IRCloggy #git 2022-10-14

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.

2022-10-14

← lucasta left00:01
→ tomboy64 joined00:03
→ lucasta joined00:03
← rama left00:04
→ rama joined00:05
→ Arokh joined00:05
← lucasta left00:08
→ TonyStone joined00:08
→ lucasta joined00:09
← atrigent left00:12
→ atrigent_ joined00:12
→ hiroot joined00:12
← gas51627 left00:13
← skylize left00:13
→ roadie joined00:14
→ arch_angel joined00:15
→ skylize joined00:16
← arch_angel left00:16
← lucasta left00:16
→ lucasta joined00:16
→ arch_angel joined00:17
← BlessJah left00:18
← roadie left00:19
→ zebrag joined00:19
→ BlessJah joined00:20
← lucasta left00:20
← rama left00:21
→ lucasta joined00:21
→ rama joined00:21
← dfee left00:24
← m0viefreak left00:24
← xx left00:26
→ gsi_ joined00:29
→ derpydoo joined00:29
← gsi left00:32
← epony left00:33
← anarcat left00:36
← mat001 left00:36
← nyah left00:37
← lucasta left00:38
→ lucasta joined00:38
→ epony joined00:48
← rama left00:50
→ rama joined00:50
→ dfee joined00:55
← Lord_of_Life left00:55
→ Lord_of_Life_ joined00:55
← lucasta left00:55
→ lucasta joined00:55
Lord_of_Life_ → Lord_of_Life00:56
→ pyeverything joined00:57
→ forgotmynick joined00:59
← gvg left01:00
→ roadie joined01:00
→ Maxattax joined01:03
← Sasazuka left01:03
← roadie left01:05
→ ajshell1 joined01:23
← Cleverness left01:25
← gsgx left01:29
← umbramalison left01:31
→ umbramalison joined01:32
← umbramalison left01:36
→ umbramalison joined01:37
← rama left01:37
→ rama joined01:38
← arescorpio left01:39
← hans_ left01:42
← rama left01:45
→ nate1 joined01:45
→ rama joined01:45
← rama left01:47
→ elastic_1 joined01:55
elastic_dog → Guest153001:55
← Guest1530 left01:55
elastic_1 → elastic_dog01:55
← pulse left02:00
← rurtty left02:05
← lucasta left02:06
→ lucasta joined02:08
ztrawhcse bad git history -- and I assume fossil as well -- includes several hundred commits in which you add, then edit, then re-edit, then remove, a bunch of debug printfs and CI workflow logging statements02:09
and I wonder, what is fossil's solution to "oops I just accidentally committed a 50mb binary 4 commits back, and I'd rather that not bloat the history and burden the object storage"02:11
it's not like you can make an editorial comment on that. And accidentally committing the wrong file isn't useful history to tell you how a feature was created "false starts and all"02:12
← lucasta left02:14
→ rama joined02:15
← rama left02:15
→ structoha joined02:17
→ thebombzen joined02:18
→ riposte joined02:19
← dfee left02:21
← ajshell1 left02:24
→ sjoshi joined02:27
← sjoshi left02:28
→ roadie joined02:30
→ Guest81 joined02:32
→ finn_elija joined02:33
← FinnElija left02:33
finn_elija → FinnElija02:33
→ sjoshi joined02:33
← sjoshi left02:34
→ YuGiOhJCJ joined02:36
→ gildasio joined02:36
→ ajshell1 joined02:38
← structoha left02:39
→ dfee joined02:39
← ajshell1 left02:40
→ gvg joined02:42
→ sjoshi joined02:42
← roadie left02:42
→ ajshell1 joined02:42
← sjoshi left02:43
← terrorjack left02:45
→ junebug joined02:46
→ terrorjack joined02:46
→ winter joined02:47
← dfee left02:48
← nate1 left02:48
→ mat001 joined02:52
→ roadie joined02:55
← Murr left02:58
→ Murr joined02:58
← Guest81 left03:00
← cbreak left03:00
← roadie left03:02
← mat001 left03:05
→ mat001 joined03:06
← derpydoo left03:08
← gast0n left03:16
→ sjoshi joined03:17
← forgotmynick left03:19
sjoshi → c0dehu1k03:20
← Betal left03:23
← hiroot left03:26
→ hiroot joined03:27
→ ProperNoun joined03:28
→ bket joined03:34
← sudoforge left03:42
← moldorcoder7 left03:45
→ Dotz0cat joined03:50
← atrigent_ left03:55
→ atrigent joined03:55
→ roadie joined04:06
→ xaxas joined04:06
← pyeverything left04:06
→ pyeveryt_ joined04:07
← bloody left04:08
→ KimK joined04:11
→ dfee joined04:11
← pyeveryt_ left04:12
intelfx Is it possible to instruct `git-send-email` to back up the final edited messages somewhere, and pick them up again when re-invoked after a network failure?04:14
E. g. I type `git send-email ... --annotate --cover-letter`, write a nice and detailed cover letter, maybe fix up some other patches, hit send and then some sort of SMTP error happens and all my edits are basically gone to /dev/null. How do I avoid that?04:14
