Perl 6 - the future is here, just unevenly distributed

IRC log for #git, 2017-03-18

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

All times shown according to UTC.

Time Nick Message
00:00 rchavik joined #git
00:00 cdown joined #git
00:01 TbobbyZ joined #git
00:05 Doginal_ joined #git
00:08 mda1 joined #git
00:10 tristanp joined #git
00:16 GodGinrai joined #git
00:18 roelmonnens joined #git
00:18 MrcRjs joined #git
00:26 brent__ joined #git
00:28 iveqy joined #git
00:33 jeffreylevesque joined #git
00:34 dreiss joined #git
00:37 TbobbyZ joined #git
00:45 Levex joined #git
00:46 durham_ joined #git
00:47 Cabanossi joined #git
00:50 jimi_sanchez joined #git
00:51 ozmage joined #git
00:53 miha_S7_ joined #git
00:54 vamiry left #git
00:55 joshszep joined #git
00:56 jstimm joined #git
00:56 MrcRjs joined #git
00:57 vamiry joined #git
00:58 yaewa joined #git
00:59 jon-mac joined #git
01:00 borei joined #git
01:00 borei hi all
01:01 borei im new one with git, need some heads-up with git flow.
01:02 borei i've init git-flow, but by misstake started branch under feature (we don't use feature at all), is there possibility to move my branch under development
01:03 borei sorry if im asking not properly
01:06 noteness joined #git
01:06 cdown joined #git
01:07 dviola joined #git
01:08 miczac joined #git
01:08 dsdeiz joined #git
01:08 dsdeiz joined #git
01:09 cdown_ joined #git
01:09 Doginal joined #git
01:10 kexmex joined #git
01:11 levex_ joined #git
01:11 cdown joined #git
01:12 tyreld joined #git
01:14 matoro joined #git
01:19 gugah joined #git
01:20 tristanp joined #git
01:20 CEnnis91 joined #git
01:26 dreiss joined #git
01:35 dreiss joined #git
01:38 Lyqyd joined #git
01:38 ReptilianBrain joined #git
01:39 arescorpio joined #git
01:41 kbs joined #git
01:46 durham joined #git
01:46 pks joined #git
01:46 durham joined #git
01:49 thiago joined #git
01:50 borei left #git
01:52 clandestino joined #git
01:53 noteness joined #git
01:54 miha_S7_ joined #git
01:56 pks joined #git
02:00 shinnya joined #git
02:00 kronos joined #git
02:00 kronos left #git
02:03 rangergord joined #git
02:06 ozmage joined #git
02:06 rangergord Hiya. How can I diff a current DIRTY file, and the same file on a different branch? I found out how to do it when I have two hashes/commits, but not for a dirty file. eg, instead of "git difftool hash1:package.json hash2:package.json" I want soemthing like "git difftool package.json hash2:packagekson".
02:06 rangergord I get "hash2:package.json no such path in the working tree"
02:07 therealgarrett joined #git
02:08 ggg joined #git
02:08 roelmonnens joined #git
02:08 matoro joined #git
02:10 MrcRjs joined #git
02:11 pks_ joined #git
02:12 therealgarrett left #git
02:12 kljfdlkj joined #git
02:12 menip joined #git
02:13 NeXTSUN joined #git
02:13 kljfdlkj left #git
02:14 kljfdlkj joined #git
02:14 MrcRjs joined #git
02:18 Rodya_ joined #git
02:18 ShadeS joined #git
02:19 MrcRjs joined #git
02:21 d^sh joined #git
02:21 ojacobson git diff hash2 -- package.json
02:22 msafi joined #git
02:22 msafi left #git
02:23 raulp joined #git
02:25 rangergord cheers ojacobson
02:26 a3Dman joined #git
02:26 Noodlewitt joined #git
02:27 Guest39635 left #git
02:27 sfix joined #git
02:31 Chinggis6 joined #git
02:35 weltall joined #git
02:38 weltall joined #git
02:40 minn I need to reconstruct commit histories in memory (following ancestors from a given revision). It would be useful although not necessary for `git rev-list --full-history` to output parents and children, so I don't need to recover children later. Is there any way to combine these options, or do I need two passes over the data?
02:46 duderonomy joined #git
02:46 aidalgol joined #git
02:46 NeverDie joined #git
02:46 DarkPsydeLord joined #git
02:46 weltall joined #git
02:46 tristanp joined #git
02:49 ilbot3 joined #git
02:49 Topic for #git is now Welcome to #git, the place for git help. Just git help, apparently. | We are aware of the SHA-1 collision. See: http://jk.gs/t/a4 | Public logs at http://jk.gs/glog | First visit? Read: http://jk.gs/git | Current stable version: 2.12.0 | Getting "cannot send to channel"? /msg gitinfo .voice | This is serious business. https://madeitwor.se/git-jokes
02:51 dreiss joined #git
02:51 eclecticjohny joined #git
02:51 Jackneillll joined #git
02:51 weltall joined #git
02:52 fstd_ joined #git
02:53 weltall joined #git
02:54 ayogi joined #git
02:54 _rgn joined #git
02:54 hyperair joined #git
02:54 ahrs joined #git
02:55 dwts joined #git
02:55 miha_S7_ joined #git
02:55 tmg joined #git
02:56 Karazhan joined #git
02:56 weltall joined #git
02:57 danmackay_ joined #git
02:57 chipotle joined #git
02:57 ribasushi joined #git
02:58 qsx joined #git
02:59 Sonderblade joined #git
02:59 ToxicFrog joined #git
03:00 bwn joined #git
03:01 cust joined #git
03:03 cyphase joined #git
03:05 ojdo joined #git
03:05 orbyt_ joined #git
03:08 roelmonnens joined #git
03:09 fahadash joined #git
03:10 raulp joined #git
03:17 Goplat joined #git
03:19 lagothri1 joined #git
03:24 dreiss joined #git
03:24 chachasmooth joined #git
03:25 tristanp joined #git
03:26 iveqy joined #git
03:32 Cabanossi joined #git
03:34 durham joined #git
03:42 amdi_ joined #git
03:42 dreiss joined #git
03:43 ertes joined #git
03:44 Mobutils_ joined #git
03:48 justan0theruser joined #git
03:49 hexagoxel joined #git
03:53 lucido-cl joined #git
03:55 Vortex35 joined #git
03:56 miha_S7_ joined #git
03:57 seni joined #git
04:01 _Vi joined #git
04:03 clandestino joined #git
04:11 venmx joined #git
04:13 mquin joined #git
04:19 pks joined #git
04:19 Rotaerk joined #git
04:28 anth0ny joined #git
04:29 hexagoxel joined #git
04:29 anth0ny I seem to be consistently running into this problem where I can’t see remote branches on my repos
04:30 cluelessperson joined #git
04:30 anth0ny I do either “git pull” or “git fetch” and the “git checkout branchname” and get the following error: error: pathspec ‘branchname’ did not match any file(s) known to git.
