Perl 6 - the future is here, just unevenly distributed

IRC log for #git, 2017-09-14

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

All times shown according to UTC.

Time Nick Message
00:02 stutter joined #git
00:06 cagomez joined #git
00:06 kerrick joined #git
00:06 cbd_ joined #git
00:10 chrisshattuck joined #git
00:10 Cabanossi joined #git
00:11 justanotheruser joined #git
00:14 EricPoe joined #git
00:14 bronson joined #git
00:18 igemnace joined #git
00:19 kerrick joined #git
00:24 tangerinecandida joined #git
00:24 tangerinecandida t
00:27 tangerinecandida I’m learning about git. The documentation is thorough & I appreciate that. I’ve a question, though, regarding best practice. If a project runs on a remote device and is being debugged remotely (Eclipse + PyDev). The source code’s copy that’s being executed therefore resides on the device, not on the computer that’s doing the editing/rewriting/etc. Would it make more sense to keep a copy of the repo on
00:27 tangerinecandida the software designer’s computer and one on the device as well, and then synchronize them both with the repo on GitHub? … Or would it make more sense to have *one* pulled copy, keep it on the device, and edit it remotely on the designer’s computer?
00:29 tangerinecandida I’m leaning towards keeping one copy on the device and then committing changes *from* the device to the GitHub repo… but I’m accustomed to committing from within the Eclipse development environment. That’s not a technical reason to keep doing it. It’s merely what I’m accustomed to doing. :)
00:31 renarc joined #git
00:32 manuelschneid3r joined #git
00:33 vjacob joined #git
00:35 rewt tangerinecandida, i'd look at having the repo locally, and pushing the code to the device to test/debug
00:36 adgtl left #git
00:38 steeze joined #git
00:41 lxnr joined #git
00:43 edvb joined #git
00:44 durham joined #git
00:45 durham joined #git
00:47 durham joined #git
00:47 durham joined #git
00:49 xaa joined #git
00:53 thiago tangerinecandida: first of all, I wouldn't put the source code on the device
00:53 thiago that's not efficient
00:53 thiago use a remote debugger so that the source code resides on the developer's computer
00:53 thiago but regardless, sincronising via GitHub is slow and inefficient too. Just synchronise locally.
00:53 thiago I would also skip git and synchronise with rsync
01:01 bitmod join ##javascript
01:02 bremner no!
01:03 bremner just kidding, I'm sure they're lovely people.
01:04 dsop joined #git
01:06 cbd_ joined #git
01:06 finalbeta joined #git
01:07 mizu_no_oto joined #git
01:07 multi_io joined #git
01:07 rchavik joined #git
01:08 dsdeiz joined #git
01:08 dsdeiz joined #git
01:10 YuGiOhJCJ joined #git
01:11 Cabanossi joined #git
01:12 bambs joined #git
01:14 Nizumzen joined #git
01:15 justanotheruser joined #git
01:21 cagomez joined #git
01:21 leprechau joined #git
01:21 EvilPenguin joined #git
01:23 nd joined #git
01:26 DIDAVISION joined #git
01:28 Murr joined #git
01:29 fscala joined #git
01:29 dsdeiz joined #git
01:29 dsdeiz joined #git
01:30 czart__ joined #git
01:32 Solvaring joined #git
01:32 CussBot_ joined #git
01:36 sunri5e joined #git
01:36 cagomez joined #git
01:36 blackwell joined #git
01:37 Elon_Satoshi joined #git
01:37 cdg joined #git
01:38 chrisshattuck joined #git
01:39 chriys joined #git
01:39 chriys joined #git
01:39 pharma_joe joined #git
01:42 dirtyroshi joined #git
01:52 hash-z3r0 joined #git
01:52 duckydan joined #git
01:55 ilbot3 joined #git
01:55 Topic for #git is now Welcome to #git, the place for git help and ><(((('> | Public logs at https://gitirc.eu/log | First visit? Read: https://gitirc.eu | Current stable version: 2.14.1 | Getting "cannot send to channel"? /msg gitinfo .voice | Have you ever seen a git eat its own HEAD?
01:57 sleepingforest joined #git
01:59 holodoc joined #git
02:05 pharma_joe joined #git
02:06 msonntag joined #git
02:06 cbd_ joined #git
02:07 svm_invictvs joined #git
02:10 dsdeiz joined #git
02:10 Cabanossi joined #git
02:12 menip joined #git
02:13 petersaints joined #git
02:15 bronson joined #git
02:19 batrick joined #git
02:23 justan0theruser joined #git
02:23 justan0theruser joined #git
02:24 Orbixx[m] joined #git
02:27 hanthings_ joined #git
02:27 orbyt_ joined #git
02:28 cagedwisdom joined #git
02:28 tunekey joined #git
02:33 jstimm joined #git
02:34 nandi_ge___ joined #git
02:38 lxnr_ joined #git
02:39 Onionnion left #git
02:40 cdg joined #git
02:40 Cabanossi joined #git
02:41 CussBot joined #git
02:41 lagothrix joined #git
02:41 chachasmooth joined #git
02:43 Vortex35 joined #git
02:47 d^sh joined #git
02:50 a3Dman joined #git
02:51 EvilPenguin joined #git
02:51 Plouescat joined #git
02:51 CubiK joined #git
02:55 dre_ joined #git
03:03 svm_invictvs joined #git
03:04 pks joined #git
03:07 tangerinecandida thiago: Thank you.
03:07 cbd_ joined #git
03:08 tangerinecandida thiago: That’s good advice. I appreciate your time and your insight.
03:13 igemnace_ joined #git
03:16 kfoonamalik joined #git
03:19 pharma_joe joined #git
03:20 diqidoq_ joined #git
03:23 pR0Ps joined #git
03:27 mizu_no_oto joined #git
03:28 Murr joined #git
03:33 michelangelo joined #git
03:35 bongo joined #git
03:36 Joel joined #git
03:36 Joel somewhat unrelated to git, github, is showing two committers, both are me, I only want to show one, any ideas where it's pul;ling the second user from? https://github.com/joelcontrastthrowaway/ops-hire-project/commit/03f9e667070fa6677c0fe582745a10f8f2d9f58a
03:39 Cabanossi joined #git
03:39 Joel git log 03f9e667070fa6677c0fe582745a10f8f2d9f58a -n1 --dense --full-diff --format='%ae %ce' email@address.com joel@deport.me
03:39 Joel looks like I need to figure out how to change that
03:45 akushner___ joined #git
03:46 chrisshattuck joined #git
03:46 Joel managed to work it out
03:46 hash-z3r0 joined #git
03:47 Bock joined #git
03:50 mkali joined #git
03:51 zumba_addict joined #git
03:52 zumba_addict i'm working on an application. I have commits. I also ran `npm version patch' which created a different commit as shown by git log. Can I combine these 2 commits into 1 before I push my feature-branch?
03:55 Joel zumba_addict git rebase -i HEAD~<number of commits to go back>
03:55 zumba_addict right now, the last commit shown was done by `npm version patch`. It just updated the version number in package.json file. My commit was the one before it
03:55 Joel I ttypically do something like git rebease -i HEAD~2
03:55 Joel then you have to squash them.
03:56 zumba_addict ok. I haven't push this feature branch yet
03:56 Joel this is a common thing. google can provide plenty of help
03:56 zumba_addict yup, i'm reading this - https://stackoverflow.com/questions/12522565/how-can-i-combine-two-commits-into-one-commit
03:59 BSaboia joined #git
04:02 xaxxon_ joined #git
04:04 zumba_addict it worked! Thanks :)
04:05 zumba_addict i need to do it often so it becomes part of my memory
04:08 netj joined #git
04:08 cagedwisdom joined #git
04:09 Cabanossi joined #git
04:14 pharma_joe joined #git
04:15 pharma_joe joined #git
04:17 bronson joined #git
04:18 kfoonamalik joined #git
04:21 EvilPenguin joined #git
04:21 jkadaver joined #git
04:25 bayman joined #git
04:26 Elon_Satoshi joined #git
04:26 jimmyrcom_ joined #git
04:27 bayman I enabled pipelines in my bitbucket repo and it added a yml file to my source. i made some changes on my local source, exec git push and it says Updates were rejected because the remote contains work that you do not have locally.
04:27 bayman what do i do?
04:28 stutter joined #git
04:28 igemnace just fetch the new yml file
04:29 igemnace if that was the only change from bitbucket, a git pull should handle it cleanly
04:29 bayman ok just ran git fetch
04:29 igemnace you did? check git log HEAD..@{u}
04:29 bayman Your branch and 'origin/master' have diverged,
04:29 bayman and have 1 and 1 different commits each, respectively.
04:29 bayman (use "git pull" to merge the remote branch into yours)
04:30 igemnace can git merge, can git rebase
04:30 igemnace doesn't really matter where it goes into history, since it's just a config file for your pipelines
04:30 bayman git log HEAD ... ->  bitbucket-pipelines.yml edited online with Bitbucket
04:31 igemnace yup
04:31 igemnace run git merge to get that commit into your local master
04:31 atomi joined #git
04:32 bayman ok i think it's fixed
04:33 bayman thanks
04:36 matoro joined #git
04:39 Cabanossi joined #git
04:40 bayman i have another issue
04:41 bayman i made some source code changes on a server that and now when i git pull from repo, it says Please, commit your changes or stash them before you can merge.