Or am I doing it wrong?04:14
bookworm you can use format-patch rather than send-email04:15
then send them manually with something that can retry04:15
there might be a better way04:16
intelfx I hope there is; I would really prefer to avoid setting up a separate command-line mailer. Surely there is a way everyone is doing it?04:17
→ Vonter joined04:19
← roadie left04:20
bookworm I mean, you can just point send-email to your preferred smtp client that has a queue04:26
local postfix say04:26
or whatever the addon of msmtp is that can queie04:26
ztrawhcse intelfx: you can run format-patch, edit your files, and then git send-email *.patch04:27
intelfx ztrawhcse: oh, that sounds workable I guess04:28
thanks04:28
ztrawhcse > Takes the patches given on the command line and emails them out. Patches can be specified as files, directories (which will send all files in the directory), or directly as a revision list. In the last case, any format accepted by git-format-patch[1] can be passed to git send-email04:28
intelfx Yeah, I completely overlooked that.04:29
ztrawhcse see, technically speaking the main use of send-email is actually running it on the files output by format-patch ;)04:29
intelfx Now that I actually think about it, you're right of course04:29
I guess I was taught to use `send-email <format-patch options>` and never stopped to see that there's another way04:30
ztrawhcse it was an eye-opening experience for me too, back in the day.04:30
that being said.04:31
it's sort of dumb that send-email doesn't save the temp directory if sending failed.04:31
the authors probably assumed that you could just re-export the patches, but this is awkward if you've added timely commentary and cover letters to them...04:32
intelfx now that I think about it a bit more, it should be a ridiculously easy thing to support, just save the temp dir and print a helpful message along the lines of "There was a problem mailing your patches out; fix it and re-run git send-email ${WORKDIR}"04:33
right?04:33
→ sudoforge joined04:36
← dfee left04:37
← fling left04:38
intelfx maybe even only do it if --annotate or --cover-letter was passed, or even hash the patches and only do it if the contents differ after invoking the editor (but this is probably too automagic)04:40
→ fling joined04:40
← clime left04:41
→ cdown joined04:41
→ roadie joined04:46
→ pyeveryt_ joined04:47
← zumba_addict left04:53
← arch_angel left04:54
← Coop left04:59
← roadie left05:01
→ madewokherd` joined05:09
→ codaraxis__ joined05:09
← madewokherd left05:12
← codaraxis___ left05:13
← jazz-o-lantern left05:21
← junebug left05:30
← zebrag left05:34
← riposte left05:37
→ riposte joined05:38
→ jiffe joined05:38
→ IRChatter joined05:41
← jiffe left05:46
→ jiffe joined05:48
← jiffe left05:53
← dsrt^ left05:54
→ jiffe joined05:54
→ srinidhi joined05:56
→ lgc joined05:57
← lgc left05:57
← gryffus left05:59
→ roadie joined06:03
← cyberpear left06:05
→ cyberpear joined06:06
→ whatsupdoc joined06:08
→ coot joined06:10
← pyeveryt_ left06:18
← mrmango17 left06:29
→ mrmango17 joined06:29
→ lucasta joined06:30
→ igemnace joined06:36
→ kenanmarasli joined06:41
← momomo left06:42
← fling left06:45
→ nate1 joined06:46
→ rfuentess joined06:46
→ pyeverything joined06:48
→ fling joined06:48
← pyeverything left06:49
→ pyeverything joined06:49
← pyeverything left06:49
→ pyeverything joined06:50
← nate1 left06:51
→ paddymahoney joined06:51
← pyeverything left06:55
← Vonter left06:55
← sudoforge left06:57
← cdown left07:02
← LuxuryMode left07:05
→ Telmud joined07:08
→ sudoforge joined07:08
→ TomyWork joined07:13
← thiago left07:13
← hamburgler left07:13
← lucasta left07:19
→ elastic_1 joined07:19
← elastic_dog left07:19
elastic_1 → elastic_dog07:19
← riposte left07:19
→ lucasta joined07:19
→ sa0 joined07:22
← zan left07:24
→ Guest3 joined07:25
mcon nedbat: Thanks. Works like a charm!07:28
→ MajorBiscuit joined07:30
← fling left07:31
→ pyeveryt_ joined07:32
→ fling joined07:32
→ agarr joined07:32
→ Major_Biscuit joined07:33