04:30 anth0ny any ideas as to how this could be happening?
04:30 anth0ny I’ve recloned the repo and it’s happened again
04:31 anth0ny no remote branches come up with “git branch —all”
04:32 Cabanossi joined #git
04:32 psprint_ joined #git
04:35 kljfdlkj left #git
04:36 root4 joined #git
04:37 dendazen joined #git
04:43 mehola joined #git
04:49 fold4 joined #git
04:49 Deruyter joined #git
04:52 diogenese joined #git
04:54 thiago anth0ny: assume git is right
04:54 thiago anth0ny: there's no branch with the name "branchname"
04:55 anth0ny thiago: no, there is… I can see it on bitbucket
04:55 anth0ny running “git ls-remote origin”, I can see it
04:56 thiago web interfaces don't count
04:56 thiago the branch doesn't exist in your repository
04:56 thiago don't argue with the evidence. It doesn't exist.
04:56 thiago now that you know it doesn't exist, try to figure out how to make it exist
04:57 miha_S7_ joined #git
04:57 thiago you fetched it from the remote?
04:57 thiago how?
04:59 anth0ny oh, you mean locally
05:00 anth0ny thiago: “git pull”
05:00 Andrew_K joined #git
05:00 anth0ny git fetch origin
05:00 Cabanossi joined #git
05:02 gkatsev joined #git
05:05 anth0ny ah, I think I’ve found it
05:05 anth0ny for some reason, [remote “origin”] in .gitconfig had “fetch = +refs/pull/*/head:refs/remotes/origin/pr/*”
05:05 anth0ny turning it to “fetch = +refs/heads/*:refs/remotes/origin/*” seems to have done the trick
05:06 dmc joined #git
05:08 mizu_no_oto joined #git
05:10 thiago indeed
05:14 mkoskar joined #git
05:27 andrew710 joined #git
05:30 Cabanossi joined #git
05:37 anth0ny joined #git
05:43 SwiftMatt joined #git
05:43 hexagoxel joined #git
05:50 jeffreylevesque joined #git
05:58 miha_S7_ joined #git
06:01 dendazen joined #git
06:01 finalbeta joined #git
06:05 ayogi joined #git
06:05 scoobertron joined #git
06:08 impliednude joined #git
06:08 pks joined #git
06:08 miha_S7_ joined #git
06:09 Derperperd joined #git
06:12 venmx joined #git
06:18 mkoskar joined #git
06:20 mkoskar joined #git
06:21 hexagoxel joined #git
06:23 freimatz joined #git
06:32 pks joined #git
06:38 dreiss joined #git
06:42 amrits joined #git
06:42 Vampire0 joined #git
06:46 hexagoxel joined #git
06:48 hashtagstrashtag joined #git
06:51 kristoff_ joined #git
06:54 khfeng joined #git
06:57 lordjancso joined #git
06:59 ankit01ojha joined #git
06:59 ToxicFrog joined #git
07:00 Cabanossi joined #git
07:03 acetakwas joined #git
07:07 freimatz joined #git
07:07 diogenese joined #git
07:08 BackEndCoder joined #git
07:10 ToxicFrog joined #git
07:13 nowhere_man joined #git
07:17 miha_S7_ joined #git
07:29 drodger joined #git
07:29 alexandre9099_ joined #git
07:37 z8z joined #git
07:38 tmg joined #git
07:39 Xtreme joined #git
07:40 Xtreme Hello guys, I want to update "all" branches with master, whenever anything is merged (ie there is a commit) to master. but merge only if there is no merge conflict.
07:40 j08nY joined #git
07:40 pks joined #git
07:40 Xtreme I am using bitbucket (on-premise)
07:40 _ikke_ You'd have to create a script that does that
07:41 Xtreme _ikke_, any pointers?
07:41 _ikke_ git for-each-ref to get a list of branches
07:41 _ikke_ git rebase to rebase onto master
07:41 Xtreme but i dont understand why bitbucket doesnt have this feature..
07:42 Xtreme _ikke_, sure about rebase and not merge?
07:42 _ikke_ Because this is not something trivial you want to to in an automatic fassion
07:42 _ikke_ Would create a messy history
07:42 _ikke_ hard to follow
07:42 Xtreme hmm. good point.
07:43 Xtreme sure about rebase though? not merge?
07:43 _ikke_ and rebasing branches that contain merges is not a good idea either imo
07:43 _ikke_ Is there a specific reason you want to do that everytime for every branch?
07:44 Xtreme remove overhead from developers to update their working branch manually.
07:45 Cabanossi joined #git
07:45 Xtreme _ikke_, how to detect if there is a merge conflict before merging?
07:45 _ikke_ Xtreme: You cannot, but you can abort when it does happen
07:45 Xtreme _ikke_, sounds like a solution.
07:45 Xtreme Thanks man. much appreciated.
07:46 _ikke_ Xtreme: But rebasing working branches of others like this can also result into problems
07:46 _ikke_ Xtreme: if they do just git pull, they would merge the old and new history together
07:46 Xtreme ahmm. true..