04:42 lxnr_ joined #git
04:42 bayman i don't care about the changes i made and i want the repo source to overwrite the changes, what should i do?
04:42 igemnace that message hints at the solution: either git commit or git stash
04:43 ti2 A 'git reset --hard HEAD' throws away all local non-committed changes.
04:43 bayman do i git stash then git reset -hard HEAD   ?
04:43 igemnace one of the two
04:44 igemnace git stash keeps it for later, git reset --hard HEAD discards it completely
04:45 geoid joined #git
04:45 bayman thanks
04:46 rkta_ joined #git
04:49 Neico joined #git
04:49 Nizumzen joined #git
04:51 jkadaver joined #git
04:51 abithvijayan[m] joined #git
04:51 dikiaap joined #git
04:52 Nizumzen joined #git
04:52 rcn joined #git
04:53 AnujSharma joined #git
04:56 cdg joined #git
04:57 _28_ria joined #git
05:00 atomi joined #git
05:06 _ng joined #git
05:06 cbd_ joined #git
05:07 dec0n joined #git
05:07 doppo joined #git
05:07 boombatower joined #git
05:08 xkr47 joined #git
05:09 Cabanossi joined #git
05:14 atesc joined #git
05:14 xPucTu4 joined #git
05:14 czart_ joined #git
05:16 fatihgenc joined #git
05:16 freimatz joined #git
05:17 roygbiv joined #git
05:18 imack joined #git
05:18 rchavik joined #git
05:21 cagomez joined #git
05:21 ZyX-I joined #git
05:23 MrMojit0 joined #git
05:23 boombatower joined #git
05:24 thierryp joined #git
05:24 pharma_joe joined #git
05:25 dreiss joined #git
05:27 armenb joined #git
05:28 Puffball joined #git
05:28 dzho joined #git
05:28 dsnow joined #git
05:28 tsunamie joined #git
05:28 jkadaver joined #git
05:28 shakalaka joined #git
05:29 vdl joined #git
05:29 durham joined #git
05:29 howdoi joined #git
05:31 jkadaver_ joined #git
05:31 ljc joined #git
05:35 thierryp joined #git
05:39 parsnip hmm, after i git submodule add mathjax --depth 1, make starts running at 99% CPU
05:40 parsnip hmm, it's ruby fsevent_watch
05:40 lss8_ joined #git
05:41 parsnip oh, it's the `jekyll serve` i left running
05:42 roelmonnens joined #git
05:43 lxnr_ joined #git
05:45 zefferno joined #git
05:49 Raging_Hog joined #git
05:49 freimatz joined #git
05:51 thierryp joined #git
05:51 EvilPenguin joined #git
05:51 _28_ria joined #git
05:53 atesc joined #git
05:56 dege joined #git
05:56 Kavothe joined #git
05:58 qwedfg joined #git
05:58 JeroenT joined #git
05:58 MACscr joined #git
05:59 _ng joined #git
05:59 Oatmeal joined #git
05:59 neil_cn joined #git
06:02 JeroenT joined #git
06:03 thierryp joined #git
06:05 imack joined #git
06:05 cdown joined #git
06:06 cbd_ joined #git
06:07 high_fiver joined #git
06:07 govg joined #git
06:07 tarkus joined #git
06:10 ppang joined #git
06:11 LeBlaaanc joined #git
06:12 ronskisika joined #git
06:14 JeroenT joined #git
06:17 pharma_joe joined #git
06:18 bronson joined #git
06:19 osse When I run git fsck I see a lot of "danglnig tag", but when I run cat-file -t on them a bunch of them are blobs, trees and commits. How is that possible?
06:22 roelmonnens joined #git
06:24 bayman joined #git
06:24 horpoppy[m] joined #git
06:27 thierryp joined #git
06:30 ash_workz joined #git
06:31 parsnip because git fsck sees all objects, and those are all objects?
06:31 parsnip "Verifies the connectivity and validity of the objects in the database."
06:31 jknetl joined #git
06:32 overlord_tm joined #git
06:33 osse But why does it think a commit is a tag ?
06:34 parsnip what are you seeing?
06:35 parsnip strange, i don't think a tag is an object, so i'm wondering how to interpret the manual
06:35 MrMojit0 left #git
06:36 parsnip i guess git fsck can additionally show tags, but i think that's a ref not an object, and not a commit.
06:37 \drew joined #git
06:39 tkore joined #git
06:39 tkore joined #git
06:40 abba joined #git
06:40 AnujSharma joined #git
06:40 Solvaring joined #git
06:43 navidr joined #git
06:43 lxnr_ joined #git
06:44 Bock joined #git
06:44 osse There is an object type called tag. One issue created when you do git tag -a
06:44 osse s/issue/is/
06:50 jsears joined #git
06:51 Commander_Cody joined #git
06:52 parsnip ah, tags are objects
06:52 ShalokShalom joined #git
06:52 Commander_Cody Upon request, I'd like to create one patch file via `git format-patch` with several commits squashed, not losing their commit messages.
06:52 cdown joined #git
06:53 Commander_Cody How can I accomplish this without soiling my commit history by *actually* squashing all those commits? I just want a patch file that looks as though I did that.
06:54 parsnip you could make a new branch ref, check that out, and squash that
06:54 parsnip that will leave the old branch ref still pointing at original commit, and thus the sequence.
06:56 Ratler joined #git
06:56 _28_ria joined #git
06:57 parsnip osse: okay, but why do you say "23:33 <osse> But why does it think a commit is a tag ?", care to share?
06:58 Thunderfunk joined #git
07:01 parsnip osse: this might be interesting to you at the moment http://eagain.net/articles/git-for-computer-scientists/
07:01 parsnip they do say tags are a "node", i'm not sure wht
07:01 parsnip *why
07:01 TomyWork joined #git
07:02 parsnip maybe in the sense that fsck sees them as objects?
07:03 Soni joined #git
07:06 cbd_ joined #git
07:06 Commander_Cody parsnip: Your suggestion worked fine, thanks.
07:07 luc4 joined #git
07:07 parsnip Commander_Cody: nice, i thought of it, wasn't sure if it was the right approach, but googling suggested it was the popular approach.
07:09 wandering_vagran joined #git
07:09 Cabanossi joined #git
07:12 juniper joined #git
07:12 jast tags are two things, kind of. generally, a tag *name* is a ref (like a branch name). the ref can point directly to a commit (which makes it a non-annotated tag) or a tag *object* (which points to another object, usually a commit) and has some metadata. tags involving tag objects are called annotated tags.
07:12 jast 'node' is not official terminology
07:13 Tobbi joined #git
07:14 EvilPenguin joined #git
07:14 parsnip it's as if he thought the hashed part of a tag was more like a commit object than a tree object is.
07:15 parsnip perhaps if we assume he was narrowing his focus to commits and refs, than sure.
07:15 parsnip which is not an impractical thing to do.
07:15 Commander_Cody joined #git
07:16 parsnip oh, can non-annotated tags and annotated tags be moved to other commits, say upon rebasing for example?
07:16 rscata joined #git
07:17 dsdeiz joined #git
07:17 iambiglooser2022 joined #git
07:17 jast tags are never changed automatically
07:18 jast filter-branch has an option to rewrite tags... rebase doesn't
07:18 jast osse: are there any replaced objects? that's the only potentially relevant thing I can see in the code at a glance
07:18 parsnip if annotated tags are hashed, than you can't move them, perhaps that's what the author was getting at.
07:19 jast not sure what you mean by 'hashed'
07:19 parsnip oh, like how commits, trees and blobs are hashed
07:19 jast tag objects are content-addressed like all other objects, so if you change where a tag object points to, the tag's SHA1 changes, too
07:19 iambiglooser2022 joined #git
07:19 theoceaniscool joined #git
07:20 valhalla_22 joined #git
07:20 valhalla_22 joined #git
07:21 valhalla_22 I was updating local dev branch with respect to master but i found i cam across some glitch , which is stopping the updation of branch , I want to understand why this is happening and how can i resolve it . here is the paste
07:21 valhalla_22 https://paste.pound-python.org/show/1KtDReOUZElstPZA1tMf/
07:22 jceb joined #git
07:25 parsnip valhalla_22: open those files in your text editor, they will have some funny lines indicating where the merge conflic occurred.
07:25 valhalla_22 yeah
07:25 valhalla_22 they had
07:26 parsnip valhalla_22: do you know what happens after that?
07:26 overlord_tm joined #git
07:26 dka joined #git
07:26 valhalla_22 not exactly but some has to be removed and some has to be left
07:26 cagomez joined #git
07:28 Murr joined #git
07:29 ExeciN joined #git
07:31 parsnip valhalla_22: you consider the merge conflict, that is, each branch has two stories on what should be there. the merge needs a decision.
07:31 valhalla_22 hmm i want to forgo the local one
07:31 parsnip valhalla_22: once you edit those lines to indicate the new end result, you remove the special markup.
07:32 valhalla_22 i just want to get updated with master one ,
07:32 parsnip ah, you could probably reset.
07:32 valhalla_22 how
07:32 parsnip hmm
07:33 parsnip um, checkout your local branch
07:33 valhalla_22 ok done
07:33 parsnip then `git reset origin/<branch>`
07:33 valhalla_22 what will this do
07:33 parsnip valhalla_22: you might be curious what reset does.
07:33 parsnip :)
07:33 valhalla_22 no
07:34 parsnip ?
07:34 valhalla_22 it reset the pointer to HEAD
07:34 parsnip you're telling me? or asking me?
