IRCloggy #git 2022-10-13

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.

2022-10-13

← TheCatBehemoth left00:02
→ hiroot joined00:07
← softirq left00:13
← nyah left00:14
→ sa0 joined00:14
→ Jong joined00:17
← moldorcoder7 left00:17
→ arch_angel joined00:24
← pyeverything left00:25
← ferdna left00:27
→ perrierjouet joined00:28
→ gsi_ joined00:30
→ softirq joined00:32
← Yruama left00:34
← FinnElija left00:40
← fling left00:40
→ fling joined00:40
→ FinnElija joined00:41
→ ALowther_ joined00:41
← softirq left00:43
← ALowther_ left00:46
→ softirq joined00:47
← softirq left00:48
→ softirq joined00:51
→ Lord_of_Life joined00:54
← arescorpio left01:02
→ skered joined01:05
→ TonyStone joined01:07
← causa_sui left01:07
← clime left01:08
← xx left01:10
← softirq left01:18
→ dsrt^ joined01:20
← Sasazuka__ left01:21
→ Ragnor joined01:21
← Stooge left01:23
← sa0 left01:29
← FinnElija left01:35
→ FinnElija joined01:36
← rama left01:43
→ junebug joined01:44
→ nate1 joined01:45
→ drbeco joined01:47
→ lucasta joined01:48
→ umbramalison joined01:59
→ gco joined02:09
← av2156 left02:13
← atrigent left02:26
→ atrigent_ joined02:26
→ finn_elija joined02:33
← FinnElija left02:33
finn_elija → FinnElija02:33
→ pyeverything joined02:38
← arch_angel left02:46
← lucasta left02:48
→ lucasta joined02:48
← nate1 left02:48
→ theobjectivedad joined02:51
→ terrorjack joined02:52
→ structoha joined02:59
← structoha left03:05
← zebrag left03:09
← junebug left03:10
→ nilradical joined03:13
nilradical hello, how can i go back to a previous commit in Github Desktop? someone has made some bad commits and i need to delete them03:13
→ fstd_ joined03:16
nilradical i feel like this should be simple but cant find any instructions on how to achieve it03:18
in other words, simply want to revert the repo to a previous state03:19
→ chexum joined03:19
← rurtty left03:19
fstd_ → fstd03:19
→ Stooge joined03:20
→ junebug joined03:20
← junebug left03:22
→ junebug joined03:22
← loulou left03:25
← lucasta left03:26
→ lucasta joined03:26
← lucasta left03:33
→ sjoshi joined03:34
→ lucasta joined03:34
← sjoshi left03:34
→ Vonter joined03:35
← gco left03:38
gnoo nilradical: git reset --hard <commit> ? this will remove all your changes after the commit so use it carefully.03:40
→ sjoshi joined03:40
→ Gamah joined03:57
→ arch_angel joined03:59
arch_angel → arch-angel03:59
→ structoha joined04:00
← lucasta left04:04
→ lucasta joined04:04
nilradical thanks gnoo04:05
i ended up using 'Revert changes on commit' in reverse chronological order to the bad commits in GUI04:05
ztrawhcse nilradical: if you need to revert a bunch of commits in one go, then the easiest might be if all those commits were merged via a merge commit and you can revert the merge04:06
but otherwise, you can do `git checkout <last-good-commit> .`04:07
this changes all files to the good version, and stages them for a new commit, so you can commit as "restore all files to the last working version" or whatever04:07
nilradical ztrawhcse: i did try the last thing first and then it said i have a 'detached head' which kinda of horrified me lol04:08
ztrawhcse well, you need the dot at the end04:08
git checkout <commit> <path>04:08
nilradical oh04:08
ztrawhcse where the path is "." for everything in the current directory04:09
← lucasta left04:09
→ lucasta joined04:09
ztrawhcse that makes checkout change your file contents, rather than move to a detached HEAD :)04:09
← junebug left04:11
→ matni joined04:12
ztrawhcse nilradical: IIRC newer versions of git offer the two modes of `git checkout` as `git switch <commit>` to go to a detached head, and `git restore -s <oldcommit> .` to change all files to the same contents as some other commit04:14
nilradical i see04:14
it would be nice if they put in the GUI just a 'restore to this state' option for dummies like me04:15
apparently if you revert things in the wrong order things can go wrong.04:15
← gast0n left04:17
ztrawhcse well, that's `git restore -s <to this state> .`04:18
← lucasta left04:18
→ lucasta joined04:19
gnoo nilradical: usually it is much better if you learn the cli tools instead of depending on gui ones. third party gui might not support everything that cli git does and might even do wrong(undesirable) things with no way to change it04:19