07:47 _ikke_ It's better to have a good workflow
07:47 JanC_ joined #git
07:48 _ikke_ Not sure if you are working with merge requests, but a good step would be to rebase before doing the pull request
07:48 _ikke_ rebase + test
07:48 _ikke_ It's better if people get familiar with the changes on master (instead of having them automatically get it)
07:49 _ikke_ and make a choice whether they need that change right now, or not (if it's an unrelated change, it's not necessary to immediately get it in your feature branch)
07:52 buzzLightBeer joined #git
07:58 Sound joined #git
07:58 lordjancso joined #git
07:59 hhee2 joined #git
08:01 oskarkv joined #git
08:04 miczac joined #git
08:04 SwiftMatt joined #git
08:04 blahdodo joined #git
08:09 NeverDie_ joined #git
08:12 venmx joined #git
08:14 Murii joined #git
08:16 Chinggis6 joined #git
08:17 kandium joined #git
08:23 senaps joined #git
08:33 venmx joined #git
08:34 Cassiopaya joined #git
08:34 cluelessperson left #git
08:38 imack joined #git
08:47 t0by joined #git
08:49 treehug88 joined #git
08:50 dvaske joined #git
08:52 imack joined #git
08:52 YuGiOhJCJ joined #git
08:58 sbulage joined #git
09:00 irqq joined #git
09:00 seishun joined #git
09:02 seishun joined #git
09:04 ok91 joined #git
09:05 atomi joined #git
09:06 hexagoxel joined #git
09:07 [Brain] joined #git
09:11 s1scha joined #git
09:15 Cabanossi joined #git
09:17 Levex joined #git
09:22 navidr joined #git
09:24 schleppel joined #git
09:25 imack joined #git
09:25 Levex joined #git
09:26 theoceaniscool joined #git
09:28 no_gravity joined #git
09:31 miha_S7_ joined #git
09:31 SteffanW joined #git
09:36 no_gravity How do you get the commit list of a file?
09:40 basiclaser joined #git
09:40 miha_S7_ joined #git
09:40 lucasem joined #git
09:41 Levex joined #git
09:46 MtWGA joined #git
09:46 Levex joined #git
09:49 afuentes joined #git
09:52 mehola joined #git
09:55 vamiry_ joined #git
09:58 scoobert1on joined #git
10:00 andrew710 joined #git
10:06 cdown joined #git
10:09 sdlnv joined #git
10:10 cdown_ joined #git
10:13 hexagoxel joined #git
10:13 choice joined #git
10:15 n1ce_ left #git
10:15 muhannad__ joined #git
10:16 babs joined #git
10:20 drodger joined #git
10:21 saul joined #git
10:23 sbasso joined #git
10:29 lucido-cl joined #git
10:30 Cabanossi joined #git
10:33 dsdeiz joined #git
10:36 shinnya joined #git
10:37 lmatteis joined #git
10:39 vuoto joined #git
10:40 seishun joined #git
10:43 rchavik joined #git
10:49 miczac joined #git
10:49 Jackneill joined #git
10:50 bariscant joined #git
10:51 vuoto joined #git
10:52 mar77i joined #git
10:52 stoopkid joined #git
10:55 Balliad joined #git
10:55 qsx is no_gravity the evil sister of grawity?
11:03 choice joined #git
11:06 hexagoxel joined #git
11:07 rivarun joined #git
11:08 howdoi joined #git
11:09 rivarun joined #git
11:10 al-damiri joined #git
11:11 rominronin joined #git
11:15 Eryn_1983_FL joined #git
11:15 Cabanossi joined #git
11:18 drodger joined #git
11:18 marianina8 joined #git
11:19 miha_S7_ joined #git
11:19 saul joined #git
11:22 ayogi joined #git
11:23 drodger joined #git
11:24 ToBeCloud joined #git
11:27 chris_wot joined #git
11:27 sbasso joined #git
11:29 miha_S7_ joined #git
11:30 alexandre9099 joined #git
11:38 Jackneill joined #git
11:39 dersand joined #git
11:42 lordjancso joined #git
11:44 plasticfish joined #git
11:45 plasticfish should merge be done on the command line or in github?
11:46 acetakwas joined #git
11:46 plasticfish and if I do it on github, what shall I do on the command line to refresh my local copy?
11:47 plasticfish also,  how do I do a pull request from command line?
11:48 satifant joined #git
11:50 kexmex joined #git
11:55 Es0teric joined #git
12:00 Cabanossi joined #git
12:01 Es0teric joined #git
12:03 alexandre9099 joined #git
12:03 choice plasticfish: The pull request is "git pull <path>"
12:03 aax joined #git
12:06 nicoulaj joined #git
12:07 mar77i joined #git
12:08 dave0x6d joined #git
12:09 d0nn1e joined #git
12:11 Jackneill joined #git
12:11 Rodya_ joined #git
12:14 borkr joined #git
12:14 brent__ joined #git
12:15 d^sh joined #git
12:19 Tobbi joined #git
12:21 kandium joined #git
12:25 IRQ` joined #git
12:30 Cabanossi joined #git
12:31 heftig joined #git
12:34 Balliad joined #git
12:34 acetakwas joined #git
12:37 a3Dman joined #git
12:39 kexmex joined #git
12:42 brent__ joined #git
12:43 sespiros joined #git
12:44 robattila256 joined #git
12:46 Miha joined #git
12:51 matoro joined #git
12:52 mar77i_ joined #git
12:56 ayogi joined #git
12:57 oleo joined #git
13:00 Cabanossi joined #git
13:03 miha_S7_ joined #git
13:05 Jackneill joined #git
13:06 Xtreme guys
13:07 Xtreme please take a look at http://pastebin.com/57df6fq1
13:07 Xtreme i am stuck at line 20.
13:07 Xtreme i want to abort merge if there is conflict.
13:07 Xtreme how can i?
13:07 Xtreme goal: merge master to all branches if there are no conflicts
13:07 sarri joined #git
13:07 sarri joined #git
13:08 Emperor_Earth joined #git
13:12 Vampire0_ joined #git
13:14 vuoto joined #git
13:15 quite joined #git
13:15 quite joined #git
13:16 muhannad__ joined #git
13:19 ce3s joined #git
13:19 sbulage joined #git
13:19 bustle joined #git
13:19 sunri5e joined #git
13:19 bustle hi, anyone use git-flow or sourcetree?
13:20 Vampire0 bustle, regarding sourcetree !gui
13:20 gitinfo bustle: Graphical user interfaces are not supported here. If you want to get support, it needs to be through the git CLI. Reasons: 1) Because very few people here use the graphical interface. 2) Because giving instructions for GUI's is difficult. 3) The command line gives you a history of what commands you have executed.