07:34 valhalla_22 i have used this a lot git reset --hard
07:34 valhalla_22 but never used git reset origin/master
07:34 parsnip valhalla_22: lets focus on `git reset --soft`, it's easier to explain this way.
07:35 valhalla_22 ok
07:35 parsnip the sequence is `--soft, --mixed, --hard`
07:35 parsnip the default is --mixed
07:35 valhalla_22 hmm so this is the philosphy
07:35 valhalla_22 ok
07:35 valhalla_22 i get it
07:35 parsnip they all do what --soft does, plus a little more
07:35 parsnip so the real idea is in --soft
07:36 parsnip `git reset --soft` moves a branch ref.
07:36 parsnip that is, it moves the branch ref that you are checked out on.
07:36 parsnip so if you are checked out on local branch, it will move that ref.
07:37 parsnip valhalla_22: does that make sense, then i say what --mixed and --hard do in addition?
07:37 valhalla_22 hmm little bit
07:37 parsnip so if you don't pass it another commit as an argument, the ref doesn't actually move.
07:38 parsnip so, now on to --mixed and --hard
07:38 valhalla_22 k
07:39 parsnip --mixed overwrites the index with whatever is in the commit we are at now
07:39 parsnip --hard additionally overwrites the working directory.
07:40 valhalla_22 oops   but for my perscpectve ,git reset --hard works fine , is that ok with it .
07:40 parsnip valhalla_22: you could, no harm in it i think
07:40 parsnip valhalla_22: you could always git stash if you're nervous
07:41 sarri joined #git
07:41 sarri joined #git
07:41 parsnip git stash just puts current status of working directory and index into a special kind of commit in a special "stash" storage.
07:41 parsnip !lol
07:41 gitinfo A nifty view of branches, tags, and other refs: git log --oneline --graph --decorate --all
07:41 valhalla_22 yeah , parsnip that sound great but how to overcome my problem
07:42 parsnip valhalla_22: do you know git pull is really just git fetch and git merge?
07:42 valhalla_22 yes
07:42 valhalla_22 it fectes and do the merge
07:42 Noodlewitt joined #git
07:42 parsnip okay, so git fetch is done, and you had problem at git merge
07:42 valhalla_22 yeah
07:43 parsnip i think you can undo the partial merge by doing `git reset --hard` on your local branch
07:43 parsnip i think that's still not quite what you want, but it won't hurt anything
07:43 valhalla_22 i did
07:43 parsnip `git reset --hard origin/<branch>`, no?
07:44 parsnip that will move the local ref
07:44 valhalla_22 git reset --hard commithash
07:44 parsnip `git reset --hard origin/commithash`, no?
07:44 valhalla_22 till here i had beautiful memory
07:44 valhalla_22 after that things got messy
07:44 parsnip i don't understand that.
07:45 lxnr_ joined #git
07:45 valhalla_22 git reset --hard 0553ac2
07:45 parsnip i'm lost
07:46 valhalla_22 why
07:46 ahr3n joined #git
07:46 valhalla_22 its just additional git reset -hard with specific cmmit number
07:46 valhalla_22 parsnip:  i think you told me solution but i have fear to implement it
07:46 parsnip :)
07:47 parsnip valhalla_22: i have an idea
07:47 valhalla_22 git reset --hard origin/dev_branch  what it will do
07:49 igemnace what was the original problem again? wasn't it just a merge conflict?
07:49 parsnip valhalla_22: it will move the branch ref
07:50 parsnip valhalla_22: if you replace --hard with --soft, that's _all_ it will do.
07:50 valhalla_22 i wont be loosing any code
07:50 valhalla_22 na
07:50 valhalla_22 i can loose this one but not 6 mnths back
07:50 valhalla_22 one
07:51 parsnip valhalla_22: here's an idea to make possible issues less difficult:
07:51 valhalla_22 ok
07:51 parsnip valhalla_22: create a temporary branch name for where local branch ref is now.
07:51 valhalla_22 ok
07:52 parsnip valhalla_22: that way, when you move your local branch ref to remote branch HEAD, you can still see full commit graph easily with `git lol`.
07:52 Tobbi joined #git
07:52 valhalla_22 ok
07:52 parsnip you can _see_ what happened, and know if you're comfortable with removing the temporary branch name.
07:53 valhalla_22 tnks but i have got the idea
07:53 valhalla_22 git is all about trees and branch
07:53 valhalla_22 so  i want to do this
07:55 Thunderfunk joined #git
07:59 parsnip valhalla_22: yeah, to me, most of my thinking is based on the commit graph, and refs, and how commits aren't really deleted (at least not immediately), you can find them if you move a branch ref. if a commit isn't pointed to by ref or a descendant that has a ref, it won't appear in git log, but you can still find it other ways.
08:00 parsnip refs move around, but commits don't really.
08:00 parsnip the diffs sort of do.
08:00 valhalla_22 ok
08:00 parsnip :)
08:00 EvilDMP joined #git
08:00 valhalla_22 i am newbie to git
08:00 valhalla_22 i went to tour on himalya for 3 months
08:00 noyb joined #git
08:00 valhalla_22 so i forgot many things  :-(
08:01 mostlybadfly joined #git
08:02 Thunderfunk joined #git
08:02 pinux joined #git
08:02 EvilDMP Is there some way to break apart a commit into two commits? I submitted a PR containing some new material and some refactoring, and I have been asked to submit this as two commits for easier reviewing.
08:03 selckin rebase -i & edit, or git reset HEAD^, to remove the commit and then you can git add -p parts
08:03 jast EvilDMP: so I'm guessing it's the topmost commit on that branch?
08:03 EvilDMP jast: yes
08:03 King_Hual joined #git
08:04 EvilDMP git add -p is new to me, I will investigate
08:04 jast yeah, I'd go with 'git reset HEAD^' and stage the two sets of changes separately with add -p
08:04 EvilDMP Thanks jast selckin!
08:04 jast add -p is one of the features I use pretty much every day :)
08:05 jast I actually use 'git commit -p' most of the time but it amounts to the same thing
08:07 osse jast: yes, I made some replacements. that could be it
08:07 osse manually stitching together something resembling a history from zip archives \o/
08:08 Solvaring joined #git
08:08 jast the type output does honour replace, the object name being output is the original
08:10 acidjnk joined #git
08:10 CheckDavid joined #git
08:13 Murii joined #git
08:14 jameser_ joined #git
08:15 Murii joined #git
08:15 EvilPenguin joined #git
08:15 LeBlaaanc joined #git
08:17 lxnr_ joined #git
08:19 bronson joined #git
08:19 Nizumzen joined #git
08:22 fatalhalt joined #git
08:25 Cabanossi joined #git
08:27 schleppel joined #git
08:28 Tobbi joined #git
08:29 frsk joined #git
08:31 Borg- joined #git
08:33 ertes-w joined #git
08:41 dimi1947 joined #git
08:43 jlebrech joined #git
08:43 jlebrech morning all
08:43 jlebrech can I ignore my dist/ folder for every user except for the CI server?
08:44 diogenese joined #git
08:45 kadoban jlebrech: What would that mean?
08:45 kadoban In git, ignore rules only have any effect on untracked content
08:45 jlebrech https://help.github.com/articles/ignoring-files/ like the 3rd section
08:46 kadoban That is a place you can put rules, in .git/info/exclude, but unless I'm misunderstanding the question, I'm having a hard time imagining how it'll be useful?
08:47 jlebrech we seem to be generating slightly different compiled files, because we have windows and mac. and it should just look the same. and only want the CI server to compile the dist folder
08:47 jlebrech we're always getting merge conflict for nothing
08:48 vjacob joined #git
08:49 pbandark joined #git
08:49 _ADN_ joined #git
08:50 jlebrech I want devs to have that folder ignored, but it to be generate by CI. but it don't think it's an issue actually, I could configure CI to compile and scp, so from the CI end it won't be using git
08:50 thierryp joined #git
08:51 thierryp_ joined #git
08:51 igemnace jlebrech: you *will* have to do that
08:52 kadoban You usually shouldn't be tracking generated files, so any way you can avoid that would be good.
08:52 igemnace jsyk you can actually just git add --force anything regardless of .gitignore
08:52 jlebrech oh, alright. seemed like a sound plan
08:52 igemnace but once you track the thing, changes will be tracked regardless of .gitignore
08:53 kadoban For pretty much the reason you're seeing, merges get weird and it'll get randomly out of sync from the codebase and etc.
08:53 jlebrech so should I remove dist first?
08:53 fatihgenc joined #git
08:54 igemnace well, yeah. much easier for everyone to just remove dist from the repo and ignore it moving forward
08:55 jlebrech ah yeah
08:55 igemnace have the CI server deploy dist/ not using git, because it'll ruin the idea of ignoring it
08:56 jlebrech haha of course, I remembered at the last minute git isn't use to deploy :D
08:56 cdg joined #git
08:56 FilipNortic joined #git
08:57 dre_ joined #git
08:57 tvw joined #git
08:57 fatihgenc joined #git
08:59 qwedfg joined #git
09:01 cstratak joined #git
09:02 venmx joined #git
09:02 venmx_ joined #git
09:06 cy1 joined #git
09:06 Chuguniy joined #git
09:06 Kavothe joined #git
09:10 Cabanossi joined #git
09:11 gxt joined #git
09:11 chll_ joined #git
09:12 achadwick joined #git
09:12 _ADN_ joined #git
09:13 cy1 I go "git log -p" and it skips showing some patches. Like when I do "git show SHA" it shows a patch (a bad merge commit of a bad merge commit), but when I do "git log -p" and go to that SHA, it says no files modified, no patch...
