Perl 6 - the future is here, just unevenly distributed

IRC log for #git, 2017-01-26

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

All times shown according to UTC.

Time Nick Message
00:00 ertes joined #git
00:00 Kaisyu joined #git
00:09 rkazak joined #git
00:13 rominronin joined #git
00:14 oaao joined #git
00:14 Gsham joined #git
00:15 Cabanossi joined #git
00:15 spacelord_ joined #git
00:18 Toordog still trying to figure out why git branch doesn't output all my branches.   running 1.9.1
00:19 kadoban Toordog: Does it output any of them?
00:24 Toordog not even master or remote origin with -r or -a
00:24 svm_invictvs joined #git
00:25 kadoban Toordog: What's 'git branch -a' output? And what's in your .git/config file I guess? Does it work if you checkout those branches that it's not showing?
00:25 Toordog i tried to create a new branch ... without success
00:25 Toordog https://hastebin.com/ebewaguxam.rb
00:25 Toordog the git branch -a return nothing
00:27 mostlybadfly joined #git
00:29 Toordog kadoban see anything suspicious at first sight?
00:29 Toordog maybe becasue the repo is empty and doesn't have data yet ?
00:30 kadoban Oh, do you have no commits at all?
00:30 Toordog no commit yet exact
00:30 kadoban That's probably why. Shit acts weird then a bit
00:30 Xionkana left #git
00:30 kadoban Orphan branches are strange
00:31 Toordog i wanted to create branch related to an environment (dev, prod) and create teh folder once checkout so when i switch from branch the folder that appear is the one related to teh environment
00:31 kadoban I'm not quite sure what you mean. But you can just do something like  'git commit --allow-empty -m "initial commit"' and try again from the start?
00:32 kadoban The empty initial commit makes a lot of git things work just better and saner. Having no commit at all is a bit of a wonky case in git.
00:32 Toordog oh boy, i created a file to commit something and it happens i did it in the dev branch and now it displays ... funny it couldn't let me know where i were before
00:33 Toordog great good to know, i wasn't aware could do an --allow-empty
00:33 kadoban It's not strictly necessary anytime I know of, but it does simplify some cases.
00:34 raspado joined #git
00:36 snowkidind joined #git
00:39 adino joined #git
00:41 dreaddymck joined #git
00:43 roelmonnens joined #git
00:44 Toordog is there a way to make sure we don't overwrite a branch with another branch ?  (we don't merge it)
00:44 KV joined #git
00:44 Toordog if they need to exist in total independance and protect integrity
00:47 masuberu joined #git
00:51 askb joined #git
00:52 Toordog time to take a break from work.  good night !
00:52 mizu_no_oto joined #git
00:52 _geoff joined #git
00:57 texinwien joined #git
01:00 Vampire0 anddam, after the rebase of course. Or in which respect does it not look how you want?
01:04 ome joined #git
01:07 CheckDavid joined #git
01:14 oven joined #git
01:15 Cabanossi joined #git
01:24 Hudu joined #git
01:29 jameser joined #git
01:30 daynaskully joined #git
01:30 adino joined #git
01:30 mischat joined #git
01:31 SwiftMatt joined #git
01:31 xissburg joined #git
01:34 dirtyroshi joined #git
01:35 lucasem joined #git
01:37 s2013 joined #git
01:44 Zeranoe joined #git
01:44 rominronin joined #git
01:45 Zeranoe How do I discard all changes made to a git repo (uncommited), and reset the state to master?
01:45 KV joined #git
01:47 lucasem Zeranoe:  you could `git stash` if you want to keep your changes. If not, I usually just do a hard reset with `git reset --hard`
01:48 Zeranoe lucasem: It looks like that doesn't remove newly created files?
01:49 lucasem If they aren't being tracked by git, you should remove them without using git
01:49 publio joined #git
01:51 filcab joined #git
01:51 osse Despite that, git clean exists
01:51 osse Might be of use
01:51 idmistir joined #git
01:51 idmistir joined #git
01:51 multi_io joined #git
01:52 adino joined #git
01:52 reactormonk joined #git
01:53 cmalmber_ joined #git
01:53 alesan joined #git
01:54 isbotnetcool joined #git
01:58 BSAlb joined #git
02:02 Gsham joined #git
02:06 bgerber joined #git
02:10 Wind0r joined #git
02:15 Cabanossi joined #git
02:21 Derperperd joined #git
02:21 menip joined #git
02:21 rt joined #git
02:22 drbean joined #git
02:23 finalbeta joined #git
02:24 pks joined #git
02:26 btree joined #git
02:27 btree left #git
02:27 Mavericks joined #git
02:27 Mavericks The -f (force flag) is recommended in order to avoid conflicts with other developers’ pushes. Since you are not using Git for your revision control, but as a transport only, using the force flag is a reasonable practice.
02:27 Mavericks at https://devcenter.heroku.com/articles/git
02:27 Mavericks I don't understand that . How's that a reasonable practice ?
02:27 salamanderrake_ joined #git
02:28 kadoban Because heroku is stupid as fuck
02:28 Mavericks Wouldn't that break the linear history that one can get with a proper rebase ?
02:28 Mavericks What exactly happens when -f flag's used ?
02:28 Mavericks kadoban:  ok
02:28 kadoban Mavericks: They're not completely wrong, the part I disagree with is them using git like this.
02:29 kadoban Mavericks: But, they're just using git as a transport/deployment mechanism, the thing you deploy ... doesn't really correspond to a sane git history necessarily.
02:30 Mavericks kadoban: ok
02:31 mischat joined #git
02:33 Mavericks i read about the git push -f flag at https://git-scm.com/docs/git-push
02:34 Mavericks it's quite dangerous in the sense that commits are lost!
02:34 kadoban Mavericks: Oh, yes so what it does is simply overwrite the remote ref with whatever you tell it to. It's quite dangerous in a shared history sense, but what you deploy to heroku probably shouldn't correspond to an actual shared history you use with others anyway.
02:35 Mavericks What happens is I've a->b->c->d (remote) and someone tries to push a->b->c'->d'->e (local ref) with -f
02:35 ojacobson if the remote permits it, the branch becomes a->b->c'->d'->e
02:36 ojacobson if not, nothing
02:36 Mavericks then the new remote will be a->b->c'->d'->e with dangling branch (c->d)
02:36 ojacobson git-push copies refs exactly
02:36 ojacobson unless d has some other name, c and d will become unreachable and mostly unfetchable
02:36 Mavericks so commits c and d are lost meaning they can be seen using git reflog or something
02:37 ojacobson they could be seen in the reflog on the remote, if it's enabled and if you have access to it
02:39 Mavericks kadoban:  ojacobson:  i see
02:40 kadoban It helped me when I stopped thinking of heroku deployment as actually using git and just considered it like ... a replacement for sending a tarball. It should really be pretty separate from your actual development history.
02:40 ojacobson (On Heroku, I believe it's not stored. However, you could use heroku-repo to fetch the git repo as a tarball and dissect it.)
02:40 ojacobson Sure. Heroku's git server is a frontend for Heroku's sources and builds APIs, nothing more
02:41 ojacobson you can ignore it and call those APIs directly, they're fairly straightforward
02:41 Mavericks if the new remote's a->b->c'->d->e due to a new collaborator's push to remote. i can still do a rebase on my local repo. and've a->b->c->d->c'->d'->e instead (assume that it's desired order)
02:41 kadoban I didn't know they had another API to do that stuff, that's good. Never saw any writeups when I was using it except the ones that used git.
02:46 ojacobson kadoban: my team maintain both the APIs and the git server :)
02:46 kadoban :)
02:46 Mavericks and do a git-push -f for remote to retain those commits
02:46 ojacobson we should do a better job of promoting the API!
02:46 ojacobson I've been meaning to sketch out a CLI command that submits your local tree to a review app
02:46 pity joined #git
02:46 Mavericks kadoban: i see. makes more sense that way
02:46 kadoban Possibly, but it's been a while since I used heroku, so my experience might be quite out of date.
02:46 Mavericks ojacobson: i will note that down. i'd use heroku-repo more often
02:46 KV joined #git
02:49 ilbot3 joined #git
02:49 Topic for #git is now Welcome to #git, the place for git help and a fresh start on a new branch | Public logs at http://jk.gs/glog | First visit? Read: http://jk.gs/git | Current stable version: 2.11.0 | Getting "cannot send to channel"? /msg gitinfo .voice | This channel may contain users known to the state of California to cause cancer.
02:50 thiago joined #git
02:51 Vampire0_ joined #git
02:51 catsup joined #git
02:52 raspado joined #git
02:54 mizu_no_oto joined #git
02:55 d^sh joined #git
03:03 mephistorium joined #git
03:04 thiago joined #git
03:04 bluepixel joined #git
03:07 sunri5e joined #git
03:10 XenophonF joined #git
03:11 donught joined #git
03:12 blunaxela joined #git
03:12 donught joined #git
03:12 Omnipotent joined #git
03:12 Omnipotent if I do a git reset --hard origin/master then running a git pull origin master should not pull anything more, correct?