13:20 quite joined #git
13:20 Vampire0 bustle, regarding git-flow you might simply state your question (of course you might state your question for both, but for sourcetree it is less likely that you get an answer)
13:21 bustle if im working with another dev, and he uses sourcetree and i use git-flow cli, will that be compatable?
13:22 grawity as far as Git itself is concerned, yeah
13:22 mar77i joined #git
13:22 grawity they all use the same kind of Git repository
13:22 bustle does sourcetree eg enforce naming constraints for feature branches, or in some other way break or make compatability error prone for those using the cli?
13:22 t0by joined #git
13:22 grawity no, as far as I know it's a completely generic Git client
13:23 grawity tbh I though it's git-flow that enforces weird naming constraints
13:24 bustle it does for soem things fwict, but eg for feature branches you can call them anything with gitflow
13:24 Vampire0 Xtreme, `git merge --no-edit master || git merge --abort`
13:28 Chinggis6 joined #git
13:29 pks joined #git
13:30 Captain_Beezay joined #git
13:31 tmsmith joined #git
13:32 IRQ` joined #git
13:33 Jackneill joined #git
13:36 miha- hi, can anyone help me with this: fatal: cannot create directory at 'node_modules/nodemailer-mailgun-transport/node_modules/mailgun-js/node_modules/proxy-agent/node_modules/pac-proxy-agent/node_modules/pac-resolver/node_modules/degenerator/node_modules/escodegen/node_modules/esutils/lib': No such file or directory
13:37 miha- i am spinning and do now know what could be an issue. I enabled long paths
13:37 miha- but the same
13:37 miha- tnx
13:37 suroegin joined #git
13:38 suroegin_ joined #git
13:38 finalbeta1 joined #git
13:39 dendazen joined #git
13:43 bremner miha-: what is that message from? what is trying to create the directory?
13:49 miha- bremner, this is the whole msg
13:49 miha- fatal: cannot create directory at 'node_modules/nodemailer-mailgun-transport/node_modules/mailgun-js/node_modules/proxy-agent/node_modules/pac-proxy-agent/node_modules/pac-resolver/node_modules/degenerator/node_modules/escodegen/node_modules/esutils/lib': No such file or directory
13:49 miha- warning: Clone succeeded, but checkout failed.
13:49 miha- You can inspect what was checked out with 'git status'
13:49 miha- and retry the checkout with 'git checkout -f HEAD'
13:50 miha- if i do git status i get this: .8) has joined
13:50 miha- * finalbeta1 (~finalbeta@ptr-e2rg2
13:50 miha- deleted:    node_modules/express/lib/application.js
13:50 miha- for a lot of files
13:51 miha- public folder and not whole node_modules folder was not cloned
13:51 miha- but the thing is this happens only on windows
13:51 miha- if i use osx it is ok
13:51 bustle left #git
13:51 vamiry_ joined #git
13:53 andrew710 joined #git
13:56 acetakwas joined #git
13:56 lucido-cl joined #git
13:57 m0viefreak joined #git
13:57 gugah joined #git
13:59 Xtreme Vampire0, thanks
13:59 Vampire0 Xtreme, yw
14:00 govg joined #git
14:00 livingstn joined #git
14:03 livingstn joined #git
14:05 hotpancakes joined #git
14:09 CussBot joined #git
14:09 livingstn joined #git
14:10 n-st joined #git
14:13 sbasso joined #git
14:14 amdi_ joined #git
14:15 Cabanossi joined #git
14:16 miha_S7_ joined #git
14:20 am2on joined #git
14:20 saintaquinas[m] joined #git
14:20 flo[m] joined #git
14:20 KevinMGranger[M] joined #git
14:20 Elliott[m]1 joined #git
14:20 M-meznak joined #git
14:20 dyce[m] joined #git
14:20 unclechu joined #git
14:20 dikiaap joined #git
14:20 musicmatze[m] joined #git
14:20 Remramm joined #git
14:20 aviraldg joined #git
14:20 monomon[m] joined #git
14:20 SpEcHiDe joined #git
14:20 UncleCJ joined #git
14:20 patrickr[m] joined #git
14:20 hiq[m] joined #git
14:20 elwan7[m] joined #git
14:20 Retropikzel[m] joined #git
14:22 livingstn joined #git
14:23 Murii joined #git
14:25 Derperperd joined #git
14:28 dsdeiz joined #git
14:28 dsdeiz joined #git
14:35 no_gravity joined #git
14:36 gugah joined #git
14:36 chipotle joined #git
14:40 nowhere_man joined #git
14:40 miha_S7_ joined #git
14:41 govg joined #git
14:41 Xtreme joined #git
14:46 dreiss joined #git
14:50 ok91 joined #git
14:50 anth0ny joined #git
14:51 andrew710 joined #git
14:55 Derperperd joined #git
14:55 bremner ah windows. and they left.
14:55 kpease joined #git
15:01 fahadash joined #git
15:01 no_gravity Doing "/list" gives me "
15:01 nowhere_man joined #git
15:01 invisbl_ joined #git
15:02 Gustavo6046 joined #git
15:03 King_Hual joined #git
15:05 bremner no_gravity: try /msg alis help
15:05 madewokherd joined #git
15:08 govg joined #git
15:08 nikitasius joined #git
15:10 marianina8 joined #git
15:12 publio joined #git
15:14 inflames joined #git
15:14 impliednude joined #git
15:15 skalpin geez, I just tried /list and it crashes glowing-bear
15:15 skalpin give me a command and I'll try it like the reckless developer I am :)
15:18 VicMackey joined #git
15:18 lordjancso joined #git
15:18 bremner rm - oh nvm
15:20 VicMackey Hello everybody. I need some help with commit messages. I have been reading about them, but when I am to write one I spend too much time in thinking in a good commit messsage. I know I have to use the present simple, like "Add X feature" "Remove this thing". But everytime I try to write one I can't help but feel like I am writing a bad commit message. For example, I have to commit a piece of code that adds walls to acertaing cell -mazes-, and I am
15:20 VicMackey thinking of a message like the following: "Add east and west walls generation". But I don't really know. Can you guys help me out in not screwing up?
