Perl 6 - the future is here, just unevenly distributed

IRC log for #git, 2016-08-11

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

All times shown according to UTC.

Time Nick Message
00:01 SteveHolt joined #git
00:01 firetree joined #git
00:01 Doginal joined #git
00:01 MattMaker joined #git
00:02 cqi joined #git
00:03 docnvk joined #git
00:03 tesuji_ joined #git
00:05 Noldorin joined #git
00:05 qsx joined #git
00:05 Mo0O joined #git
00:06 arescorpio joined #git
00:13 axl__ joined #git
00:14 Khisanth joined #git
00:18 Rodya_ joined #git
00:18 fuchstronaut joined #git
00:18 fuchstronaut joined #git
00:19 boombatower joined #git
00:22 vassagus joined #git
00:28 t-mart_ joined #git
00:29 dreiss joined #git
00:29 danslo joined #git
00:30 sullyj3 joined #git
00:31 byteflame joined #git
00:31 jimi_ joined #git
00:32 axl_ joined #git
00:32 MattMaker joined #git
00:33 lugzer joined #git
00:37 therealGent joined #git
00:37 therealGent joined #git
00:38 kulelu88 joined #git
00:38 ksolv joined #git
00:41 Somelauw joined #git
00:41 Somelauw is there a way to make git diff open a pager? without doing git diff | less and losing colour?
00:43 ericP joined #git
00:44 ToxicFrog Somelauw: it'll activate the pager by default if it produces more than one screen of output.
00:45 ToxicFrog If you want it to paginate unconditionally, "git --paginate diff ..."
00:46 ericP i accidentally merged in the wrong direction and have a repo with 27 commits that have not been merged into origin/master (a github repo)
00:46 ericP how do i undo that giant merge?
00:46 ericP reset HEAD~1?
00:46 Somelauw ToxicFrog: not for me, not for diff
00:46 ericP or does that only bring me back one commit?
00:47 sebascar joined #git
00:49 Manuel15 joined #git
00:49 Manuel15 joined #git
00:51 Somelauw ToxicFrog: this is what my config looks like https://pastee.org/mthfb
00:52 Rodya_ joined #git
00:52 PeterRabbit ericP: You can always look in your reflog to see what you were at before
00:52 ToxicFrog Someguy123: what is "colordiff"? Have you tested that command to make sure it works as expected?
00:52 ToxicFrog Er
00:53 ToxicFrog Somelauw: ^
00:53 Somelauw well, colordiff is just input to my pager, right?
00:53 ToxicFrog What? No.
00:53 ToxicFrog you set pager.diff = 'colordiff'
00:54 Somelauw but it's like diff expect that it does colour
00:54 ToxicFrog What that means (per 'git help config') is "paginate the output of "git diff" by piping it through the "colordiff" command"
00:54 Doginal joined #git
00:54 ToxicFrog If colordiff is not a pager, it won't paginate.
00:55 Somelauw so i need something like pager.diff = 'colordiff $1 | less -RX'
00:55 zivester joined #git
00:55 ToxicFrog Or you could just set pager.diff = true because git diffs are already coloured
00:56 ToxicFrog Which means it will respect your environment variables/other config options with respect to which pager to use
00:56 ToxicFrog (if your diffs are not coloured but you want them to be, look at the color.diff option)
00:56 Rodya_ joined #git
00:56 d4rklit3 joined #git
00:56 ToxicFrog Also, I'm pretty sure you can't pass args like $1 to it, but I could be wrong about that.
00:58 Somelauw ehm, you are right git is already coloured by default, i don't know why i ever put that line there
00:59 Somelauw instead of setting it to true i could just remove the whole line
00:59 ToxicFrog IIRC, removing the whole line will make it autopaginate for output > 1 screen, setting it to true will make it paginate always
00:59 ToxicFrog So it depends on which of those behaviours you want
01:00 Somelauw nice, thank you
01:02 tesuji joined #git
01:03 Rodya_ joined #git
01:06 clemf joined #git
01:08 blockzombie left #git
01:09 wazowski joined #git
01:09 MattMaker joined #git
01:10 thiago joined #git
01:13 axl_ joined #git
01:13 aidalgol joined #git
01:15 Skipp_OSX joined #git
01:16 Skipp_OSX is there a way you can pull down a specific tag from a remote?
01:16 Dreamer3 joined #git
01:16 Skipp_OSX I don't want to clone, I just want to get the tag itself
01:17 Skipp_OSX and I don't want to pull down all of the tags, because too many
01:17 JanC_ joined #git
01:17 Diabolik joined #git
01:17 axl_ joined #git
01:18 fuchstronaut joined #git
01:18 Diabolik joined #git
01:19 Skipp_OSX nm... (I got it)
01:19 Diabolik joined #git
01:19 Skipp_OSX git pull remote <tagname>
01:19 Diabolik joined #git
01:21 d^sh joined #git
01:22 pks joined #git
01:23 Skipp_OSX wait that seemed to work but didn't actually add the tag
01:23 Skipp_OSX is there anybody even here?
01:23 vassagus joined #git
01:25 Rodya_ joined #git
01:28 lugzer joined #git
01:29 Doginal joined #git
01:29 iamdevnul joined #git
01:29 hagabaka joined #git
01:31 unreal joined #git
01:31 CEnnis91 joined #git
01:33 therealGent joined #git
01:33 axl_ joined #git
01:35 eka joined #git
01:37 ShekharReddy joined #git
01:37 Tee_Pee joined #git
01:38 Guest21717 joined #git
01:38 jost_ joined #git
01:40 ytl joined #git
01:40 MattMaker joined #git
01:46 SenpaiSilver joined #git
01:46 sangy joined #git
01:46 DieguezZ joined #git
01:46 Dougie187 joined #git
01:48 ilbot3 joined #git
01:48 Topic for #git is now Welcome to #git, the place for git help and doorbells | Public logs at http://goo.gl/BuUi5o | Current stable version: 2.9.2 | First visit? Read: http://jk.gs/git | Getting "cannot send to channel"? /msg gitinfo .voice | Knock knock. Who's there? Git. Git-who? Sorry, 'who' is not a git command - did you mean 'show'?
01:48 Goplat joined #git
01:50 linuxmint joined #git
01:54 MattMaker joined #git
01:54 romerocesar joined #git
01:56 realz__ joined #git
01:57 johnny56_ joined #git
01:58 harish_ joined #git
01:59 dreiss joined #git
02:02 Rodya_ joined #git
02:02 pikajude joined #git
02:02 pikajude joined #git
02:05 realz__ joined #git
02:06 Doginal joined #git
02:08 MattMaker joined #git
02:09 dhollinger joined #git
02:15 p4trick joined #git
02:19 fuchstronaut joined #git
02:23 cqi joined #git
02:24 diogenese joined #git
02:26 l4v2 joined #git
02:26 cqi joined #git
02:26 lugzer joined #git
02:28 MattMaker joined #git
02:29 vrishab joined #git
02:29 Rodya_ joined #git
02:29 SwiftMatt joined #git
02:30 MaBunny joined #git
02:30 danslo joined #git
02:30 Doginal joined #git
02:31 MattMaker joined #git
02:31 MaBunny Is this correct:  git --pull -v --repo=myrepo
02:34 Rodya_ joined #git
02:36 MaBunny I forgot the commands,am on mobile now
02:38 Rodya_ joined #git
02:40 fairuz joined #git
02:42 lagothrix joined #git
02:45 tchan joined #git
02:45 MattMaker joined #git
02:47 hexagoxel joined #git
02:51 skylite joined #git
02:54 cdg joined #git
02:56 bq_ joined #git
02:58 sabgenton joined #git
02:58 bq_ i just cloned a git repo and then something changes on remote repo so I pulled the changes. then i made some change on one file. now i need to commit and push. how do i do that? long time no use git for me
02:59 Rodya_ joined #git
02:59 sabgenton is there  away to git checkout otherbranch .      without overriding untracked files?
03:00 aavrug joined #git
03:01 stuartrexking joined #git
03:02 aavrug joined #git
03:02 lulzylulzer joined #git
03:04 cdown joined #git
03:05 bq_ how to add a rule in .gitignore ignoring all .swp files including all subfolders?
03:05 csd_ joined #git
03:07 fairuz joined #git
03:08 bq_ how to uncommit the most recent commit?
03:08 rewt git reset --hard @^
03:09 rewt as long as you haven't pushed it
03:11 Rodya_ joined #git
03:11 sabgenton bq_: don't use --hard if you want to keep the work you didn't want to commit
03:14 bq_ sabgenton: oh my god. all work are gone. how do i recover it?
03:14 sabgenton you can't blame rewt
03:14 mohabaks joined #git
03:15 sabgenton if you left out --hard  it would leave the new work   but --hard restores to the old commit
03:15 sabgenton sorry
03:15 sabgenton yes you can
03:15 rewt if you have the commit's hash in your terminal history, you can still check it out
03:15 sabgenton um just  reset hard  again to the old commit
03:15 sabgenton git reflog
03:16 Rodya_ joined #git
03:16 fstd_ joined #git
03:19 sabgenton bq_: normally git reset --hard HEAD@{1}
03:19 bq_ sabgenton: are you sure if I did 'git reset --hard @^' again, it will recover?
03:19 sabgenton yeah if you didn't do anything else
03:19 sabgenton check git reflog
03:19 sabgenton first
03:19 lugzer joined #git
03:20 bq_ HEAD@{1} == @^?
03:20 rewt ^ is parent
03:20 rewt meaning 1 commit before
03:20 rewt @ is HEAD
03:20 bq_ yes git reflog shows that.
03:21 rewt @^ will go back 1 more commit, not to the one that was discarded
03:23 Rodya_ joined #git
03:23 bq_ rewt: yes it didn't. how do i recover?
03:23 sabgenton rewt: no he's seeing the log of when you first told him to type @^
03:24 sabgenton bq_: yes that is the one you want
03:24 sabgenton git reset --hard HEAD@{1}
03:25 sabgenton again type git reflog one more time to make sure it still says that first
03:25 bq_ i did twice 'git reset --hard @^'. nothing else. that is not what i work.
03:26 bq_ i want to get back my work
03:26 jkr joined #git
03:26 vassagus joined #git
03:26 sabgenton bq  if you look thru the reflog and can see the sha1  of where you were   when you first did the reset hard   reset hard to that
03:27 sabgenton then    follow  rewt's orriganal addvice without  --hard
03:27 sabgenton and you get what you want
03:29 sabgenton bq_: basicly you need to restore the commit you didn't want first becasue it had you data in it
03:29 sabgenton if you did even more work since that commit then its gone   sorry
03:30 sabgenton gota go
03:30 sabgenton hope that helps
03:31 Raed|Laptop joined #git
03:32 sabgenton joined #git
03:34 bq_ ok. get back
03:34 sabgenton bq_: ok yeah my laptop was going flat @^ would be to new would be one lower in the reflog
03:34 sabgenton bq_: ok great
03:34 sabgenton if your not shaken    git reset   without --hard  is your goto
03:35 sabgenton --hard is the only option that writes the commit back to you hardrive project root
03:35 bq_ sabgenton: does .gitignore work on both tracking(add) and commit?
03:36 sabgenton ok .gitignore is evil (I found this out the hard way)
03:37 sabgenton .gitignore means  "git act as if the files wern't there"
03:37 sabgenton so yay it doesn't add files  (cause there not there remember)
03:37 sabgenton but if the files ever got into git before and and you check them out.....
03:38 sabgenton boom they overide what was of them in you project root
03:38 danimal2 joined #git
03:38 dsantiago joined #git
03:38 fahadash joined #git
03:39 bq_ how to check what files are committed on current commit?
03:39 sabgenton bq_: gitignore is only really safe you set it up at the start  so the files never go into git to come back and clober you later
03:40 bq_ sabgenton: yeah. so i need to uncommit, untrack, then add a rule in .gitignore. then add and commit again. does this work?