nilradical ok04:20
→ vanessa joined04:25
← structoha left04:37
← pyeverything left04:40
→ pyeveryt_ joined04:41
← lucasta left04:45
← pyeveryt_ left04:45
→ nate1 joined04:45
→ lucasta joined04:45
← bloody left04:46
← dermoth left04:47
→ dermoth joined04:47
→ skylize joined04:49
← nate1 left04:50
← Stooge left04:52
← lucasta left04:52
→ lucasta joined04:52
→ bloody joined04:55
← bloody left04:58
→ srinidhi joined05:06
→ lgc joined05:07
← lucasta left05:09
→ lucasta joined05:09
← lgc left05:10
← squirrel left05:10
→ bgs joined05:13
← dermoth left05:15
→ dermoth joined05:15
← sudoforge left05:16
→ powftw joined05:17
→ madewokherd` joined05:17
← madewokherd left05:21
← dsrt^ left05:36
← nilradical left05:40
→ structoha joined05:41
→ carl- joined05:44
→ carl-- joined05:46
← adanwan left05:47
← arch-angel left05:47
→ adanwan joined05:48
← carl- left05:48
← chimbosonic left05:54
→ chimbosonic joined05:54
← hbautista left05:57
← lucasta left05:59
→ lucasta joined06:00
→ igemnace joined06:10
→ rfuentess joined06:20
→ theoceaniscool_ joined06:21
→ theoceaniscool joined06:21
← jazz-o-lantern left06:30
← thiago left06:36
← R2robot left06:40
→ R2robot joined06:41
→ aidalgol joined06:46
→ mexen joined06:47
→ rom1v joined06:52
→ mncheck joined06:52
→ Telmud joined06:56
→ kenanmarasli joined06:59
→ menace joined07:00
→ IRChatter joined07:05
← Vonter left07:06
→ trace1918 joined07:08
→ VaniaPy joined07:09
← fling left07:11
→ fling joined07:12
→ gryffus joined07:12
→ sa0 joined07:14
→ RiFo joined07:15
→ MajorBiscuit joined07:17
← menace left07:20
→ menace joined07:20
menace → Deknos07:20
← trace1918 left07:28
→ trace1918 joined07:30
→ dionysus69 joined07:30
→ DibbTubberOne joined07:36
→ rosco joined07:42
← fling left07:44
→ Bilge joined07:44
→ fling joined07:44
→ TomyWork joined07:45
→ mns joined07:48
← fling left07:49
← voldial left07:49
→ fling joined07:49
→ Vonter joined07:50
→ clime joined08:03
← wagle left08:04
→ wagle joined08:05
→ feriman joined08:07
← fling left08:16
← hiroot left08:17
→ fling joined08:17
← kenanmarasli left08:31
← skapata left08:36
← rosco left08:37
→ peirik_ joined08:44
← Telmud left08:44
→ mobidrop joined08:45
→ nate1 joined08:47
← nate1 left08:52
→ rosco joined08:58
→ makara joined09:01
→ alfredb joined09:02
→ makara1 joined09:02
→ moldorcoder7 joined09:05
← t-mart left09:07
← makara left09:11
← makara1 left09:11
→ nyah joined09:12
→ makara joined09:12
→ akimbo533430847 joined09:13
← momomo left09:15
→ makara1 joined09:15
→ cdown joined09:16
→ momomo joined09:16
→ gxt joined09:20
→ lacrymology joined09:22
→ bkircher joined09:25
Voxel Hey all! ^^09:35
osse Hey one09:36
← VaniaPy left09:37
Voxel For example, there are 2 commits. The first has two files chaged (e.g. file1, file2). The second has modified 'file2' only. Does Rebase default Squash these 2 commits will make a single commit with two files 'file1' and the new 'file2' from the second commit?09:37
← Vonter left09:38
apo rebase doesn't squash by default, no09:38
Voxel I mean, I just did squash it interactively and thought that it would ask what lines to choose from the both commits in file2, yet it just returned a combined commit09:39
So, it just chose the new file2?09:39
← _xor left09:41
osse Voxel: it applied the firt commit's changes to both file1 and file2, then without committing first it applied the changes from commit2 to file2, and commited that whole thing09:43
→ _xor joined09:43
osse that's all09:43
no more (or less) magic09:43
Voxel I see. So, in the result, there are file1 from commit1 and file2 from commit2 since it got modified09:44
Magnificent!09:44
git <309:44
→ battos joined09:45
osse note that it doesn't necessarily have to be exactly the same09:45
if eg you do git rebase -i origin/master09:45
← _xor left09:46
osse then you might combine the results but on top of any changes from upstream09:46
→ gas51627 joined09:46
Voxel I am sorry, but I did not get it09:46
→ john_johnk joined09:46
Voxel What is not the same?09:46
← powftw left09:47
Voxel Wouldn't it check the order and apply the changes similarly?09:47
Like, apply local commits (commit2) to fetched origin/master (commit1) -> rebased?09:48
If squashed*09:48
Or, am I mistaken here?09:48
Perhaps, I just didn't get "have to be exactly the same"09:48
→ _xor joined09:53