03:12 Omnipotent or it just overwrites my chances with my "local copy of" origin/master?
03:12 eroussel joined #git
03:13 BackEndCoder joined #git
03:13 kadoban It does a fetch and then a merge, it depends if the fetch does anything.
03:13 Omnipotent you are talking about pull right?
03:14 kadoban Yes
03:14 Omnipotent my question is more about, if I do git reset --hard origin/master  it should just overwrite all my changes with the latest commit on origin/master
03:14 Omnipotent right?
03:14 Omnipotent so immediately doing a git pull origin master should say "updated with master" am I right?
03:15 Omnipotent "Already Updated with Master" or something like that
03:15 brent__ joined #git
03:15 kadoban It will change your current branch to be exactly equal to whatever your local idea is of what origin/master looks like, yes. Including history.
03:15 kadoban It does a fetch and then a merge, it depends if the fetch does anything.
03:15 kadoban Where those "It"s are different
03:15 hahuang65 joined #git
03:16 Omnipotent lol
03:16 rominronin joined #git
03:16 Omnipotent I see.. so that's what "Local idea of origin/master" looks like
03:17 Omnipotent so doing a git --reset hard wouldn't rewrite all the changes as with the actual origin/master the server until I pull on it afterwards
03:17 Gsham joined #git
03:18 Vortex34 joined #git
03:19 kadoban 'git reset' doesn't do anything over the network, it just looks at your remote tracking branches
03:19 kadoban So yeah, it doesn't do a fetch, which is the part that updates those.
03:20 d1z joined #git
03:23 a_thakur joined #git
03:23 chachasmooth joined #git
03:24 fstd_ joined #git
03:29 mehola joined #git
03:32 mischat joined #git
03:33 Omnipotent kadoban, thanks!
03:33 Omnipotent understood why I saw different behaviours on different machines, I did a git reset --hard origin/master on one machine and then git pull origin master, which basically pulled the newer changes after reset because my local origin/master wasn't updated
03:33 Omnipotent but on the second machine, I did git pull origin master first... and then git reset --hard origin/master after which I again followed a git pull origin master because I wasn't sure... but it said "Already updated with origin master" so I was confused.
03:33 Omnipotent but since on second machine origin/master was already fetched ... it simply overwrote my master branch with origin/master and viola it was already updated.
03:34 kadoban Ah, sounds right
03:34 kadoban And yeah, anytime.
03:34 cjohnson cello it was already updated
03:34 lukky513 joined #git
03:34 cjohnson amiright
03:34 cjohnson I'll see myself out
03:34 freekevin joined #git
03:34 danielmahon joined #git
03:35 fahadash joined #git
03:37 govg joined #git
03:38 salamanderrake joined #git
03:43 dsdeiz joined #git
03:44 lb1c joined #git
03:44 plasticle joined #git
03:46 rivarun joined #git
03:46 KV joined #git
03:48 hwrdprkns joined #git
03:49 raspado joined #git
03:52 a_thakur joined #git
03:56 adino joined #git
04:04 sunri5e joined #git
04:04 hexagoxel joined #git
04:04 chipotle joined #git
04:04 fatalhalt joined #git
04:10 mizu_no_oto joined #git
04:10 boombatower joined #git
04:10 boombatower joined #git
04:11 roelmonnens joined #git
04:17 bcc joined #git
04:17 bcc joined #git
04:17 raspado joined #git
04:18 donught joined #git
04:22 d1z joined #git
04:23 miczac joined #git
04:24 ntnn joined #git
04:26 ecuanaso joined #git
04:29 Cabanossi joined #git
04:32 robattila256 joined #git
04:32 jameser joined #git
04:33 mischat joined #git
04:35 safe joined #git
04:36 bilb_ono joined #git
04:37 pur3eval joined #git
04:41 Muzer joined #git
04:43 halirutan joined #git
04:44 halirutan I have tagged and pushed a commit to GitHub. Now I saw that my built is broken. Is it possible to include further changes in the tagged commit?
04:45 bocaneri joined #git
04:45 kadoban No
04:46 kadoban Not without modifying history, and deleting/recreating the tag.
04:47 arnas joined #git
04:47 KV joined #git
04:48 halirutan kadoban: Damn, I was afraid of this. So I delete the tag local and on origin, make a further commit and re-tag?
04:49 rominronin joined #git
04:49 kadoban Well, that's a bit of a sketchy thing to do, because if anyone is using the history you published you're making their life quite annoying.
04:49 kadoban But if that doesn't apply or if you don't care, then yes.
04:49 kadoban The other option is just make a new commit and a different tag, which is fine really.
04:50 halirutan OK, no one except of me is pushing. Still, lets just create a new tag.
04:50 halirutan kadoban: Thanks!
04:50 diogenese joined #git
04:50 kadoban It really more matters if anyone but you is reading it. Even if you're the only one pushing, it can still greatly annoy people.
04:50 kadoban 'welcome
04:54 jameser joined #git
04:57 catsup joined #git
04:58 ecuanaso joined #git
04:58 arescorpio joined #git
04:58 jmpp joined #git
04:58 jmpp joined #git
05:03 oaaao joined #git
05:06 mmattice joined #git
05:06 pur3eval joined #git
05:07 HandheldPenguin joined #git
05:07 rebrec joined #git
05:07 McKraken_zzzzz joined #git
05:08 davidstrauss joined #git
05:08 _W_ joined #git
05:08 simon816 joined #git
05:12 pur3eval joined #git
05:14 milki joined #git
05:15 sostenuto joined #git
05:15 Vaelatern joined #git
05:16 GGMethos joined #git
05:19 Dreamer3 joined #git
05:20 Dreamer3 what is the with-blk-sha1 compile option?
05:20 johnny56_ joined #git
05:25 Goplat joined #git
05:27 gugah joined #git
05:28 pur3eval joined #git
05:28 nevodka joined #git
05:29 hahuang65 joined #git
05:34 jost joined #git
05:34 mischat joined #git
05:34 ndim joined #git
05:36 netj joined #git
05:36 ckruczek joined #git
05:36 z3t0 joined #git
05:38 _ikke_ Dreamer3: Where did you find it?
05:38 TomTomTo1 joined #git
05:38 navidr joined #git
05:40 a_thakur joined #git
05:41 Derperperd joined #git
05:42 a_thakur_ joined #git
05:43 neunon joined #git
05:48 KV joined #git
05:48 Omnipotent left #git
05:52 tirej joined #git
05:58 King_Hual joined #git
05:58 Dreamer3 BLK_SHA1 it's a env that you can set when compiling
05:58 Dreamer3 it's in the git source
05:59 tristanp joined #git
05:59 grawity so is the description, no?
05:59 thiago you can git grep for it to find out where it gets used
06:00 Noodlewitt joined #git
06:02 jmpp joined #git
06:02 jmpp joined #git
06:03 finalbeta joined #git
06:06 _ikke_ It's apparently a sha1 implementation shipped with git
06:07 _ikke_ "Define BLK_SHA1 environment variable to make use of the bundled optimized C SHA1 routine. "
06:07 _ikke_ in the Makefile
06:08 rscata joined #git
06:12 dec0n joined #git
06:13 masuberu joined #git
06:13 Guest82 joined #git
06:18 Dreamer3 just seems weird it's not a default
06:19 grawity afaik the default is to use OpenSSL's optimized C SHA1 routine
06:19 rominronin joined #git
06:19 overlord_tm joined #git
06:20 jaafar joined #git
06:21 _ikke_ correct
06:27 mephistorium joined #git
06:29 jmpp joined #git
06:29 jmpp joined #git
06:30 ismithers joined #git
06:31 donught joined #git
06:32 thiago_ joined #git
06:32 sparr joined #git
06:32 duderonomy joined #git
06:35 mischat joined #git
06:37 z3t0 joined #git
06:37 z3t0 joined #git
06:40 Nilesh_ joined #git
06:44 z8z joined #git
06:45 Dreamer3 i read that it can be slow to spoil up all the OpenSSL stuff
06:46 osse how does one spoil an .so or .dll
06:46 brillenfux joined #git
06:47 sbulage joined #git
06:48 Lauven joined #git
06:49 KV joined #git
06:50 ThomasLocke joined #git
06:52 zeroed joined #git
06:52 zeroed joined #git
06:53 dreiss joined #git
06:53 qt-x joined #git
06:57 Raging_Hog joined #git
06:59 Cabanossi joined #git
07:00 freimatz joined #git
07:00 cyan__ joined #git
07:01 cyan__ joined #git
07:02 cdown joined #git
07:03 cyan__ joined #git
07:03 zeroed joined #git
07:03 zeroed joined #git
07:04 zefferno joined #git
07:04 adino joined #git
07:06 ihradek joined #git
07:06 cyan__ joined #git
07:07 cyan__ joined #git
07:08 cyan__ joined #git
07:15 mda1_ joined #git
07:16 roelmonnens joined #git
07:18 ome joined #git
07:20 rominronin joined #git
07:21 johnny56 joined #git
07:22 JeroenT joined #git
07:23 JeroenT joined #git
07:26 svm_invictvs joined #git
07:28 nidr0x joined #git
07:30 n-st joined #git
07:32 jmpp joined #git
07:35 Eugene Spill beer on it
07:35 mischat joined #git
07:36 suy joined #git
07:36 themill joined #git
07:37 ome_ joined #git
07:38 pur3eval joined #git
07:39 osse If I want to send a patch on behalf of another person I add a new "From:" line in the body of the message, right?