09:15 venmx joined #git
09:15 venmx_ joined #git
09:15 jast cy1: as briefly mentioned in the manpage, 'git show' defaults to using --cc which affects how merge commits are displayed. by default (with -p which doesn't automatically imply -c or --cc), merges don't display a diff.
09:16 cy1 Oh, I thought --cc  was more terse, though?
09:16 cy1 Am I getting that backwards?
09:16 osse jast: goddammit
09:17 osse git fsck | awk '$2 = "tag" {print $3}'
09:17 osse git fsck | awk '$2 == "tag" {print $3}'
09:17 osse (also parsnip)
09:17 dsdeiz joined #git
09:17 jast hah
09:18 lxnr_ joined #git
09:18 cy1 well, adding --cc seems to show the patches in git log. Thank you!
09:18 jast cy1: terser than -c, not terser than the default (nothing) :)
09:19 wxg joined #git
09:21 ahr3n joined #git
09:22 hanthings_ joined #git
09:22 JeroenT joined #git
09:25 lxnr_ joined #git
09:25 imack joined #git
09:28 Murr joined #git
09:29 kurkale6ka joined #git
09:29 gusnan joined #git
09:29 gusnan joined #git
09:30 bergice_ joined #git
09:31 belmoussaoui_ joined #git
09:32 freimatz3 joined #git
09:32 czart__ joined #git
09:33 magicturtle joined #git
09:34 mkali joined #git
09:35 Floflobel joined #git
09:36 cagedwisdom joined #git
09:41 kurkale6ka joined #git
09:42 cagomez joined #git
09:43 freimatz joined #git
09:43 AnujSharma joined #git
09:44 ArcaTech joined #git
09:45 EvilPenguin joined #git
09:46 [worksti] left #git
09:47 wxg joined #git
09:48 enkrypt joined #git
09:49 hash-z3r0 joined #git
09:52 freimatz3 joined #git
09:58 cagomez joined #git
10:04 cstratak joined #git
10:05 bybb joined #git
10:06 Richard_Cavell Hello all.  I have a git repo.  I see that I can get a commit ID from git log.  I want to see all the changes made to a file between one commit and another commit.  How do I do that?
10:06 jast git diff commit1 commit2 -- file
10:07 Richard_Cavell And in git log where it says "commit d462e84181c534d73658b367f872628f7dc36994
10:07 Richard_Cavell " is that the string that I'm looking for?
10:07 jast or as separate diffs for each commit in the range: git log -p commit1..commit2 -- file
10:08 jast that's the commit ID, yes. you can often shorten it to a few characters (e.g. d462e) and get the same result
10:08 jast depends on how big your repo is, a prefix of eight characters will work virtually all the time and 4-5 characters are typically fine for smaller repos
10:08 rchavik joined #git
10:12 manuelschneid3r joined #git
10:15 mentazoom joined #git
10:19 someone235 joined #git
10:19 sarri joined #git
10:19 sarri joined #git
10:20 bronson joined #git
10:21 someone235 Hi, I did a "git clone git@github.com:bitcoin/bitcoin.git", but I want to switch to branch 0.10. I did "git branch -a", but I can't see it, although it appears in the branch list here: https://github.com/bitcoin/bitcoin/
10:21 someone235 someone knows what am I doing wrong?
10:23 maxb Having cloned, it'll be there as "remotes/origin/0.10"
10:25 Cabanossi joined #git
10:25 someone235 maxb, "git checkout remotes/origin/0.10" doesn't work
10:26 Nizumzen joined #git
10:26 ShalokShalom_ joined #git
10:26 maxb "doesn't work" is not very descriptive, if you give the exact error message, it would make it easier to help you
10:27 freimatz joined #git
10:29 someone235 maxb, error: pathspec 'remotes/origin/0.10' did not match any file(s) known to git.
10:31 HolodocPrime joined #git
10:32 maxb I'm not sure what you're doing then, because running the exact commands you've given above in double-quotes works for me
10:33 ljc joined #git
10:33 venmx joined #git
10:34 venmx_ joined #git
10:34 enkrypt joined #git
10:34 jast try a 'git fetch', maybe the clone was incomplete for some weird reason
10:35 j08nY joined #git
10:35 JeroenT joined #git
10:36 Oatmeal joined #git
10:38 Anja_ joined #git
10:39 bitmod joined #git
10:39 thinkpad joined #git
10:42 Anticom joined #git
10:42 osse someone235: what does 'git config remote.origin.fetch' print ?
10:43 someone235 osse, +refs/heads/master:refs/remotes/origin/master
10:44 osse someone235: run   git config remote.origin.fetch "+refs/heads/*:refs/remotes/origin/*"
10:44 osse then git fetch. now you'll see it
10:46 bybb joined #git
10:47 Elon_Satoshi joined #git
10:49 rnsanchez joined #git
10:51 bomb-on joined #git
10:51 Anticom joined #git
10:51 jast the situation looks as though --single-branch was passed when cloning
10:54 osse I concur with this assessment
10:54 osse Furthermore, Carthage must burn.
10:55 someone235 thx
10:56 venmx joined #git
10:56 venmx_ joined #git
11:00 Celelibi Is there an efficient way to version large binary objects and tie them to commits?
11:00 cagomez joined #git
11:00 osse git-annex, git-lfs, git-slave
11:00 Celelibi Thanks, I'll look at that.
11:00 Celelibi I tried to just commit my sqlite database, but some simple git commands take very long.
11:04 jast !annex
11:04 gitinfo git-annex, git-media, and Git LFS are some solutions to the !binary problem. They work by keeping the blobs outside of the repo, storing a reference to the blob in the repo instead. http://git-annex.branchable.com https://github.com/alebedev/git-media http://git.io/git-lfs
11:04 jast gitslave is for managing groups of repos, not related to this IMO
11:07 SteffanW joined #git
11:10 Cabanossi joined #git
11:14 osse Ah, my bad
11:15 EvilPenguin joined #git
11:16 cagomez joined #git
11:17 kurkale61 joined #git
11:18 dsdeiz joined #git
11:18 thierryp joined #git
11:21 theoceaniscool joined #git
11:23 Chuguniy joined #git
11:24 arbes joined #git
11:25 arbes left #git
11:25 adioe3 joined #git
11:25 adioe3 hi folks
11:25 adioe3 somebody seems to have mucked up my repo, when i clone it i get:
11:25 adioe3 fatal: Invalid revision range 0000000000000000000000000000000000000000..53441323826d5d716274f6a79063e4b2081e952a
11:26 adioe3 any idea how to fix that or what the problem is?
11:26 watabou joined #git
11:26 clmsy joined #git
11:28 kurkale6ka joined #git
11:28 Murr joined #git
11:30 Soni joined #git
11:31 JeroenT joined #git
11:33 Darcidride joined #git
11:33 jmccarthy joined #git
11:36 jmccarthy If I have a list of old commits (non-contiguous) which are missing a Change-Id - what is a handy way to script running the commit hook against them to add one in ? Can this be done in some way that doesn't require a force push ?
11:38 bybb joined #git
11:38 grawity no, it cannot
11:38 sekmo joined #git
11:40 jmccarthy May with some sort of rebase magic ? no ?
11:40 grawity no
11:40 grawity changing old commits requires a force push, end of story
11:42 ppang joined #git
11:42 jmccarthy :/
11:43 glad joined #git
11:45 peff joined #git
11:46 Chuguniy joined #git
11:46 bayman joined #git
11:46 jmccarthy Is filter-branch really that bad ? It says 'don't use it' - but in a test I did, I was able to doctor up the commits just fine (in my local workspace)
11:47 sonOfRa You will still need to force-push
11:49 jmccarthy Ok, yep I have read that in many places - just checking :) ! Or if maybe there is a preferred way besides filter-branch ?
11:50 dan2k3k4 joined #git
11:51 ShalokShalom joined #git
11:52 jast not for the kind of thing you're trying to do... and even if there were, it would still be impossible to do it without a force push
11:52 jmccarthy Ok, I believe ye :)
11:54 jmccarthy Thanks :) !
11:58 dave0x6d joined #git
12:02 synthroid joined #git
12:03 jmccarthy Follow up question: How does this differ from 'reword', if I do interactively, I think I can add Change-Ids and push up a review ok, is that right ?
12:03 jast sure... but if you pushed the original commit earlier, it requires a force push
12:04 jast the main difference between filter-branch and rebase is that rebase works on patches, i.e. when re-creating the commit it re-applies the corresponding patch
12:04 jast when all you're doing is changing the commit message, that difference is mostly irrelevant
12:05 str joined #git
12:06 sz0 joined #git
12:09 dzho joined #git
12:10 Cabanossi joined #git
12:11 JeroenT joined #git
12:13 Gurkenglas joined #git
12:13 Acerific joined #git
12:13 Acerific joined #git
12:16 epta joined #git
12:16 sucode_ joined #git
12:17 sucode_ Hi! everyone, I want to apply for co-mentor in the outreachy program.
12:17 jceb joined #git
12:18 epta Why do git ask for password when doing 'git clone https://github.com/a/b' even if the repo is public?
12:19 tobiasvl epta: are you sure you're using the HTTPS URL for cloning?