osse Voxel: it would apply the changes similarly09:58
but if you actually re-base (as in change the base of the commits) then file1 and file2 might be different from when you made the commits09:58
Maybe you know all this, but I just reacted to "file2 from commit2" as if file2 was *exactly the same* as it was in commit2 before the rebase as after the rebase09:59
that's not always the case09:59
← john_johnk left10:01
← fling left10:04
→ fling joined10:05
→ zeenk joined10:07
→ Vonter joined10:11
← mncheck left10:12
Voxel osse, you mean an interactive rebase with manual changes in between?10:14
→ oprypin joined10:14
Voxel i.e. 'edit' instead of 'pick' in an interactive rebase?10:15
← DibbTubberOne left10:20
← YuGiOhJCJ left10:27
← vishal left10:29
← oprypin left10:29
→ oprypin_ joined10:29
← rosco left10:30
→ vishal joined10:31
← Vonter left10:32
← trace1918 left10:33
→ trace1918 joined10:35
← Jong left10:39
← trace1918 left10:39
← MajorBiscuit left10:41
← lucasta left10:42
→ vitali64 joined10:44
→ DaNeenjah joined10:44
→ Vonter joined10:45
→ elastic_1 joined10:46
elastic_1 → elastic_dog10:46
→ f451 joined10:49
osse Voxel: No, I mean any rebase in general11:01
→ xx joined11:04
← makara left11:05
→ makara joined11:05
→ makara_ joined11:06
← makara1 left11:06
→ MajorBiscuit joined11:09
→ arch_angel joined11:14
→ thuna` joined11:18
→ Fernando-Basso joined11:26
→ iomari891 joined11:26
← bgs left11:29
→ john_johnk joined11:30
← vitali64 left11:31
← Vonter left11:31
→ DibbTubberOne joined11:33
→ Vonter joined11:33
→ VaniaPy joined11:34
VaniaPy hi guys, after merging my branch to master i should to the master correct?11:35
→ CarlFK joined11:35
→ kuhilas joined11:35
→ paperino42[m] joined11:35
→ heftig joined11:35
→ jmcantrell joined11:35
→ yuu[m] joined11:35
→ Celmor[m] joined11:35
→ jinsun_ joined11:35
→ jcgruenhage joined11:35
→ identitas joined11:35
→ ForeverNoob[m] joined11:35
→ graham joined11:35
→ cp1313[m] joined11:35
→ vitali64[m] joined11:35
→ Eryn83FL_Matrix[ joined11:35
→ MassiveBox[m] joined11:35
→ Helmholtz joined11:35
→ morguldir[m] joined11:35
→ nazarewk[m] joined11:35
→ r00ster91 joined11:35
→ T_UNIX[m] joined11:35
→ dunpeal joined11:35
→ morguldir joined11:35
→ tuxayo joined11:35
→ tiziodcaio joined11:35
→ stretch1 joined11:35
→ kadoban joined11:35
VaniaPy i should push*11:35
→ grahamperrin joined11:35
→ DX099 joined11:35
→ matthewcroughan- joined11:35
→ defolos joined11:35
→ underpantsgnome[ joined11:35
→ psydroid joined11:35
→ bluepenquin joined11:35
→ Dyluster[m] joined11:35
ikke yes11:36
→ ben[m] joined11:36
→ artemon joined11:36
ikke I mean, it's not required, but would make sense11:36
→ klopsch[m] joined11:36
→ KREYREN joined11:36
→ oooo[m] joined11:36
→ hepatocyte[m] joined11:36
→ rlgo[m] joined11:36
→ cpackham[m] joined11:36
ikke If you want to publish that merge11:36
VaniaPy yep, but i'm getting rejected main -> main (fetch first)11:37
ikke It means main got changes you do not have locally yet11:37
VaniaPy should i fetch ?11:38
ikke fetch and then merge what was fetched or rebase your changess on top of it11:38
← atrigent_ left11:39
→ atrigent joined11:39
VaniaPy ok11:39
ty11:39
damn now i'm getting non-fast forward11:39
← Deknos left11:45
→ Hylden joined11:46
VaniaPy ikke when i'm trying to merge it opens a window with MERGE_MSG11:47
Hylden Ok, simple question: I checkout a previous commit, now how do I make it the master?11:47
VaniaPy am it on the right way?11:47
Hylden *checked out11:47
IRChatter VaniaPy, it's often preferable to merge into main/master through PRs/MRs. And if your branch is behind, instead of merging to your topic branch, rebase your topic branch11:47
ikke VaniaPy: it's expected if you merge diverged history (which has happened)11:48
→ pulse joined11:49
IRChatter Hylden, instead of `checkout` use `reset --hard` to move the branch to that commit11:49
VaniaPy IRChatter i'm merging to a dev branch which is for dev purposes before merge to actual master11:49
← DibbTubberOne left11:51
→ Schrostfutz joined11:51
← john_johnk left11:52
IRChatter VaniaPy, do you really need a long-running dev branch?11:53
VaniaPy it's not my choice11:54
unfortuantely11:54
→ DibbTubberOne joined11:54
IRChatter VaniaPy, well, I'd still try to rebase that dev branch instead of merging it with your topic branch11:55