07:40 donught joined #git
07:41 vuoto joined #git
07:42 chele joined #git
07:42 jnavila joined #git
07:44 Eugene Reply-To
07:44 Eugene >_>
07:44 osse <_<
07:44 osse also, I don't get it
07:44 ianmethyst joined #git
07:46 Alienpruts joined #git
07:46 roelmonn_ joined #git
07:46 barteks2x joined #git
07:47 elect joined #git
07:47 pity joined #git
07:47 buzzLightBeer joined #git
07:48 thiago_ osse: send where?
07:49 KV joined #git
07:50 osse thiago_: to a mailing list (not git's)
07:50 osse specifically, zsh-workers
07:51 osse ahh, from man git-am: ""From: " and "Subject: " lines starting the body override the respective commit author name and title values taken from the headers."
07:51 gitinfo the git-am manpage is available at http://jk.gs/git-am.html
07:53 Singmyr joined #git
07:56 Hobbyboy joined #git
07:57 JeroenT joined #git
07:58 JeroenT_ joined #git
07:59 JeroenT_ joined #git
08:00 Tobbi joined #git
08:01 johnny56_ joined #git
08:04 donught joined #git
08:04 JeroenT joined #git
08:05 m4sk1n joined #git
08:05 ThomasLocke_ joined #git
08:05 ThomasLocke_ joined #git
08:09 JeroenT joined #git
08:14 Cabanossi joined #git
08:14 Hudu joined #git
08:15 philipsd6 joined #git
08:23 mischat joined #git
08:24 mathematic-alpha joined #git
08:24 themill joined #git
08:25 suy joined #git
08:26 Hobbyboy joined #git
08:26 mehola joined #git
08:27 quadz joined #git
08:33 cdown joined #git
08:34 m1m3-50 joined #git
08:35 mathematic-alpha left #git
08:35 barteks2x joined #git
08:35 jknetl joined #git
08:37 rominronin joined #git
08:38 Limes joined #git
08:38 jsolano joined #git
08:38 GT4066 joined #git
08:38 Snugglebash joined #git
08:38 PioneerAxon joined #git
08:40 rominron_ joined #git
08:42 Balliad joined #git
08:43 PioneerAxon joined #git
08:44 Lauven joined #git
08:45 PioneerAxon joined #git
08:46 Lunatrius joined #git
08:47 XenophonF joined #git
08:48 mehola joined #git
08:49 pur3eval joined #git
08:49 jagob joined #git
08:49 Dreamer3 left #git
08:50 netj joined #git
08:50 KV joined #git
08:52 g_zen joined #git
08:58 justanotheruser joined #git
08:58 memfrob joined #git
08:59 memfrob Hi all, pulling an all nighter, 3am and feel awesome. Git time!
09:00 memfrob What's this "This channel may contain users known to the state of California to cause cancer" comment all about?
09:00 grawity https://github.com/EugeneKay/git-jokes
09:00 memfrob California is the state of high performance cars with also the strictest emissions laws. So ironic!
09:00 grawity hmm that's rather out-of-date
09:01 jmpp joined #git
09:01 jmpp joined #git
09:01 memfrob Everyone owns a 69 Shelby GT500 but nowhere to drive it XD
09:02 aard_ joined #git
09:02 Particle1olem left #git
09:02 sea-gull joined #git
09:02 ParticleGolem joined #git
09:03 kurkale6ka joined #git
09:04 armyriad joined #git
09:07 frobnic move over to germany.  there is about ~120 km Autobahn without speed limit right here, behind the next hill :)
09:09 Tobbi joined #git
09:09 stroybat joined #git
09:12 marcogmonteiro joined #git
09:13 bongjovi joined #git
09:15 pluszak left #git
09:17 blackwind_123 joined #git
09:21 MineCoins joined #git
09:21 gottaname|wurk joined #git
09:21 gottaname|wurk anyone knowes of a good git client with UI that supports git LFS?
09:23 memfrob linux from scratch?
09:23 osse Large File Storage
09:24 gottaname|wurk ^
09:24 memfrob ah.
09:24 gottaname|wurk I can'
09:24 gottaname|wurk I can't seem to find any docs saying tortise git supports it
09:24 gottaname|wurk only a feature request
09:24 * gottaname|wurk is planning on using git-labs and tortise git for a self hosted solution
09:25 osse i assume github's own client does
09:25 davimore_ joined #git
09:27 sfix joined #git
09:28 roelmonnens joined #git
09:28 jsolano joined #git
09:28 mischat joined #git
09:29 gottaname|wurk osse, can it support external git servers from github?
09:29 mathematic-alpha joined #git
09:30 osse i don't know
09:30 mischat joined #git
09:30 z3t0 joined #git
09:31 mischat joined #git
09:31 texinwien joined #git
09:33 mathematic-alpha left #git
09:33 b100s joined #git
09:33 clmsy joined #git
09:34 Hudu joined #git
09:36 nkuttler joined #git
09:37 finalbeta joined #git
09:42 Tobbi joined #git
09:43 chipotle joined #git
09:48 m1m3-50 joined #git
09:48 ShapeShifter499 joined #git
09:48 memfrob Hey whats the way to generate a patch using git but keeping the commit summary in the patch? git diff HEAD~ doesn't keep the summary
09:49 memfrob Entire commit message.
09:49 chll_ joined #git
09:49 Vgr_ joined #git
09:49 Vgr_ joined #git
09:49 Vgr_ joined #git
09:49 Vgr_ joined #git
09:49 mehola joined #git
09:49 Vgr_ joined #git
09:49 Vgr_ joined #git
09:50 selckin man git-format-patch
09:50 gitinfo the git-format-patch manpage is available at http://jk.gs/git-format-patch.html
09:50 Vgr_ joined #git
09:50 Vgr_ joined #git
09:50 memfrob git format-patch --stdout HEAD~
09:51 KV joined #git
09:51 a3Dman joined #git
09:52 theoceaniscool joined #git
09:54 memfrob Reason to use git apply over patch?
09:56 jnavila joined #git
09:57 Vampire0 memfrob, to apply the patch from format-patch?
09:58 Vampire0 memfrob, you should use man git am then, not apply
09:58 gitinfo memfrob: the git-am manpage is available at http://jk.gs/git-am.html
09:58 mikecmpbll joined #git
09:58 Limes joined #git
09:58 suy gottaname|wurk: atlassian's client also has LFS support
09:59 Vampire0 memfrob, If you have a plain patch without commit message, author etc, apply still makes more sense than patch, especially if you use additional parameters like --3way which a normal patch binary cannot provide as it does not understand Git objects
10:00 memfrob Ah, yeah, git am keeps the summary intact for git log etc
10:00 memfrob why does git apply even exist if git am is better?
10:01 memfrob git apply is basically just patch :/
10:01 Vampire0 memfrob, it is not better, but different
10:01 memfrob It's superior in the sense that it actually integrates the patch into your git tree.
10:01 memfrob Not just applying it to objects.
10:02 Vampire0 memfrob, git am is for applying a patch created with format-patch. It will create multiple commits from multiple format-patch formatted files, preserving the commit message and author information
10:02 djb-irc joined #git
10:02 memfrob Yeah and git apply doesn't do that
10:02 Vampire0 memfrob, git apply just applies a patch, even if it is not formatted like format-patch does
10:03 Vampire0 If you have a simple patch from git diff or svn diff or whatever, you have no chance to create a proper commit from it, as you have no author information and no commit message
10:03 Vampire0 It is simply a totally different use-case
10:03 zhalla joined #git
10:03 Vampire0 They just happen to both involve applying some changes to some files
10:04 memfrob is there a cleaner way to use git format-patch to output to a file than -k --stdout HEAD~ &> foo.bar ?
10:04 memfrob Maybe I missed a spot in the man doc?
