→ BackupBackupNick joined20:29
gehn rafasc, looks like --squash and --no-ff are mututally exclusive? would it ever be possible e.g. for a single commit branch to accidentally fast forward?20:29
rafasc can you run with --show-origin or --show-scope?20:29
gehn or does --squash also always prevent fast forward?20:29
→ st-gourichon-fid joined20:29
rafasc gehn: what you are asking makes no sense. Why do you care about --no-ff; if you're not even making a merge?20:30
When you're just using git merge to the the files?20:30
If that's not the case, I completely missed your intention.20:31
gehn https://git-scm.com/docs/git-merge because "Note that fast-forward updates do not create a merge commit and therefore there is no way to stop those merges with --no-commit. Thus, if you want to ensure your branch is not changed or updated by the merge command, use --no-ff with --no-commit."20:31
← arcatech left20:32
rafasc --squash already implies --no-commit;20:32
← st-gourichon-fid left20:32
gehn ok so then that begets "Thus, if you want to ensure your branch is not changed or updated by the merge command, use --no-ff with --no-commit."20:32
← Aki-lucky left20:32
gehn so does --squash also imply --no-ff20:33
rafasc --no-ff makes no sense for --squash.20:33
what are you trying to do>20:33
?20:33
→ st-gourichon-fid joined20:33
gehn git checkout -b my-feature; echo "some change" >> README.txt; git add README.txt; git commit -m "my msg";20:34
now I want to change back to master, and get the changes from my-feature as if they were unstaged and I hadn't made a branch nor committed them20:34
like git stash apply20:34
but I'd like to "apply" a branch20:35
rafasc git checkout master; git restore --source=my-branch --worktree;20:37
correction: git checkout master; git restore --source=my-branch --worktree :/ ;20:37
with older versions of git, you can: git chekout master; git checkout my-branch -- :/ ; git reset20:37
or yet another one: git checkout master; git merge --squash my-branch; git reset;20:38
gehn ok20:40
→ duderonomy joined20:40
→ SunWuKung joined20:41
rafasc there are multiple ways to do it, you just have to understand how the different commands affect the worktree, the staging area and HEAD.20:41
← init_ left20:41
gehn is the ":/ ;" a required part of the command or is :/ an emoticon?20:41
← tiin57 left20:41
rafasc it's not an emoticon, :/ the root of the repository.20:42
gehn ah ok so that would be the <pathspec> then?20:42
rafasc correct. It's a pathspec for the root of the repository.20:43
man git glossary20:43
gitinfo the git manpage is available at https://gitirc.eu/git.html20:43
rafasc man gitglossary20:43
gitinfo the gitglossary manpage is available at https://gitirc.eu/gitglossary.html20:43
→ arcatech joined20:44
← ghost43 left20:44
→ ghost43 joined20:44
→ tiin57 joined20:45
gehn which chapter+section in https://git-scm.com/book/en/v2 covers worktrees?20:45
← st-gourichon-fid left20:46
rafasc git worktree is relatively recent, I'm not sure if the book covers it.20:46
gehn hmm, so what is a worktree?20:47
← Jackneill left20:47
rafasc hmm wait.20:47
worktree can either refer to: man git worktree20:47
gitinfo the git-worktree manpage is available at https://gitirc.eu/git-worktree.html20:47
rafasc or your directory.20:47
→ st-gourichon-fid joined20:48
rafasc the place where git materializes the files that exist on the repo. The directory that usually contains the .git/ directory.20:48
gehn I'm interested in https://gitirc.eu/git-worktree.html20:48
it's informative, but not very descriptive20:48
e.g. it never defines what a worktree is or what the use case is20:48
rafasc you cannot explain every single term every time it occurs.20:50
working tree20:50
→ Habbie joined20:50
rafasc The tree of actual checked out files. The working tree normally contains the contents of the HEAD commit’s tree, plus any local changes that you have made but not yet committed.20:50
the glossary I linked earlier contains the definition.20:50
→ ch joined20:50
rafasc and the description contains some use cases.20:51
gehn there's no description in the glossary20:52
ch hi!20:52
gitinfo ch: 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.20:52
gehn unless you mean the DESCRIPTION section at the top of this https://git-scm.com/docs/git-worktree20:52
which I will go ahead and disagree that it has any verbiage around describing use cases20:53
rafasc gehn: there's a description on the glossary of what a worktree means, there's a description of the usescases of git worktree on the manpage, along with a example on the botton.20:53
gehn by manpage do you mean this? https://git-scm.com/docs/git-worktree20:53
rafasc yes20:53
gehn no use cases in that description20:53
rafasc In its simplest form, git worktree add <path> automatically creates a new branch whose name is the final component of <path>, which is convenient if you plan to work on a new topic. For instance, git worktree add ../hotfix creates new branch hotfix and checks it out at path ../hotfix. To instead work on an existing branch in a new working tree, use git worktree add <path> <branch>. On the other hand,20:54