12:19 kurkale61 joined #git
12:19 tobiasvl epta: if you're using an SSH URL instead (or git@github.com with no protocol in front) it probably doesn't ask for a password, but your SSH key's passphrase?
12:20 dzho epta: I often encounter that error when I have a typo in my URL. It can't find the project, and so then must authenticate me to check to see if a private repo of that name exists.
12:20 jast sucode_: probably a typo in the URL. on github, trying to clone a non-existing repo results in a password prompt
12:20 jast whoops
12:21 jast epta: probably a typo in the URL. on github, trying to clone a non-existing repo results in a password prompt
12:21 tobiasvl aah, good call.
12:21 bronson joined #git
12:22 epta jast: whoa, it was a typo, thanks
12:22 jast happens to the best of us :)
12:22 YuGiOhJCJ joined #git
12:22 epta left #git
12:24 Paker joined #git
12:25 plexigras joined #git
12:26 bongo joined #git
12:26 kurkale6ka joined #git
12:27 noecc joined #git
12:29 plexigras how can i set the default merge message?
12:30 jeffreylevesque joined #git
12:30 jkadaver joined #git
12:31 Led_Zeppelin left #git
12:35 hash-z3r0 joined #git
12:35 cstratak joined #git
12:36 southernirc joined #git
12:36 lempface joined #git
12:37 jast plexigras: I think the only way to override it is the '-m' command line option
12:38 jast which I guess will skip the message editor, too...
12:38 a3Dman joined #git
12:38 lempface So I've had this repository for a couple years, and no issues. I created a feature branch and made one commit just fine. I've made a bunch more changes and want to start adding commits but now when I try to git add any files I get an error: "fatal: internal error: ce_mode is 0" I can't seem to find a whole lot related to this error. Any ideas? I'm a little nervous my repo is borked!
12:38 ShalokShalom joined #git
12:40 kurkale6ka joined #git
12:41 up_the_irons joined #git
12:41 jast lempface: your index may be corrupt, try 'git reset' -- this will forget which changes you have staged
12:41 Fissure joined #git
12:41 lempface any risk of losing my work with that?
12:41 lempface I'll back it up first
12:42 sucode_ joined #git
12:42 jast it affects only git's knowledge of what is staged and what isn't. commits and your working tree are left alone if you don't add any more options
12:42 jast but a backup never hurts :)
12:42 MoaMoaK left #git
12:43 ShalokShalom joined #git
12:43 lempface @jast - much love friend, that fixed it WHEW!
12:43 jast great
12:44 MattMaker joined #git
12:45 MattMaker joined #git
12:45 EvilPenguin joined #git
12:46 MattMaker joined #git
12:50 plexigras i want the default merege message of branch `hotfix/something` to be `fix: something` and of `feature/something`, `feat: something`
12:50 mrkake joined #git
12:51 jlebrech how do i checkout a branch of the same project into a submodule?
12:52 diqidoq joined #git
12:55 rawste joined #git
12:56 mkali joined #git
12:58 renarcii joined #git
12:58 bruce_lee joined #git
12:58 bruce_lee joined #git
12:59 mrkake joined #git
13:01 pinux joined #git
13:04 lempface @jast - After I committed my changes, somehow one of my submodules was deleted. Best route to fix?
13:04 lempface should just reinit be fine?
13:05 renarc joined #git
13:06 jast huh. worth a try, certainly...
13:06 lempface I have no idea how it became all wonky, I didn't do anything out of the ordinary and I'm the sole author on this project
13:08 lempface Reinit seems to have brought it back but it is set to HEAD instead of master. Still fairly new to submodules so maybe this is expected
13:09 lempface Nevermind, that is a tag gitkraken auto created
13:09 Cabanossi joined #git
13:12 str joined #git
13:12 chriys joined #git
13:14 murii joined #git
13:15 cdg joined #git
13:16 igemnace joined #git
13:18 dsdeiz joined #git
13:18 dsdeiz joined #git
13:21 FatKiwi joined #git
13:21 cdg_ joined #git
13:22 gugah joined #git
13:25 jeffreylevesque_ joined #git
13:28 Murr joined #git
13:29 Anticom joined #git
13:29 Tobbi joined #git
13:31 govg joined #git
13:31 str joined #git
13:33 plexigras how do i change the default merge message
13:34 Cavallari joined #git
13:35 tinanoo joined #git
13:35 Es0teric joined #git
13:36 Anja_ joined #git
13:36 mescalinum joined #git
13:37 bybb_ joined #git
13:38 peacememories joined #git
13:38 jast plexigras: I've already answered...
13:39 jast about an hour ago
13:39 Cabanossi joined #git
13:40 osse prepare-commit-msg would do the trick
13:40 plexigras no -m changes the merege messeage i want to change the default merge message
13:40 plexigras ok i will take a look at that
13:41 Tobbi joined #git
13:42 ahr3n joined #git
13:44 bitonic joined #git
13:45 jast man githooks for reference
13:45 gitinfo the githooks manpage is available at https://gitirc.eu/githooks.html
13:45 bitonic is it possible to check if there are any conflicts between two branches without attempting a merge?
13:45 bitonic i need this to perform some sanity checks in a ci script
13:45 jast bitonic: not really
13:45 bitonic jast: that's a bummer
13:45 jast about the best you can do is redirect the index and working tree to a temporary location to try the merge
13:46 osse how about checking if merge-tree prints anythign
13:46 nettoweb__ joined #git
13:46 osse (haven't done this myself. it's just as much a question as an answer)
13:47 jast I think merge-tree doesn't run the whole gamut of automatic merging
13:47 MattMaker joined #git
13:47 jast it does a trivial merge, meaning it doesn't perform automatic merging within a file
13:47 lebwoski joined #git
13:47 osse ah
13:48 balance joined #git
13:48 balance hi
13:48 jast but doing this and then trying manual merges on individual stages in the index would remove the requirement of creating a complete temporary working tree
13:48 lebwoski joined #git
13:49 bayman joined #git
13:49 jast so I guess you can do a decently efficient job simulating a merge but it takes a fair bit of familiarity with the concepts
13:49 ahr3n joined #git
13:49 balance How come that git pull origin master doesn't actually "update" a file? I cloned the repo, made a change to file X but didn't commit it and then just wanted to get to the newest state by doing pull origin master but the file keeps the same.
13:49 osse balance: because if it did then it would wipe your uncommitted changes
13:50 osse git doesn't like that
13:50 jast 'git pull' is for fetching updates, it doesn't overwrite your changes because it considers them valuable
13:50 jast if you want to get rid of uncommitted changes, check the output of 'git status' for useful hints
13:50 tobiasvl balance: git should have told you this when pulling
13:50 rewt it should tell you what it's doing ... yes, it's a lot of text, but it has important info
13:50 jast or use 'git reset --hard' for tabula rasa
13:50 balance osse so what happens in this case and why do I get "already up to date"?
13:51 jast because 'git pull' didn't have any updates to give you
13:51 rewt what does `git status` say?
13:51 jast you're already up-to-date, you just have some uncommitted changes
13:51 jast pulling never means "throw away things I did", it means "combine my things with upstream things"
13:52 robotroll joined #git
13:52 balance jast osse tobiasvl ok thanks - gonna reread some tutorial about it. status shows the uncommited changes.
13:53 AaronMT joined #git
13:53 rewt pull = fetch + merge
13:53 jast yeah, and it also shows you how to get rid of your changes :)
13:53 JeroenT_ joined #git
13:54 balance jast yeah, solved it. thanks!
13:55 lebwoski Hi, I want to rebase a fork to the original vanilla repository, but I only have a snapshot(not git files). I tried to create a repository with git init and rebase with adding a new remote (upstream) and 'git rebase' but I get conflicts with two files 'both added' by both repositories. Is there a way around patching the files manually? Thx!
13:55 bitmod joined #git
13:56 schasi joined #git
13:57 Guest16340 joined #git
13:57 seds joined #git
13:59 JeroenT joined #git
13:59 schasi Can anyone tell me why there is no "force" switch to clone into an non-empty directory?
13:59 zefferno joined #git
14:00 osse My guess is that no one has bothered making it
14:01 schasi Hm. But the bothered not allowing cloning into non-empty directories.
14:01 seds joined #git
14:01 osse because that's easier to implement :p
14:02 osse git init && git remote add origin .... && git fetch origin && git checkout -b master origin/master
14:02 osse there. dirty clone
14:02 schasi Heh. Well thanks X)
14:03 seds joined #git
14:04 NeverTired joined #git
14:04 snowkidind joined #git
14:06 manuelschneid3r joined #git
14:06 freimatz3 joined #git
14:08 nettoweb__ joined #git
14:08 stutter joined #git
14:09 phluks joined #git
14:09 erhandsome joined #git
14:11 Es0teric joined #git
14:12 guardian joined #git
14:14 Thunderfunk joined #git
14:14 kpease joined #git
14:15 EvilPenguin joined #git
14:16 erhandsome joined #git
14:17 MattMaker joined #git
14:18 dsdeiz joined #git
14:18 erhandsome joined #git
14:19 cbd_ joined #git
14:20 synthroid joined #git
14:20 erhandsome joined #git
14:21 kurkale61 joined #git
14:22 bladdezz left #git
14:22 bronson joined #git
14:23 hash-z3r0 joined #git
14:23 nick-mbp joined #git
14:23 Gustavo6046 joined #git
14:25 Cabanossi joined #git
14:25 dsdeiz joined #git
14:25 FrostCandy joined #git
14:25 ForeignBiscuit joined #git
14:25 Plouescat joined #git
14:26 al-damiri joined #git
14:26 JanC_ joined #git
14:26 erhandsome joined #git
14:26 cagomez joined #git
14:27 LeBlaaanc joined #git
14:29 knudLeopold joined #git
14:29 Jokler joined #git
14:31 kurkale6ka joined #git
14:34 nick-mbp left #git
14:34 Prisoner-2460_1 joined #git
14:34 Prisoner-2460_1 how might you go about dropping multiple stashes at once?