VaniaPy hmm Current branch dev is up to date.11:55
i' m getting11:55
→ Ingvix joined11:57
← CarlFK left11:57
IRChatter VaniaPy, am I misunderstanding something? You have a master, a dev and a topic branch, am I right?11:58
→ user joined11:58
VaniaPy yep11:58
→ CarlFK joined11:59
IRChatter VaniaPy, ok, so, you said that hard merged your topic branch with dev locally and when trying to push you couldn't because there were changes remotely that you hadn't (in dev). If so, my way of dealing with it, would be to undo the merge you did, then fetch the changes, rebase (so your branch starts at the last commit), and then you can merge and12:00
push without conflict12:00
you said that you had* merged12:01
← user left12:01
← Hylden left12:01
VaniaPy how can i undo the merge?12:05
← Schrostfutz left12:05
→ leah2 joined12:07
→ user joined12:08
IRChatter VaniaPy, reset the local dev branch to the previous commit. `git checkout dev`, `git reset --hard HEAD^`12:08
→ Schrostfutz joined12:08
→ kaisyu joined12:09
IRChatter After that, pull the dev changes (which can be done via `git pull` or `git fetch` + `git merge`). And after that, you'll be able to rebase and push without issue12:10
VaniaPy ok ok ty12:11
mackerman I encourage seperate fetch and merge, until you understand them.12:11
← FinnElija left12:11
VaniaPy i did git reset --hard HEAD and git pull12:11
and then merged again12:11
and pushed12:12
→ FinnElija joined12:12
IRChatter VaniaPy, that works, but the history is not as clean as with git rebase12:13
← Vonter left12:15
← Schrostfutz left12:20
→ Schrostfutz joined12:22
→ Vonter joined12:22
VaniaPy ok12:25
→ Rashad joined12:30
← VaniaPy left12:36
→ Schrostfutz_ joined12:37
← Schrostfutz left12:39
← sa0 left12:39
→ sa0 joined12:40
← peirik_ left12:40
→ gooble_gobble joined12:46
→ nate1 joined12:48
← pickanick left12:50
→ derpydoo joined12:51
→ jetchisel joined12:52
← nate1 left12:54
→ zumba_addict joined12:54
← battos left12:56
→ VaniaPy joined12:57
→ bloody joined12:58
← Murr left12:58
← feriman left12:58
→ Murr joined12:58
→ vicecea joined13:01
→ hoochmonger joined13:03
→ pickanick joined13:06
← fling left13:09
→ fling_ joined13:09
→ dfee joined13:10
← carl-- left13:16
fling_ → fling13:16
← Xenguy left13:17
→ arescorpio joined13:19
← KimK left13:21
← oprypin_ left13:25
← dfee left13:27
→ dfee joined13:28
← moldorcoder7 left13:31
→ moldorcoder7 joined13:32
← fling left13:34
→ fling joined13:35
← DibbTubberOne left13:36
← cyberpear left13:39
→ rosco joined13:42
→ carl- joined13:43
← derpydoo left13:46
→ carl-- joined13:47
→ gast0n joined13:47
← Vonter left13:50
← carl- left13:50
← structoha left13:51
→ hamburgler joined13:53
← drbeco left13:54
→ drbeco joined13:55
→ oprypin joined13:56
→ sudoforge joined13:59
← rosco left14:01
→ madewokherd joined14:01
→ roadie joined14:01
← carl-- left14:01
← MassiveBox[m] left14:01
← oooo[m] left14:01
← paperino42[m] left14:01
← matthewcroughan- left14:01
← stretch1 left14:01
← makara_ left14:01
← xx left14:01
← mobidrop left14:01
← R2robot left14:01
← chimbosonic left14:01
← dermoth left14:01
← sjoshi left14:01
← daru_ left14:01
← wilz left14:01
→ davve joined14:01
→ DreamweaverN joined14:01
→ Marmotte joined14:01
→ poucet joined14:01
→ leftyfb joined14:01
→ goblin joined14:01
→ Arsen joined14:01
→ wilsonjholmes joined14:02
→ txtsd joined14:02
→ gjnoonan joined14:02
→ mobidrop joined14:02
→ jumpcutking joined14:02
→ doppo joined14:02
→ nyuszika7h joined14:02
→ carl-- joined14:02
→ farn joined14:02
→ jakesyl____ joined14:02
→ wilz joined14:02
→ jelmer joined14:02
→ dupondje joined14:02
→ k-man joined14:02
← mobidrop left14:02
→ mobidrop joined14:02
→ Xeago joined14:02
→ Quietus joined14:02
→ ephemer0l joined14:02
→ osse joined14:02
→ biapy joined14:02
→ coldmountain joined14:02
← nyuszika7h left14:02
→ nyuszika7h joined14:02
→ hnOsmium0001 joined14:02
← k-man left14:02
→ k-man joined14:02
← leftyfb left14:03
→ leftyfb joined14:03
→ stretch1 joined14:03
← hnOsmium0001 left14:03
→ hnOsmium0001 joined14:03
→ JanC joined14:03
→ Voxel joined14:03
→ FFY00 joined14:03
→ dermoth joined14:03
→ makara_ joined14:03
→ xx joined14:03
→ GingerGeek joined14:03
→ tcurdt joined14:03
→ anotherNightmare joined14:03
anotherNightmare → another14:03