10:04 PioneerAxon joined #git
10:04 Vampire0 memfrob, do not use --stdout
10:05 Vampire0 memfrob, creating files is the standard behavior
10:05 nowhereman joined #git
10:06 ankit01ojha joined #git
10:07 memfrob Ah got it. -o ../ :) Thanks Vampire0
10:07 Vampire0 memfrob, but you will then get one file per commit, which in your case (`HEAD~`) is only one anyway
10:08 memfrob Yeah I'm doing it one by one to avoid big patch blob
10:09 Serpent7776 joined #git
10:09 Vampire0 Big patch blob?
10:09 memfrob Otherwise it'd be several MB
10:09 memfrob I'm kernel hacking hardcore XD
10:09 Vampire0 If you don't do --stdout, you get one file per commit anyway
10:09 Vampire0 memfrob, maybe you should skim through man git format-patch
10:09 gitinfo memfrob: the git-format-patch manpage is available at http://jk.gs/git-format-patch.html
10:11 memfrob Problem though is that if I don't use it, then it won't output the last commit which I'm already up to date with.
10:12 memfrob If I do git commit -as and make a commit, git format-patch then becomes useless without HEAD~ because HEAD is now my modified tree.
10:13 memfrob git format-patch defaults to HEAD, not the commit before it, so because it's in sync and up to date, there is no patch to generate ;)
10:14 rnsanchez joined #git
10:14 memfrob Oh.. I see what you're saying. If I wanted to run something like git format-patch HEAD~3 then it'd generate a file per commit.
10:15 gottaname|wurk suy, sourcetree? will it work with external projects?
10:15 Vampire0 You said you create patches commit by commit to not generate too big patch files. But if you do 5 commits and then say `git format-patch -k @~5`, you get 5 patch files for 5 commits
10:15 Vampire0 exactly
10:15 memfrob @ = HEAD good to know.
10:16 manuelschneid3r joined #git
10:16 Vampire0 indeed
10:16 Vampire0 If your git is new enough
10:16 howitdo joined #git
10:16 howitdo joined #git
10:16 memfrob Pretty new :P
10:17 suy gottaname|wurk: no idea, I don't use it. I just saw it mentioned in several LFS-related pages from atlassian claiming that it has good support.
10:17 gottaname|wurk suy, oh
10:17 gottaname|wurk meh
10:18 suy I suppose it makes sense to spport any git repo, not just their own. Otherwise, many projects with submodules that point to 3rd party sites not under their control would not work
10:18 Vampire0 gottaname|wurk, Sourcetree is just a GUI, you can use it on any repo
10:20 durham joined #git
10:22 mar77i joined #git
10:22 the_cybersphinx joined #git
10:23 gottaname|wurk Vampire0, yeah but the install makes me try to login to my atlassion acc
10:23 yanome joined #git
10:23 Vampire0 gottaname|wurk, I never used it and basically !gui ;-)
10:23 gitinfo gottaname|wurk: 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.
10:23 nostrora joined #git
10:24 gottaname|wurk Vampire0, non technical types using it
10:24 gottaname|wurk that's why the GUI
10:24 gottaname|wurk oh
10:24 Vampire0 gottaname|wurk, but besides that, is there no "Skip" button probably? Maybe they have a tighter integration for their hosted BitBucket stuff, but I don't think it should be necessary
10:24 andschwa joined #git
10:25 SlashLife^work joined #git
10:25 SlashLife^work joined #git
10:26 ronny joined #git
10:28 narendraj9 joined #git
10:30 Vampire0 gottaname|wurk, ah, no, you need to have an account to use SourceTree according to its documentation
10:30 mikecmpbll joined #git
10:31 robotrolll joined #git
10:32 WakiMiko joined #git
10:33 Vampire0 gottaname|wurk, but still, it should work with all repositories
10:35 alex1a joined #git
10:35 afuentes joined #git
10:35 JeroenT joined #git
10:41 KV joined #git
10:43 almy joined #git
10:43 jmpp joined #git
10:43 jmpp joined #git
10:46 marcogmonteiro joined #git
10:47 chipotle joined #git
10:47 k5tux joined #git
10:48 tirej joined #git
10:48 masuberu joined #git
10:50 chipotle_ joined #git
10:51 selckin joined #git
10:52 PioneerAxon joined #git
10:52 _joes__ joined #git
10:59 vuoto joined #git
11:00 bejeb joined #git
11:03 Xethron joined #git
11:03 finalbeta joined #git
11:05 lb1c38 joined #git
11:05 ShalokShalom_ joined #git
11:12 TomyWork joined #git
11:13 bruce_lee joined #git
11:13 bruce_lee joined #git
11:13 TomyWork just curious... for git rev-list, what's the difference between --max-age=<timestamp> and --since=<date>?
11:14 chipotle joined #git
11:16 finalbeta joined #git
11:19 YuGiOhJCJ joined #git
11:19 Limes joined #git
11:21 eduardas_m joined #git
11:23 m1m3-50 joined #git
11:23 mathematic-alpha joined #git
11:24 Snugglebash joined #git
11:24 _ikke_ TomyWork: I guess that they just take timestamps, not a textual description of the date
11:24 _ikke_ (it's only for plumbing)
11:24 stroybat joined #git
11:25 _ikke_ https://github.com/git/git/commit/fcf​da02bc7e3a9dacf63fa43df8eee530a471f96
11:28 Cabanossi joined #git
11:31 z3t0 joined #git
11:34 mwlang joined #git
11:35 mindfart joined #git
11:36 JeroenT joined #git
11:36 mwlang I just did a git pull and have a few files that conflicted with “both were modified”  In all conflict cases, I want to keep the files as they were before I pulled latest.  Is there an easier way to do this quickly rather than going through each file with editor and fixing the <<< … >>> sections?
11:37 ugjka joined #git
11:39 _ADN_ joined #git
11:39 _ikke_ git checkout --ours <file>
11:39 _ikke_ But why merge then?
11:39 Tobbi joined #git
11:40 mda1 joined #git
11:43 Hudu joined #git
11:48 mwlang _ikke_:  thanks.  As for why merge, I’m not sure what happened but I made some changes to about 10 files and committed to main branch, but when I switched to feature branch (by other developer), those ten files essentially looked like they were before my changes when reading through the conflicts.  the other developer says he rebased to the main branch, so…no idea why my changes didn’t flow to his branch.
11:48 m1m3-50 joined #git
11:48 mwlang my guess: error on his part and easiest way seems to just be to accept “ours” as you suggested and move on.
11:49 Spherical joined #git
11:52 ertesx joined #git
11:55 CussBot joined #git
12:01 memfrob How to restrict git diff to a file?
12:01 Tobbi joined #git
12:01 kettlewell joined #git
12:02 memfrob Ideally how do I diff a file from a certain commit to another commit?
12:03 memfrob Ah git diff <commit> file works XD nooiiccee
12:03 memfrob You guys did a kick ass job writing git.
12:03 cdown joined #git
12:04 tobiasvl that will compare the <commit> with the working directory, you can do "git diff <commit1> <commit2> file" if you want to diff between two commits
12:04 memfrob Beautiful!
12:05 memfrob You guys thought of everything!
12:05 tobiasvl hehe! thanks, but I didn't write git ;) there are surely some contributors in here, but many are just users like you
12:10 Sonderblade does git support having global ignores? that is applied for all projects you work with?
12:11 _ikke_ Sonderblade: git config --global core.excludesFile <path>
12:11 Vampire0 Sonderblade, yes
12:11 Vampire0 _ikke_, you don't need to configure it, do you? There is a standard place too, isn't it?
12:12 tvw joined #git
12:12 _ikke_ "Defaults to $XDG_CONFIG_HOME/git/ignore"
12:13 Anja joined #git
12:13 Vampire0 or ~/.config/git/ignore if XDG_CONFIG_HOME is not set, yes, just found it in the man page :-)
12:13 Vampire0 Sonderblade, man gitignore
12:13 gitinfo Sonderblade: the gitignore manpage is available at http://jk.gs/gitignore.html
12:13 Cabanossi joined #git
12:14 Sonderblade aha i didn't get that that was the path to it
12:14 zeroed joined #git
12:14 zeroed joined #git
12:14 adamru joined #git
12:15 Vampire0 Sonderblade, yes, `ignore` is the actual file.
12:15 hobodave joined #git
12:18 DrSlony joined #git
12:18 marcogmonteiro joined #git
12:19 seni joined #git
12:20 suttonwilliamd joined #git
12:21 DrSlony Hi, what versioning scheme would you recommend for a program which has two stable branches and several feature branches?
12:22 ckruczek two stable branches and several feature branches
12:22 ckruczek and maybe a development branch
12:22 DrSlony One of the stable branches will be killed off soon, if that matters. But it will always have a few feature branches.
12:22 ckruczek where everything gets merged into and on release it get merged into the appropriate stable branch
12:24 DrSlony The stable branches (gtk2 and gtk3) get tagged, but in the meanwhile builds from the feature branches are made so that users can help test the new features. They will get merged into the stable branch once ready.