14:36 Rubin joined #git
14:36 dsdeiz joined #git
14:37 dsdeiz_ joined #git
14:38 cha18 joined #git
14:41 stutter joined #git
14:41 freimatz joined #git
14:42 Thunderfunk joined #git
14:43 AfroThundr joined #git
14:43 chriys joined #git
14:43 jast Prisoner-2460_1: I don't think there's a friendly way to do it, but you can manually edit .git/logs/refs/stash (newest stashes at the bottom)
14:45 sz0 joined #git
14:46 peacememories joined #git
14:47 thiago joined #git
14:47 Prisoner-2460_1 jast: lovely! also seems handy for renaming
14:49 chriys joined #git
14:49 jesst joined #git
14:50 Gsham joined #git
14:52 npgm joined #git
14:55 crose joined #git
14:56 netj joined #git
14:58 peacememories joined #git
15:02 roelmonn_ joined #git
15:02 Vortex34 joined #git
15:02 marinin joined #git
15:03 roelmonn_ joined #git
15:03 mkali joined #git
15:04 shdw joined #git
15:08 Phylock joined #git
15:09 steeze joined #git
15:11 jceb joined #git
15:12 chriys joined #git
15:12 chrisshattuck joined #git
15:15 nettoweb__ joined #git
15:17 kurkale61 joined #git
15:18 MattMaker joined #git
15:20 roelmonnens joined #git
15:24 freimatz3 joined #git
15:25 synthroid joined #git
15:26 synthroi_ joined #git
15:26 seds joined #git
15:28 imack joined #git
15:29 inovas joined #git
15:30 Murr joined #git
15:31 kerrick joined #git
15:32 dviola joined #git
15:32 cagomez joined #git
15:33 Es0teric joined #git
15:34 meLon joined #git
15:34 meLon joined #git
15:34 edvb joined #git
15:35 seds joined #git
15:35 sunri5e joined #git
15:35 boombatower joined #git
15:36 SwiftMatt joined #git
15:37 imrg90 joined #git
15:39 kurkale6ka joined #git
15:40 Cabanossi joined #git
15:40 igemnace joined #git
15:41 dreiss joined #git
15:42 nettoweb__ joined #git
15:43 sim590 joined #git
15:43 kaldoran joined #git
15:44 chriys joined #git
15:44 imrg90 joined #git
15:46 vush joined #git
15:46 bomb-on joined #git
15:47 madewokherd joined #git
15:47 SerusWork joined #git
15:51 imrg90 joined #git
15:51 gugah_ joined #git
15:52 antfoo joined #git
15:54 musicmatze[m] left #git
15:54 robotroll joined #git
15:55 Plouescat joined #git
15:55 j08nY joined #git
15:56 a3Dman joined #git
15:56 deepakbndeepu joined #git
15:57 overlord_tm joined #git
15:57 EvilPenguin joined #git
15:57 qpdb joined #git
15:57 Gurkenglas joined #git
15:57 TomyLobo joined #git
15:58 nettoweb__ joined #git
15:59 cstratak joined #git
15:59 murii joined #git
16:00 cstratak joined #git
16:02 navidr joined #git
16:05 CheckDavid joined #git
16:07 inovas joined #git
16:10 Cabanossi joined #git
16:10 balance1 joined #git
16:11 hash-z3r0 joined #git
16:12 Guest16340 joined #git
16:13 z1haze joined #git
16:13 durham joined #git
16:14 z1haze when doing an interactive rebase.. what is the 'onto' and what is the 'from'? I have done them before from the command line (still not really knowing what im doing) but these terms confuse me. My goal is to take this branch im working on, and squash a few commits that are stupid insignificant changes. Can anyone advise?
16:16 imack joined #git
16:16 shdw- joined #git
16:16 Kavothe joined #git
16:19 EricPoe joined #git
16:19 MattMaker joined #git
16:19 aFeijo joined #git
16:19 Tobbi joined #git
16:20 synthroid joined #git
16:20 qwedfg joined #git
16:21 orbyt_ joined #git
16:22 thiago_ joined #git
16:22 wcpan joined #git
16:23 SwiftMatt joined #git
16:23 erhandsome joined #git
16:23 bronson joined #git
16:24 tarkus joined #git
16:25 LeBlaaanc joined #git
16:27 Plouescat joined #git
16:28 upperdeck joined #git
16:30 Brain joined #git
16:34 Noldorin joined #git
16:36 Noldorin joined #git
16:37 cdown joined #git
16:42 hash-z3r0 joined #git
16:42 hdqn joined #git
16:42 hdqn how do I gitignore a whole directory except a chose few files?
16:44 theoceaniscool joined #git
16:45 kfoonamalik joined #git
16:45 bayman joined #git
16:45 Es0teric joined #git
16:46 mostlybadfly joined #git
16:49 Seveas hdqn: something like this in .gitignore:
16:49 Seveas foo/
16:49 Seveas !foo/bar
16:49 gitinfo bar
16:49 parsnip !ssh
16:49 gitinfo Please use SSH to talk to remote repos. http:// and git:// are  completely unsecured and http:// suffers from smart webservers trying to implement policy. You can try to do "smart" https://, but it can be  troublesome to configure if you do not know how (see !web_repos for simple instructions) unless your company implements a MITM attack and prevents end-to-end crypto.
16:50 rk joined #git
16:50 Seveas https is just fine, gitinfo
16:50 Seveas z1haze: https://git.seveas.net/rebasing-illustrated.html#rebasing-illustrated may help
16:50 jstimm joined #git
16:55 hahuang65 joined #git
16:56 v4ult joined #git
16:56 durham joined #git
16:56 Gurkenglas joined #git
16:57 noecc joined #git
16:57 cdown joined #git
16:58 chriys joined #git
16:58 z1haze I'm just trying to understand some terminology Seveas : normally if I rebase interactively i do 'git rebase -i master' and then i pick/squash my commits, whatever .. my question is though for terminology on my GUI thing.. it has dropdown for branches labeled 'onto' and 'from'
17:01 Seveas z1haze: 'from' is the oldest commit to be rebased, 'onto' the commit to transplant them onto
17:02 z1haze so i guess in translating 'git rebase -i master' what would be onto what what would be from? Nice website btw
17:03 dsdeiz_ joined #git
17:04 imack joined #git
17:05 roelmonnens joined #git
17:05 bronson joined #git
17:07 a3Dman joined #git
17:07 peacememories joined #git
17:08 cdown joined #git
17:08 ArcaTech joined #git
17:10 Cabanossi joined #git
17:10 Zoxc joined #git
17:13 iNCoNFuN joined #git
17:16 feodaron joined #git
17:18 ShalokShalom joined #git
17:19 cyberz joined #git
17:20 MattMaker joined #git
17:20 kyan joined #git
17:20 UrsoBranco joined #git
17:24 high_fiver joined #git
17:24 manuelschneid3r joined #git
17:26 roelmonnens joined #git
17:26 Mattiaslndstrm joined #git
17:26 Seveas onto would be master@{u}, from would be the first commit after that and to would be master
17:27 aard_ joined #git
17:28 Paker joined #git
17:28 svm_invictvs joined #git
17:31 kerrick joined #git
17:33 a3Dman joined #git
17:34 stutter joined #git
17:34 parsnip on mnemonic i have is, the branch that is currently checked out, is the only ref that will move.
17:34 fahadash joined #git
17:35 parsnip that seems to be true for many git commands, like [commit, rebase, cherry-pick, merge, reset]
17:35 parsnip s/^on /one /
17:37 cyberz joined #git
17:39 hgrall joined #git
17:39 a3Dman joined #git
17:39 a3Dman joined #git
17:39 glad_ joined #git
17:41 Seveas yup
17:41 z1haze thanks Seveas
17:41 Seveas I think only push and fetch update multiple refs at once
17:42 konsolebox joined #git
17:42 parsnip okay, fetch updates multiple remotely controlled branches
17:43 parsnip doesn't the merge part of pull only act on currently checked out branch?
17:43 parsnip s/branches/refs/
17:45 grawity it does
17:46 somiaj joined #git
17:47 fatihgenc joined #git
17:47 somiaj I can't seem to figure out what changed, but on repos that were working fine (and work fine from antother computer) when I try 'git pull' or 'git fetch' (which use to work) I get the following error and nto finding much info on google about it. "fatal: Malformed response in ref list: <html>"
17:47 retropikzel joined #git
17:48 APic Can You name the affected Repositories?
17:48 OMGOMG are you by any chance on a free wifi
17:49 Seveas somiaj: that smells like http:// cloning and a bad MITM proxy. Try https:// or ssh:// cloning
17:50 Seveas or a git host that's broken and serves error responses as html without giving it a non-200 return code
17:50 somiaj I'm using https://, I upgraded the git server I was using, and it has caused a few other problems, maybe I should download it.