← c0dehu1k left07:34
→ Ugrastil277 joined07:35
← lucasta left07:36
→ lucasta joined07:36
← MajorBiscuit left07:37
→ Guest82 joined07:37
→ Vonter joined07:38
→ Guest6246 joined07:40
← Guest6246 left07:40
→ zan joined07:40
← pyeveryt_ left07:41
← roadie left07:43
← YuGiOhJCJ left07:43
← lucasta left07:45
→ feriman joined07:45
→ lucasta joined07:45
→ YuGiOhJCJ joined07:46
← lucasta left07:50
→ lucasta joined07:52
→ VaniaPy joined07:54
→ riposte joined07:54
← vdamewood left07:55
→ vinleod joined07:55
→ roadie joined08:08
→ carl- joined08:09
← lucasta left08:11
← adanwan left08:13
→ adanwan joined08:14
→ pyeveryt_ joined08:15
← riposte left08:15
→ lucasta joined08:15
← roadie left08:16
← skapata left08:16
→ riposte joined08:16
→ __xor joined08:22
← pyeveryt_ left08:22
← lucasta left08:23
← hiroot left08:25
→ lucasta joined08:25
→ DibbTubberOne joined08:25
→ mastarija joined08:27
mastarija Is there a way to check if a git repository is empty? When I clone an empty git repo I get "You appear to have cloned an empty repository."08:28
How can I test for that in my script?08:28
→ dionysus69 joined08:30
osse mastarija: How Git interally does it is to check whether HEAD points to a valid branch08:30
mastarija Is there a command for that check? Something I can test?08:31
osse o something like git rev-parse --verify $(git symbolic-ref HEAD) I guess08:31
mastarija Ok, thanks!08:32
→ gas51627 joined08:32
→ nyah joined08:32
osse Or you can test if git clone prints that message? :P08:33
→ iomari891 joined08:33
mastarija I need to test it after cloning08:35
← DonRichie left08:40
→ roadie joined08:41
→ IRChatter0 joined08:43
← lucasta left08:43
→ lucasta joined08:44
← IRChatter left08:45
IRChatter0 → IRChatter08:45
→ rsx joined08:46
← KimK left08:46
← roadie left08:46
→ murii joined08:47
← dionysus69 left08:48
→ dionysus69 joined08:48
→ Masklin__Gurder joined08:49
→ Anticom joined08:49
← Vonter left08:51
→ clime joined08:51
← Masklin left08:52
Masklin__Gurder → Masklin08:54
← Telmud left08:55
→ DonRichie joined08:58
← Murr left08:58
→ Murr joined08:58
→ roadie joined08:59
→ pyeveryt_ joined08:59
→ Flow joined08:59
← DonRichie left08:59
← lucasta left09:00
→ lucasta joined09:00
→ DonRichie joined09:00
← roadie left09:05
← pyeveryt_ left09:07
→ travisghansen joined09:15
← lucasta left09:15
→ lucasta joined09:17
→ roadie joined09:18
← roadie left09:20
→ roadie` joined09:20
← travisghansen left09:22
← reset left09:22
→ travisghansen joined09:23
→ Vonter joined09:24
← Vonter left09:32
← dob1 left09:33
→ dob1 joined09:33
← sudoforge left09:34
← DibbTubberOne left09:38
← Masklin left09:39
← Guest82 left09:39
→ pyeveryt_ joined09:40
← roadie` left09:41
→ roadie joined09:42
→ lu joined09:46
→ Masklin__Gurder joined09:47
← pyeveryt_ left09:47
Masklin__Gurder → Masklin09:53
← theobjectivedad left09:56
→ theobjectivedad joined09:58
← YuGiOhJCJ left10:00
→ masber joined10:02
← roadie left10:04
→ zeroes joined10:07
← lucasta left10:09
→ lucasta joined10:10
→ Vonter joined10:14
→ skfax joined10:15
→ DibbTubberOne joined10:16
← mastarija left10:19
→ rosco joined10:21
→ pyeveryt_ joined10:22
→ theoceaniscool joined10:22
→ theoceaniscool_ joined10:23
← hoochmonger left10:29
← pyeveryt_ left10:29
→ eroux joined10:29
→ thuna` joined10:30
← __xor left10:30
→ vitali64 joined10:32
← skfax left10:33
← Guest3 left10:34
← rosco left10:39
← DibbTubberOne left10:40
→ pulse joined10:45
← iomari891 left10:45
→ nate1 joined10:47
→ iomari891 joined10:51
← nate1 left10:52
← Anticom left10:55
← Murr left10:57
→ Murr joined10:57
→ Cienisty joined10:59
→ pyeverything joined11:02
← lu left11:02
zeroes I tried to resolve conflicts by executing "git mergetool", but forgot to specify a tool for it. So I exitted vim editor, and tried "git mergetool --tool=vimdiff3", but nothing happended "No files need merging".11:05
→ Anticom joined11:06
zeroes How I can re-enter vim for resolving conflicts?11:06
I could undo changes, and do the same changes again. Looking for more elegent solution.11:07
farzat My first guess would be delete the .orig files11:09
Just a guess though11:09
→ cbreak joined11:10