if you just plan to make some experimental changes or do testing without disturbing existing development, it is often convenient to create a throwaway working tree not associated with any branch. For instance, git worktree add -d <path> creates a new working tree with a detached HEAD at the same commit as the current branch.20:54
ch i'm often in the situation that i want to swap the order of the two last commits (so HEAD~1). is there a nice/fast way of doing this, without rebase -i and an editor?20:54
← Cabanossi left20:54
← opal left20:54
gehn well I think we are somehow not looking at the same document20:54
because that text does not occur in the "DESCRIPTION" section of https://git-scm.com/docs/git-worktree20:55
← wootehfoot left20:55
gehn actually nowhere on the page even20:55
→ Cabanossi joined20:55
rafasc gehn: https://gitirc.eu/git-worktree.html20:55
gehn why is there a separate set of docs for IRC?20:56
→ Jackneill joined20:56
_raymond → raymond20:57
rafasc The docs on gitirc are generated from source by jast, looks like there's a mismatch for some reason.20:57
gehn: https://git-scm.com/docs/git-worktree#_examples20:58
but there's still an example at the end.20:58
← benfelin left20:59
rafasc ch: why are you trying to avoid rebase -i/editor?20:59
ch: for just two commits, you could so something like: git reset --hard HEAD~2; git cherry-pick <commit> <commit>21:00
ch rafasc: i do this so often, its annoying to wait for vim to start, manipulate the lines and quit again21:00
← BackupBackupNick left21:01
rafasc ch: I have a trick for you, wait a sec.21:02
ch rafasc: i found this on stackoverflow, but ... there has to be a better way -> https://stackoverflow.com/a/33388211/229838621:02
rafasc ch: I was thinking something similar: GIT_SEQUENCE_EDITOR=tac21:06
Habbie tac is nicer than sed :)21:07
← feriman left21:07
← bomb-on left21:09
← arcatech left21:09
rafasc Habbie: tac doesn't work. It needs to save the file, not pipe it.21:10
Habbie right21:10
← tiin57 left21:11
rafasc for a quick swap, It's usually pretty quick: git rebase -i HEAD~2; ddp:wq21:13
→ mozgalom joined21:14
→ tiin57 joined21:15
← sudoforge left21:17
rafasc GIT_SEQUENCE_EDITOR='todo=$(tac "$1"); echo "$todo" >"$1"' git rebase -i HEAD~2;21:22
something like that.21:23
ch heh, alright21:23
rafasc You need to read and write on separate invocations, as reading and writing to the same file can have unintended effects.21:24
ch thanks for taking the time. i'll see about making a script sometime soon :)21:24
← vdamewood left21:25
← oatmealraisin left21:28
← matelot left21:30
→ matelot joined21:34
→ arcatech joined21:36
rafasc GIT_SEQUENCE_EDITOR='f(){ mapfile <"$1"; printf '%s' "${MAPFILE[@]}" | tac >"$1" ;};f'21:37
→ Irish joined21:38
← mirrorbird left21:39
← Kaivo left21:40
← tiin57 left21:41
→ silverwhitefish joined21:45
→ tiin57 joined21:45
→ mns joined21:57
← rafasc left21:59
← KnoP left22:00
→ stormsz joined22:02
gitinfo set mode: +v22:03
← irc_user left22:04
→ s2013 joined22:04
← carlesc left22:04
← rustyshackleford left22:04
← OvermindDL1 left22:04
← dmlloyd left22:04
← afisher left22:04
← SrPx left22:04
← itok left22:04
← hussam left22:04
→ OvermindDL1 joined22:04
→ itok joined22:05
← Anticom left22:05
→ hussam joined22:05
→ davispuhh joined22:05
→ SrPx joined22:05
→ carlesc joined22:05
→ irc_user joined22:05
→ rustyshackleford joined22:06
→ dmlloyd joined22:06
→ afisher joined22:06
← davispuh left22:06
→ Xenguy joined22:06
← ch1astre left22:07
→ ch1astre joined22:09
← tiin57 left22:11
← milkt left22:12
→ tiin57 joined22:15
← mns left22:15
← Irish left22:23
← CodeSlingerPaul left22:33
← leeN left22:34
← arcatech left22:35
← tiin57 left22:41
→ tiin57 joined22:45
← thiago left22:48
← dqx_ left22:48
← Xenguy left22:54
→ Xenguy joined22:58
→ mns joined23:00
→ thiago joined23:06
← s2013 left23:07
→ mat001 joined23:08
← PJBoy left23:08
← mat001_ left23:09
← gehn left23:10
→ gehn joined23:11
← tiin57 left23:11
→ Oatmeal joined23:12
← mns left23:13
→ tiin57 joined23:15
→ hineios joined23:16
← calcul0n_ left23:20
→ jaziz joined23:21
→ opal joined23:25
→ arcatech joined23:28
→ DiffieHellman joined23:29
← gehn left23:29
← mozgalom left23:29
→ mozgalom joined23:31
← kent\n left23:31
← rkta left23:32
→ rkta joined23:32
→ ycon_ joined23:37
→ init_ joined23:41
← tiin57 left23:41
→ tiin57 joined23:45
← bamdad left23:46
→ bamdad joined23:47
→ BUSY joined23:49
← hineios left23:52
→ hineios joined23:55
ycon_ Hi, I'd like to know what I'm doing wrong here with sub modules23:57
I keep having to run "git submodule update --recursive" . Is that because I've used a push force-with-lease?23:57
← Aquazi left23:57
ycon_ (reason I have to run update, is it constantly shows (new commits) in the submodules23:57

Logs Search ←Prev date Next date→ Channels Documentation