03:40 sabgenton bq_: well  if you delete you .gitignore prior to checkout  git will not clober anything and just warn you
03:41 sabgenton then restore the .gitignore from the commit checkout
03:42 sabgenton bq_: don't worry to much everybody does it you could just  sanitise your branch tip commits  then only checkout branches (not detatched head)
03:43 sabgenton you could clone the  repo   by cd ing out a dir   , clones are linked up
03:43 sabgenton change the clone (totaly safe) then when your happy  push --force  it into your main
03:45 sabgenton you may need to git checkout -b dummy first  in the main one   so it is not on  a  branch your using
03:45 dsdeiz_ joined #git
03:45 sabgenton bq_: if you only have one ignore  file  just delete it before checking out   to see if it's a problem
03:47 tinanoo joined #git
03:47 sabgenton bq_: git people that to like c and stuff only use .gitnore for stuff like the compiled exe's  stuff the don't care about that was it's original intended design AFAIK
03:47 elmcrest joined #git
03:48 unreal joined #git
03:48 harv joined #git
03:49 benbergman joined #git
03:49 mrkurtz joined #git
03:49 joki joined #git
03:49 liuzhen joined #git
03:49 Peuc joined #git
03:50 timmu joined #git
03:50 sabgenton bq_: be nice if they made a .keep_precious file for everyone else but your stuck with .gitignore
03:50 tfnico joined #git
03:51 AustinMatherne joined #git
03:51 hfp_work joined #git
03:51 mehola joined #git
03:51 SleePy joined #git
03:51 SleePy joined #git
03:52 Cabanossi joined #git
03:52 Ratler joined #git
03:52 sullyj3 joined #git
03:53 mohabaks joined #git
03:53 M-shine joined #git
03:53 BrianBlaze420 joined #git
03:53 ben__ joined #git
03:53 Takumo joined #git
03:54 sunny256 joined #git
03:54 kyan joined #git
03:54 woodruffw joined #git
03:54 astaves joined #git
03:54 Takumo joined #git
03:54 Jonuz joined #git
03:54 woodruffw joined #git
03:55 woodruffw joined #git
03:55 hdon joined #git
03:55 [Brain] joined #git
03:55 aidalgol joined #git
03:55 go1111111 joined #git
03:55 jaequery joined #git
03:56 monty joined #git
03:56 gfixler joined #git
03:56 cdown_ joined #git
03:57 akkad joined #git
03:57 Ratler joined #git
03:57 fairuz joined #git
03:58 cdown__ joined #git
03:59 harish_ joined #git
04:02 thecomedian joined #git
04:03 Doginal joined #git
04:05 bronson joined #git
04:07 a_thakur joined #git
04:07 cyan__ joined #git
04:07 bronson_ joined #git
04:09 fairuz joined #git
04:10 bq_ how to show which files are going to be committed if I type git commit?
04:10 Raed|Laptop git status
04:11 Raed|Laptop bq_: ^
04:13 WizJin joined #git
04:14 safe joined #git
04:15 _tgunr joined #git
04:16 Cyclohexane joined #git
04:16 mattcen joined #git
04:18 byteflame joined #git
04:19 fairuz joined #git
04:19 jstimm joined #git
04:20 dsantiago joined #git
04:20 fuchstronaut joined #git
04:23 cagomez joined #git
04:24 slayerjain joined #git
04:25 Doginal joined #git
04:27 vassagus joined #git
04:29 Cabanossi joined #git
04:40 sangy joined #git
04:44 MrWoohoo joined #git
04:46 SuperPhly joined #git
04:46 pwnz0r joined #git
04:46 SuperPhly This is more of a question about gitlab... I need to be able to add a dependency to package.json (npm) so that it can download a private repo from GitLab.
04:47 MacWinne_ joined #git
04:47 SuperPhly I'm playing around with every token/stackexchange recommendation and can't seem to figure out a way to use a token to get a repo.
04:47 jeffreylevesque_ joined #git
04:47 nilg joined #git
04:47 SuperPhly http://blah/group/repo.git?token=XXXXXX doesn't work
04:51 SuperPhly I keep getting 401's at best.
04:52 sagerdearia joined #git
05:00 subhojit777 joined #git
05:00 moei joined #git
05:01 Jupelius joined #git
05:01 filcab joined #git
05:01 PasiFist joined #git
05:01 matthiaskrgr joined #git
05:01 Ratler joined #git
05:02 Manuel15 joined #git
05:02 Manuel15 joined #git
05:02 satya4ever joined #git
05:02 a_thakur joined #git
05:03 Brando753-o_O_o joined #git
05:04 BrianBlaze420 joined #git
05:05 Raging_Hog joined #git
05:10 fairuz joined #git
05:12 leeN joined #git
05:15 bocaneri joined #git
05:20 basiclaser joined #git
05:22 fuchstronaut joined #git
05:24 fairuz joined #git
05:27 hahuang61 joined #git
05:28 vassagus joined #git
05:36 JustAnotherIdiot joined #git
05:36 rik316 joined #git
05:36 JustAnotherIdiot would adding the line "modes/*/config.json" to .gitignore be valid?
05:37 grawity yes
05:37 JustAnotherIdiot ok
05:39 Rodya_ joined #git
05:40 TomyLobo joined #git
05:42 jas4711 joined #git
05:42 moei joined #git
05:42 Derperperd joined #git
05:42 theresajayne_ joined #git
05:43 harish_ joined #git
05:43 Michaela joined #git
05:43 Doginal_ joined #git
05:43 elijah_ joined #git
05:43 inflames joined #git
05:43 wizonesolutions_ joined #git
05:43 JackMc_ joined #git
05:44 Anthaas_ joined #git
05:44 yena_ joined #git
05:45 ionelmc_ joined #git
05:45 robogoat_ joined #git
05:46 rpd_ joined #git
05:46 tvl joined #git
05:46 elementalest joined #git
05:48 ueberall joined #git
05:48 ueberall joined #git
05:48 themill_ joined #git
05:48 Lea2 joined #git
05:48 tesuji joined #git
05:48 dhollin3 joined #git
05:48 leeN joined #git
05:48 djb-irc joined #git
05:48 troyt joined #git
05:48 creppe joined #git
05:48 Jonuz joined #git
05:48 thejoecarroll joined #git
05:49 tcurdt joined #git
05:49 pks joined #git
05:49 EvilJStoker joined #git
05:49 alcohol joined #git
05:49 nioncode joined #git
05:49 drkhsh joined #git
05:49 m1dnight_ joined #git
05:49 jwest joined #git
05:49 aspiers joined #git
05:50 Nothing4You joined #git
05:50 forrestv joined #git
05:50 dhollinger joined #git
05:51 Doginal joined #git
05:54 kini joined #git
05:58 ketas joined #git
05:58 MrWoohoo joined #git
05:59 Repox joined #git
06:02 vuoto joined #git
06:10 beyertech joined #git
06:12 eka joined #git
06:12 rscata joined #git
06:12 beyertech joined #git
06:14 beyertech joined #git
06:15 freimatz joined #git
06:16 beyertech joined #git
06:17 roelmonnens joined #git
06:18 cqi joined #git
06:19 zeroed joined #git
06:22 fuchstronaut joined #git
06:22 fuchstronaut joined #git
06:23 Macaveli joined #git
06:26 br4d joined #git
06:26 br4d left #git
06:28 hahuang61 joined #git
06:28 govg joined #git
06:29 vassagus joined #git
06:29 eka joined #git
06:29 tinanoo1 joined #git
06:31 fairuz joined #git
06:34 ketas joined #git
06:35 fairuz joined #git
06:38 ki0 joined #git
06:41 Sceorem joined #git
06:42 alansaul joined #git
06:44 fairuz joined #git
06:46 SlizardWizard joined #git
06:46 SlizardWizard say i switched my branch to a new branch, I make a bunch of changes i haven't commited the changes yet.
06:46 SlizardWizard The new branch i switched to was the wrong branch
06:47 SlizardWizard can i switch the branch, then commit the changes?
06:47 SlizardWizard or will it get reset to whatever that branch was at
06:47 ShekharReddy joined #git
06:48 danslo joined #git
06:49 sargas joined #git
06:50 zeemz joined #git
06:50 Raed joined #git
06:53 lb SlizardWizard you can just !float over the changes
06:53 gitinfo If you have made a change in your working directory and have NOT YET COMMITTED, you may "float" that change over to another (`git checkout oldbranch`) or new (`git checkout -b newbranch`) branch and commit it there.  If the files you changed differ between branches, the checkout will fail.  In that case, `git stash` then checkout, and `git stash apply` and go through normal conflict resolution.
06:55 submitnine joined #git
06:55 Balliad joined #git
06:59 sargas joined #git
07:00 ShapeShifter499 joined #git
07:01 kronos joined #git
07:01 harish_ joined #git
07:02 kronos left #git
07:02 romerocesar joined #git
07:02 Darcidride joined #git
07:03 pwnz0r joined #git
07:03 SlizardWizard lb: awesome
07:03 SlizardWizard thanks
07:04 lb yw
07:04 howdoi joined #git
07:04 howdoi is there no way to add a per-commit hook to the entire team, as in only once to a repo and it must get fetch on the next pull?
07:05 lb pull includes fetch
07:06 thiago howdoi: adding a hook is an action taken on each and every repository, manually
07:06 howdoi lb, yup but it doesn't pull hooks
07:06 thiago howdoi: at best, you can automate it by creating a skeleton hook dir, so when git creates a repository (init or clone), it will install that hook for you
07:06 csd_ joined #git
07:06 thiago but it needs to be done by each and every user, willingly and explicitly
07:06 howdoi oh ho
07:07 jaafar joined #git
07:07 howdoi I agree, if it pull a hook as well someone could just do a rm -rf /
07:07 howdoi *if it were to
07:07 thiago right
07:07 lb hooks are not getting pulled
07:08 thiago lb: did you read what I and howdoi just said?
07:08 lb thiago i did after i pressed enter :/
07:08 thiago ah, ok
07:08 howdoi lb: pull includes fetch, yes I know, pull = fetch + merge
07:08 thiago race condition :-)
07:09 howdoi will go with https://www.npmjs.com/package/pre-commit
07:09 lb thiago still recovering from alcohol intoxication...
07:09 thiago lb: don't drink and code
07:09 lb i bet you'll race me again today ^^
07:10 thiago been there, done that, suffered the effects
07:10 thiago (mostly, "wtf was I thinking?")
07:10 jguillen joined #git
07:10 lb thiago only doing docs today. i guess they will be pretty hilarious for the next dev ^^
07:15 howdoi :D
07:15 TomyWork joined #git
07:16 red_ joined #git
07:18 mariuscc joined #git
07:18 liuzhen joined #git
07:18 fairuz joined #git
07:19 beyertech joined #git
07:19 DevAntoine joined #git
07:20 beyertech joined #git
07:21 ome joined #git
07:23 fuchstronaut joined #git
07:25 dsantiago joined #git
07:29 abyx joined #git
07:29 vassagus joined #git
07:31 tlolczyk What are the ways I can sync up a local repository with a repository on another computer?
07:32 ahmedelgabri joined #git
07:33 garethdaine joined #git
07:34 alcohol tlolczyk: rsync, scp, tar, git, pidgeon carrier, several days of copy-pasting
07:34 LiohAu joined #git
07:34 tlolczyk alcohol, pidgeon carrier?
07:34 tlolczyk scp not ssh?
07:35 alcohol ssh doesn't transfer files like scp does
07:35 tlolczyk I want git to do the transfering not some other program.
07:37 alcohol i want to swap all my working days for holidays and holidays for working days.
07:40 lb tlolczyk https://en.wikipedia.org/w​iki/IP_over_Avian_Carriers
07:41 lb or more formal: https://tools.ietf.org/html/rfc2549
07:41 alcohol formal original: https://tools.ietf.org/html/rfc1149
07:43 lb correct
07:43 lb that must have been alcohol intoxication as well
07:47 leo_ joined #git
07:47 lugzer joined #git
07:49 jceb joined #git
07:49 ams__ joined #git
07:51 leo_ left #git
07:51 mohabaks joined #git
07:52 encod3 joined #git
07:53 nuc joined #git
07:53 beyertech joined #git
07:54 tlolczyk And I thought Jon bentley was only joking.