← pyeverything left11:10
← Anticom left11:10
zeroes farzat: Just removed *.orig files, but still "No files need merging"11:15
farzat Ok idk then11:16
→ memoryleak joined11:17
J_Darnley Abort the merge and redo11:21
→ c0dehu1k joined11:22
→ transhumanist joined11:27
→ mastarija joined11:29
← vitali64 left11:30
→ sd5 joined11:34
→ moldorcoder7 joined11:35
→ pyeverything joined11:39
→ phylaz joined11:43
→ jinsun joined11:44
→ fflam joined11:45
→ Coop joined11:48
← pyeverything left11:49
← nullroute left11:56
→ peterhil joined11:57
→ structoha joined11:57
→ Anticom joined12:00
← TonyStone left12:00
← murii left12:02
← indy_ left12:09
← feriman left12:12
→ nullroute joined12:12
← nullroute left12:12
→ nullroute joined12:12
→ pyeveryt_ joined12:17
→ spacenautx joined12:20
← pyeveryt_ left12:23
→ pyeveryt_ joined12:23
→ rustyshackleford joined12:24
→ cebor joined12:25
← cebor left12:25
→ cebor joined12:26
← lucasta left12:26
→ lucasta joined12:26
← pyeveryt_ left12:29
vinleod → vdamewood12:31
→ xx joined12:32
→ peirik_ joined12:34
← lucasta left12:35
→ lucasta joined12:36
← carl- left12:36
← mastarija left12:37
← lucasta left12:37
→ pyeveryt_ joined12:38
→ Visor joined12:40
← pyeveryt_ left12:43
← peirik_ left12:44
← jetchisel left12:50
← agarr left12:51
→ jetchisel joined12:53
→ derpydoo joined12:56
← Murr left12:57
→ cdown joined12:57
→ Murr joined12:58
→ odoood joined13:02
← spacenautx left13:04
→ ozoned joined13:06
→ mw_ joined13:06
← odoood left13:09
→ pyeverything joined13:11
← pyeverything left13:22
← Xenguy left13:23
← ajshell1 left13:37
→ ajshell1 joined13:39
→ atrigent_ joined13:40
← atrigent left13:41
→ pyeverything joined13:51
→ hoochmonger joined13:54
← ajshell1 left13:57
→ hololeap joined13:58
→ ajshell1 joined13:59
→ trace1918 joined14:00
← trace1918 left14:01
← pyeverything left14:02
← thuna` left14:03
→ trace1918 joined14:05
→ gh34 joined14:08
← rsx left14:12
← Vonter left14:14
← Deknos left14:19
→ rsx joined14:25
← derpydoo left14:27
→ junebug joined14:28
← GNUmoon left14:33
→ GNUmoon joined14:34
← masber left14:38
→ impermanence joined14:39
→ diverdude joined14:40
diverdude hello, i get the following when i do a pull: https://dpaste.org/BcYWs Which branches is referred to when it says: "You have divergent branches" ?14:42
→ IRChatter8 joined14:42
→ Visne joined14:42
← impermanence left14:44
→ arch_angel joined14:45
← IRChatter left14:46
IRChatter8 → IRChatter14:46
→ nate1 joined14:49
→ pyeveryt_ joined14:50
← nate1 left14:54
← Murr left14:57
→ thiago joined14:58
→ Murr joined14:58
→ jazz-o-lantern joined15:00
→ rama joined15:01
← Voxel left15:05
→ Voxel joined15:05
→ Vonter joined15:06
→ rgrinberg joined15:07
→ delay joined15:11
→ impermanence joined15:13
← pyeveryt_ left15:15
→ lucasta joined15:15
→ madewokherd joined15:19
← madewokherd` left15:21
→ Rashad joined15:22
→ vitali64 joined15:23
→ rurtty joined15:25
← junebug left15:28
→ tflow joined15:28
← aidalgol left15:29
→ derpydoo joined15:31
→ junebug joined15:32
→ aidalgol joined15:32
→ atrigent joined15:32
← atrigent_ left15:32
← delay left15:32
→ delay joined15:33
← Coop left15:33
← coot left15:37
← lucasta left15:37
→ lucasta joined15:38
→ NSOL joined15:40
← NSOL left15:40
← lucasta left15:42
→ ajshell12 joined15:42
← ajshell1 left15:42
ajshell12 → ajshell115:42
andi_ diverdude: local and remote15:43
→ lucasta joined15:44
→ coot joined15:45
→ NSOL joined15:46
← coot left15:46
← VaniaPy left15:49
← rama left15:49
→ rama joined15:49
← rfuentess left15:49
→ ajshell17 joined15:50
→ pyeveryt_ joined15:50
← ajshell1 left15:51
ajshell17 → ajshell115:51
← rama left15:51
→ rama joined15:51
← madewokherd left15:52
→ feriman joined15:55
→ ajshell13 joined15:59
→ hbautista joined15:59
← ajshell1 left15:59
ajshell13 → ajshell115:59
← rama left15:59
→ indy joined16:00
→ rama joined16:00
← delay left16:01
← Anticom left16:01
→ delay joined16:01
← lucasta left16:05
→ lucasta joined16:05
← delay left16:08
← rama left16:09
→ rama joined16:09
← lucasta left16:09
→ lucasta joined16:10
→ ajshell11 joined16:16
← ajshell1 left16:16
ajshell11 → ajshell116:16
← dionysus69 left16:20
← lucasta left16:21