→ matthewcroughan- joined14:03
→ R2robot joined14:03
→ daru joined14:04
→ paperino42[m] joined14:04
← madewokherd` left14:04
→ oooo[m] joined14:04
→ derpydoo joined14:04
→ A_Dragon joined14:05
← roadie left14:06
→ MassiveBox[m] joined14:07
→ sjoshi joined14:07
→ DonRichie joined14:07
→ tuxifreund joined14:09
← Arsen left14:13
← kaisyu left14:18
→ Arsen joined14:19
→ vdamewood joined14:23
← mobidrop left14:25
← umbramalison left14:28
← zumba_addict left14:28
→ thiago joined14:30
→ roadie joined14:32
← carl-- left14:32
→ drogas joined14:35
← atrigent left14:36
→ gh34 joined14:37
← fling left14:38
← roadie left14:39
→ roadie joined14:39
→ fling joined14:39
← gast0n left14:41
→ atrigent joined14:42
→ blaklistd joined14:45
← dupondje left14:46
→ dupondje joined14:46
→ nate1 joined14:49
→ rurtty joined14:53
← nate1 left14:54
→ umbramalison joined14:54
← R2robot left14:55
→ cloaker joined14:56
← Murr left14:58
→ Murr joined14:59
→ jazz-o-lantern joined15:00
← Rashad left15:01
→ gast0n joined15:04
→ vitali64 joined15:07
→ nullroute joined15:09
← nullroute left15:09
→ nullroute joined15:09
A_Dragon → Awoobis15:10
← cloaker left15:21
← Voxel left15:23
→ Voxel joined15:24
← VaniaPy left15:30
← atrigent left15:33
→ atrigent_ joined15:33
→ petaflot joined15:34
→ feriman joined15:36
petaflot hello! I sumbitted a feature request a while ago (< 30 days AFAIK) but did not subscribe to the mailing list.. now I'm trying to see if anyone responded but I am unable to find anything (I tried searching for my e-mail on https://lore.kernel.org/git/)15:37
← RiFo left15:38
→ mven joined15:38
petaflot is it possible to get some assistance on this?15:41
selckin whats your email or subject etc15:42
→ Vonter joined15:43
→ cyberpear joined15:51
← Schrostfutz_ left15:52
← karabaja4 left15:55
→ NSOL joined15:57
→ skapata joined16:00
→ rama joined16:01
→ menace joined16:05
← menace left16:06
→ menace joined16:06
menace → Deknos16:06
← rfuentess left16:07
← dfee left16:09
→ lagash joined16:10
→ Jong joined16:11
← gryffus left16:14
→ dfee joined16:15
← thuna` left16:16
← MajorBiscuit left16:20
→ MajorBiscuit joined16:22
→ delay joined16:22
→ R2robot joined16:24
← dionysus69 left16:24
← MajorBiscuit left16:27
→ gryffus joined16:32
← DaNeenjah left16:37
→ softirq joined16:40
→ gnoo joined16:43
← feriman left16:44
→ crb joined16:44
→ Guest3 joined16:50
→ mat001 joined16:52
another → anotherNightmare16:52
← delay left16:54
→ powftw joined16:55
← rama left16:56
→ rama joined16:56
← mat001 left16:56
→ mat001 joined16:57
→ delay joined16:57
→ rsx joined16:57
← fling left16:59
← delay left17:00
← coot left17:00
→ fling joined17:00
→ PlasmaHH joined17:02
← sa0 left17:07
→ zan joined17:10
→ Sasazuka joined17:11
← zmt00 left17:12
← oprypin left17:14
← rama left17:14
→ rama joined17:14
→ cdown_ joined17:15
← cdown left17:17
cdown_ → cdown17:17
← rsx left17:25
← vitali64 left17:26
← user left17:27
→ thuna` joined17:27
← softirq left17:27
→ softirq joined17:29
→ machaddr joined17:32
→ oprypin joined17:32
← skylize left17:33
→ Betal joined17:33
→ GNUmoon joined17:33
→ sa0 joined17:35
← igemnace left17:36
← rama left17:36
→ rama joined17:37
← atrigent_ left17:39
→ delay joined17:44
→ Telmud joined17:48
→ t3nj1n joined17:52
→ wootehfoot joined17:57
← machaddr left18:01
→ atrigent joined18:02
→ hays joined18:04
→ njt joined18:06
← delay left18:06
→ cdown_ joined18:06
← Coop left18:07
← rama left18:07
→ rama joined18:08
← cdown left18:09
cdown_ → cdown18:09
→ int3r3st3d joined18:11
← int3r3st3d left18:12
← softirq left18:12
→ trace1918 joined18:14
→ delay joined18:16
→ user joined18:16
→ Coop joined18:20
← delay left18:20
→ zer0bitz joined18:22
→ feriman joined18:22
→ coot joined18:28
← coot left18:28
→ coot joined18:28
→ dionysus69 joined18:29
← t3nj1n left18:31
← NSOL left18:33
→ NSOL joined18:33
← NSOL left18:33
← Vonter left18:34
→ NSOL joined18:35
→ skylize joined18:35
← gsi_ left18:35
→ gsi joined18:36
← gsi left18:42
→ gsi joined18:43
→ green_squirrel joined18:45
→ reset joined18:45
← Guest3 left18:48
← rama left18:48
→ rama joined18:48
green_squirrel Hi, I want to achieve this using git: provided: [a list of <directories>, a <first> commit, a <last> commit], I want to forge a git command that will return the list of git commits between <first> and <last> with at least 1 modification touching a file contained in <directories>, ordered by the number of files modified contained in <directories>.18:49