15:20 j416 looks good to me
15:21 j416 VicMackey: why do you feel it is bad?
15:22 inflames joined #git
15:22 n1ce joined #git
15:22 VicMackey To be honest is a mixture of not being a native speaker and not having much experience with git. Therefore I rewrite the commit messages a thousand times because I am scared of making a typo or writing a message that does not follow correctly the sentence structure
15:23 VicMackey j416
15:23 j416 ah
15:24 j416 try to think of it like this. There are some commits (patches) and the commit message of each describes the action one takes by applying it
15:24 j416 applying that commit means "add wall generation"
15:25 linuxmodder joined #git
15:25 j416 as for the body, I keep it either the same form as the subject, or I just describe the patch itself "this commit does blabla"
15:26 VicMackey j416 I understand. Thank you for your advice. That makes me feel a little bit more comfortable now :)
15:26 VicMackey I'll try to apply your advice in my commit messages
15:26 j416 often I would describe the current state "X is broken in that blabla" and what this does to fix it "This fixes it by doing blabla so that blabla"
15:27 j416 a common mistake is to mix in past tense, avoid that
15:27 j416 see it as "current state" is whatever is before your commit
15:28 j416 then describe how your commit changes the current state
15:28 skalpin VicMackey: I think the fact that you're even thinking about it is a good sign
15:28 dersand joined #git
15:28 j416 ^ it's awesome
15:28 j416 I wish more people cared about writing proper commit messages
15:29 skalpin me too, which is probably why those articles get written in the first place
15:30 VicMackey [< j416 >] Thank you. I am just writing your advices down in order to not forget them. I have been reading a lot about the topic, and even have been visiting many popular repos to see which kind of commit messages people write. But whenever I had to face writing one I felt like super insecure about it. This helps a lot, seriously. [< skalpin >] Thank you for those kind words! Nonetheless not everyone knows your backgournd, and seeing what kind of replies
15:30 VicMackey Linus Torvalds give about incorrect messages scared the hell out of me lol
15:30 j416 you sparked my motivation to write another one; there can never be too many articles on proper commits :D
15:31 VicMackey Happy to read that :=)
15:31 drodger joined #git
15:32 j416 give me a year or two. :P
15:32 VicMackey Holly <3
15:32 j416 I'd have to start an article collection first
15:32 j416 anyway. Thanks
15:33 VicMackey Good luck with that then, and thank YOU, seriously :)
15:34 _Cyclone_ joined #git
15:35 j416 most welcome
15:36 j416 ask again if there is anything else
15:36 durham joined #git
15:36 govg joined #git
15:37 VicMackey b(^ω^)d
15:38 seishun joined #git
15:39 kbs joined #git
15:43 einnjo joined #git
15:44 a_thakur joined #git
15:45 Cabanossi joined #git
15:46 menip joined #git
15:47 xissburg joined #git
15:48 Vampire0 joined #git
15:48 robattila256 joined #git
15:50 peterbecich joined #git
15:50 skalpin If I have a feature branch A, and feature branch B. If i want some of the changes from B in A I can 'git merge --no-ff --no-commit' -> remove the things I don't want, and then commit them. Then I merge A back into master. Then I merge B back into master. Will the changes I removed during the B -> A merge be in master?
15:50 miczac joined #git
15:50 sotoz joined #git
15:51 hasc joined #git
15:52 amdi_ joined #git
15:53 hasc joined #git
15:55 d0nn1e joined #git
15:57 muhannad__ joined #git
16:00 MapMan left #git
16:01 kbs not an expert on this :) but it sounds like the merge-base for the B->master merge would be the same as the mergebase for the A->master merge?
16:02 skalpin kbs yes A and B are both branched from the same commit in master
16:04 Noldorin joined #git
16:04 einnjo joined #git
16:04 kbs I said it poorly - just looking at it as a diagram, would the merge-base for the B->master merge become the parent of the B->A merge on the B branch?
16:06 wooy left #git
16:06 seishun joined #git
16:07 skalpin :) I'm getting out my pencil and paper one sec
16:07 kbs at any rate... aiui at least, merge only considers three snapshots - the merge-base, and the two candidate branches; so at first blush it does seem to me that it would add those items, but maybe someone wiser here might point out git does something smarter
16:08 govg joined #git
16:09 sbasso joined #git
16:09 mizu_no_oto joined #git
16:14 Khisanth joined #git
16:15 Cabanossi joined #git
16:16 miha- joined #git
16:19 skalpin kbs thanks! drawing helped me figure this one out.
16:19 kbs :-) glad to be a sounding board, if nothing else
16:21 endiruna joined #git
16:21 Gustavo6046_ joined #git
16:22 skalpin You're right, it's not a problem unless merge A->B becomes the merge-base for merge B->master.
16:24 Impaloo joined #git
16:24 kbs and pragmaically, git merge-base would probably quickly clarify what to expect
16:26 mar77i_ joined #git
16:27 kbs (one of these days, would love to listen to a darcs expert around their theory about merging - it looks a lot like rebase for the tree merge bits, and I wonder how that compares to a simple 3-way merge.)
16:27 Chinggis6 joined #git
16:29 oleo joined #git
16:30 Gustavo6046 joined #git
16:31 nevodka joined #git
16:32 iamthad joined #git
16:32 sotoz joined #git
16:33 dviola joined #git
16:34 halftroll left #git
16:36 acetakwas joined #git
16:38 govg joined #git
16:41 orbyt_ joined #git
16:43 jason237 joined #git
16:43 cdg joined #git
16:45 govg joined #git
16:50 ankit01ojha joined #git
16:50 cdg joined #git
16:51 tmsmith joined #git
16:51 govg joined #git
16:52 LeBlaaanc joined #git
16:56 mar77i joined #git
16:59 boboc_ joined #git
16:59 govg joined #git
16:59 boboc joined #git
17:00 raulp joined #git
17:00 Vampire0_ joined #git
17:03 boboc_ joined #git
17:09 cdown joined #git
17:10 boboc joined #git
17:10 Sound joined #git
17:11 sotoz joined #git
17:11 govg joined #git
17:11 shgysk8zer0_ joined #git
17:13 multi_io joined #git
17:14 sarbs joined #git
17:16 avar Is there some option I'm missing to make git diff follow symlinks?