07:55 reda joined #git
07:58 filcab joined #git
07:59 marcogmonteiro joined #git
08:00 zeroed joined #git
08:02 bongjovi joined #git
08:07 fished1 joined #git
08:07 p4trix joined #git
08:09 saedelaere joined #git
08:13 tomog999 joined #git
08:13 troulouliou_div2 joined #git
08:13 reda joined #git
08:18 reda joined #git
08:18 GavinMagnus joined #git
08:18 xaa joined #git
08:19 biertie joined #git
08:19 biertie joined #git
08:22 jagob joined #git
08:23 t0by joined #git
08:24 fuchstronaut joined #git
08:28 aidalgol joined #git
08:28 ki0 joined #git
08:30 vassagus joined #git
08:30 hahuang61 joined #git
08:30 ki0 joined #git
08:32 eijk joined #git
08:32 alansaul joined #git
08:33 mikecmpbll joined #git
08:34 ferr joined #git
08:35 bexelbie joined #git
08:35 bexelbie left #git
08:35 fished joined #git
08:43 lordjancso joined #git
08:43 eijk_ joined #git
08:44 rorro joined #git
08:46 slayerjain joined #git
08:47 eijk_ joined #git
08:47 bakibour joined #git
08:48 bakibour left #git
08:48 bourbaki joined #git
08:49 liuzhen joined #git
08:49 bourbaki hiho, i woult like to know if there are new commits on a specific branch to pull from a server
08:49 bourbaki to build something like a build trigger for a light weight buildserver
08:52 yuhlw_ joined #git
08:52 fished joined #git
08:53 canton7 bourbaki, the cheapest way would be to 'git ls-remote origin', find the hash for your branch, and compare. Although if you are then going to fetch those new commits, you might as well just do a 'fetch' to start with
08:54 heftig can i hook something to run on rebases? both explicit and internally triggered. i want to print a diffstat for before..after, similar to what merges produce
08:54 canton7 although, from experience, I'd recommend against trying to write your own build server. There are plenty of free half-decent pre-existing ones, which will do the job better than a home-grown solution for a fraction of the effort
08:54 torbjorn joined #git
08:55 bourbaki canton7: did you try to write one :)?
08:55 Ezzy joined #git
08:56 zeroed joined #git
08:56 canton7 I know someone who did
08:57 bourbaki Im shifting more and more of the build stuff into F# FAKE and it seems to me once i am done all i do in the end is calling that script in jenkins
08:57 fuchstronaut joined #git
08:57 netj joined #git
08:57 ki0_ joined #git
08:58 nilg joined #git
09:00 davisonio joined #git
09:01 wrouesnel1 joined #git
09:03 a_thakur joined #git
09:04 CussBot joined #git
09:06 dsantiago joined #git
09:08 star_prone joined #git
09:08 nanodrone joined #git
09:09 romerocesar joined #git
09:10 fuchstronaut joined #git
09:12 guardian hello, when I do 'git submodule' git lists submodule and annotates commits with tag, e.g. SHA1 path (tag)
09:12 guardian except for some submodules, there's no (tag) part in the output
09:12 guardian what could be the reason?
09:14 PCatinean joined #git
09:14 whatevsz joined #git
09:15 Takumo Hi all, is there a way to rebase/reset/re-reference the commit tree to basically prune all previous commits
09:15 Takumo and treat the current HEAD as the first commit?
09:16 tobiasvl Takumo: fastest would probably be to !rebase_interactive
09:16 tobiasvl !interactive_rebase
09:16 gitinfo Interactive rebase sounds similar to rebase but has completely different abilities. It can do this to commits: change the order, squash some of them together, remove some, add random existing commits from other branches, split them, and more... it's very powerful. Documentation is in the section "Interactive Rebase" in 'man git-rebase'.
09:16 Takumo oh awesome
09:16 nanodrone joined #git
09:16 tobiasvl and just squash everything together (visual block in vim?)
09:16 notebox joined #git
09:16 tobiasvl probably easier ways to do it too though
09:16 fairuz joined #git
09:17 lb Takumo the fastest would be, delete the .git directory and do a fresh git init & git add . & git commit -m "first commit"
09:18 lb bam all history gone :D
09:18 i7c Takumo, git checkout --orphan does exactly that, if I understand you correctly.
09:19 i7c And I bet my ass that it’s faster than the other solutions here. :P
09:20 lb i don't want to posess i7c ass
09:20 Takumo depends on the ass
09:20 i7c lb, don’t you want your life to be meaningful?
09:20 Takumo if it can carry stuff
09:20 i7c ;)
09:20 i7c well played
09:21 lb ^^
09:23 MrWoohoo joined #git
09:24 rahtgaz joined #git
09:25 robsco left #git
09:25 phanimahesh joined #git
09:27 Takumo and yes i7c checkout --orphan was definately quick
09:28 i7c \o/
09:29 Olipro joined #git
09:31 vassagus joined #git
09:32 fuchstronaut joined #git
09:37 auzty joined #git
09:39 alansaul joined #git
09:40 Shentino joined #git
09:40 playfullyExist joined #git
09:40 zeroed joined #git
09:42 astrofog joined #git
09:42 elementalest joined #git
09:44 nilg joined #git
09:47 harish_ joined #git
09:48 satifant joined #git
09:50 ferr joined #git
09:51 MattMaker joined #git
10:01 delboy1978uk joined #git
10:01 cd-rum joined #git
10:02 fairuz joined #git
10:03 cd-rum_ joined #git
10:06 eijk joined #git
10:07 nbdy joined #git
10:09 delboy1978uk left #git
10:11 interrobangd joined #git
10:13 eijk_ joined #git
10:13 cyan__ joined #git
10:17 Jonuz I accidantially made all my commits about one issue in master branch, how can I move them to new branch from certain point?
10:18 grawity create a new branch off current master
10:19 grawity then reset master back to before your commits
10:19 Jonuz Sounds good, thanks
10:19 grawity i.e. `git checkout master` & `git branch foo` & `git reset --hard <whatever_commit>`
10:19 choki joined #git
10:23 Jellyg00se joined #git
10:23 fuchstronaut I have created a worktree, did my work for a hotfix and pushed the commits. How do I clean up a worktree?
10:24 _ikke_ fuchstronaut: clean up what exactly?
10:26 fuchstronaut i have the main project folder, lets call it projects/my_project and I have a folder for the worktree projects/my_project-hotfix (this was created using 'git worktree add'). Can I simply 'rm -rf projects/my_project-hotifx' and than cd into projects/my_projects and keep on doing what I was doing?
10:26 fuchstronaut I thought worktree does some magic linking these two folders
10:26 _ikke_ Ah, a worktree like that
10:27 _ikke_ there is git worktree prune
10:27 _ikke_ (I haven't worked with them yet)
10:27 djb-irc joined #git
10:27 freimatz joined #git
10:27 _ikke_ man git worktree says that removal of worktrees is currently a manual process
10:27 gitinfo the git-worktree manpage is available at http://jk.gs/git-worktree.html
10:27 Seveas yeah, rm as you did and then git worktree prune
10:28 Seveas that does the trick
10:28 Seveas worktree prune deletes those linking bits
10:29 fuchstronaut Seveas: thank you, so I am not going to screw anything up doing that? I found this: https://stacktoheap.com/blog/2016/01/​19/using-multiple-worktrees-with-git/, but it sounds weird to me, that I can either do git prune or let git garbage collect do that
10:30 fuchstronaut never worked with worktree before (had git 2.4) and as always I am afraid I might break something (usually it is hard to screw things up with git, but you never know)
10:30 eijk joined #git
10:30 dangerousdave joined #git
10:30 tlolczyk I'm tryig to sync up two repositories on different machines. I don't want to setup a "full blown" server ( ie gitorius, gitlab ) I jsut want to st up a simple erver for sync.How can I do that?
10:30 grawity tlolczyk: git init
10:30 tlolczyk Or better, just point me to a web page that tells me how.
10:30 grawity well, git init --bare, that is
10:31 hahuang61 joined #git
10:31 grawity that's mostly it – you create a bare repo somewhere, then access it via ssh:// or user@host:/path
10:31 tlolczyk grawity, and if the repositories already exist and I just want to push opr pull from one to the other?
10:31 grawity same
10:32 grawity though pushing to non-bare repos is a bit more troublesome
10:32 vassagus joined #git
10:32 tlolczyk Not the 'git" protocol?
10:33 grawity i.e. you can fetch/pull, or push into a separate branch, but not push into the current branch
10:33 grawity git:// is generally read-only
10:33 grawity you *can* activate pushing (upload) with it, but it has nothing like authentication
10:33 grawity so IMHO stick with ssh
10:34 tlolczyk And I don't need to do anything on the machine I pushing.pulling to?
10:34 tinajohnson___ joined #git
10:34 grawity um, which machine
10:34 beyertech joined #git
10:34 grawity "the machine I'm pushing/pulling to" kinda describes two different ends at once
10:35 djb-irc joined #git
10:36 tlolczyk So let's say I'm just starting out on Machine B. On Machine A I have a complete repository ( lots of commits, branches ... ) So the first thing I do is git clone ssh://... on machine B. Do I have to do something on machine A so that I can connect?
10:36 jhass if ssh machine_a works, no
10:37 maxb You need a server you can SSH to, with the command line Git tools installed. You don't need anything Git-specific running on the server.
10:37 grawity you have to install a ssh server on machine A, if there isn't any
10:37 tlolczyk Same for user@host://  ?
10:37 grawity (that usually means openssh-server, and one out of two systems already have that installed)
10:37 grawity yeah, that's just a different way of writing it
10:38 grawity host:/path or user@host:/path is rcp/scp style, ssh://host/path or ssh://user@host/path is URL style
10:38 tlolczyk So if I can  login via ssh I don't have to do anything more?
10:38 grawity no
10:38 tlolczyk Cool.
10:39 interrobangd_ joined #git
10:39 eijk joined #git
10:41 eijk_ joined #git
10:45 eijk joined #git
10:46 johnny56 joined #git
10:46 sebascar joined #git
10:47 eijk_ joined #git
10:51 eijk_ joined #git
10:53 Kaisyu joined #git
10:53 fractalizator2 joined #git
10:55 asd5a joined #git
10:56 nuc joined #git
10:57 dangerousdave Can someone advise if this is possible, and the right approach please...
10:57 dangerousdave I want to produce a git combined diff, for a single person, between two dates, however I want to ignore all changes where the only change is a case change or a number change...
10:58 dangerousdave i can do the diff, between the dates, for a single person, but the other requirements are more tricky. I was thinking of running the diff through a ruby script to remove the case changes and number changes
10:58 dangerousdave think this is a good approach?
10:58 bremner I don't think git can help you with case or numbers.
10:59 grawity maybe run both inputs through a script instead, e.g. to lowercase everything
11:00 danslo joined #git
11:01 fees joined #git
11:02 stamina joined #git
11:03 Thorn__ joined #git
11:03 psprint_ joined #git
11:06 dangerousdave grawity thanks, but I don't want to see changes that consist entirely of a case change
11:07 jor dangerousdave: gnu diff has a -i to ignore case (can use git difftool to call the external diff)... but ignore number change: never heard anybody doing that
11:07 dangerousdave jor, thanks
11:08 dangerousdave jor, the problem is VB6 editor makes random layout and case changes, and I want to filter these out
11:08 bremner layout changes? like breaking lines?
11:09 jor dangerousdave: but number changes? what is this, old school BASIC with linenumbers?
11:11 dangerousdave jor, the vb6 editor makes random changes to font sizes, and layout after each compilation. these can't be removed before commit, because a binary is created each time that has to be kept in sync with the source.
11:11 reda joined #git
11:12 ki0 joined #git
11:12 eijk_ joined #git
11:12 fairuz joined #git
11:12 jeffreylevesque joined #git
11:14 bremner font sizes!
11:14 bremner truly MS is a different universe
11:14 nuc joined #git
11:15 dangerousdave yes, it's a sorry state of affairs
11:15 bremner it kindof sounds like a binary file
11:15 bremner does diff even work at all?
11:16 cdown joined #git
11:16 mohabaks joined #git
11:18 dangerousdave bremner , not on the binary, but on it's associated source file it does. Both need to be kept in sync, and both need to be in version control
11:18 ojdo joined #git
11:19 bremner ok, but the source file has font sizes?
11:19 Bennux joined #git
11:19 eijk_ joined #git
11:19 bremner also keeping binaries in source control is wacky
11:20 dangerousdave yes, source file has font sizes
11:20 bremner ok, then I repeat my question, does git diff work at all on them?
11:21 dangerousdave bremner, there are changes you make in VB6, and they are only persisted in these binary files
11:21 dangerousdave bremner http://stackoverflow.com/questions/322​488/how-do-you-deal-with-visual-basic-​6-frm-and-frx-files-in-source-control
11:21 eijk_ joined #git
11:22 a_thakur joined #git
11:22 star_prone joined #git
11:23 bremner that suggests they are _not_ text files.
11:24 a_thakur joined #git
11:25 Anoia hi allll
11:26 dangerousdave bremner. In VB6, when you make a form change, the .frm and .frx are both updated. One is a binary, one is text based. Both need to be in sync for the application to run. You can diff the text file, but not the binary.
11:26 dangerousdave Anoia hi
11:26 * Anoia runs away screaming
11:26 Anoia I thought I'd escaped VB6!
11:27 dangerousdave Anoia you're fortunate then!