The use case is: I am tasked to produce a changelog for an application in a monorepo.18:49
If someone has some pointers how I could achieve this, Id appreciate!18:50
→ nate1 joined18:50
← mat001 left18:50
← trace1918 left18:51
← iomari891 left18:52
→ structoha joined18:55
← nate1 left18:56
← roadie left18:56
cbreak green_squirrel: have you looked at git rev-list?18:57
in particular, git rev-list a..b -- path18:57
also, consider making individual repositories for unrelated projects, that'll save you a lot of trouble18:57
→ zmt00 joined18:58
green_squirrel Oh nice, thanks! And yes I agree for individual repos... didn't make that decision :P19:01
← TomyWork left19:02
→ mat001 joined19:02
← mat001 left19:02
→ mat001 joined19:03
→ lucasta joined19:05
cbreak green_squirrel: git log supports the same syntax19:09
if you want the log messages and not just the revisions19:09
or git log -p if you want the whole non-binary patches19:10
→ kenanmarasli joined19:11
← Sasazuka left19:15
→ roadie joined19:21
→ hans_ joined19:22
← roadie left19:26
← rama left19:27
→ trace1918 joined19:30
PlasmaHH Hi, is there a way to automatically have one file being ignored during a merge from one branch to another?19:30
ikke no19:31
→ rama joined19:31
→ codaraxis___ joined19:34
← powftw left19:35
→ roadie joined19:38
← codaraxis__ left19:38
← lucasta left19:42
→ lucasta joined19:43
← roadie left19:44
← zeenk left19:45
→ roadie joined19:47
→ delay joined19:47
← Telmud left19:49
→ Sasazuka joined19:52
← delay left19:53
IRChatter PlasmaHH, what's the reason behind your question?19:54
← roadie left19:57
← trace1918 left19:58
→ delay joined20:00
← delay left20:00
→ roadie joined20:02
→ delay joined20:03
← delay left20:08
green_squirrel cbreak: ok nice thanks20:13
PlasmaHH IRChatter: I would like to kind of document the branch purpose in a certain file thats always the same. Think of githubs README.md or so. It will be displayed in a webinterface and I want to tell the users a bit more about that branches purpose and status than just the branch name20:14
→ LuxuryMode joined20:15
cbreak PlasmaHH: README.branchname.md20:16
then you won't ever have merge conflicts20:16
PlasmaHH cbreak: thats nice if the company webinterface would support that...20:16
cbreak ?20:16
you can't pick the name of files?20:17
PlasmaHH no, it will display only one certain file, Readme.wiki (even the case must fit)20:18
cbreak then just document all branches in that file20:18
← TonyStone left20:19
→ delay joined20:20
PlasmaHH cbreak: that doesn't give me the functionality I need, I need different contents of the file. Among other things, the first sentence/heading is used in menus on the webinterface20:20
cbreak then you'll have to merge manually.20:21
you can resolve merge conflicts with git checkout --ours or --theirs20:22
← wootehfoot left20:22
← feriman left20:22
PlasmaHH cbreak: yeah I know... I was hoping for something there that would essentially use this to ignore the file...20:23
→ Tangent-Man joined20:26
← Tangent-Man left20:27
→ Tangent-Man joined20:28
mackerman Make a nice commit message on the first commit of the branch.20:29
Hmm, doesn't work as in the web UI preview.20:30
← thuna` left20:31
← coot left20:31
← fling left20:36
→ fling joined20:37
PlasmaHH hm, could it work that way...? when the file in question is deleted in master but present in the branch, that should lead to a merge conflict, and by chosing ours for that, it essentially ignores the file... and then maybe set an attribute for that file and use a custom merge driver for it... but it seems that .gitattributes doesnt work in this case since the file isn't there...20:39
← rama left20:39
← lucasta left20:40
← kenanmarasli left20:40
→ rama joined20:40
→ lucasta joined20:40
← srinidhi left20:41
PlasmaHH indeed when the file is there, this works and just ignores the changes from the other file, that could be something to work with20:43
← Sasazuka left20:44
→ Sasazuka joined20:45
→ m0viefreak joined20:48
← delay left20:49
← sjoshi left20:53
← PlasmaHH left20:56
← jinsun left20:58
← cdown left20:58
← Murr left20:58
→ Murr joined20:58