17:16 avar I.e. show -foo +bar on this: git diff --no-index <(echo foo) <(echo bar)
17:17 grawity can't remember there being one
17:17 grawity use regular `diff -u` maybe
17:17 grawity pipe it to tig or colordiff
17:19 dfas joined #git
17:23 jimi_sanchez joined #git
17:23 Noldorin joined #git
17:25 lucido-cl joined #git
17:26 govg joined #git
17:29 muhannad__ joined #git
17:32 govg joined #git
17:34 jimi_sanchez joined #git
17:34 davimore joined #git
17:36 warthog9 joined #git
17:37 peterbecich joined #git
17:40 govg joined #git
17:40 Derperperd joined #git
17:41 Silenced joined #git
17:42 Silenced joined #git
17:42 dirtyroshi joined #git
17:45 acetakwas joined #git
17:46 govg joined #git
17:47 thiago joined #git
17:47 Rotaerk joined #git
17:50 Sound joined #git
17:53 Balliad joined #git
17:54 ayogi joined #git
17:56 mar77i_ joined #git
17:59 borkr joined #git
18:00 Cabanossi joined #git
18:03 bariscant joined #git
18:04 govg joined #git
18:11 ertesx joined #git
18:12 govg joined #git
18:16 navidr joined #git
18:17 ToBeCloud joined #git
18:21 govg joined #git
18:21 kbs joined #git
18:25 miha- joined #git
18:28 govg joined #git
18:31 kexmex joined #git
18:34 Puffball joined #git
18:35 govg joined #git
18:40 amdi_ joined #git
18:41 aw1 joined #git
18:41 astrofog joined #git
18:42 clandestino joined #git
18:44 muhannad__ joined #git
18:46 chitopunk joined #git
18:47 Spinkzeit joined #git
18:48 TbobbyZ joined #git
18:49 TbobbyZ joined #git
18:50 Derperperd joined #git
18:50 govg joined #git
18:50 TbobbyZ joined #git
18:51 TbobbyZ_ joined #git
18:51 wootehfoot joined #git
18:52 mizu_no_oto joined #git
18:54 Mobutils_ joined #git
18:56 Glooomy joined #git
18:56 jstimm joined #git
18:57 govg joined #git
18:59 ayogi joined #git
18:59 jason237 joined #git
18:59 ronny joined #git
19:05 sbasso joined #git
19:06 YuGiOhJCJ joined #git
19:06 kbs joined #git
19:08 govg joined #git
19:08 aw1 joined #git
19:09 einnjo joined #git
19:10 irqq joined #git
19:13 jimi_sanchez joined #git
19:15 Cabanossi joined #git
19:18 Derperperd joined #git
19:18 miczac joined #git
19:21 govg joined #git
19:27 govg joined #git
19:36 Seveas avar: no, there's not. I have a patch series for it, I need to send v3 to the list
19:37 Seveas avar: dk/diff-no-index-v2 on github.com/seveas/git.git
19:38 govg joined #git
19:45 wootehfoot joined #git
19:45 variable joined #git
19:46 endiruna joined #git
19:46 variable howdy all; is there a way to ask git if the diff is "empty" post conflict resolution ?
19:46 avar sweet
19:46 variable iow http://sprunge.us/YbdX
19:46 variable this diff is empty
19:46 variable since I resolved all the conflicts, but I didn't git add files
19:46 variable git diff-index --quiet HEAD --
19:46 variable returns 1 though
19:47 endiruna joined #git
19:47 sbasso joined #git
19:51 govg joined #git
19:52 eroussel joined #git
19:54 Atm0spher1c joined #git
19:55 miha- joined #git
19:55 miha- hello. this issue with long paths on windows, i have set it to true, but still the same thing for me. It done not clone all dir. Can I do something? tnx
19:58 lucido-cl joined #git
19:58 chitopunk joined #git
19:59 Goplat joined #git
19:59 dreiss joined #git
20:01 jozwior joined #git
20:03 Derperperd joined #git
20:07 jnewt joined #git
20:10 govg joined #git
20:10 mehola joined #git
20:10 mehola joined #git
20:10 choice joined #git
20:11 choice How do you list all commits that changed a certain file with git?
20:11 avar git log --all -- <path-to-file>
20:11 bariscant joined #git
20:12 choice avar: What is the "--" ?
20:12 avar just disambiguates file paths from branches or whatever, I don't know if you called your file "master" or "--ZoMG-hacks.txt" :)
20:12 BlackMaria joined #git
20:13 avar Anyway, writing it out like that is just in my muscle memory..
20:13 Atm0spher1c joined #git
20:13 choice I don't understand.
20:13 aw1 joined #git
20:14 choice I understand that "log" is a command, "--all" is a flag and <path-to-file> is an argument. But what is "--"?
20:14 avar On unix systems in general, if you do e.g. "ls *" if you have a file called "-l" in your directory, and another one called "foo" you'll just run the command "ls -l foo", which is not what you'd expect
20:14 btree joined #git
20:14 avar So to disambiguate this pretty much every command-line program on your system takes the form "ls --some --flags -- <list of files>" instead
20:15 avar But for git log in particular "git log <thing>" will try to resolve <thing> as a branch, or a file, or a tag or whatever
20:15 avar doing -- means you mean a file
20:15 choice I don't understand.
20:15 choice * is expanded by the shell.
20:15 avar I have this in my finger memory because if you type "git log <TAB>" with bash completion you'll get the full list of branches and whatnot, but "git log -- <TAB>" gives you the file
20:16 dreiss joined #git
20:16 avar Yes exactly, it's expanded by the shell, not the command, so the command just gets "-l foo" as an argument, it's indistinguishable from you having typed "ls -l foo" yourself
20:17 choice I still have no idea what "--" is.
20:17 Jellyg00se joined #git
20:17 choice "ls -- *" seems to do exactly the same as "ls *"
20:17 avar It's what command-line programs use to mean "anything before the -- was our list of options, if any, what follows is a list of files"
20:17 avar $ mkdir /tmp/test && cd /tmp/test && touch -- -l foo && ls *
20:17 avar -rw-r--r-- 1 avar avar 0 Mar 18 20:16 foo
20:17 avar $ ls -- *
20:17 avar foo  -l
20:18 avar What OS are you on? Maybe BSD ls doesn't do anything special because "foo" will sort before "-l"
20:18 choice linux
20:19 avar right, if you run that on linux you'll see that ls eats up command-line arguments from a file you have in your directory
20:19 govg joined #git
20:19 choice "ls *" does exactly the same.