11:27 dangerousdave we are rewting our VB6 app, but it's a over million lines of code
11:28 Anoia ours was huge as well, and management refused to starta  rewrite
11:28 Anoia so I left
11:28 Anoia I new use perl :)
11:28 Anoia now
11:28 dangerousdave Anoia: wise ;-0
11:29 Anoia and after 15 years, I was a bit fed up of it
11:31 Thorn__ joined #git
11:31 nuc joined #git
11:33 dangerousdave Anoia, problem I have is frm and frx dont play well with git http://stackoverflow.com/questions/1064858/​stop-visual-basic-6-from-changing-my-casing and http://stackoverflow.com/questions/​31028314/how-can-i-prevent-random-n​umeric-code-changes-in-visual-basic
11:34 dangerousdave so I have load of random case changes and font size changes that I want to filter from a diff
11:34 Anoia yeah...
11:35 Anoia not git specific, but down to VB
11:35 Anoia I was very specific about maintain case, and standardising it
11:35 Anoia date pickers as well had a "formet" property that changed randomly on save
11:36 Anoia and depending on how you saved it, it would randomly change the case of the _ExtentX/Y properties too :()
11:37 Anoia and if I removed somethign from the frx but saved too soon, it would leave a blank space that somethign else would remove on the next save, decreasing all the frx offsets
11:37 Anoia #fun
11:37 Faylite joined #git
11:39 varnie joined #git
11:39 varnie good day.
11:39 Anoia hi
11:40 Anoia dangerousdave: to standardise them all, I created a variable with the correct case, then force it to save every file (normally by search and replace of "End" with "End")
11:41 Anoia then commit and it makes it much easier to tidy up when they're accidentally changed
11:41 eijk_ joined #git
11:41 MattMaker joined #git
11:42 varnie something went wrong on my git locally. I have file A, it shows me content, let it be for simplicity "A". now I go to the remote server, check this file and it shows me content "B". both, local computer and remote server at the same branch, namely "master".
11:42 jceb joined #git
11:42 fuchstronaut joined #git
11:42 fuchstronaut joined #git
11:42 tobiasvl git status?
11:42 dangerousdave Anoia that sounds like a good idea, however I am sure we have seen case changes when there has only been a single occurence of a variable
11:43 varnie locally it says: everything is clean
11:43 dangerousdave Anoia. > and if I removed somethign from the frx but saved too soon, it would leave a blank space that somethign else would remove on the next save, decreasing all the frx offsets  -  exactly!
11:43 dangerousdave very frustrating
11:43 varnie i.e. "on branch master. Your branch is up-tyo-date with 'origin/master'. nothing to commit, working directory clean"
11:44 stuartrexking joined #git
11:44 varnie same for remote server.
11:44 tobiasvl but are master on the same commits locally and remotely?
11:44 tobiasvl git fetch
11:44 varnie git fetch on my local computer?
11:44 tobiasvl yes
11:44 tobiasvl the fact that master is up-to-date with your local copy of origin/master doesn't mean much
11:44 tobiasvl git fetch updates origin/master
11:44 varnie I've got forced update
11:45 tobiasvl well then
11:45 varnie okay, let me recheck that file now...
11:45 tobiasvl well you didn't merge, so the file shouldn't have changed
11:45 Anoia dangerousdave: http://stackoverflow.com/a/38895471/588306
11:45 varnie nope, the content remained "invalid"
11:45 varnie git merge?
11:46 tobiasvl varnie: after fetching, what does git status say about master vs origin/master?
11:46 rivarun joined #git
11:46 varnie tobiasvl, it said: "From git.ls1.ru:tophotels-group/new-tophotels-ru
11:46 varnie + 72b372b...8768b05 release    -> origin/release  (forced update)
11:46 varnie "
11:46 varnie I don't get "origin/release" <--
11:46 tobiasvl well release isn't master
11:46 tobiasvl you don't get "origin/release"?
11:46 Anoia dangerousdave: regarding hte single use, this can happen with a name in a class library/control
11:47 ChristopheT joined #git
11:47 ChristopheT left #git
11:47 varnie I don't understand why it says "origin/release"
11:47 Anoia msxml is notorious at lower casing the first letter
11:47 varnie is it correct?
11:47 varnie why not "origin/master"?
11:47 tobiasvl varnie: well you fetched a branch called "release". it's a different branch from master.
11:47 dangerousdave Anoia that first answer in your link is worth a try, thanks
11:47 eijk joined #git
11:48 varnie but why should I fetch a release branch into my master one?
11:48 varnie some of my colleagues says he f*cked up today a little
11:48 nuc joined #git
11:48 tobiasvl I dunno if you want to do that or not, but you didn't
11:48 varnie and messed up the master branch.
11:48 Sceorem joined #git
11:48 tobiasvl he might have force pushed to it then
11:48 varnie but now I would like to understand how do I fix that on my local computer
11:49 tobiasvl ok, so do you just want to throw away your local state? throw away file A and replace it with B?
11:49 varnie he suggests that I checked out all the project locally again and it would fix that.
11:49 tobiasvl varnie: to throw away master and replace it with origin/master: git reset --hard origin/master
11:50 varnie the problem is that it relates not only to one file, chances are several files were altered.
11:50 stuartrexking joined #git
11:50 tobiasvl varnie: do you have any local stuff that you want to keep?
11:50 tobiasvl or do you just want to get up to date with the remote
11:50 varnie hmm, I guess now at this moment
11:50 varnie I guess no*
11:50 varnie I have no local changes
11:50 tobiasvl OK, then git reset --hard origin/master
11:51 varnie hmm, and what if I want to, as you said, "to get up to date with the remote"?
11:52 varnie I would be grateful for more information.
11:52 daynaskully joined #git
11:52 Jellyg00se joined #git
11:52 tobiasvl you fetched, right? so origin/master is up to date with the remote's master, presumably.
11:52 tobiasvl did you ever run git status again after that like I said?
11:53 varnie correct, I performed fetch.
11:53 tobiasvl what does git status say about master and origin/master now?
11:53 varnie git status says that I am on branch master and there's nothing to commit, working directory clean
11:53 tobiasvl nothing else?
11:54 varnie hmm, that's all.
11:54 tobiasvl earlier you said it said "Your branch is up-tyo-date with 'origin/master'", it doesn't say anything like that now?
11:54 varnie correct
11:55 anuxivm joined #git
11:55 rbern joined #git
11:55 tobiasvl sounds weird. what does git log master..origin/master say?
11:56 varnie it shows no output.
11:56 tobiasvl anyway, if you're sure you have no local changes, do a "git reset --hard origin/master"
11:56 tobiasvl might want to "git stash" first just in case you have some local changes
11:57 varnie okay, I did that.
11:57 tobiasvl and how does the working dir look?
11:57 varnie now I see the "right" content of the file A, specifically.
11:57 tobiasvl ok, good.
11:57 tobiasvl then you're in sync with the remote again.
11:57 eijk_ joined #git
11:57 varnie thank you, tobiasvl!
11:57 tobiasvl np. tell your coworkers not to force push.
11:58 tobiasvl !rewrite
11:58 gitinfo Rewriting public history is not recommended. Everyone who has pulled the old history will have to do work (and you'll have to tell them to), so it's infinitely better to just move on. If you must, you can use `git push --force-with-lease <remote> <branch>` to force (and the remote may reject that, anyway). See http://goo.gl/waqum
11:58 tobiasvl which is what I assume someone has done
11:59 GavinMagnus joined #git
12:00 dsdeiz joined #git
12:00 aguimaraes1986 joined #git
12:00 paul424 joined #git
12:00 Anticom Hi all. How can i check, whether my working copy is dirty similar to `git describe --dirty` but without the "description part" ?
12:01 mingrammer joined #git
12:01 eijk_ joined #git
12:02 grawity git diff --quiet
12:02 grawity git-prompt uses `git diff --no-ext-diff --quiet && git diff --no-ext-diff --cached --quiet`
12:03 xaa joined #git
12:04 Anticom hm i get [0] for both commands you suggested although my wc isn't clean
12:04 Anticom ^ grawity
12:05 eijk_ joined #git
12:05 nuc joined #git
12:05 ahmed_elgabri joined #git
12:06 synthroid joined #git
12:06 Anticom oh wait, i haven't staged the file
12:06 Anticom is there any way to detect that aswell?
12:07 synthroi_ joined #git
12:07 notebox joined #git
12:07 piattino joined #git
12:08 danslo joined #git
12:10 eijk joined #git
12:16 eijk joined #git
12:16 jceb joined #git
12:17 circ-user-sSHeT joined #git
12:17 Maple__ joined #git
12:18 flavius joined #git
12:20 TooLmaN joined #git
12:20 eijk_ joined #git
12:21 dangerousdave is there a format you can output a "git log" that will allow a third party diff viewer to interpret, and display the diffs correctly
12:21 dangerousdave ?
12:22 j416 dangerousdave: what third party diff viewer do you have in mind?
12:23 nuc joined #git
12:23 rwb joined #git
12:24 Anticom dangerousdave: `git difftool --cached` ?
12:24 dwmw2_gone joined #git
12:24 konrados joined #git
12:24 dangerousdave any with a graphical output
12:25 j416 dangerousdave: it seems you are confused by what you want, or I don't understand
12:25 j416 dangerousdave: are you looking for a log viewer?
12:25 daniel_rose joined #git
12:25 konrados Morning :)
12:25 Anticom Think difftool is what he wants...
12:25 j416 dangerousdave: try gitk (cross-os gui), tig (terminal), gitx (os x only), for instance.
12:25 Anticom howdy o/
12:26 konrados Fine :)
12:26 dangerousdave j416 i have output a "git log" to a file, and performed some filtering on it. I would now like to see the results in a sise by side diff
12:26 HanClinto joined #git
12:26 star_prone joined #git
12:26 j416 dangerousdave: what kind of filtering?
12:27 Munt joined #git
12:27 j416 dangerousdave: you want to diff two log output text files?
12:27 konrados Actually I have a question regarding git-svn not git itself, but I hope I'll get help anyway (I can't see a channel for git-svn). I modified a few files but I want to commit just one of them, how to do that?
12:27 j416 konrados: commit as usual, then send that over to svn using dcommit
12:28 j416 konrados: nothing special.
12:28 ShekharReddy joined #git
12:29 j416 konrados: git-svn is merely an interface between git and svn; you would use svn as usual and git as usual (but avoiding merges)
12:29 qasaur joined #git
12:29 konrados j416,  But how, according to the docs ( https://git-scm.com/docs/git-svn ) the dcommit option doesn't have a --file_path option or something like that
12:30 j416 konrados: it doesn't work like that. You commit in git, then you use dcommit to commit those commits as svn commits.
12:30 sdothum joined #git
12:31 bruce_lee joined #git
12:31 bruce_lee joined #git
12:31 konrados Is 'add' the command which commits to local? If so then I can't see it...
12:31 qasaur joined #git
12:31 j416 konrados: 'commit'.
12:32 hahuang61 joined #git
12:33 selckin should probably learn some git basics before complicating it with svn in the mix
12:33 perlpilot joined #git
12:33 konrados oh, 'commits' repeats 141 times on the site :)
12:34 vassagus joined #git
12:34 spudowiar joined #git
12:35 konrados selckin, j416  actually I used to use git, I remember having 'commit' and 'push' commands, but I used Netbeans to do that, and I could right-click on a file to perform those actions. Ok, I'll try. Is it git svn commit 'path_to_file'?
12:35 j416 konrados: selckin has a good point. Your question is a pure git question. Add svn to the mix later.
12:35 Rish joined #git
12:35 MrMojito joined #git
12:35 MrMojito joined #git
12:36 j416 konrados: I would recommend learning native git (commandline) instead of using [random GUI].
12:36 konrados OK, I'll try to find a good resource, thanks j416 & selckin !
12:36 j416 konrados: skip svn for now. And, there is no 'git svn commit'.
12:36 j416 konrados: 'git commit'.
12:36 j416 konrados: !book
12:36 gitinfo konrados: There are several good books available about git; 'Pro Git' is probably the best: http://git-scm.com/book but also look at !bottomup !cs !gcs !designers !gitt !vcbe and !parable
12:37 konrados j416 thanks again j416 :)
12:37 zivester joined #git
12:38 nilg joined #git
12:40 NightStrike joined #git
12:40 dangerousdave j416, i am manually removing some changes based on regular expression
12:43 vuoto joined #git
12:44 fuchstronaut joined #git
12:44 a_thakur joined #git
12:45 roelmonnens joined #git
12:45 cyan__ joined #git
12:48 chachasmooth joined #git
12:48 j416 dangerousdave: reason?