→ lucasta joined16:22
→ Ram-Z joined16:22
← rsx left16:26
← lucasta left16:27
→ lucasta joined16:29
← rama left16:29
→ rama joined16:30
→ subopt joined16:31
← lucasta left16:32
→ lucasta joined16:34
→ Sasazuka joined16:38
→ Coop joined16:39
→ nhartman__ joined16:43
nhartman__ Is there a good workflow for edit a commit in my history with changes to multiple files? I.e. I have commit A in my history. I modify files 'a' and 'b'. I want to amend commit A with my changes to file 'a'. Normally I would: git stash, interactive rebase, edit commit A, git stash pop. But this causes problems if file b has conflicts with the git stash pop. Is there a better way to do this?16:44
← pyeveryt_ left16:45
→ ick joined16:45
nhartman__ s/edit/editing/16:45
← zmt00 left16:45
ikke nhartman__: use git commit --fixup + git rebase -i --autosquash16:45
you'd limit each fixup commit to the changes you want to the commit you want to fix16:46
← lucasta left16:46
→ coot joined16:46
→ lucasta joined16:46
← pickanick left16:48
nhartman__ ikke: Wow, that's perfect. Thanks16:49
→ realies joined16:50
→ paul424 joined17:01
paul424 Hello, how do I unstash the command git stash ?17:02
ajak `git stash -h`17:02
← leftyfb left17:03
← rama left17:03
→ rama joined17:03
→ leftyfb joined17:03
paul424 ajak, and ?17:04
then the most recent one right ?17:04
ajak hm? maybe `man git stash` is what you want17:05
gitinfo the git-stash manpage is available at https://gitirc.eu/git-stash.html17:05
← rurtty left17:06
← Visne left17:09
← sa0 left17:10
→ sa0 joined17:10
← junebug left17:11
paul424 error: Your local changes to the following files would be overwritten by merge: source/render/ODFrameListener.cpp17:11
Please commit your changes or stash them before you merge.17:11
→ Sasazuka__ joined17:12
→ Visne joined17:12
ajak yeah you want to stash your changes17:13
← Sasazuka left17:14
→ pyeverything joined17:14
paul424 I just reverted back, and the git stash pop works as I wanted17:14
many thakns17:14
← lucasta left17:16
→ lucasta joined17:17
← pyeverything left17:19
← vdamewood left17:23
→ Narrat joined17:24
← Major_Biscuit left17:25
→ madewokherd joined17:25
→ john_johnk joined17:26
→ magic_ninja joined17:27
← Rashad left17:28
← adanwan left17:31
→ han-solo joined17:31
→ adanwan joined17:32
→ zebrag joined17:34
Sasazuka__ → Sasazuka17:34
→ tirnanog joined17:36
← paul424 left17:37
→ skapata joined17:40
→ thuna` joined17:41
→ hamburgler joined17:41
→ DoubleJ joined17:41
→ zmt00 joined17:42
← lucasta left17:44
→ lucasta joined17:46
← structoha left17:48
→ roadie joined17:48
→ Betal joined17:51
→ pyeverything joined17:51
← lucasta left17:51
→ lucasta joined17:52
nedbat i want to create some files while on the main branch, then add and stash the changes, switch to another branch, and "stash pop" the changes. but it's a merge conflict during the pop. I want to just clobber the files that are there. "stash pop" doesn't have a --hard switch. What should I do?17:55
← pyeverything left17:56
nedbat (looking at the scrollback, this seems very similar to the question an hour ago)17:57
(or maybe not)17:57
← Visor left18:00
← indy left18:00
→ Visor joined18:01
→ Telmud joined18:01
→ indy joined18:02
ajak no, i think you understand how stashing works (but i don't know if there's a good solution for your situation)18:05
waterkip I would just do `git stash apply` fix the conflict and `git stash drop` and be done with it18:05
NSOL nedbat: you could also use a patch18:06
nedbat on stackoverflow I found git stash show -p | git apply18:06
except for some reason it tells me: <stdin>:114: trailing whitespace.18:08
for a few lines18:08
IRChatter nedbat, if what you want is to have the very same file you could `git checkout <stash> file`18:08
nedbat IRChatter: where is "git checkout stash" documented?18:09
← srinidhi left18:10
IRChatter nedbat, when you stash something, git creates essentially a commit that gets discarded after you pop the changes. If you show your git log without poping you should see the checksum of such commit. If that's too much work, you can just use `git checkout stash@{0} <file>`18:13