20:19 choice as "ls -- *"
20:19 d1z joined #git
20:19 choice even if there is a file "-l" in the current dir.
20:19 choice ah no
20:19 imMute choice: try /bin/ls instead of just ls - maybe you have it aliased to something.
20:20 kbs choice: not to chase squirrels here :) but "--" is just another parameter to the git log command. Git commands generally use it as a "separator parameter", so the commands before the "--" are applied only to the files specified after the "--"
20:20 Atm0spher1c joined #git
20:21 irqq joined #git
20:21 choice Ok, I can see that "ls *" and "ls -- *" behave differently now.
20:21 oleo joined #git
20:21 choice I still don't get what "--" is.
20:21 trout choice "--" is often used to separate "flags" from "value paramaters"
20:21 imMute choice: it's an indicator to ls (or whatever command) that what follows shouldn't be interpreted as parameters/switches/options/whatever
20:22 choice imMute: Ah!
20:22 trout choice as an example if you have a file called "--diff"
20:22 trout the shell expands '*'
20:22 trout and then git gets it
20:22 choice So the "ls" command has that built in? The author of the "ls" command added the handling of "--"?
20:22 trout it doesn't know that it was previously a glob and should be a file
20:22 trout correct
20:22 imMute choice: yes.
20:23 trout though its fairly standard and common
20:23 choice I see.
20:23 choice How about just not having files starting with -?
20:23 davidptc joined #git
20:24 imMute choice: in the case of git, it's used to disambiguate between files and other things that might be named like files (such as branch and tag names)
20:24 choice So the command to list commits that involve a file is "git log --all <file>"?
20:26 imMute choice: assuming that <file> isn't the name of a branch or tag, sure.
20:26 mda1 joined #git
20:26 choice I see. And puttin "--" before the file would remove that ambiguity.
20:26 imMute correct.
20:27 avar Seveas: Looks like the conclusion from the thread is a lot of back & forth, but ultimately "just submit it with docs", right?
20:27 choice Wouldn't it make more sense like this? git log --all --file=<file>
20:27 govg joined #git
20:28 Seveas avar: I did do that as a v2, which got some feedback from Junio which I needed to address
20:28 Seveas running make test on the result now. Will send it out in a bit
20:28 avar Ah, gmail managed to not find the v2..
20:29 avar choice: If we're going to discuss unfortunate git cli interface choices we'll be here all night :)
20:29 imMute choice: but then I have to type 4 more characters.  and 7 more for every additional file.
20:29 avar But in general, it's obviously a trade-off between verbosity and accuracy
20:29 Seveas avar: 20170113102021.6054-1-dennis@kaarsemaker.net
20:30 avar yup, found it
20:30 Cabanossi joined #git
20:30 choice avar, imMute: ok.
20:30 choice "git log" shows all commits ever done in a repository?
20:30 avar Much better would be to change every *nix command to accept arguments as schema'd and versioned XML on stdin as an argument
20:31 avar git log --all --full-history -- <file> should give you all commits ever to a file in any ref, as long is it hasn't been renamed
20:31 avar Then you need --follow, -M<num>% etc.
20:32 TbobbyZ joined #git
20:32 Glooomy joined #git
20:33 choice What is the "--all" for?
20:33 btree left #git
20:33 Glooomy joined #git
20:34 Atm0spher1c joined #git
20:34 Glooomy joined #git
20:35 Glooomy joined #git
20:35 imMute avar: +1   ;)
20:35 Vampire0 joined #git
20:36 Glooomy joined #git
20:36 choice I get the same output with and without "--all".
20:37 SwiftMatt joined #git
20:37 avar Have you read the description of --all in "git help log" and don't get it, or are we just going to be paraphrasing the documentation for you?:)
20:38 choice I have read it.
20:38 choice I have no idea what "refs" are.
20:39 avar The most common refs are branches, --all would just give you both commits to <file> if you'd made two changes to it, one on each branch
20:39 avar but you likely don't have that, so in your case --all is just the same as without --all
20:40 choice If I only have one branch (the master branch) I never have to care about --all?
20:40 avar yes
20:40 choice Awesome.
20:40 * choice makes a not to never get into branches.
20:41 choice s/not/note
20:41 avar A branch fell on a distant relative of mine and he died, better to stay away
20:41 imMute choice: uh... well I suppose if you want to avoid a really powerful feature of git.
20:41 govg joined #git
20:42 jozwior joined #git
20:42 choice imMute: It's very rare that I feel the urge to have a branch of something.
20:42 hfp joined #git
20:42 choice I have been coding more then fulltime for 30 years now. And I think there is exactly one project where I have 2 branches.
20:43 hfp_work joined #git
20:44 avar If you cloned a git repo from somewhere you already have at least 2 branches right now!
20:44 choice Well, I just call them "repos".
20:44 choice "my repo", "his repo" etc.
20:44 choice But I haven't used Git much.
20:45 choice I wrote my own version control over 10 years ago and it's still my favorite.
20:45 imMute choice: to each their own.  I have a project I work on full time and I have 13 branches just for me.   75 between the other developers.
20:46 choice imMute: I see. That's a lot.
20:46 choice I just started a new startup with another guy. And because he already knows Git, we will use it.
20:46 choice To my horror, today I discovered that he knows almost nothing about Git :)
20:47 choice So now I have to learn it so I can teach it to him ...
20:48 hfp joined #git
20:48 robattila256 joined #git
20:51 avar You should work at a Fortune 500 company like I do instead, then you could have hundreds, nay thousands of collegues who use git and know nothing about it, except enough to get themselves into trouble!
20:51 avar <- the pain is real :)
20:51 choice I love startups.
20:52 choice I like that it's a multidisciplinary challenge.
20:52 choice And I like it that I am God in my own little universe.
20:53 choice And the reward, when it takes off.
21:01 Seveas avar: v3 sent.