12:24 Vampire0 DrSlony, ckruczek maybe you are after !gitflow?
12:24 gitinfo DrSlony: The description of the gitflow branch workflow model is at http://nvie.com/posts/a-succ​essful-git-branching-model/ while a tool to help implement this workflow is at https://github.com/petervanderdoes/gitflow  See http://sethrobertson.github.c​om/GitBestPractices/#workflow for other workflow suggestions/references
12:24 DrSlony thanks Vampire0 will read it
12:24 ckruczek Vampire0: yes, thats what I had in mind
12:25 leeN joined #git
12:26 mathematic-alpha joined #git
12:29 adamru joined #git
12:30 DrSlony My question is how would you recommend we version our builds when they come from a feature branch for example. That gitflow article doesn't really answer that.
12:30 adamru joined #git
12:31 Vampire0 Maybe just postfix the feature branch name to the version?
12:31 adamru joined #git
12:31 DrSlony and how will users tell several builds from the same feature branch?
12:31 _ikke_ Why version feature branches?
12:32 _ikke_ Only when the features are merged in they are usally released
12:32 Vampire0 e. g. 1.2.3-feature1.1, 1.2.3-feature1.2, 1.2.3-feature1.2? or maybe , 1.2.3-feature1-build.1, 1.2.3-feature1-build.2, 1.2.3-feature1-build.3?
12:32 Vampire0 _ikke_, as they described, they release feature branch preview builds to users for preliminary testing of the new feature
12:33 Vampire0 DrSlony, actually this question has not really anything to do with Git. :-D
12:33 DrSlony All I personally care about is a commit hash, but users who help test development builds from feature branches may want some kind of version which tells them whether the build they're installing is newer than the one they already have. So do we also make tags in feature branches? Or do we just append a commit hash to the tag because thats the only reliable info?
12:33 _ikke_ DrSlony: what about git describe?
12:34 BackEndCoder joined #git
12:34 DrSlony _ikke_ yes that's one option
12:34 _ikke_ and otherwise I would use something Vampire0 described
12:34 DrSlony the format of git describe's output depends on what you checkout, making it unreliable to parse
12:34 DrSlony check out a tag, you get <tag>
12:35 DrSlony check out a branch, you may get just <tag>, or you may get <tag>-<revs>-g<commithash>
12:36 Vampire0 DrSlony, you can do either, append the SHA to the version or create a tag in the feature branch, that's totally up to you. You can even do both.
12:37 Vampire0 DrSlony, be aware that the SHA cannot be found maybe **if** you rewrote the history after the release in the feature branch
12:37 _ikke_ DrSlony: yeah, it describes commits, not the branch you happen to be on
12:38 _ikke_ and if there is a tag pointing to the commit, then just giving the tag is specific enough
12:39 yqt joined #git
12:40 achadwick joined #git
12:41 adamru joined #git
12:42 DrSlony _ikke_ is there a way to make it print <tag>-<revs>-g<commithash> even when i checkout a tag? it would help with script parsing
12:43 DrSlony s/script/cmake/
12:44 notebox joined #git
12:45 re1 joined #git
12:46 pR0Ps joined #git
12:46 sdothum joined #git
12:47 sarri joined #git
12:47 sarri joined #git
12:47 Vampire0 DrSlony, man git describe tells you to use --long
12:47 gitinfo DrSlony: the git-describe manpage is available at http://jk.gs/git-describe.html
12:51 nd_ joined #git
12:51 memfrob 1248 lines so far of security features backported from 4.9 to 4.4 :D thank you git
12:51 grayjoc joined #git
12:52 memfrob I need a break, been staring at C the past several hours
12:52 mehola joined #git
12:53 memfrob Do you guys get overworked in a short time frame when you're doing complicated things?
12:53 canuk joined #git
12:53 moritz yes. That's what pauses are for
12:53 moritz and a healthy mental regards towards work
12:54 memfrob what does the latter mean?
12:54 moritz I highly recommend becoming a parent; that really puts things into perspective, priority-wise
12:54 memfrob I'm getting a vasectomy to ensure that won't happen.
12:54 ckruczek lol
12:55 moritz it means not working overtime to chase after arbitrary deadlines, just because your boss set that deadline
12:55 moritz and then finding out that the project blocks on an external contractor that won't be finished for two weeks after the deadline after all
12:55 elsevero joined #git
12:55 ckruczek moritz: I can approve this!
12:55 memfrob XD
12:56 DrSlony Vampire0 thanks
12:56 memfrob I'm doing this for fun mostly and also a bit of publicity.
12:58 memfrob Lots of people have been asking for it and nobody will, so I am :)
12:59 ojdo joined #git
12:59 Vampire0 DrSlony, yw
13:01 Fiouz joined #git
13:01 byte512 joined #git
13:03 memfrob Well I'm gunna go on The Crew (amazing open world street racing game) take care all! Thank you all again for your help!
13:04 Derperpe1 joined #git
13:06 robogoat joined #git
13:06 tomek joined #git
13:06 tomek joined #git
13:08 ckruczek any ideas how it is possible that github thinks that folder is a submodule instead of being a folder with normal content? I already checked if any .gitmodules file exist, but its not.
13:10 ckruczek so following happend: I created a folder, cloned a empty repo, copied some files to the repo and pushed everything to github, now github thinks the one folder I added is a submodule
13:10 osse ckruczek: which folder?
13:11 moritz ckruczek: check the top-level directory for a .gitmodules file
13:11 pR0Ps joined #git
13:11 osse ckruczek: do   git ls-tree HEAD:parentfolder
13:11 ckruczek on toplevel there is also no .gitmodules I already checked this at first
13:12 ckruczek doing it osse.
13:12 osse see what type it says next to the directory
13:13 synthroid joined #git
13:13 ckruczek by parent folder you mean the real folder name osse?
13:14 osse yes
13:14 osse if git thinks  foo/bar is a submodule do  git ls-tree HEAD:foo
13:16 ckruczek no output just an error
13:16 ckruczek 'no valid object name'
13:16 osse which error
13:16 osse uhh
13:16 osse !repro
13:16 gitinfo [!transcript] Please paste (using https://gist.github.com/ or similar) a transcript ( https://git.io/viMGr ) of your terminal session so we can see exactly what you see
13:18 ckruczek some colleque of mine will come soon and will continue this discussion, I have to go now. thx osse for your help, I will investigate further.
13:18 Alex__ joined #git
13:18 noecc joined #git
13:19 seni joined #git
13:20 osse ckruczek: I am wondering whether it says "commit" next to it instead of "tree". If it does that it *is* a submodule. This can happen if you copy a directory that is a git repo into another and do git add path/dir
13:20 osse Even if there is no .gitmodules
13:20 Alex__ Hi everyone, I believe by me saying I'm having the same problem as ckruczek, you'll know what the problem is?
13:20 ckruczek yes they already know it. osse would you be so kind, to continue with Alex?
13:22 Tobbi joined #git
13:24 skunkz joined #git
13:25 jast basically if, from the root of the repo, the submodule in question is at foo/bar, do: git ls-tree HEAD:foo, and see which type (commit/tree/...) it outputs next to the submodule location
13:26 ThomasLocke_ joined #git
13:26 ThomasLocke_ joined #git
13:27 Alex__ When I try that it states that the submodule is not a tree object
13:27 mostlybadfly joined #git
13:30 cyberz hi guys
13:31 HardlySeen joined #git
13:31 gitinfo [!welcome] Welcome to #git, a place full of helpful gits. If you have a question,  feel free to just go ahead and ask—somebody should answer shortly.  For more info on this channel, see http://jk.gs/git/  Take backups (type !backup to learn how) before taking advice.
13:31 HardlySeen joined #git
13:31 gitinfo cyberz: [!welcome] Welcome to #git, a place full of helpful gits. If you have a question,  feel free to just go ahead and ask—somebody should answer shortly.  For more info on this channel, see http://jk.gs/git/  Take backups (type !backup to learn how) before taking advice.
13:32 cyberz is there any way to get `commit fixup file` to fixup the last commit of file without having me to go through the logs?
13:32 UTAN_dev_ joined #git
13:32 Vampire0 !snack
13:32 gitinfo [!botsnack] Om nom nom
13:32 bernardio joined #git
13:32 Vampire0 cyberz, what is `commit fixup file`?
13:33 cyberz Vampire0 it's commit --fixup sha file
13:34 cyberz where sha is the last commit of file, which is usually the 90% of the time I use fixup
13:34 Vampire0 Ah, nice, I didn't know --fixup yet, I always made the messages manually
13:34 dogmatic69 joined #git
13:36 Vampire0 cyberz, `git commit --fixup="$(git log -1 --pretty=format:%H file)" file`
13:36 Vampire0 cyberz, you can make this work as git alias of course
13:38 cyberz Vampire0 yes, that can be easily done but I was just wondering if there's a builtin way to do so, maybe even interactive
13:39 cyberz at least for me it's a super common scenario -  so I was wondering
13:39 Vampire0 `git config alias.fixup '!fixup() { echo git commit --fixup="$(git log -1 --pretty=format:%H $1)" $1; }; fixup'`
13:39 Vampire0 And then `git fixup file`
13:39 Vampire0 At least not that I know of
13:40 cyberz nice!