12:50 tvw joined #git
12:51 davisonio joined #git
12:53 jimi_ joined #git
12:53 jimi_ joined #git
12:54 nuc joined #git
12:55 Beaker__ joined #git
12:55 mingrammer joined #git
12:55 fuchstronaut joined #git
12:55 fuchstronaut joined #git
12:56 veilig2000 joined #git
12:58 tlaxkit joined #git
12:58 dangerousdave j416, I want to produce a git combined diff, for a single person, between two dates, however I want to ignore all changes where the only change is a case change or a number change...
12:58 dangerousdave j416, i can do the diff, between the dates, for a single person, but the other requirements are more tricky. I was thinking of running the diff through a ruby script to remove the case changes and number changes
13:00 anuxivm joined #git
13:00 dangerousdave j416 we are using VB6 with git, and introduces random case and number changes (fonts, control positions etc) that can't be removed before commiting. It is theefore necessary to filter them after
13:03 j416 dangerousdave: you can ignore certain files, for instance, if that helps
13:04 spudowiar joined #git
13:04 Manuel15 joined #git
13:04 Manuel15 joined #git
13:04 csd_ joined #git
13:04 dangerousdave j416, I need to ignore changes that have been introduced by the editor on line by line basis
13:05 lukebyrne joined #git
13:05 j416 dangerousdave: seems you have a workflow problem there :D
13:05 j416 dangerousdave: I've no good suggestion then
13:05 nfk joined #git
13:05 dangerousdave j416 i was trying to write a script that would delete these lines, but then want to open the result in a visual diff tool
13:05 ksolv joined #git
13:06 j416 dangerousdave: why does the editor change case and numbers in your source files?
13:07 j416 dangerousdave: if it is form layout files, just ignore the files altogether? not that interesting to look at anyway
13:07 al-damiri joined #git
13:08 dangerousdave j416  because VB6 is poop
13:08 j416 the editor really changes your code?
13:08 WalrusPony joined #git
13:08 j416 I don't believe that it changes case and adds numbers
13:08 dangerousdave j416 the form binaries are tied to the form source, they need to stay in sync or the application won't compile
13:08 j416 also, of course you can choose to not commit those changes
13:08 cdg joined #git
13:09 dangerousdave j416 http://stackoverflow.com/questions/​31028314/how-can-i-prevent-random-n​umeric-code-changes-in-visual-basic http://stackoverflow.com/questions/1064858/​stop-visual-basic-6-from-changing-my-casing
13:09 dangerousdave j416 have to commit them, not an option
13:09 j416 yeah so don't commit the frm file unless it changed?
13:10 cdg joined #git
13:10 dangerousdave j416 but when it is changed, you have to commit both frm and frx
13:10 j416 I used to be a Delphi developer, there were similar issues. I solved it by only committing the parts that I actually changed. A little bit of a hassle, but it works.
13:10 j416 is the frx file generated from the frm file?
13:11 dangerousdave j416 we tried that, but frm gets out of sync with frx, and compilation is no longer possible
13:11 j416 oh, it's like side-loaded binary content
13:11 j416 huh.
13:11 j416 so, do as I suggested perhaps.
13:11 dangerousdave j416 yes!
13:12 j416 great
13:12 nuc joined #git
13:13 airdisa joined #git
13:15 lukebyrne joined #git
13:16 robogoat joined #git
13:17 sanketdg joined #git
13:17 lukebyrne hi all
13:17 gitinfo lukebyrne: hi! I'd like to automatically welcome you to #git, a place full of helpful gits. Got a question? Just ask it — chances are someone will answer fairly soon. The topic has links with more information about git and this channel. NB. it can't hurt to do a backup (type !backup for help) before trying things out, especially if they involve dangerous keywords such as --hard, clean, --force/-f, rm and so on.
13:18 lukebyrne if i have a master branch and sub branches of that called master-foo and master-bar, is there a way that I can prevent master-foo or master-bar being merged back into master
13:18 lukebyrne i.e. basically only a downstream merge
13:19 ExoUNX joined #git
13:21 star_prone joined #git
13:21 Tsutsukakushi joined #git
13:22 Anoia hi all
13:22 phanimahesh joined #git
13:22 Anoia following on form a conversation last night regarding a commit "missing" from a branch
13:23 Anoia I have a bit more info now
13:23 alansaul joined #git
13:23 Anoia I have a commit, git show mentions two files
13:23 Anoia git log for one of them shows the commit
13:23 Anoia git log for the other doesn;t
13:24 Anoia wth am I missing
13:24 Anoia I see no merges that have remove my changes
13:24 Anoia just no sign of them in the history at all
13:25 j416 Anoia: is the repo available to clone?
13:26 Anoia I can clone it (it's private)
13:26 j416 I meant, can I clone it
13:26 Anoia nope, sorry
13:26 j416 alright.
13:26 paws_ joined #git
13:26 j416 no time to investigate then, sorry
13:28 NeXTSUN joined #git
13:28 Anoia that's fine, thanks though
13:28 sanketdg joined #git
13:29 nuc joined #git
13:29 sanketdg joined #git
13:30 Ezzy Anoia: Maybe, "git log --full-history" will show this commit ? ("git log --full-history --simplify-merges" should work too, but with less garbage)
13:31 treehug88 joined #git
13:31 nurey joined #git
13:32 choki joined #git
13:32 sanketdg joined #git
13:33 sanketdg joined #git
13:34 clemf joined #git
13:34 sanketdg joined #git
13:34 vassagus joined #git
13:35 Faylite_ joined #git
13:35 fmcgeough joined #git
13:35 Ezzy joined #git
13:35 sanketdg joined #git
13:36 daynaskully joined #git
13:36 synthroid joined #git
13:36 sanketdg joined #git
13:36 LionsMane joined #git
13:38 quenode joined #git
13:38 phanimahesh joined #git
13:40 konrados Hello again. I'm on this: https://git-scm.com/book/en/v2 but I have a problem. We have a policy in our company that we have to include a ticket number in the commit message. I did git add xyz.php then git commit xyz.php I entered the right message, and then git svn dcommit - and the svn hook returned me a message that the commit message is wrong, what can I do about it?
13:41 byteflame joined #git
13:43 DJ_Spies joined #git
13:43 phanimahesh joined #git
13:44 ilmari konrados: git commit --amend
13:44 konrados ilmari, sec...
13:44 Rish_ joined #git
13:45 Dougie187 joined #git
13:45 f0rblaze joined #git
13:46 Anoia Ezzy: that DOES show my commit
13:46 Anoia but is no longer limited to that file
13:47 Anoia note that I can see my commit on another branch too
13:48 phanimah1sh joined #git
13:48 talin hmm, i have a problem with git these days... every now and then, when i checkout another branch, fetch/rebase, then i get "not a talin.git repo. could not detach HEAD"
13:49 konrados ilmari I don't want to fix/remove last commit. I want to commit once again, but "git svn" doesn't seem to send a message along with the commit, so the commit is blocked.
13:49 talin and then i try to check out master again, but next time i rebase it's the same
13:49 gravitation joined #git
13:49 mingrammer joined #git
13:51 ilmari konrados: git svn dcommit does include the commit message (at least it did when I last used it 3.5 years ago)
13:51 zacts joined #git
13:52 averell joined #git
13:53 konrados ilmari -Thank you. I tried twice :) Nvm tomorrow my cooworker comes back from a vacation, he already figured out.
13:53 bremner yes, it does
13:53 ash_workz joined #git
13:53 bremner I mean, I would be a pretty big missing feature if it dropped all commit messages
13:53 konrados The exactlty same message I entered using git commit, bremner?
13:54 konrados Or does it modify it somehow?
13:54 bremner i think it adds git-svn metadata, not sure now
13:54 xpk joined #git
13:54 ilmari iirc the metadata is in the git commits it creates from the svn commits, not vice versa
13:54 rorro joined #git
13:55 bremner that sounds right. I've never looked at the svn side
13:55 Ezzy Anoia: Do you mean, "git log --full-history -- <file path here>" shows commits that do not change this file ? I've thought, your problem is with "git commit && git revert HEAD~1 && git merge master" sequence, that might hide these two commits from history. (See TREESAME part of the help)
13:55 konrados OK, thank you, I think I'll just wait for my cooworker, thanks bremner & ilmari !
13:56 Somelauw joined #git
13:57 Anoia Ezzy: yes, it seems to
13:58 Anoia and I don;t know how that sequence is involved
13:58 Anoia I can;t see any merges that would have "undone" my commit
13:59 playfullyExist joined #git
14:00 AaronMT joined #git
14:02 kaldoran joined #git
14:02 playfullyExist joined #git
14:03 devxom joined #git
14:04 HanClinto joined #git
14:04 beyertech joined #git
14:04 dhollinger joined #git
14:04 Fwcaw joined #git
14:04 sangy joined #git
14:04 davisonio joined #git
14:04 cdown joined #git
14:07 fuchstronaut joined #git
14:08 The_Phoenix joined #git
14:08 The_Phoenix joined #git
14:09 MattMaker joined #git
14:10 beyertech joined #git
14:10 abyx joined #git
14:10 JrCs joined #git
14:11 beyertech joined #git
14:11 axl_ joined #git
14:12 beyertech joined #git
14:14 fuchstronaut joined #git
14:14 fuchstronaut joined #git
14:14 ResidentBiscuit joined #git
14:16 nuc joined #git
14:16 cdg_ joined #git
14:16 l4v2 joined #git
14:18 nikivi joined #git
14:18 sanketdg joined #git
14:20 beyertech joined #git
14:21 CuriousMind joined #git
14:21 synthroid joined #git
14:21 CuriousMind I'm trying to set up a remote to a repo using an SSH key I believe?
14:21 CuriousMind Forgive me if the terminology I am using is incorrect
14:21 MattMaker joined #git
14:25 starkhalo joined #git
14:25 MattMaker joined #git
14:25 nikivi joined #git
14:26 Darren_ joined #git
14:26 topdownjimmy joined #git
14:26 topdownjimmy Hi everyone
14:26 gitinfo topdownjimmy: hi! I'd like to automatically welcome you to #git, a place full of helpful gits. Got a question? Just ask it — chances are someone will answer fairly soon. The topic has links with more information about git and this channel. NB. it can't hurt to do a backup (type !backup for help) before trying things out, especially if they involve dangerous keywords such as --hard, clean, --force/-f, rm and so on.
14:27 romerocesar joined #git
14:28 canton7 CuriousMind, it's not clear what your question is?
14:28 danslo joined #git
14:28 Narwhaal !backup
14:28 gitinfo Worried about your data while trying stuff out in your repo? The repository in its entirety lives inside the .git directory in the root of your work tree so to backup everything `cp -a path/to/workdir path/to/backup` or equivalent will suffice as long as the repo is not modified during backup. See also http://sethrobertson.github.c​om/GitBestPractices/#backups
14:28 Rish joined #git
14:28 CuriousMind canton7: Nevermind I was able to set a remote for the repo
14:29 canton7 cool!
14:29 CuriousMind I appreciate you helping me though lol it wasn't hard
14:30 notebox joined #git
14:30 MattMaker joined #git
14:30 abyx joined #git
14:31 fuchstronaut joined #git
14:32 star_prone joined #git
14:32 hahuang61 joined #git
14:33 algae joined #git
14:33 gechr joined #git
14:33 topdownjimmy I made a commit (call it ab12) to an old branch (call it v1); meanwhile, branch v2 was being being updated with more commits that diverged from mine. I want to replay my commit ab12 onto the newer branch v2, without using any of the code from ab12 that is now outdated. Is there a way to do that?
14:33 nebg what can give me a fast intro to git ?
14:34 logikos joined #git
14:35 logikos i'm trying to use submodules with git but my submodules always seem to be detatched rather than on master
14:35 danslo nebg: https://try.github.io/levels/1/challenges/1
14:35 logikos i can go to them and checkout master but then when i do a git submodule update .. they go back to some detatched branch
14:35 logikos i do have branch = master in my .gitmodules
14:35 vassagus joined #git
14:35 logikos though i did not have it before .. i added it after researching this issue
14:36 logikos however the issue persists  .. any advice please?
14:36 fuchstronaut joined #git
14:36 Vinnie_win joined #git
14:37 settermjd joined #git
14:39 danslo topdownjimmy: it's a single commit? just git cherry-pick ab12
14:39 danslo (while on branch v2)
14:39 topdownjimmy danslo: thanks!