17:50 Seveas yeah, broken git server. Check the logs
17:51 somiaj I can't seem to clone any repos either. The machine that is having troulbe is centos6, using git 1.7, the machine that works has git 2.1.4, think I just need to downgrade the gitserver I'm using.
17:51 a3Dman joined #git
17:53 nuked0ne joined #git
17:53 Gurkenglas joined #git
17:54 bayman joined #git
17:55 somiaj don't see anything unsual in the logs. Anyways, I will just have to downgrade...thanks
17:57 orbyt_ joined #git
18:00 Es0teric joined #git
18:01 manuelschneid3r joined #git
18:02 Sasazuka joined #git
18:02 kerrick joined #git
18:03 venmx joined #git
18:03 venmx_ joined #git
18:09 chriys joined #git
18:10 Cabanossi joined #git
18:10 seds joined #git
18:13 Argylelabcoat joined #git
18:13 rk joined #git
18:15 dmc joined #git
18:15 m0viefreak joined #git
18:16 upperdeck_ joined #git
18:17 chriys joined #git
18:17 rkta joined #git
18:18 selckin joined #git
18:19 hash-z3r_ joined #git
18:19 def_jam joined #git
18:20 MattMaker joined #git
18:23 cdg joined #git
18:29 upperdeck joined #git
18:31 basiclaser joined #git
18:31 netj joined #git
18:34 dersand joined #git
18:36 grazfather joined #git
18:36 grazfather hey guys, how  can I create a branch without checking it out? I just want to make a new branch point to a certain commit, but not suffer the time it takes to change over to it
18:37 grawity `git branch <name> <base>`
18:38 grazfather wow
18:38 grazfather ty
18:38 peacememories joined #git
18:38 cdg_ joined #git
18:38 ankit01ojha joined #git
18:39 grazfather oh you can even move the branch you're not on
18:39 grazfather i assumed it was always working on your current branch
18:40 cdg__ joined #git
18:40 phluks joined #git
18:41 Ldip joined #git
18:41 peacememories joined #git
18:42 grazfather left #git
18:42 stutter joined #git
18:47 MattMaker joined #git
18:49 belmoussaoui_ joined #git
18:49 Murii|osx joined #git
18:50 peacememories joined #git
18:52 Gurkenglas joined #git
18:55 ka joined #git
18:56 bitbit joined #git
18:56 mescalinum joined #git
18:56 hash-z3r0 joined #git
18:59 cagedwisdom joined #git
19:00 doener joined #git
19:02 nixjdm joined #git
19:02 tvw joined #git
19:02 raynold joined #git
19:03 olalonde joined #git
19:03 synthroid joined #git
19:03 mar77i_ joined #git
19:05 Ladus joined #git
19:06 marinin joined #git
19:07 thiago joined #git
19:08 EvilPenguin joined #git
19:09 jimsio joined #git
19:10 salamanderrake joined #git
19:10 peacememories joined #git
19:11 jnavila joined #git
19:11 mr_sm1th joined #git
19:12 Dougie187 joined #git
19:14 Bishop_Bautista joined #git
19:16 pks joined #git
19:23 _Patrick_ joined #git
19:24 peacememories joined #git
19:28 Murr joined #git
19:29 skalpin When I do a merge from two branches and resolve the conflicts, it results in no new merge commit because resolving the conflicts keeps the files as is. Is there a way to resolve these conflicts permanently? I keep getting the same conflict on each merge.
19:31 perlpilot skalpin: enable rerere ?
19:36 _Patrick_ joined #git
19:37 prg318 joined #git
19:37 skalpin oh, I'm dumb. after completing the merge I needed to do a `git merge --continue` to create the merge commit.
19:38 skalpin I just didn't think there was anything to commit because posh-git was telling me nothing had changed. But I was still in a merging state.
19:42 orbyt_ joined #git
19:44 seds joined #git
19:44 kerrick joined #git
19:47 nixjdm joined #git
19:48 pks joined #git
19:49 vjacob joined #git
19:50 Gurkenglas joined #git
19:50 cdown joined #git
19:52 yqt joined #git
19:53 ozcanesen joined #git
19:56 kadoban joined #git
19:56 bomb-on joined #git
19:57 nobso joined #git
20:00 Baughn joined #git
20:01 seds joined #git
20:03 steeze joined #git
20:07 King_Hual joined #git
20:07 ZyX-I joined #git
20:07 marinin joined #git
20:08 cdown joined #git
20:08 Hello71 joined #git
20:08 kerrick joined #git
20:09 geekhands joined #git
20:10 synthroid joined #git
20:14 rkta joined #git
20:15 Es0teric joined #git
20:16 EvilPenguin joined #git
20:16 thiago joined #git
20:18 cdg joined #git
20:19 seds joined #git
20:20 kerrick joined #git
20:21 Anja joined #git
20:24 cdg joined #git
20:29 relipse___ joined #git
20:30 Guma joined #git
20:30 rk joined #git
20:34 Phylock joined #git
20:34 stethewwolf joined #git
20:35 beerdrop joined #git
20:36 mustmodify_ joined #git
20:37 mustmodify This is probably crazy, but can I ask git whether the word "box-shadow" has been removed from my repo in the last ... say.... 2 months?
20:37 mattcen joined #git
20:39 _ikke_ mustmodify: git log --since='2 months ago' -S 'box-shadow' --all
20:39 _ikke_ something like that
20:39 mustmodify -S ...
20:39 sbeyer joined #git
20:39 mustmodify I'll check it out. Thanks.
20:39 _ikke_ (it whould also show commit that added it, add -p to see the diff)
20:39 _ikke_ yes
20:39 _ikke_ -S shows you commits that either added or removed that string
20:40 kerrick joined #git
20:40 mustmodify HOLY CRAP IT WORKED. That's amazing. Thank you. And I'm going to just send a thank you out to Torvalds et al.
20:40 _ikke_ hehe
20:40 kasual joined #git
20:41 seds joined #git
20:44 vush joined #git
20:44 xaxxon joined #git
20:47 nixjdm joined #git
20:54 Gurkenglas joined #git
20:58 kerrick joined #git
20:58 anakinz23E joined #git
20:59 tijensse joined #git
20:59 torshido joined #git
21:00 tijensse joined #git
21:02 Marvel joined #git
21:03 steeze joined #git
21:03 King_Hual joined #git
21:03 King_Hual joined #git
21:05 EvilPenguin joined #git
21:06 mescalinum joined #git
21:07 markus13 joined #git
21:07 bronson joined #git
21:09 tijensse joined #git
21:11 Xethron joined #git
21:11 Gustavo6046 joined #git
21:14 Rc43 joined #git
21:14 Rc43 Hi guys
21:15 Rc43 I am just wondering what I am doing wrong with git-subtree
21:15 paguy61 joined #git
21:15 cdown joined #git
21:15 Rc43 Trying to merge history of other repo in to my main repo
21:15 Rc43 But got `git-subtree cannot be used without a working tree`, though .git is present in current folder
21:16 preyalone joined #git
21:18 justan0theruser joined #git
21:19 orbyt_ joined #git
21:22 tijensse joined #git
21:26 Rc43 Even on freshly cloned repo with origin
21:28 tijensse joined #git
21:28 Murr joined #git
21:29 chriys joined #git
21:30 mr_sm1th joined #git
21:30 Mindi joined #git
21:30 utrack joined #git
21:30 tijensse joined #git
21:32 xaxxon Rc43 did you init your submodules?
21:32 xaxxon Rc43 hrmm, nm
21:32 seds joined #git
21:33 Elon_Satoshi joined #git
21:33 xaxxon Rc43 did you follow the instructions here: https://git-scm.com/book/en/v1/Git-Tools-Subtree-Merging ?
21:34 relipse___ left #git
21:34 phinxy joined #git
21:35 relipse joined #git
21:35 relipse I have a coworker who made some commits NOT on a branch, how do I pull those changes into a branch i have the commit hash
21:35 phinxy left #git
21:36 kadoban relipse: Do you just want to give them a name, like a new branch name? Then 'git branch omgimabranch whateverSHAid'
21:36 Rc43 xaxxon, probably this is the key: "git checkout -b rack_branch rack_remote/master"
21:37 Rc43 I followed SO answer, it misses it
21:37 kadoban Or you can just merge them into something also if you want, just using the sha-id.
21:37 relipse yes kadoban
21:37 dsdeiz_ joined #git
21:40 noobineer joined #git
21:41 xaxxon_ joined #git
21:43 Rc43 xaxxon, actually this link solves the problem without merging histories, git read-tree I mean
21:43 Rc43 oh he left
21:45 xaxxon_ oh he's back
21:45 tijensse joined #git
21:46 dviola joined #git
21:47 Rc43 xaxxon_ :) strange that I can found nothing about git-subtree
21:47 tijensse joined #git
21:47 nixjdm joined #git
21:48 dsdeiz_ joined #git
21:48 TomyLobo joined #git
21:50 jeffreylevesque joined #git
21:50 Rc43 except git --help of course
21:51 Rc43 this weird error message ..
21:51 Rc43 could it be because of windows?
21:52 seds joined #git
21:55 orbyt_ joined #git
21:58 Sootherer joined #git
21:59 iamthad joined #git
21:59 Gurkenglas joined #git
22:00 m0viefreak joined #git
22:06 jeffreylevesque joined #git
22:08 bayman joined #git
22:10 jeffreylevesque_ joined #git
22:10 energizer joined #git
22:11 jnewt joined #git
22:14 AfroThundr joined #git
22:16 mel00010 joined #git
22:20 edvb joined #git
22:20 ozcanesen joined #git
22:22 jeffreylevesque_ joined #git
22:28 angular_mike___ is it possible to have a hook be called on remote side after push even if everything is up to date?