13:42 lindenle joined #git
13:42 osse you can also do  git commit --fixup:/'part of commit message'
13:43 marcogmonteiro joined #git
13:44 Vampire0 osse, really? Where is this documented? I didn't find this
13:45 osse Vampire0: the fixup argument takes all the same things that rev-parse understands.
13:45 osse man gitrevisions
13:45 gitinfo the gitrevisions manpage is available at http://jk.gs/gitrevisions.html
13:45 JeroenT joined #git
13:46 Vampire0 osse, ah, ok, so it actually is <commit-ish>, not <commit>, I see, thx
13:47 mischat_ joined #git
13:47 Vampire0 osse, or did the docu convention change in this regard?
13:48 osse Vampire0: i dunno
13:50 Vampire0 osse, but anyway, your syntax would not really help. He doesn't know which commit it is. He wants "the last commit that touched file X".
13:50 osse yes, it was more just a "by the way"
13:51 osse Because often don't know what commit it is either but I remember that I used "coffee" in the commit message, so I do --fixup=:/coffee
13:51 Vampire0 definitely something I should remember :-)
13:51 osse of course "coffee" is a palceholder for some relevant word
13:52 fahadash joined #git
13:52 osse if the commit message is "Add check for event = null"  then I do :/event
13:52 osse or something
13:52 jast so you mean I shouldn't literally type "coffee"? darn
13:53 Vampire0 jast, no, you should drink coffee
13:54 JanC joined #git
13:54 Vampire0 Except if you ask gitinfo about !coffee
13:54 gitinfo Drink Earl Grey Tea. It’s better.
13:54 Vampire0 sssh gitinfo, take a !beer
13:54 gitinfo Beer! It's what's for dinner!
13:56 DrSlony left #git
13:58 Hudu joined #git
13:58 lss8 joined #git
13:58 lss8 joined #git
13:58 HaywardGB joined #git
13:59 Cabanossi joined #git
13:59 mischat joined #git
14:00 vuoto joined #git
14:00 Derperpe1 joined #git
14:01 eduardas_m joined #git
14:03 jimi_ joined #git
14:04 seni joined #git
14:04 finalbeta joined #git
14:04 rwb joined #git
14:05 TheMontyChrist joined #git
14:05 mp4 joined #git
14:06 TheMontyChrist joined #git
14:06 TheMontyChrist joined #git
14:06 TheMontyChrist how do "revert"?
14:06 TheMontyChrist from the command line?
14:06 osse TheMontyChrist: !revert
14:06 gitinfo TheMontyChrist: That's a rather ambiguous question... options: a) make a commit that "undoes" the effects of an earlier commit [man git-revert]; b) discard uncommitted changes in the working tree [git reset --hard]; c) undo committing [git reset --soft HEAD^]; d) restore staged versions of files [git checkout -p]; e) move the current branch to a different point(possibly losing commits)[git reset --hard $COMMIT]?
14:07 TheMontyChrist I'm thinking git reset --hard
14:07 TheMontyChrist from the looks of it.
14:07 TheMontyChrist I want to checkout a branch but the working dir has a bunch of changes that need to be tossed.
14:07 TheMontyChrist that's the scenario
14:07 TheMontyChrist clean up -> then checkout
14:08 osse reset --hard is for you
14:08 Gsham joined #git
14:09 TheMontyChrist thanks
14:09 Tobbi joined #git
14:10 TheMontyChrist I'm new to git but come from mercurial, so met me ask
14:10 TheMontyChrist for b)
14:11 TheMontyChrist discard uncommited means: "staged or modified"  So modified means, it's modified but hasn't been staged.  In fact, staging is new for me.
14:11 TheMontyChrist staged means - added, deleted, or modified
14:12 davimore joined #git
14:12 TheMontyChrist how what's the command to stage a modified file?
14:12 TheMontyChrist git add - for new file
14:12 TheMontyChrist git rm - to remove file
14:12 TheMontyChrist is it git add too for modified files?
14:12 kadoban joined #git
14:12 osse TheMontyChrist: yes
14:13 canton7 TheMontyChrist, yes. 'git status' tells you what command to use, if you need reminding
14:13 osse git add means "add this stuff to my next commit"
14:13 osse "this stuff" can both be modified or new content
14:16 GT4066 joined #git
14:19 _geoff joined #git
14:19 CEnnis91 joined #git
14:20 marcogmonteiro joined #git
14:22 Tobbi joined #git
14:22 albel727 joined #git
14:26 ocbtec joined #git
14:27 durham_ joined #git
14:27 govg joined #git
14:28 Snugglebash joined #git
14:28 tirej_ joined #git
14:29 Derperpe1 joined #git
14:29 grayjoc joined #git
14:29 ash_workz joined #git
14:31 ShekharReddy joined #git
14:31 mischat_ joined #git
14:31 todd_dsm joined #git
14:32 zhalla joined #git
14:34 finalbeta joined #git
14:35 GodGinrai joined #git
14:36 dsdeiz joined #git
14:36 dsdeiz joined #git
14:36 davimore joined #git
14:39 sarri joined #git
14:39 sarri joined #git
14:39 marcogmonteiro joined #git
14:41 zhalla- joined #git
14:41 thebinary joined #git
14:41 ikelso joined #git
14:42 denete joined #git
14:42 durham joined #git
14:43 zhalla joined #git
14:43 denete_ joined #git
14:43 Synthbread joined #git
14:44 mnaser joined #git
14:44 mnaser i have a folder containing patches i'd like to apply on a repo, I'm running the following: `git apply --directory=. horizon-patches/*`
14:45 mnaser a patch is failing with "fatal: invalid path './doc/source/topics/settings.rst'" .. however, that file does exist
14:45 mnaser am i misinterpreting the error? :X
14:46 a_thakur joined #git
14:47 seni joined #git
14:48 cdg joined #git
14:48 roentgen joined #git
14:50 cdg joined #git
14:54 Ediz joined #git
14:55 Tobbi joined #git
14:55 zhalla joined #git
14:55 chardan joined #git
14:58 crose joined #git
14:59 Cabanossi joined #git
15:00 mischat joined #git
15:00 netj joined #git
15:00 bonzibuddy joined #git
15:02 jaeckel joined #git
15:03 rscata joined #git
15:07 Tobbi joined #git
15:10 Dougie187 joined #git
15:11 snowkidind joined #git
15:12 Es0teric joined #git
15:13 dsdeiz joined #git
15:13 dsdeiz joined #git
15:14 snowkidi_ joined #git
15:16 shinnya joined #git
15:18 cyberz hi guys
15:18 sdothum joined #git
15:19 polyrob joined #git
15:19 cyberz I can't understand this situation: I have a branch, and If I do git rebase -i HEAD~30 and I save without touching anything I get rebase conflicts anyway
15:19 cdg_ joined #git
15:20 a_thakur joined #git
15:20 Es0teric joined #git
15:20 Vampire0 cyberz, I'd guess you have a merge in the rebased range
15:21 Dougie187 ^^
15:21 Dougie187 Between HEAD~30...HEAD
15:21 Vampire0 cyberz, rebase doesn't handle merges too nice. By default it just leaves them out and flattens the history
15:21 cyberz yes I do, but I can't understand why that causes rebase issues
15:21 Vampire0 cyberz, maybe --preserve-merges is what you need. read more at man git rebase
15:21 gitinfo cyberz: the git-rebase manpage is available at http://jk.gs/git-rebase.html
15:21 cyberz thanks guys
15:21 Dougie187 But --preserve-merges gets weird with -i, doesn't it?
15:22 WayToDoor joined #git
15:22 gitinfo cyberz: This channel tracks karma based on who has gotten lots of thanks for being helpful. If you want to help someone reach karmic nirvana, please mention their name when thanking them with "thank you", "thankyou", "thanks", "thx" or "cheers". Try ".karma <nick>" or ".topkarma" to show karma status of a person.
15:22 Vampire0 Dougie187, does it?
15:22 Dougie187 I'm pretty sure the man page says not to use them both
15:22 cyberz thank you Vampire0
15:22 Dougie187 Because it can't convey to you that there's a merge in the editor when you're modifying commits.
15:23 zhalla left #git
15:23 cyberz preserve-merges did the trick, great!
15:23 texinwien_ joined #git
15:23 Dougie187 Vampire0: "This uses the --interactive machinery internally, but combining it with the --interactive option explicitly is generally not a good idea unless you know what you are doing (see BUGS below)."