14:40 thebloggu joined #git
14:42 ChoHag_ left #git
14:44 kpease joined #git
14:44 mingrammer joined #git
14:45 Vinnie_win joined #git
14:45 MattMaker joined #git
14:47 roelmonnens joined #git
14:47 corpsicle joined #git
14:47 corpsicle argh
14:48 osse now what
14:48 corpsicle i detached my head, how do i get back to the latest on the remote
14:48 _ikke_ !detached
14:48 gitinfo A detached HEAD(aka "no branch") occurs when your HEAD does not point at a branch. New commits will NOT be added to any branch, and can easily be !lost. This can happen if you a) check out a tag, remote tracking branch, or SHA; or b) if you are in a submodule; or you are in the middle of a c) am or d) rebase that is stuck/conflicted. See !reattach
14:48 osse if you do git checkout origin/master you do get detached
14:48 osse you can get back to the latest on the remote, but you'll still be detached
14:48 corpsicle !reattach
14:48 gitinfo Letters refer to !detached. (a) and (b): 'git checkout branchname' to continue working on another branch, or 'git checkout -b branchname' to start a new one here; (c) git am --continue; (d) git rebase --continue
14:49 corpsicle osse: ah, it was that simple, thanks a lot =) i blame late thursday afternoon
14:49 _ikke_ It's forgiven
14:50 nanga joined #git
14:50 nanga joined #git
14:52 vassagus joined #git
14:53 HappyLoa- joined #git
14:54 clemf joined #git
14:55 daniel_rose joined #git
14:56 foist joined #git
14:56 h0schi joined #git
14:57 JustAnotherIdiot joined #git
14:57 nanga joined #git
14:57 nanga joined #git
14:57 Nex6 joined #git
14:58 daniel_rose joined #git
14:59 Nex6 joined #git
14:59 kballou joined #git
14:59 jon_ross joined #git
15:01 zivester joined #git
15:02 nanga joined #git
15:03 docnvk joined #git
15:05 perlpilot joined #git
15:05 nArk joined #git
15:07 nilg joined #git
15:08 jon_ross Hey! I'm reading up on git's object model and I have a question. Internally git blobs and trees don't know their own name, right?
15:08 perlpilot joined #git
15:08 rudycane joined #git
15:09 _ikke_ a blob doesn't have a name
15:09 _ikke_ trees map hashes to names
15:09 MattMaker joined #git
15:09 jon_ross but only of the children
15:09 Lomex joined #git
15:10 _ikke_ Yes
15:10 _ikke_ The root tree does not even have a name
15:10 jon_ross makes sense, just wanted to double check
15:10 jon_ross thanks!
15:11 _ikke_ This way, a blob can map to multiple files if the content is the same
15:11 osse trees too!
15:11 jon_ross same for trees
15:11 osse but you can't hardlink directories
15:11 osse funny, thta
15:11 * osse <3 git
15:13 kristofferR joined #git
15:15 Manuel15 joined #git
15:15 Manuel15 joined #git
15:15 Jellyg00se joined #git
15:16 jon_ross I was building my own little (encrypted) merkle DAG and my objects knew their full name. That made the lookup based on the path easy. But the downside is that you have to "generate new objects" when you move stuff.
15:16 dmto joined #git
15:16 osse is that the same as a german chansellor DAG ?
15:17 osse oh I see
15:18 jon_ross directed acyclic graph :)
15:18 bensarz joined #git
15:18 osse I was just making a funny based on merkle <-> merkel
15:20 spinningarrow joined #git
15:21 danslo joined #git
15:22 daguz joined #git
15:22 cyan__ joined #git
15:23 JrCs_ joined #git
15:23 romerocesar joined #git
15:23 phaleth joined #git
15:23 vuoto joined #git
15:23 Lomex joined #git
15:24 _ikke_ It's hashes all the way down
15:25 liamonade joined #git
15:26 vuoto joined #git
15:26 Rodya_ joined #git
15:26 daguz I have a read/write deploy key.  old git version 1.8.3.  How can I tell git to use this key when pushing changes?? (without using the .ssh/config, nor agents)
15:27 daguz I hope there is a setting in .git/config that I can use but cannot find reference.
15:27 nikivi joined #git
15:27 moritz daguz: you can set GIT_SSH_COMMAND env variable
15:27 moritz to "ssh -i thekeyfile" or something like that
15:28 marc_v92 joined #git
15:28 moritz or GIT_SSH to a small shell script which constrains ssh -i theykeyfile "$@"
15:28 fairuz joined #git
15:28 moritz if your git doesn't yet support GIT_SSH_COMMAND
15:28 timmu joined #git
15:29 JrCs_ joined #git
15:29 MattMaker joined #git
15:29 caffeinatedcode joined #git
15:29 caffeinatedcode joined #git
15:30 axisys joined #git
15:31 ahmed_elgabri joined #git
15:31 frobnic joined #git
15:31 daguz mortiz: thanks - I'm trying to do this within a python function that uses subprocess.  I've not been able to get the environment variables to work even outside python.
15:32 cdown joined #git
15:32 JrCs_ joined #git
15:32 tyreld joined #git
15:32 HappyLoa- joined #git
15:32 osse FYI: git rebase makes patches of everythign
15:33 osse I knew that, but it still came as a surprise that Git didn't appreaciate me telling it to rebase a commit that removes hundres of MB of binary files
15:34 star_prone joined #git
15:35 cYmen *headache*
15:35 cYmen so I committed something
15:35 cYmen and pulled and somebody else had committed something
15:35 cYmen I don't want a merge commit because the things are unrelated
15:35 cYmen so I want to rebase my commit on that other commit
15:36 johnny56 joined #git
15:36 le_melomane joined #git
15:36 cYmen but I cannot rebase because I have unstaged and uncommitted changes
15:36 osse that's what git stash is for \o/
15:36 osse man git-stash
15:36 gitinfo the git-stash manpage is available at http://jk.gs/git-stash.html
15:36 cYmen yeah I know it
15:36 osse git stash; do whatever; git stash pop
15:36 slayerjain joined #git
15:37 ahmedelgabri joined #git
15:37 cYmen but I just want to change the history not anything about the current working copy
15:37 cYmen why do I have to stash stuff?
15:37 osse because you have unstaged and uncommitted changes
15:37 osse you need to get rid of them
15:37 thiago_ joined #git
15:38 InfoTest joined #git
15:38 osse git rebase might need to modify those files
15:39 perlpilot cYmen: the working copy is just as much for git's benefit as it is for yours :)
15:39 roelmonnens joined #git
15:39 Darren_ joined #git
15:40 cqi joined #git
15:41 spinningarrow cYmen: you could enable the 'rebase.autoStash' setting: git config --global rebase.autoStash true
15:41 JrCs_ joined #git
15:41 spinningarrow then you won't have to manually stash and unstash your changes
15:42 cYmen oh god
15:42 cYmen how did I manage do detach my head again, now
15:43 MattMaker joined #git
15:43 JrCs_ joined #git
15:43 _ikke_ Put it onto your body again
15:43 _ikke_ It happens when you checkout anything other than a branch
15:44 cYmen I think the rebase detached it
15:44 AaronMT_ joined #git
15:44 _ikke_ Then the rebase is not yet finished
15:45 _ikke_ What does git status say?
15:45 synthroid joined #git
15:45 cYmen rrtauinedun
15:45 JrCs_ joined #git
15:45 cYmen gna!
15:46 cYmen git rebase origin/master HEAD
15:46 cYmen what is wrong with that line? :)
15:46 _ikke_ except that HEAD is redundant, nothing
15:47 cYmen are you sure? because it behaves like it should be git rebase origin/master master and the other will detach head
15:48 MattMaker joined #git
15:49 roelmonnens joined #git
15:50 JrCs joined #git
15:50 rorro joined #git
15:50 _ikke_ cYmen: You are correct, it does detach HEAD
15:50 _ikke_ I think it interprets it as you providing a commit
15:51 subhojit777 joined #git
15:51 _ikke_ instead of a branch
15:51 flavi0 joined #git
15:52 fairuz joined #git
15:53 daniel_rose joined #git
15:54 GavinMagnus left #git
15:58 mingrammer joined #git
16:03 daniel_rose joined #git
16:03 le_melomane joined #git
16:04 cYmen well at least my mistake was not totally stupid then ;)
16:05 johnny56 joined #git
16:07 MattMaker joined #git
16:07 JrCs joined #git
16:08 rubyonrailed joined #git
16:10 pioiytr_ joined #git
16:10 vuoto joined #git
16:10 nitric joined #git
16:11 nikivi joined #git
16:12 sangy joined #git
16:12 jaequery joined #git
16:13 veilig2000 joined #git
16:14 EvilPeng1 joined #git
16:14 madewokherd joined #git
16:15 GavinMagnus joined #git
16:15 fairuz joined #git
16:16 nikivi joined #git
16:16 ajf- joined #git
16:16 JrCs joined #git
16:16 t0by joined #git
16:17 MattMaker joined #git
16:18 JrCs joined #git
16:20 JrCs joined #git
16:20 fuchstronaut I am in a rebase right now and need to extract a file from my current branch, how do i do this?
16:22 danslo joined #git
16:22 HappyLoa- joined #git
16:23 zacts joined #git
16:24 pioiytr_ joined #git
16:24 kadoban joined #git
16:25 JrCs joined #git
16:27 blackwind_123 joined #git
16:27 jagob joined #git
16:28 p1k I seem to end up with somewhat fragmeneted validation logic between model and serializers
16:28 p1k do most users only use serializer validations and skip django full_clean() entirely?
16:29 pioiytr_ joined #git
16:29 p1k it would be much cleaner to code validations in only one place rather than two :S
16:29 JrCs joined #git
16:30 nitric fuchstronaut: git checkout $branch -- $file
16:31 Electrometro joined #git
16:31 comps fuchstronaut: (or, if you need it under a different name, git show branch:path/to/file > file.txt)
16:32 JrCs joined #git
16:32 p1k also if you want different fields to be allowed /  read_only on create/update do you tend to slit into two serializers with the same model  ?
16:32 nitric p1k: those would be great questions for, uh, #django?  ;-P
16:33 nicksloan joined #git
16:33 leeN joined #git
16:33 WalrusPony left #git
16:33 TomyLobo joined #git
16:34 p1k nitric: I must be asleep =/
16:34 p1k -_-
16:34 nicksloan if I have branches A and B, where B is a few commits ahead of A, and then I create branch C from B and add a few more commits, then add a few commits to A and rebase B on A, what will happen if I rebase C on B?
16:34 abyx joined #git
16:35 JrCs joined #git
16:35 star_prone joined #git
16:35 p1k nicksloan: rebase will change the commit sha's of b, so c will have duplicate commits with b that have different hashes
16:35 nicksloan p1k: that's what I thought
16:35 p1k nicksloan: generally if you haven't altered history you can just prune the commits in c from b
16:35 nicksloan ah
16:36 p1k nicksloan: but if you did stuff like commit --amending B's commits in C, then you can/will have some work to do
16:36 nicksloan so those would be empty commits most likely
16:36 nicksloan gotcha
16:36 nicksloan I think prune should work
16:36 p1k **(Prune the commits from C which came from B)
16:36 p1k since rebase will pick them up
16:37 nicksloan yep
16:37 nicksloan perfect
16:37 nicksloan p1k: so I prune using git prune? it doesn't seem like those commits are unreachable?
16:38 JrCs joined #git
16:39 nicksloan or do I manually specify the duplicate commits (which are hopefully empty)
16:40 MattMaker joined #git
16:41 JrCs joined #git
16:41 mohabaks joined #git
16:42 p1k nicksloan: I do it in rebase -i and delete the lines which is probably not the optimal way to do it
16:42 JrCs joined #git
16:45 rahtgaz When I close down a pull request on github, I go to master on my local repo and git fetch and git pull. I then delete the branch that was merged into master on github. However I always get a warning that I am deleting a branch that has not been merged to HEAD. How do I properly manage my local repo after a pull request?
16:45 bq_ joined #git
16:46 rahtgaz To be clear, after a pull request is accepted and its branch deleted on origin
16:47 _pl_ joined #git
16:49 nedbat i have a branch (dogwood), and I want to know what work is on it that isn't on master. there's been a lot of cherry-picking, so it's not as simple as looking for commits. "git checkout dogwood; git cherry -v master" shows me information, but includes some cherry-picked commits. is there a better way?
16:49 sargas joined #git
16:49 XDS2010 joined #git
16:50 mingrammer joined #git
16:51 MattMaker joined #git
16:53 nitric rahtgaz: that sounds odd, do you have squash merging enabled in the repo?