22:31 rk joined #git
22:31 kadoban Hm, I suspect not. I don't think the usual ones even run
22:33 ertes joined #git
22:34 angular_mike___ I'm setting up a hook to redeploy a docker service when I push to dev branch on the repo on docker host
22:35 angular_mike___ but being able to just restart that way as well would be convenient
22:35 kadoban Ya, this is the kind of thing you run into when you try to use git as a deployment whatever, IMO. Then you have to find weird ways to signal things through git that kind of don't make sense to git, since they're not content they're just commands to the deployment system or something
22:36 rk joined #git
22:36 ovidnis joined #git
22:39 Oatmeal joined #git
22:40 Cabanossi joined #git
22:41 oskarkv joined #git
22:41 Nizumzen joined #git
22:43 jfeets joined #git
22:47 ozcanesen joined #git
22:47 nixjdm joined #git
22:49 dsdeiz joined #git
22:49 dsdeiz joined #git
22:51 dave0x6d joined #git
22:53 courrier________ joined #git
22:56 LuksLux joined #git
22:58 Gurkenglas joined #git
23:00 xaxxon joined #git
23:04 ozcanesen joined #git
23:05 EvilPenguin joined #git
23:07 angular_mike___ git is a filesystem
23:08 angular_mike___ which makes deployment one of its primary use cases
23:08 angular_mike___ maybe I can do an empty amend on last commit orsmth?
23:08 blaeks joined #git
23:08 angular_mike___ since there's no issue with doing force push and force checkout
23:09 feodoran_ joined #git
23:09 kadoban Git isn't a filesystem? And if it were, I am not sure how that makes deployment something it must do. *shrug*
23:09 renarc joined #git
23:10 angular_mike___ it is according to its author Linux Torva
23:10 preaction it started as one, sure
23:10 roelmonnens joined #git
23:10 preaction if you want, you could proxy the git communication (like gitolite does) to do what you need
23:11 rameerez joined #git
23:11 angular_mike___ I wanna avoid installing jenkins
23:12 preaction was that a response to me? because i'm not sure what that was a response to
23:13 matthiaskrgr effects like pain? maybe 10 minutes
23:13 matthiaskrgr oops
23:13 matthiaskrgr wrong channel
23:15 jimmyrcom_ joined #git
23:16 angular_mike___ preaction: maybe
23:16 Hello71 "Linux Torva"
23:16 preaction alright, well, i can't help you until you're clear about what you're asking
23:16 Hello71 I exhaled more deeply than usual.
23:16 preaction so, good luck
23:17 xaxxon joined #git
23:18 angular_mike___ preaction: execute post-receive or similar hook after I push and it is already up-to-date
23:18 angular_mike___ hmm
23:18 angular_mike___ tho maybe, I could just create a new branch and push it?
23:18 preaction angular_mike___: you aren't listening to me, so i don't know why you keep pulling me into the conversation. there is no conversation if you're not listening to me.
23:19 angular_mike___ cause I set up the script to check out the branch that has been pushed
23:19 preaction it's just you saying things at me. that's not a conversation
23:19 angular_mike___ preaction: I'm just saying what seems most important
23:19 preaction i again ask: Why are you saying it TO ME?
23:19 angular_mike___ preaction: idk, do you know how to use git?
23:20 mizu_no_oto joined #git
23:20 angular_mike___ before you complained i wasn't talking to you enough
23:20 preaction yes. i, in fact, know how to solve your problem
23:20 angular_mike___ girl, stop these hot n cold games
23:21 preaction [18:10:48]  <preaction>if you want, you could proxy the git communication (like gitolite does) to do what you need
23:21 dsdeiz joined #git
23:21 angular_mike___ idk what that means
23:21 angular_mike___ why are you so mysterious
23:22 ozcanesen joined #git
23:22 preaction you didn't ask. if you want to know what something means, you need to ask
23:22 preaction look at how gitolite works, then you can see how you can intercept git communication to do what you need
23:23 angular_mike___ I don't wanna install another tool
23:23 angular_mike___ my setup is already complex enough
23:24 preaction which is why i said _look_, and not _install_
23:24 AfroThundr joined #git
23:24 nadley joined #git
23:25 angular_mike___ hm, it uses 't' for test directory name
23:25 Cabanossi joined #git
23:25 angular_mike___ oh shit ,its perl
23:26 preaction yes, "t" as a test directory is common for perl code
23:26 relipse joined #git
23:27 Elon_Satoshi joined #git
23:28 relipse__ joined #git
23:28 Murr joined #git
23:28 angular_mike___ https://i.ytimg.com/vi/LR8fQiskYII/hqdefault.jpg
23:30 preaction yes, that's lwall, the developer of perl
23:33 kerrick joined #git
23:34 mattcen joined #git
23:35 idmistir hey guys.. I could really use some help. I think I messed with the timeline ^^
23:35 Puffball_ joined #git
23:35 edvb joined #git
23:35 idmistir I think, about a month ago, I commited to a repo while being checked out on an earlier version and effectively altering the course of history since then
23:36 idmistir can I recover the commits on the parallel universe I cut loose?
23:36 preaction git rebase is a good tool for this, yes, idmistir
23:37 idmistir preaction; i can't see those commits at all.. I don't know their hashes or anything..
23:37 idmistir I've been looking for them with git fsck
23:37 preaction git fsck doesn't look for things though. maybe you meant git reflog?
23:37 idmistir i did that too
23:37 wxg joined #git
23:37 idmistir i couldn't see anything that seemed relevant
23:38 DIDAVISION joined #git
23:38 idmistir preaction; Should I be able to view the commits I can't "see" in my normal history?
23:39 plasticle joined #git
23:39 idmistir (btw, this is all on one branch, master)
23:40 preaction i mean, if you can't find them, and you don't know their commit hashes, they're likely gone
23:40 preaction if the commit isn't reachable from master, it won't show up in the log
23:40 ozcanesen joined #git
23:40 idmistir I think I just skipped a beat
23:40 idmistir and not in a good way :/
23:41 preaction did git fsck report removing any unreachable things?
23:41 preaction how did you merge the upstream's master?
23:41 someone235 joined #git
23:41 idmistir I see three unreachable things, and some dangling blobs, but nothing weird. What IS weird is that I see one commit being referenced like 8 times as "amended" (I never amended that commit)
23:41 idmistir which is very strange
23:43 preaction how did you merge the upstream's master?
23:44 idmistir preaction; I'm sorry, I don't quite understand that.
23:45 idmistir not very good with git terminology
23:45 preaction idmistir: how did the commits disappear? how did you learn that the upstream had moved on? if it's all the same branch, you  must have done something locally (git pull perhaps) that added their commits to your branch
23:47 idmistir preaction; I'm extrapolating based on what I see. There is definitely work lost, and this is the only thing that makes sense. Since the repo that is missing commits is a submodule of a main repository, the working theory is that something like "git push origin HEAD:master" happened while in a detached HEAD state
23:47 idmistir which, in all honesty, I am not sure if it would cause this
23:47 preaction a push would not cause this, no
23:47 idmistir what would?
23:47 bronson joined #git
23:48 preaction committing to a detached head and then doing git submodule update on the parent would
23:48 brent__ joined #git
23:48 idmistir committing without having pushed?
23:49 preaction sure. you can't push if you don't have a branch
23:49 preaction and submodules only recently started tracking branches at all
23:49 shdw joined #git
23:49 idmistir Fair enough.. but this doesn't make sense in the context of how we work.. consider dev machines, and a remote staging server. the staging had the changes that I am currently missing. they're not on there now either
23:49 idmistir what you're saying would make sense if we lost the changes on local only?
23:50 idmistir Our remote is on bitbucket
23:51 idmistir ^ remote git is on bitbucket*
23:51 preaction you're saying this code was committed and was in bitbucket and now it's gone?
23:51 durham_ joined #git
23:51 idmistir allegedly, yes.
23:51 preaction what do you mean allegedly?
23:52 idmistir well, I have yet to prove this. Just the only thing that makes sense so far.
23:52 idmistir could I have messed with the HEAD and and then garbage collection cleaned up any references?
23:52 preaction so you did some work. you committed that work. you pushed that up to bitbucket. and... then what?
23:52 idmistir aight, let me take it from the top
23:53 idmistir I have a guy I work with.. let's call him "Smith". Smith did some work, committed and pushed that work, and that work was pulled on the staging server from the bitbucket repo
23:53 idmistir About a month ago, I added a few things to the repo. Now, changes Smith made before I did my additions (on completely unrelated files, btw) got lost
23:54 Gurkenglas joined #git
23:54 idmistir Consequently, a git pull was made on the staging server
23:54 idmistir The staging server, of course, is now up to date with master, which does not have said commits on log
23:54 idmistir Makes sense?
23:55 idmistir "Now, some changes Smith made"*
23:55 idmistir some being the key word
23:58 preaction how did you push your changes to the server?
23:58 idmistir to the staging server?
23:59 preaction to bitbucket, first of all
23:59 idmistir just a git push from phpstorm
23:59 preaction and phpstorm does not add --force?
23:59 preaction did you add --force?

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