15:23 Vampire0 yep, so know what you are doing and you are fine :-D
15:24 Dougie187 Yeah
15:24 Es0teric joined #git
15:25 JrCs joined #git
15:26 denete_ joined #git
15:26 Toordog joined #git
15:28 e14 joined #git
15:29 cyberz indeed preserve-merges freaks out with interactive
15:29 HarryHallman joined #git
15:30 Vampire0 cyberz, read the man page ;-)
15:31 Dougie187 cyberz: It'd have to. It can't really draw a graph in that context, and ordering is really weird as well when it needs to draw commits from two branches.
15:31 dersand joined #git
15:32 cyberz well, I'm trying to reword a couple of commits down there - this should be fine
15:33 cyberz the man says this is supposed to be ok but It is not really the case
15:34 ResidentBiscuit joined #git
15:36 philipsd6 joined #git
15:38 aep joined #git
15:38 aep on our ci, git describe fails because of no tags. i tried git fetch --tags, which suceeds, but .git/refs/tags is still empty
15:38 memfrob joined #git
15:38 aep any idea what that is?
15:38 memfrob what what is? sorry i just got here
15:39 Gsham joined #git
15:39 aep oh, let me repeat :D
15:39 _ikke_ aep: use git describe --always
15:39 aep on our ci, git describe fails because of no tags. i tried git fetch --tags, which suceeds, but .git/refs/tags is still empty
15:39 aep _ikke_: that doesnt behave the way i'd expect it
15:39 aep it'll obviously not have the tag names
15:39 aep there must be something wrong with git fetch --tags here, i think
15:40 memfrob broken version of git? :/
15:40 _ikke_ aep: does git tag list tags?
15:41 apeiros joined #git
15:41 aep _ikke_: yes
15:41 aep which is weird
15:41 memfrob so the only problem is that the tags don't show up in .git/refs/tags?
15:41 aep yeah, i think so
15:42 _ikke_ they can be packed
15:42 _ikke_ so looking in there does not say much
15:42 memfrob i never browse the .git folder XD
15:42 aep ah!
15:42 memfrob so no idea..
15:42 aep can i 'unpack' them just for testing?
15:42 _ikke_ aep: git for-each-ref refs/tags
15:43 _ikke_ Just use git tools to query them
15:43 aep _ikke_: yep shows me the correct tags and their shas as expected
15:43 aep k
15:43 texinwien joined #git
15:43 byjwcs joined #git
15:43 memfrob k?
15:43 mischat_ joined #git
15:43 aep *ok
15:43 anuxivm joined #git
15:44 memfrob XD sorry I only hear that on facebook
15:44 d1z joined #git
15:44 Es0teric joined #git
15:44 mischat_ joined #git
15:45 aep oh... here's the problem
15:45 aep i just did a git log on the ci's checkout, and it ends after 10 commits
15:45 aep the ones before that are all missing
15:45 aep and the tag is 36 commits away from head
15:45 troulouliou_div2 joined #git
15:46 _ikke_ so it did a shallow clone probably
15:46 memfrob --depth=10?
15:46 bauruine joined #git
15:46 aep dunno what it does. its hidden D:
15:46 _ikke_ it only fetches the last x commits
15:46 mischat_ joined #git
15:46 aep i mean, dunno what the ci does
15:46 aep probably. it would make sense
15:47 aep there we go, git fetch --depth=100 fixed it
15:47 aep thanks!
15:47 _ikke_ while you're at it
15:48 _ikke_ git fetch --unshallow
15:48 aep fatal: --unshallow on a complete repository does not make sense
15:48 aep odd
15:48 aep its not complete
15:48 _ikke_ how many commits are there?
15:48 aep ah... 90
15:48 aep so it was complete :D
15:49 aep cool thanks
15:50 leftriver41 joined #git
15:50 dreiss joined #git
15:53 stoopkid joined #git
15:54 Ranmaru joined #git
15:55 lordjancso joined #git
15:55 tubbo left #git
15:57 Tobbi joined #git
15:57 Derperpe1 joined #git
15:57 jsolano_ joined #git
15:59 avrdude joined #git
16:00 avrdude Hi, I'm trying to set up a project on Gitlab, with tortoisegit on my machine. When I go to push, I only see this: http://i.imgur.com/CBLgYyT.png (I can't click OK). What am I doing wrong? I've never done this before
16:01 ResidentBiscuit Do you need to set your pushed branch?
16:01 mischat joined #git
16:02 Vampire0 cyberz, then it maybe is a bug you should report on the Git !mailinglist
16:02 avrdude To be honest, I don't know what that means, I'm very new to using git
16:02 mischat joined #git
16:02 cyan__ joined #git
16:02 ResidentBiscuit I don't think this is a bug with git, or with anything
16:02 Vampire0 cyberz, !mailing_list I meant
16:02 gitinfo cyberz: The mailing list can be reached via git@vger.kernel.org. You don't need to subscribe to the list, you will always be put in cc on reply. Read archives at http://j.mp/gitlist
16:02 ResidentBiscuit The first drop down, select your branch you want to push
16:03 ResidentBiscuit Or check the box if you wanna push everything
16:03 ResidentBiscuit (I've never used tortoisegit, just guessing here)
16:03 avrdude ResidentBiscuit: there is nothing in either of the dropdowns :/
16:03 Vampire0 avrdude, !gui I'm afraid
16:03 gitinfo avrdude: 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.
16:04 ResidentBiscuit Yeah don't know what to tell you. As the handy bot has pointed out, the cli is source of all truth
16:04 ResidentBiscuit IMO there's not really a great GUI for git out there
16:05 memfrob the CLI is very easy to use, it's not cvs or svn at all.
16:06 mischat_ joined #git
16:06 memfrob It's like a better version of mercurial and bazaar combined, it's AWESOME.
16:06 chele how can i get a repo config?
16:07 LeBlaaanc joined #git
16:07 memfrob !git-repo-config ?
16:07 memfrob fail. man git-repo-config
16:08 memfrob how come it doesn't work for me when i do it? -_-
16:08 nickabbey joined #git
16:08 thebinary joined #git
16:08 ResidentBiscuit !git-repo-config
16:09 ResidentBiscuit !help
16:09 gitinfo What do you need help with? Something with git? Tell us what's going on, and please be specific. For information on how to abuse me (I'm a bot!), see http://jk.gs/git/bot . Please be gentle.
16:09 Vampire0 memfrob, It would be man git repo-config
16:09 gitinfo memfrob: the git manpage is available at http://jk.gs/git.html
16:09 memfrob ahhh
16:09 Vampire0 memfrob, but as you see, there is no command called repo-config
16:10 thebinary joined #git
16:10 Vampire0 memfrob, btw. you sound like Git came after hg and bzr which is not true. :-P That it is better than both is true of course :-D
16:10 Vampire0 chele, what exactly do you mean?
16:10 Vampire0 chele, `git config -l`?
16:10 memfrob git is long before bzr and hg
16:10 WayToDoor joined #git
16:11 Vampire0 memfrob, well git and hg started at the same time, bzr came after them
16:11 svm_invictvs joined #git
16:11 memfrob bzr is a git rip off honestly
16:12 brent__ joined #git
16:12 emg joined #git
16:12 memfrob there used to be a git repo-config btw that did what he asked.
16:13 Vampire0 ah, ok, never heard of
16:13 Vampire0 what did it do?
16:13 monokrome joined #git
16:13 jeffreylevesque joined #git
16:14 seni joined #git
16:14 Vampire0 Ah, predecessor of git config: https://git-scm.com/docs/git-repo-config
16:15 memfrob I only use git config to set up my name and email tbh
16:15 Hudu joined #git
16:16 Vampire0 I use it for setting all configuration options I want to set
16:16 Nilesh_ joined #git
16:16 memfrob git config --global user.name "Cottage Cheese" ftw
16:17 m4sk1n joined #git
16:17 Es0teric joined #git
16:18 Vampire0 aliases, difftool, mergetool, autosquash, push default, ...
16:18 Vampire0 Git has many useful options :-)
16:19 qqx joined #git
16:20 avrdude oh.. so i need to commit before i push?
16:21 avrdude also, i've forgotten my passphrase for my private ssh-key, how fucked am i?
16:22 chele Vampire0 the configuration option of the remote bare repo
16:22 chele options
16:22 reynierpm joined #git
16:23 _ikke_ avrdude: It's not rescuable
16:23 nowhereman joined #git
16:23 _ikke_ avrdude: If you cannot find it again, you need to generate new keys
16:23 Vampire0 chele, `git remote -v`? `git config --get-regexp remote.*`?
16:24 al-damiri joined #git
16:24 GodGinrai joined #git
16:24 chele Vampire0 this show local config
16:25 Vampire0 chele, go to the remote and do it there
16:25 avrdude thats not really a big problem i guess
16:25 WayToDoor joined #git
16:25 Vampire0 chele, you cannot get the config of a remote repo afaik
16:25 chele I see.