user i am the git master21:02
osse no i am21:10
gtfowtmfbs21:11
→ trace1918 joined21:11
← roadie left21:14
→ PlasmaHH joined21:22
← Tangent-Man left21:23
← gh34 left21:29
→ intelfx joined21:31
← PlasmaHH left21:32
← rama left21:33
→ rama joined21:33
→ mcon joined21:34
← intelfx left21:36
→ intelfx joined21:36
mcon is it possible to do a `git rebase -i "since last push"`? or do I need to know how many unpushed commits are there?21:36
nedbat mcon: try this: git rebase -i @{u}21:38
← derpydoo left21:38
← oprypin left21:38
→ roadie joined21:40
→ oprypin joined21:40
← lucasta left21:43
mcon nedbat: "fatal: no upstream configured for branch 'lua'" Apparently I never pushed my current branch; what is the spell to `git rebase -i` since branch creation? (i can count 8 commits in `git log`21:45
→ lucasta joined21:45
← roadie left21:48
gsi "upstream" is not about having pushed, it's about tracking another branch21:48
mcon nedbat: ... so I can do `git rebase -i HEAD~8`, but I am still curious.21:48
gsi see checkout --track or branch --upstream21:48
and yes you can always specify which commit to use21:49
intelfx Hi.21:49
Just a quick sanity check before I send a longer mail to the git ML and/or submit a bug report: is there any reason why `git fetch` (more specifically, `git fetch --all -j$(nproc)`) would sometimes (very often) touch some of the existing packs, without actually modifying their content?21:49
gsi @{u} just is the default when you omit the hash21:49
→ cdown joined21:51
→ gsgx joined21:51
← structoha left21:52
gsgx I often make the mistake of staging particular files or changes with `git add -p`, but then accidentally commit with `git commit -am "commit message"`, which commits all changes. Currently I just `git reset HEAD~1` and then repeat the process, but I'm wondering if there's any better way to get back to the state before committing, where only certain files were staged?21:54
← lucasta left21:56
→ av2156 joined21:56
BtbN There is reset --soft, but I don't think it does quite exactly that21:57
← av2156 left21:57
BtbN Since it will consider the files "-a" added as staged I'm pretty sure21:57
gsgx hmm, I thought `git reset` was the same as `git reset --soft`?21:57
→ lucasta joined21:57
BtbN no, the default is --mixed21:57
gsgx TIL, thanks21:58
← epony left21:58
→ epony joined21:59
← drbeco left22:00
← green_squirrel left22:01
BtbN I don't think there is a good solution to that issue. Other than somehow banning commit -a22:01
← rama left22:01
← mexen left22:01
→ rama joined22:02
→ piwu joined22:06
← lucasta left22:07
← trace1918 left22:08
→ Cleverness joined22:11
→ delay joined22:12
← piwu left22:12
→ roadie joined22:14
nedbat mcon: if you haven't pushed yet, then: git rebase -i $(git merge-base @ main) # I hope you aren't on windows22:14
← roadie left22:19
→ piwu joined22:19
← delay left22:20
→ delay joined22:20
← piwu left22:21
→ piwu joined22:21
← f451 left22:26
← NSOL left22:27
← rama left22:28
← piwu left22:29
→ rama joined22:29
→ piwu joined22:29
→ pyeverything joined22:30
→ roadie joined22:30
← roadie left22:35
← piwu left22:37
cbreak I'd recommend against using git commit -a, ever22:37
→ piwu joined22:38
cbreak also -m is bad, since it encourages bad commit messages22:38
← cdown left22:38
→ lucasta joined22:38
← Jong left22:38
← dionysus69 left22:40
EdFletcher `git commit -am '.'` -- my most-commonly executed git command. fortunately for me, my colleagues, and my dignity, `git rebase -i ...` is second on that list, so no one ever knows. oh no, except now you all do! 😆22:42
→ Jong joined22:43
→ Xenguy joined22:44
← rama left22:44
EdFletcher but in all seriousness, `-a` is actually quite dangerous and my muscle memory of it has gotten me into trouble many times. probably best to heed cbreak's advice and never start with it in the first place.22:44
→ rama joined22:45
EdFletcher but also: learn the rebase workflow. making many, many small commits can really save your bacon in those times where you're experimenting a lot. ctrl+z is no substitute.22:46
→ roadie joined22:46
← arch_angel left22:48
← roadie left22:51
mackerman I'm more offended by the use of rebase than -am, honestly.22:51
→ nate1 joined22:52
EdFletcher what the what?! the rebase workflow is ❤️22:53