stash@{n} (n = the index in the stash, 0 being the top/latest change stashed)18:15
← lucasta left18:17
nedbat i guess i didn't realize that "stash" was a ref to the stash18:17
→ lucasta joined18:18
IRChatter If you run `git stash list` you'll see each entry listed that way18:19
← rama left18:19
→ rama joined18:20
← tflow left18:20
mackerman I prefer copying changes to another branch with cherry-pick, but to each their own18:20
← ozoned left18:21
→ ozoned joined18:21
→ structoha joined18:22
IRChatter mackerman, yeah, that's an alternative. And if you'd like to keep working on it and avoid creating a new commit, is as simple as adding -n (--no-commit) to cherry-pick18:24
← nullroute left18:30
→ pyeverything joined18:30
← rgrinberg left18:31
justHaunted → ffffffradiofree18:32
ffffffradiofree → jusFffffffradiof18:32
→ nate1 joined18:33
← rama left18:33
→ rama joined18:33
→ srinidhi joined18:36
← lucasta left18:36
→ nullroute joined18:38
← nullroute left18:38
→ nullroute joined18:38
← nate1 left18:38
→ lucasta joined18:38
← jetchisel left18:39
→ velix joined18:43
velix Is this bad or right? https://stackoverflow.com/a/640734218:43
oops, it's from 2011 :D18:43
ikke It's still accurate18:43
→ Rashad joined18:43
← pyeverything left18:47
velix okay. So "never use git pull"?18:47
ikke You _can_ use git pull18:48
→ loulou joined18:48
ikke You just need to be aware of what it does18:48
Note that personally I always use git pull --rebase18:48
→ nvmd joined18:49
velix I think I need to do some experiments to understand the differences18:49
→ forgotmynick joined18:49
→ jetchisel joined18:51
→ FFY00_ joined18:55
← iomari891 left18:56
← Murr left18:57
→ Murr joined18:57
→ dom_ joined18:58
← FFY00 left18:58
→ KimK joined18:58
← dom_ left18:58
→ wallabra_ joined18:59
→ dom_ joined18:59
wallabra_ → wallabra19:01
← thebombzen left19:02
← dom_ left19:05
← roadie left19:07
→ ___nick___ joined19:09
← structoha left19:10
← lucasta left19:11
→ wootehfoot joined19:14
← Narrat left19:15
→ g0zart joined19:15
jusFffffffradiof → justThumbs19:17
← john_johnk left19:18
← Rashad left19:18
→ pyeveryt_ joined19:19
justThumbs → justVain19:20
IRChatter velix, branches are just pointers that point to a specific point in history (aka commit). Your local repo and the remote one, they both have their own independent set of branches that most of the times get named the same way (because it's easier to associate that way). Their history may deviate while working in teams, so to get updates on what the19:20
state of the remote branch is, one runs `git fetch`. That only updates the reference of the remote branch (e.g origin/my-branch). If you want to include/merge those new changes into your local version of the branch, you can then run `git merge origin/my-branch`. If you simply were behind in history, then git will try to just fast-forward your local19:20
branch and point to the same commit. On the other hand, if there's commits in your local branch that are not remotely, then you won't be able to fast-forward and you are presented with 2 choices: either a non fast-forward merge `git merge origin/my-branch` that creates a commit, or `git rebase origin/my-branch` that moves the start of your branch19:20
to the last commit of the remote one. In both cases (merge or rebase) if there's any conflicts, they'll need to get resolved before continuing. This process of fetching and merging is so common that git devs included `git pull` which essentially does `git fetch` and `git merge`. Alternatively, you can run `git pull --rebase` which does `git fetch`19:20
and `git rebase`19:20
velix IRChatter: rebase is still a mystery to me SOMETIMES :D sometimes it does what I need, sometimes it does not ,)19:22
← rama left19:23
→ rama joined19:23
→ __nick__ joined19:25
← ___nick___ left19:25
← pyeveryt_ left19:28
IRChatter velix, rebase, re-base, set the base of your branch elsewhere in relation to where it came from. That's in a nutshell the idea. It can get a bit more complex if you throw in things like squashing, but it's definitely not as scary as it seems19:28