16:53 fished joined #git
16:53 clemf joined #git
16:53 rahtgaz yes. I squash indeed to avoid bringing the whole local tree of changes to origin
16:54 Noldorin joined #git
16:54 fished joined #git
16:55 jaafar joined #git
16:56 lightstalker joined #git
16:56 daynaskully joined #git
16:58 nitric rahtgaz: i *think* that's how squash merging works: it's a trick that git has, and one of the drawbacks is that the branch doesn't look merged in the history
16:58 nitric so just be aware of that, and use -D instead of -d to delete the branch locally after the PR is closed
16:58 oxsyn joined #git
16:58 dreiss joined #git
16:58 rahtgaz excellent nitric. thank you
16:59 SteffanW joined #git
16:59 fate_ joined #git
16:59 AaronMT joined #git
17:03 Silox| joined #git
17:03 sdothum joined #git
17:03 GavinMagnus left #git
17:05 jfran joined #git
17:05 cdg joined #git
17:10 jimi_ joined #git
17:14 MattMaker joined #git
17:15 charlesr joined #git
17:16 InfoTest1 joined #git
17:16 fuchstronaut joined #git
17:20 btree joined #git
17:20 dreiss joined #git
17:21 Somelauw what is the best way to contribute if you want to make multiple contributions? Do you take the effort of spreading them over branches? What if they are small ones? What if you as an implementer want to have them all enabled at the same time?
17:21 HedgeWork Somelauw: It depends a great deal on the size and complexity of the contributions in question, and the coding practices of the project being contributed to.
17:21 HedgeWork Somelauw: if you provide some context, we can try to give you our take on best practices.
17:21 sargas joined #git
17:22 thiago joined #git
17:22 btree left #git
17:24 nedbat Somelauw: it's generally a good idea to make distinct pieces of work separate contributions, though if you have a handful of tiny changes, one contribution for all of them would probably be ok
17:24 JrCs joined #git
17:24 Derperperd joined #git
17:25 algae joined #git
17:25 MattMaker joined #git
17:28 davisonio joined #git
17:28 liuzhen joined #git
17:29 mikecmpbll joined #git
17:30 ShekharReddy joined #git
17:31 iskander joined #git
17:35 mjones joined #git
17:35 JrCs joined #git
17:35 Guest5 joined #git
17:36 MattMaker joined #git
17:36 johnmilton joined #git
17:39 Didac joined #git
17:40 eze joined #git
17:40 lugzer joined #git
17:41 eze I am rebasing and I found a conflict. I solved the conflict, should I commit my resolution before git rebase --continue?
17:41 sagerdearia joined #git
17:42 osse eze: it's enough to git add it
17:42 osse i think it works either way, but I know just adding definitely works
17:43 Dazzyp joined #git
17:43 Didac joined #git
17:43 eze osse: Ok, I will try that then.
17:44 Darren_ joined #git
17:45 roelmonnens joined #git
17:45 roelmonnens joined #git
17:45 Somelauw HedgeWork, nedbat: thanks
17:46 osse 6
17:46 osse oops
17:46 eze osse: It seems to work. I will run my tests now. Thank you
17:47 romerocesar joined #git
17:48 clemf joined #git
17:49 roelmonnens joined #git
17:49 ProbabilityMoon joined #git
17:51 roelmonnens joined #git
17:52 clemf joined #git
17:52 roelmonnens joined #git
17:53 Sasazuka joined #git
17:55 t-mart joined #git
17:55 roelmonnens joined #git
17:56 osse praise gaben
17:56 star_prone joined #git
17:59 MattMaker joined #git
18:00 Harzilein joined #git
18:03 roelmonnens joined #git
18:03 _Patrick_ joined #git
18:03 mibr0 joined #git
18:04 jantje joined #git
18:05 roelmonnens joined #git
18:05 Rodya_ joined #git
18:06 roelmonnens joined #git
18:10 settermjd joined #git
18:10 GodGinrai joined #git
18:11 vuoto joined #git
18:11 JrCs joined #git
18:11 Rodya_ joined #git
18:11 vuoto joined #git
18:12 vuoto joined #git
18:14 BSaboia joined #git
18:14 JrCs joined #git
18:16 jagob joined #git
18:19 rorro joined #git
18:20 daniel_rose joined #git
18:22 dreiss joined #git
18:23 MattMaker joined #git
18:25 abyx joined #git
18:25 Fissure joined #git
18:25 vdv joined #git
18:25 drkhsh left #git
18:28 Vgr joined #git
18:29 Vgr joined #git
18:29 cyan__ joined #git
18:31 gravitation joined #git
18:32 robotroll joined #git
18:33 fuchstronaut joined #git
18:33 JrCs joined #git
18:34 hahuang61 joined #git
18:34 alansaul joined #git
18:35 lugzer joined #git
18:36 JrCs joined #git
18:37 mohabaks joined #git
18:37 bear24rw joined #git
18:38 JrCs joined #git
18:38 pR0Ps joined #git
18:41 Vgr_ joined #git
18:42 veilig2000 joined #git
18:42 JrCs joined #git
18:45 livingstn joined #git
18:45 math386 joined #git
18:46 mingrammer joined #git
18:47 beyertech_ joined #git
18:47 daniel_rose joined #git
18:48 daniel_rose joined #git
18:48 digimer joined #git
18:48 davisonio joined #git
18:48 Ranhir joined #git
18:50 yan joined #git
18:50 JrCs joined #git
18:51 yan I happened to accidentally create one large commit of a lot of changes I was planning to split up into multiple commits. What'd be the best way to go back to HEAD^1 and break up all the changes into multiple commits. detaching head, then readding, then rewriting what head points to?
18:54 JrCs joined #git
18:54 Pulp joined #git
18:54 catag87 joined #git
18:55 paul424 joined #git
18:55 vassagus_ joined #git
18:56 MattMaker joined #git
19:00 Limes_ joined #git
19:02 choki joined #git
19:03 roelmonnens joined #git
19:04 gumnos joined #git
19:05 gumnos Is there a way to push a subset of commits?  I have A->B->C->D and upstream is at B, and I want to push up C but not D (yet).  If I git-push, it will do D before it's ready to deploy
19:05 star_prone joined #git
19:05 osse gumnos: you can do git push origin D:master
19:06 osse gumnos:  git push origin HEAD~1:master for example, but any sha1 will work
19:06 osse or any commit really
19:06 gumnos ah, perfect. Thanks!
19:06 nwkj86 joined #git
19:06 Seveas yan: (assuming no uncommitted changes) git reset --hard HEAD^ ; git cherry-pick -n HEAD@{1}; git add -p; git commit; git add -p; git commit; git add -p; git commit;  etc...
19:06 digimer left #git
19:07 sangy joined #git
19:08 zmachine joined #git
19:08 gravitation joined #git
19:08 Guest5_ joined #git
19:08 roelmonnens joined #git
19:08 devster31 joined #git
19:09 devster31 joined #git
19:09 devster31 joined #git
19:12 Oog joined #git
19:15 jeremyabel joined #git
19:16 uwjesq joined #git
19:16 uwjesq Hello. There is a changed directory that contains a submodule that does not work to add anymore.
19:16 uwjesq git status shows it as changed
19:17 jeremyabel hello git friends. So I had a bit huge repository I forked (Unreal Engine), and I was working with it in VS 2015, and everything was just suuuuuper slow, slow editing and way slow compiling. I did some research, read some system logs, discovered that VS was reading from various files in my .git directory just CONSTANTLY
19:17 sbeller_ uwjesq: maybe you need to fix the .git git dir and workdir for the submodule manually :(
19:17 uwjesq But when I try git add auto complete on the dir name does not work and when I type it manually after git add it still shows as changed and not added
19:18 uwjesq sbeller_: And by "fix" you mean what?
19:18 jeremyabel so I said screw it and just removed the entire .git directory and continued working cause I was tired of dealing with the slowdown. Now I'd really like to figure that whole situation out but I have no idea how to hook all my stuff back up to the fork I made
19:18 govg joined #git
19:18 abyx joined #git
19:18 jeremyabel anybody have any advice?
19:18 faux___ joined #git
19:18 despoil joined #git
19:19 sbeller_ uwjesq: I think I misread your question/situation
19:19 ojacobson "I wouldn't start from here" :)
19:19 ojacobson git init && git remote add origin url-of-your-fork && git fetch will get you _started_
19:19 uwjesq sbeller_: Ok.
19:19 jeremyabel ojacobson: that for me?
19:19 uwjesq sbeller_: I guess git just quit service on me.
19:20 rpd_ left #git
19:20 ojacobson I would `git checkout -B my-detached-workspace && git add --all && git commit` to get a versioned copy of your current workspace before doing much else
19:20 ojacobson yes
19:20 rpd_ joined #git
19:20 jeremyabel cool, thanks
19:20 ojacobson then maybe `git checkout -B master origin/master && git rm -rf :/ && git checkout my-detached-workspace -- :/` to stage the workspace as a new descendent of master
19:20 ojacobson this is one of those "you need a good grounding in git's theory and in its terrible UI" problems though
19:21 jeremyabel god I do not want to talk about it's terrible ui lol
19:21 jeremyabel that recent xkcd comic summarizes my relationship to git
19:21 jeremyabel memorize commands, curl into ball and cry when things don't work
19:22 GenZai joined #git
19:22 ojacobson theory will help with the latter
19:22 ojacobson man pages with the former
19:22 rpd_ left #git
19:22 msonntag joined #git
19:22 ojacobson you do kind of have to know what you're looking for unfortunately
19:23 jeremyabel yeah that's the problem I've most run into
19:23 ojacobson the broad strokes version of what I described is "take a snapshot of the files, then recreate that snapshot so you can commit it as if it happened off of master"
19:23 evanvarvell joined #git
19:23 ojacobson underlying theory: git is purely snapshot-based
19:23 rpd joined #git
19:23 ojacobson if you have the files, you can construct the history
19:23 filadome joined #git
19:24 UTAN_dev joined #git
19:24 AaronMT_ joined #git
19:24 ojacobson it's not always _easy_ to construct the right history, but `git checkout REVISION -- :/` will stage and place in the work tree any files that existed in REVISION, verbatim
19:24 jeremyabel so I reinit the detached workspace, make a branch and add everything there, and then restage it under master, yeah?
19:24 ojacobson the git rm step is to remove (from staging, from the work tree) all pre-existing files, so that you don't end up with any strays that weren't in REVISION but were in master
19:24 filadome why does echo "Apple Pie" | git hash-object --stdin produce a different sha1 than normal?
19:25 ojacobson filadome: what's "normal"?
19:25 oio joined #git
19:25 filadome http://www.sha1-online.com/ 8d69b7365f59237d3fb052c1f2f15ea33457fe51
19:25 leeN joined #git
19:25 ojacobson Did you include the object header when you pasted the file into that site?
19:25 osse filadome: printf 'blob 9\0Apple Pie' | sha1sum
19:25 ojacobson the string "blob ", the decimal encoding of the file's length, and a nil byte?
19:25 Darcidride joined #git
19:26 Guest5 joined #git
19:26 filadome sorry, i'm watching a tutorial video and it hasn't explained that part yet
19:26 osse oops
19:26 osse filadome: printf 'blob 10\0Apple Pie\n' | sha1sum
19:26 osse filadome: is it the one by Michael McCullough ?
19:26 dsantiago joined #git
19:26 rahtgaz joined #git
19:26 filadome plural sight, how git works, paolo perrotta
19:27 osse ah ok
19:27 filadome http://imgur.com/a/H2l1I
19:28 MattMaker joined #git
19:28 rubyonrailed joined #git
19:28 osse There's on by Michael McCullough where he does this experiment and he gets it wrong even when it should have been right
19:29 uwjesq left #git
19:29 codeman joined #git
19:29 osse You can tell he's slightly perturbed but just glosses over it. It's due to him pasting some shell commands and then OSX does it's annoying thing with smart quotes instead of straight quotes :3
19:30 jeremyabel ojacobson: now if I wanted to commit those changes under a different branch instead of master do I just replace origin/master with origin/my-cool-branch?
19:30 ojacobson what commit do you want them to appear to be based on?
19:30 ojacobson Check that commit out.
19:30 lugzer joined #git
19:30 ojacobson `origin/master` is something I assumed :)
19:30 jeremyabel ok
19:31 jeremyabel so `git checkout -B my-cool-branch origin/master && git rm -rf :/ && git checkout my-detached-workspace -- :/`
19:31 zacts joined #git
19:31 ojacobson yup. !backup just in case, if you like.