mackerman The author of fossil can explain it better than I: https://fossil-scm.org/home/doc/trunk/www/rebaseharm.md22:54
Its not git, sure, but I think the opinion of a person whose tools cannot rebase is interesting.22:55
← lucasta left22:55
← theoceaniscool left22:56
cbreak rebase is quite essential in git22:56
→ lucasta joined22:56
cbreak it allows crafting a readable and explainable history, while still working with checkpoints22:56
← theoceaniscool_ left22:57
EdFletcher ^that's a bingo22:57
mackerman * essential to some people's git workflows22:57
← nate1 left22:57
cbreak not using rebase in git is like never deleting any character in a text editor: Silly. :)22:58
git is not only about recording history, it's about crafting history, about cooperating. Understanding this was one of the keys in understanding git for me.22:59
mackerman Its like needing to rewrite the draft of a book, and needing to burn all copies of version 1, without calling it version 2.22:59
cbreak with rebase, you can create a new draft of a book. Without, you're forced to write corrections into the margins or between the lines, or maybe chroma-multiplex fixes with red ink on top23:00
because erasing what is already there would be oh-so-terrible, wouldn't it?23:01
gsi when a book gets published it's called the first edition, right? regardless of how many iterations the author internally required23:01
← lucasta left23:01
cbreak git history is the product, not just a record of the path a developer took.23:01
mackerman git is every draft the editor saw, not just the end result.23:01
cbreak not really23:02
that's what the reflog is :)23:02
mackerman Reflog is the recycle bin before it gets emptied.23:02
→ lucasta joined23:03
mackerman If you can always rebase before other people get your history, sure it works without much problem.23:04
cbreak good git history tells a story, that of added features, fixed bugs, basically history. Bad git history is just a sequence of random snapshots during development, with shitty comments like "fixed bug" or "wip 3: feature x"23:04
good git history can be bisected automatically, and can be read to study how development progressed. Bad history might not even compile all the time.23:05
mackerman Maybe tools to see the big picture need to be better.23:05
cbreak The text editor I use shows for every line the commit it came from.23:06
it annoys me greatly if it tells me that it's from "fixed bug" commit.23:06
git history similarly useful as much documentation as the source code comments themselves. Or even more useful.23:07
well, anyway, obviously many people don't have the skill, motivation or even any reason to think about this as much as I do, but I recommend giving it a try :)23:08
mackerman Maybe it needs to tell you more about the branch that was merged in from.23:09
← sa0 left23:09
→ sa0 joined23:10
cbreak it does.23:11
You don't write merge commit messages?23:11
usually, when I merge feature branches, I give a short description of the overall feature23:12
that way, git log --first-parent shows an overview, and without it, I see all the details23:12
→ zumba_addict joined23:13
→ BrianBlaze420 joined23:13
← rama left23:14
→ rama joined23:15
← sa0 left23:16
← GNUmoon left23:16
→ GNUmoon joined23:17
gsi haven't read all of it yet, but cannot agree with the fossil presentation of why rebase is harmful23:20
can agree that rebase _can_ cause harm when misused, but when used properly it's very useful a feature23:21
and can agree that not everybody "needs" it in their daily routine, while others like it very much and use it often23:21
and no its main intention (as I see it) is not to fake a perfect creation that's never real, but instead to end up with something that's more maintainable23:22
→ dsrt^ joined23:25
← delay left23:30
gsi yep, that fossil page takes misuse for granted, concentrates on the excessive and misled forms, and completely ignores the useful use of rebase (AFAICS)23:37
so another ACK on misused rebase is harmful, that was never the question :)23:37
← lucasta left23:39
→ lucasta joined23:39
← IRChatter left23:44
← lucasta left23:46
→ lucasta joined23:46
← pikapika_lunar left23:49
→ pikapika joined23:50
← lucasta left23:50
→ lucasta joined23:54
← pyeverything left23:55

Logs Search ←Prev date Next date→ Channels Documentation