velix IRChatter: I have to say, I'm using gitkraken. It helps me, but it feels like fraud to me. Normally I'm a shell guy. But I'm too afraid to mess things up.19:30
git is like the box of pandorra.19:30
and full of toxic snakes19:30
indiana jones wouldn't be happy about that.19:30
→ reset joined19:30
→ roadie joined19:32
← coot left19:33
IRChatter velix, no need to worry, once you've committed changes, it's unlikely that you're going to lose them. Even if you reset branches and a commit doesn't show up in your log, it's still there in gits database, and it can be recovered19:34
→ coot joined19:36
← roadie left19:37
→ Rashad joined19:39
IRChatter velix, try creating a new repo, committing a few files, `git reset --hard HEAD~`, and finally `git fsck --lost-found`19:39
→ rurtty joined19:40
→ roadie joined19:41
velix IRChatter: ok ;)19:41
What really confuses me is that some rebases create "remote-tracking branches". Maybe this is displayed in GitKraken only. I actually see no use in this.19:43
← rama left19:43
→ rama joined19:43
← srinidhi left19:44
IRChatter velix, I don't see how's that possible. A rebase doesn't create new branches19:46
← loulou left19:46
velix Gitkraken allows to ways to push a rebase: "Force push" (no tracking) and "fast forwared" (with tracking)19:47
Let's say, "main" has some hotfixes, that "feature" doesn't have. I'll then rebase "feature" on "main", won't I? "Github Desktop" does a "merge" when you update "feature" on "main".19:48
→ rama_ joined19:52
← rama left19:52
← rama_ left19:52
← g0zart left19:53
→ rama joined19:53
IRChatter velix, you can do both things to resolve that divergence in history, yes. The difference is that the merge will create a commit, whereas the rebase will simply put the feature branch on top of main19:54
velix In PRs, I often read "please do a rebase". So I think "rebase" is used more often.19:54
IRChatter velix, it definitely should be used more often, but my experience tells me otherwise19:55
velix A frontend guy told me, he doesn't like to rebase, since he often does many commits into a feature, which messes up the history. So he either squashes or merges.19:56
→ loulou joined19:56
← loulou left19:56
→ loulou joined19:56
→ pyeverything joined20:02
IRChatter velix, I don't see how rebasing messes up the history. Also, rebasing doesn't mean not merging. When you rebase the feature branch you'll still be merging it with the main one. You can decide if you want to just fast-forward main or create a commit, but a merge needs to happen. (I personally prefer to create a commit even when fast-forward is20:05
possible: `git merge --no-ff feature-branch`)20:05
velix IRChatter: I need to read what's the sense of that tracking :D20:06
← han-solo left20:06
→ thebombzen joined20:07
← nvmd left20:07
velix The big benefit of merging main back into the feature is: a nice graph. you can see where it came from :D20:08
→ odoood__ joined20:09
→ TonyStone joined20:09
← atrigent left20:09
→ atrigent_ joined20:10
← ozoned left20:11
← __nick__ left20:12
IRChatter velix, the point of git is not to track how things get developed, but what the exact steps to do it need to be. That "nice graph" that you mention is ok if you have 1 or 2 short-lived topic branches, but I've seen way too many topic branches that take weeks to merge back into main, and that "nice graph" turns into a labyrinth pretty quickly20:13
→ ___nick___ joined20:14
velix makes sense20:15
← ___nick___ left20:15
→ ozoned joined20:16
→ ___nick___ joined20:17
← odoood__ left20:19
velix Ah, when doing "Pull (fast-forward if possible)", then Gitkraken creates a tracking thing.20:19
When doing "push", it doesn't do.20:19
erm20:19
When doing "force push", it doesn't do.20:19
Github Desktop also does a "force push"20:19
→ odoood joined20:20
→ Crispy joined20:20
← ozoned left20:22
← wallabra left20:22
→ wallabra joined20:23
→ ss4 joined20:23
IRChatter velix, "tracking" is just the action of associating a remote branch with a local one. When running the pull (in Gitkraken), since a remote branch is being merged with a local one, it's fair to assume that you're going to want to push that local branch to the remote one that got merged. So to avoid having to explicitly state the name of the remote20:26