19:31 gitinfo Worried about your data while trying stuff out in your repo? The repository in its entirety lives inside the .git directory in the root of your work tree so to backup everything `cp -a path/to/workdir path/to/backup` or equivalent will suffice as long as the repo is not modified during backup. See also http://sethrobertson.github.c​om/GitBestPractices/#backups
19:33 rossome joined #git
19:33 BSaboia joined #git
19:33 nilg joined #git
19:34 BSaboia joined #git
19:37 daniel_rose joined #git
19:37 LBV_User joined #git
19:37 nilg joined #git
19:38 stamina joined #git
19:40 ixian joined #git
19:40 firetree_ joined #git
19:40 mingrammer joined #git
19:41 BSaboia joined #git
19:44 blaman joined #git
19:44 styx_ joined #git
19:44 diego2 joined #git
19:44 blaman anyone using git thru a socks proxy?
19:45 blaman I'm trying to connect to an https repo through the proxy, but I get Failed to connect to localhost port 443: Connection refused for some reason
19:45 blaman even though I set https.proxy='socks://localhost:9870' and http.proxy='socks://localhost:9870'
19:46 dminca joined #git
19:46 dminca joined #git
19:47 blaman on
19:47 blaman oh
19:47 blaman $ GIT_CURL_VERBOSE=1 git remote show origin
19:47 blaman * No valid port number in proxy string (9870')
19:47 lugzer joined #git
19:48 t0by joined #git
19:48 ksolv joined #git
19:49 fuchstronaut joined #git
19:50 paul424 joined #git
19:50 tom[] joined #git
19:52 BSaboia joined #git
19:52 BSaboia joined #git
19:52 tom[] for testing http-based services, there are convenient one-liner server invocations such as https://gist.github.com/willurd/5720255  does anything like that exist to create an ad hoc http git server for a single repo?
19:53 osse tom[]: git instaweb
19:53 BSaboia joined #git
19:53 tinajohnson___ joined #git
19:53 tom[] osse: interesting. tnx
19:53 osse not particularly smooth, though
19:54 qsx and it won't create a "git clone"able server
19:54 osse it won't? :O
19:54 qsx `git serve' is aliased to `!sh -c 'git daemon --reuseaddr --verbose "$@" --base-path=. --export-all ./.git' sh'
19:55 CheckDavid joined #git
19:55 tom[] right. i want to be able to fetch/clone from it
19:55 vuoto_ joined #git
19:56 nicksloan joined #git
19:56 qsx yeah, the alias i just posted will
19:56 tom[] qsx: i'm trying it :D
19:57 osse qsx: but that doesn't have pretty pictures!
19:57 qsx osse: pity.
19:57 qsx tom[]: it won't be http though
19:57 tom[] protocol doesn't actually matter to me. i just assumed http would be the easy way
19:58 _ikke_ imo, ssh is even easier
19:58 tom[] i see now it was a dumb assumption
19:59 _ikke_ Not really
20:02 MattMaker joined #git
20:02 kn-928 joined #git
20:03 eivarv joined #git
20:06 tom[] qsx: what shell is that alias from?
20:07 qsx tom[]: git.
20:07 qsx git help config -> /alias
20:07 qsx derp.  man git-config
20:07 gitinfo the git-config manpage is available at http://jk.gs/git-config.html
20:07 tom[] ah!
20:09 m0viefreak joined #git
20:09 djb-irc joined #git
20:13 NeonSynth joined #git
20:14 MattMaker joined #git
20:15 Derperperd joined #git
20:15 jeremyabel ojacobson: so I did all the stuff we talked about, and I did `git add .` and then `git commit -m "whatever"` and then I went to do `git push` but it says the tip of my current branch is behind the remote counterpart
20:15 jeremyabel but I did `git pull` and it says I'm on the latest
20:16 starkhalo joined #git
20:17 kbs joined #git
20:17 jeremyabel --force comes to the rescue again
20:18 tom[] Ahh! tremenjous, qsx, simply tre-menjous
20:21 sargas joined #git
20:21 ferr joined #git
20:23 jstein_ joined #git
20:24 Raed|Mobile joined #git
20:26 vassagus joined #git
20:26 lugzer joined #git
20:27 nb_bez___ joined #git
20:27 dreiss joined #git
20:28 Guest5 joined #git
20:32 nicksloan joined #git
20:35 hahuang61 joined #git
20:37 rik316 joined #git
20:37 nicksloan joined #git
20:38 Faylite joined #git
20:41 MattMaker joined #git
20:42 blaman1 joined #git
20:44 Xizor joined #git
20:45 Viewtiful joined #git
20:45 nicksloan joined #git
20:48 skylite joined #git
20:48 nicksloan joined #git
20:52 fuchstronaut joined #git
20:56 star_prone joined #git
20:57 cyan__ joined #git
20:58 sangy joined #git
20:59 YourTheBombNow23 joined #git
21:01 Xizor joined #git
21:02 vassagus joined #git
21:04 fairuz joined #git
21:06 merced joined #git
21:06 Guest5 joined #git
21:08 mingrammer joined #git
21:10 nicksloan joined #git
21:11 Manuel15 joined #git
21:11 Manuel15 joined #git
21:12 abyx joined #git
21:12 eivarv joined #git
21:13 nicksloan is there any way to commit and tag simultaneously?
21:13 dsdeiz joined #git
21:13 vktec nicksloan: Why can't you just commit then tag?
21:13 nicksloan vktec: just trying to save some keystrokes
21:14 vktec You tag _that often_?
21:15 GodGinrai honestly, tagging and committing shouldn't really be done in the same process anyway
21:15 nicksloan vktec every time a release is staging-bound. It gets a version number. Version number means updating setup.py and committing, and creating a tag
21:15 GodGinrai you should commit
21:15 GodGinrai then review that it is worthy of a tag
21:15 GodGinrai then tag
21:15 vktec nicksloan: And how often is a release staging-bound?
21:15 GodGinrai why hardcode the version numbers?
21:15 vktec That too
21:15 terminal_echo joined #git
21:16 nicksloan that's just how Python packaging works. I haven't seen an alternative. Personally I don't love that either
21:16 vktec Right
21:17 vktec Anyway, how often do you need to do this?
21:18 nicksloan a few times a week across several repositories, but occasionally if we find bugs on staging, a few times in quick succession after the first one.
21:18 osse git config alias.doboth '!git commit -m "$2" && git tag -a -m "$2" "$1" && :'
21:18 * kbs wonders - the "you're holding it wrong" thoughts aside :) fewer keystrokes to have just scripted it by now maybe?
21:18 nicksloan (well, per repository)
21:18 kbs haha
21:18 osse usage:  git doboth v1.2.3 "This is a message"
21:18 vktec Hmm.. there's a very relevant xkcd that I can't seem to find
21:18 vktec Ah, got it
21:19 vktec https://xkcd.com/1205/
21:19 nicksloan kbs, osse: totally fair. I just wanted to make sure I wasn't overlooking something. Worth the effort to do it properly if it exists.
21:19 merced that xkcd is the best one against automation out there
21:20 vktec :)
21:20 vassagus joined #git
21:20 nicksloan the bigger gain in this is in no longer forgetting to do the tagging
21:21 lugzer joined #git
21:22 nicksloan it's not a huge deal either way, but it seemed like something that might have come up before so I thought I'd ask.
21:23 kbs osse gets the cigar for actual solutions, and I get to learn another useful trick :) all good
21:26 LionsMane joined #git
21:27 Shentino joined #git
21:27 Sample joined #git
21:27 merced left #git
21:28 Vinnie_win_p joined #git
21:29 Rodya_ joined #git
21:31 mellernoia joined #git
21:32 GodGinrai vktec: judging by that chart, he likely has already passed the time alotted
21:32 vktec :)
21:33 clemf joined #git
21:34 urda joined #git
21:34 Mannn joined #git
21:34 ddavidd joined #git
21:35 ddavidd Is there a way to par down on the .git directory?
21:35 amir107 joined #git
21:35 amir107 guys check out my friend video its about windows 10 it help you using it and fixing problems you never knew about https://www.youtube.com/watch?v=​X1nRXSxdnTo&amp;feature=youtu.be
21:35 amir107 guys check out my friend video its about windows 10 it help you using it and fixing problems you never knew about https://www.youtube.com/watch?v=​X1nRXSxdnTo&amp;feature=youtu.be
21:35 amir107 left #git
21:35 osse kbs: is it cuban?
21:35 blaman left #git
21:36 kbs osse: any kind you prefer :)
21:37 Blkt joined #git
21:38 ddavidd left #git
21:38 ddavidd joined #git
21:41 Dougie187 left #git
21:42 starkhalo joined #git
21:42 aidalgol joined #git
21:45 dsantiago joined #git
21:46 fuchstronaut joined #git
21:49 rubyonrailed joined #git
21:49 fahadash joined #git
21:53 cd-rum joined #git
21:56 nikivi joined #git
21:57 b1tchcakes joined #git
21:57 rkazak joined #git
22:02 p4trix joined #git
22:03 romerocesar joined #git
22:05 gravitation joined #git
22:06 chris2 joined #git
22:07 Sample joined #git
22:12 _nwkj86 joined #git
22:16 lugzer joined #git
22:17 clemf joined #git
22:18 Synthead joined #git
22:25 l4v2 joined #git
22:28 Electrometro joined #git
22:29 MattMaker joined #git
22:30 madewokherd joined #git
22:31 johnny56_ joined #git
22:31 rubyonrailed joined #git
22:32 fairuz joined #git
22:34 spudowiar joined #git
22:35 starkhalo joined #git
22:36 hahuang61 joined #git
22:37 red_ joined #git
22:37 tristanp joined #git
22:39 alansaul joined #git
22:47 reim joined #git
22:47 fuchstronaut joined #git
22:51 MattMaker joined #git
22:56 LeBlaaanc joined #git
22:56 stuartrexking joined #git
22:57 mischat joined #git
23:00 vdv joined #git
23:03 bluezone joined #git
23:06 EvilPenguin joined #git
23:07 Rodya_ joined #git
23:08 PabloR joined #git
23:10 SwiftMatt joined #git
23:10 lugzer joined #git
23:10 reim joined #git
23:11 endiruna joined #git
23:11 Ntemis joined #git
23:11 Ntemis hi
23:11 terminal_echo joined #git
23:12 Ntemis i have installed a new os and i dont have my github key right now
23:12 Ntemis i have my fingerprint though
23:12 Ntemis how i can add it to git so i can work atm?
23:13 Ntemis my Fingerprint is like c3:v5:etc
23:13 Ntemis any help?
23:13 shgysk8zer0 joined #git
23:17 bernardio joined #git
23:17 gkatsev why not generate a new key and add it to your account?
23:18 zmachine joined #git
23:19 MattMaker joined #git
23:20 PabloR When starting a new repo in github it asks you if you want to add a .gitignore file. Is this file any different and in addition to the global .gitignore I create locally on my computer? Should I add it for potential users that may want to use the repo even though I have the global .gitignore because they can’t access it?
23:21 mattcen joined #git
23:21 happy-dude joined #git
23:22 EvilPenguin joined #git
23:22 magicTurtle joined #git
23:23 Ntemis gkatsev: do i need two?
23:24 The_Phoenix joined #git
23:24 gkatsev Ntemis: it doesn't hurt. I generally have one per machine I use, fwiw.
23:25 ojacobson Ntemis: things you can recover your ssh keys with: a copy of your ssh keys
23:25 ojacobson things you _can't_ recover your ssh keys with: the public keys, the fingerprints, the label
23:25 ojacobson I generally make a new key on each machine, it's the easiest way and then I can revoke them individually if I need to, too
23:25 gkatsev PabloR: besides that, it is also often used for ignoring repo/project specific stuff and makes it work for other users as well
23:26 gkatsev PabloR: built files are often ignore in a local gitignore
23:26 gkatsev and dependencies that should be re-installed by the user
23:26 jon_ross joined #git
23:28 kn-928 joined #git
23:29 PabloR gkatsev so at minimum it should have all the files and type of files that my local .gitignore, correct?
23:32 t-mart joined #git
23:34 jimi_ joined #git
23:35 dsdeiz joined #git
23:44 despoil joined #git
23:45 vassagus joined #git
23:49 bbhoss_ joined #git
23:49 Ntemis made a new key and it works
23:49 Ntemis thanks
23:49 Ntemis left #git
23:51 YourTheBombNow23 joined #git
23:52 zivester joined #git
23:54 janx_ joined #git
23:54 cstrahan_ joined #git
23:58 kal0pr joined #git

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