21:01 govg joined #git
21:01 Seveas avar: at least you can share the pain with me. And I keep locking it down to keep them out of trouble :)
21:02 Seveas Also, damn, we're fortune 500... when I joined we werent even fortune 1000 yet I think.
21:04 avar Fortune 308 according to fortune.com, "years on list = 4"
21:04 iveqy joined #git
21:04 d1z joined #git
21:07 tojo joined #git
21:12 mar77i joined #git
21:13 jost_ joined #git
21:20 chipotle joined #git
21:21 hfp joined #git
21:22 hfp_work joined #git
21:27 aw1 joined #git
21:28 govg joined #git
21:28 hfp joined #git
21:29 hfp_work joined #git
21:30 Atm0spher1c joined #git
21:32 boombatower joined #git
21:32 bekks hi all
21:32 choice hi bekks
21:33 bekks When having a repo, grown over quite a long time, and now deciding that everything older than 3 years is utterly irrelevant, is there a way to somehow "drop" all things > 3 years, and "compress" the repo?
21:34 bekks Like, merging everything to be not older than 3 years, dropt history > 3 years, etc.?
21:35 choice How much memory would that save you?
21:36 bekks Several GB, like 3-4GB.
21:37 choice One of the things I find questionable about git is that the whole structure is so complex/opaque.
21:38 choice There probably is no easy way to delete commits older then x.
21:38 Deruyter left #git
21:38 bekks Hmm, hmmm.
21:39 choice I think I tried to understand gits structure when Linus just released it.
21:39 choice But I couldn't really grasp it quickly.
21:39 choice I might give it another try these days.
21:40 cagedwisdom joined #git
21:40 CrystalMath joined #git
21:40 CrystalMath hi
21:40 gitinfo CrystalMath: 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.
21:40 kadoban If you want to change history like that, it's a matter of just creating a commit that squashes exactly the state at the cut-point you want into one commit and then rebasing.
21:41 jagob joined #git
21:41 CrystalMath is there any good version of git that works on windows XP?
21:41 CrystalMath by version i mean client for, sorry
21:41 choice kadoban: What would be the commands to do so?
21:41 CrystalMath any good client for git that works on windows XP?
21:43 vuoto joined #git
21:43 kadoban choice: checkout the commit you want, create a new orphan branch there, commit, then rebase the original. I'm not really willing to give exact commands because it's tricky and I'm distracted, basically guaranteed to screw it up. But the only uncommon command you need is 'git checkout --orphan' really, the rest is a fairly normal rebase I believe
21:44 hfp_work joined #git
21:44 hfp joined #git
21:44 garo joined #git
21:44 choice kadoban: Well, over my head at the moment. I don't even what "checkout" means.
21:44 govg joined #git
21:44 choice don't even *know*.
21:45 kadoban man git checkout <-- that
21:45 gitinfo the git-checkout manpage is available at http://jk.gs/git-checkout.html
21:45 kbs joined #git
21:45 CrystalMath nevermind, looks like version 1.8 works
21:45 kadoban But if you're not that comfortable, better to skip it for now, yeah.
21:45 CrystalMath we need this for ReactOS
21:45 choice kadoban: Well, it wasn't me who wants to delete old commits. It was bekks.
21:45 CrystalMath and... this is off topic here, as i'd like to say this to all open source projects: please think before you drop windows XP support...
21:45 choice I'm just interested in Git in general.
21:46 kadoban Oh
21:46 CrystalMath thanks anyway
21:46 CrystalMath left #git
21:46 kadoban Pro Git is a decent start, or hanging out here certainly helps
21:47 wcpan joined #git
21:49 Vampire0_ joined #git
21:52 govg joined #git
21:54 hasc_ joined #git
21:55 tmg joined #git
21:57 BSaboia joined #git
22:04 hfp joined #git
22:05 hfp_work joined #git
22:05 tax joined #git
22:05 fedepad joined #git
22:07 govg joined #git
22:08 Vampire0 joined #git
22:09 Cassiopaya joined #git
22:09 tristanp joined #git
22:14 sz0 joined #git
22:15 minn joined #git
22:15 Cabanossi joined #git
22:16 Vampire0_ joined #git
22:16 minn Is it possible to show blame attribution for lines with git log or git diff, or do I also need to look at git blame output?
22:17 _ikke_ joined #git
22:19 Cork joined #git
22:21 endiruna joined #git
22:21 hexa- joined #git
22:25 govg joined #git
22:26 neunon joined #git
22:28 btree joined #git
22:31 medeirosthiiago joined #git
22:33 Cork joined #git
22:35 Seveas heh, windows XP.
22:35 Seveas sod that.
22:36 btree left #git
22:38 IRCFrEAK joined #git
22:39 robattila256 joined #git
22:40 johnny56 joined #git
22:41 aielima joined #git
22:42 mar77i joined #git
22:42 davimore joined #git
22:43 Derperperd joined #git
22:48 alezandro joined #git
22:48 kexmex joined #git
22:49 anth0ny joined #git
22:49 govg joined #git
22:52 jimi_sanchez joined #git
22:52 chipotle joined #git
22:53 alexandre9099 joined #git
22:56 j416 minn: how would you imagine blame would look for a diff? What problem is it that you are trying to solve?
22:57 acetakwas joined #git
23:00 Cabanossi joined #git
23:11 Emperor_Earth joined #git
23:11 cdown joined #git
23:12 mar77i joined #git
23:15 SwiftMatt joined #git
23:16 gunnaro joined #git
23:28 Atm0spher1c joined #git
23:32 minn j416: I'm computing churn metrics. Knowing who was responsible for each line prevents looking that information up later (this can be slow). In retrospect I want to attribute authorship across renames and copies, so just knowing who git finds responsible for each line won't cut it.
23:32 Atm0spher1c joined #git
23:33 dhollinger joined #git
23:35 ToBeCloud joined #git
23:35 shgysk8zer0_ joined #git
23:42 irqq joined #git
23:45 anuxivm joined #git
23:45 tristanp joined #git
23:47 dreiss joined #git
23:50 Atm0spher1c joined #git
23:53 Rodya_ joined #git
23:54 alezandro joined #git
23:56 leeN joined #git
23:58 learning joined #git

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