branch it was previously "tracked". If it's not tracked then you'll need to run `git push origin remote-branch`, when tracked you just need to `git push`20:26
← wootehfoot left20:26
velix IRChatter: see.. that's what I call "rocket science".20:27
← TomyWork left20:27
IRChatter velix, a piece of cake once you understand it20:27
→ rgrinberg joined20:28
→ wootehfoot joined20:28
ztrawhcse merge commits when merging a feature into master is fine. it's repeated merges from master into temporary feature branches that causes issues.20:30
IRChatter velix, the way that you manually start tracking a branch is by including the option --set-upstream (-u) to `git pull` (e.g. `git pull -u origin remote-branch`)20:30
push20:30
sorry20:31
← ss4 left20:31
IRChatter Muscle memory20:31
`git push -u origin remote-branch`20:32
→ junebug joined20:32
← loulou left20:36
← gxt left20:36
← Sasazuka left20:36
← feriman left20:38
← bkircher left20:41
→ gxt joined20:41
→ skfax joined20:42
→ Tangent-Man joined20:44
← Tangent-Man left20:44
← derpydoo left20:45
→ Tangent-Man joined20:45
→ nate1 joined20:45
→ haritz joined20:45
← haritz left20:45
→ haritz joined20:45
← vitali64 left20:45
← hololeap left20:50
→ hololeap joined20:50
← rama left20:51
→ rama joined20:51
← rama left20:54
→ rama joined20:54
← gh34 left20:55
→ loulou joined21:00
← loulou left21:00
→ loulou joined21:00
← pyeverything left21:02
AmyMalik i am an insufferable git21:04
→ dom_ joined21:05
← dom_ left21:06
← kenanmarasli left21:06
← sd5 left21:06
← roadie left21:09
→ roadie joined21:09
← junebug left21:10
→ junebug joined21:10
← odoood left21:11
IRChatter AmyMalik git out of here then21:12
AmyMalik :D21:12
← rama left21:13
→ rama joined21:13
← roadie left21:14
← mw_ left21:18
← forgotmynick left21:19
← coot left21:19
← Telmud left21:22
← hoochmonger left21:23
← Jong left21:23
← zebrag left21:33
→ zebrag joined21:33
→ arescorpio joined21:33
← pikapika left21:39
← robo left21:40
→ pyeveryt_ joined21:40
→ vertreko joined21:40
→ roadie joined21:40
← loulou left21:40
← vertreko left21:46
← pyeveryt_ left21:46
→ robo joined21:47
← nate1 left21:48
← velix left21:49
← roadie left21:50
← igemnace left21:54
← rama left21:56
→ rama joined21:56
→ pikapika joined21:59
→ gast0n joined22:00
→ rsx joined22:06
→ Cleverness joined22:07
→ z1haze joined22:10
z1haze when trying to pull from a subtree that im used to pulling from, I am now getting an error of could not rev-parse split hash -- what could cause this?22:10
→ dionysus69 joined22:13
→ roadie joined22:16
→ pyeveryt_ joined22:19
← rama left22:19
→ rama joined22:20
→ Sasazuka joined22:22
← dionysus69 left22:23
← trace1918 left22:23
← hololeap left22:24
→ mw_ joined22:24
← gas51627 left22:24
← roadie left22:28
← pyeveryt_ left22:29
← rsx left22:33
← rama left22:36
→ rama joined22:36
← rama left22:40
→ rama joined22:41
→ cvmn joined22:50
← wootehfoot left22:54
← elastic_dog left22:55
→ roadie joined22:55
→ elastic_dog joined22:56
← skfax left22:56
← arescorpio left22:59
← rama left23:03
→ jed joined23:03
→ pyeveryt_ joined23:03
→ rama joined23:03
← roadie left23:04
← hbautista left23:06
← cvmn left23:08
← rama left23:09
→ rama joined23:09
← sa0 left23:09
→ cvmn joined23:10
← pyeveryt_ left23:12
→ pyeveryt_ joined23:12
→ sa0 joined23:13
← rama left23:15
→ rama joined23:15
← rama left23:17
→ rama joined23:17
← rama left23:18
← sa0 left23:18
→ rama joined23:18
→ roadie joined23:19
← rama left23:19
→ rama joined23:19
→ loulou joined23:20
← loulou left23:20
→ loulou joined23:20
← Tangent-Man left23:23
← cvmn left23:26
← IRChatter left23:29
← roadie left23:31
→ k8yun joined23:31
← epony left23:32
← cdown left23:37
← rurtty left23:37
← impermanence left23:38
← user left23:39
→ d1b joined23:42
→ user joined23:43
→ nate1 joined23:45
← rama left23:48
→ rama joined23:48
← jed left23:49
→ epony joined23:50
← nate1 left23:50
← junebug left23:53
← rama left23:54
→ m0viefreak joined23:55
→ roadie joined23:56

Logs Search ←Prev date Next date→ Channels Documentation