16:25 Vampire0 chele, what exactly do you want to know from that remote config?
16:25 chele if denynonff is enabled
16:26 Vampire0 Well, just try to push non-ff and you will know
16:26 Vampire0 If it didn't work it is on, if it worked it is off and you can push the previous ref to undo the tests push
16:27 memfrob avrdude, pushing without a commit? there's nothing to push if there's nothing to commit
16:27 avrdude you're implying i even know what the concepts of pushing and commiting means
16:28 memfrob avrdude, as long as you can upload a new ssh key to the server so it can accept you there's no problem with forgetting your passphrase
16:28 jwest joined #git
16:28 memfrob github allows you to add and remove keys at any time via the website.
16:28 avrdude anyway, i've got it going now :)
16:28 memfrob (for exmaple)
16:28 Vampire0 avrdude, !bottom
16:28 gitinfo avrdude: [!bottomup] 'Git from the bottom up' starts with explaining the building blocks of git and proceeds to tell you how they fit together. http://ftp.newartisans.com/​pub/git.from.bottom.up.pdf (http://ikke.info/git.from.bottom.up.pdf)
16:28 memfrob HEY I GOT THIS VAMPIRE!
16:28 memfrob hes asking questions i know the answers to XD
16:28 avrdude I'm using Gitlab btw, but I assume its identical
16:29 memfrob yeah websites wont lock you out for losing your keys or whatever
16:29 svm_invictvs joined #git
16:29 memfrob (ones ive used anyway)
16:29 ResidentBiscuit The bottom up link is ded btw
16:29 memfrob HAH! so i can still help him
16:29 Vampire0 memfrob, I don't know why you shouted at me now :-/
16:29 ResidentBiscuit https://jwiegley.github.io/git-from-the-bottom-up/
16:29 memfrob XD
16:29 memfrob sorry Vampire0
16:29 Nysar joined #git
16:29 memfrob I mess around a lot.
16:30 Vampire0 memfrob, I know a page that locks you out if you forget your passphrase, as the key is the only way to auth to that site
16:30 avrdude glad i'm not using that site
16:30 Vampire0 memfrob, it's the page of the german online tax website
16:30 memfrob But your ssh key and the SSL connection to the website are two different things
16:31 memfrob and also unrelated as your browser has nothing to do with your ssh key
16:32 pR0Ps joined #git
16:32 avrdude can't believe i've never done version control before.. seriously wtf
16:32 memfrob but i'll take your word for it as you know more about git th=an me
16:32 WayToDoor joined #git
16:32 mohabaks joined #git
16:32 Vampire0 memfrob, as I said, you login to that website with your key
16:32 raspado joined #git
16:33 Vampire0 memfrob, and if it is not valid anymore or you forgot your passphrase, you are locked out
16:33 memfrob How is that possible if lets say Firefox doesnt use your ssh key as authentication?
16:33 memfrob avrdude, git is a lot of fun
16:33 Vampire0 memfrob, technical wonders of the modern times :-D
16:34 Vampire0 memfrob, https://www.elsteronline.de/eportal/Konfiguratio​nsAssistent1JS.tax?ElsterRequestKeys.KONFIGURATI​ONS_ASSISTENT_WORKFLOW=8DoDKisUeVYwGbC&amp;schri​tt=3&amp;javascript=1&amp;webStorageActive=true
16:34 CheckDavid joined #git
16:37 thebinary joined #git
16:38 grayjoc joined #git
16:38 memfrob I like it when people leave commit messages like "update stuff" "change this stupid thing" etc XD
16:39 APic
16:39 memfrob whats that face for?
16:40 _geoff joined #git
16:41 Darren_ joined #git
16:41 memfrob ☺ ☺ ☺
16:42 reynierpm hi there, I am running the following commands in a Docker container but for some reason the branch isn't checkout, why? "RUN git init && git remote add origin https://github.com/pyrocms/pyrocms.git && git fetch origin && git checkout origin/3.2" what I am missing here?
16:43 moritz reynierpm: so what's the error message?
16:43 moritz reynierpm: or the (undesired) result?
16:44 gopi1410 joined #git
16:44 hobodave joined #git
16:44 reynierpm moritz, after those commands I should run `php artisan install --ready` but I got `Could not open input file: artisan` because the file isn't checkout and I can tell you that because I am listing the directory content just right after the checkout command
16:44 moritz reynierpm: what does "git status" say?
16:44 moritz reynierpm: and why not just "git clone --branch 3.2 https://github.com/pyrocms/pyrocms.git" ?
16:44 Es0teric joined #git
16:45 memfrob or git clone https://github.com/pyrocms/pyrocms.git && git checkout 3.2
16:45 reynierpm moritz, git status `# HEAD detached at origin/3.2 nothing to commit, working directory clean`
16:46 denete_ joined #git
16:46 reynierpm moritz, memfrob let me try those instead
16:46 madewokherd joined #git
16:47 synthroid joined #git
16:47 moritz those note that this will create a pyrocms/ subdir in which the files land
16:47 tang^ joined #git
16:47 reynierpm moritz, your command is cloning inside pyrocms directory any way to avoid the creation of such directory and leave under the directory where command is being run?
16:49 moritz reynierpm: ah right, you can do a git clone $url .
16:49 moritz the trailing . causes it clone into the existing dir
16:49 reynierpm moritz, ok
16:49 GodGinrai joined #git
16:50 f0rks joined #git
16:51 ecuanaso joined #git
16:52 multi_io joined #git
16:52 Es0teric joined #git
16:53 jagob joined #git
16:53 mehola joined #git
16:54 robnester Question: Is it possible to do a sparse checkout for a single file in a git submodule?
16:54 robnester so that I only checkout one file from a repo, instead of the entire contents of the repo.
16:54 reynierpm moritz, it works thanks
16:56 svm_invictvs joined #git
16:59 Es0teric joined #git
16:59 svm_invictvs joined #git
17:02 robnester Question: Is it possible to do a sparse checkout for a single file in a git submodule?
17:02 robnester so that I only checkout one file from a repo, instead of the entire contents of the repo.
17:02 memfrob joined #git
17:02 memfrob joined #git
17:02 mischat joined #git
17:03 _ikke_ robnester: we got it the first time :-)
17:03 denisMone joined #git
17:03 robnester _ikke_: oh, my apologies, I wasn't sure if I was voiced in here :/
17:03 robnester not trying to be repetitious
17:04 danimal2 joined #git
17:06 memfrob What did he say? I missed it
17:06 robnester memfrob: at this point i'm afraid to repeat it :) lol
17:06 Keverw joined #git
17:06 memfrob ROFL
17:07 Es0teric joined #git
17:07 Ryanar joined #git
17:08 chardan joined #git
17:08 robnester memfrob: trying to figure out if / how I can do a sparse checkout of a git submodule so that I checkout only a single file, and not the other 500+mb of the repo which I don't need.
17:09 Keverw Hello. I been using SourceTree for a while, but I've not able to get it to work with a private server for HTTP Auth running Gitea...  It seems BitBucket and GitHub are only supported by that client. I asked in that channel and they recommended switching to a open source visual one instead. Any recommendations for one that uses the official Git HTTP? I also how it also lets you select parts of a file, etc. Very visual but looking for a
17:09 Keverw open replacement. I'm using macOS but cross platform would be neat too.
17:10 Eugene I'm not sure what you mean by "the official Git HTTP"? git doesn't care what your backend is, you generally feed it a username+pasword and it goes
17:10 Eugene I haven't had a chance to use Gitea yet; is it doing something dumb for authentication?
17:10 memfrob Keverw, gitk?
17:10 jaafar joined #git
17:10 Eugene And for my 2c, I like(and paid for) SmartGit.
17:10 jast robnester: even if you don't checkout everything, you do have to download everything. the only other option is if the server supports remote 'git archive' and even then the best you can do is download a single (sub)dir. if you want to access a single file, ideally you'll find some non-Git way to do it (e.g. github allows you to download single files via HTTP)
17:11 jccn joined #git
17:11 Lauven joined #git
17:11 jast and regarding checkout specifically, submodules only support checking out the full tree
17:11 robnester jast: unfortunately this is running on an internal git repo :/
17:11 pypypypy joined #git
17:12 robnester jast: i had ran across this: https://gist.github.com/johnhunter/3333533 which looked promising.
17:12 memfrob tcl/tk is multi platform and gitk is extremely powerful and complete as far as git protocol goes
17:12 ecuanaso joined #git
17:12 Keverw It works with Git in the command line. But not with Source Tree. I was thinking there was a official GIT HTTP spec that the terminal version followed, and maybe SourceTree didn't work together.
17:12 Topic for #git is now Welcome to #git, the place for git help and alternative commits | Public logs at http://jk.gs/glog | First visit? Read: http://jk.gs/git | Current stable version: 2.11.0 | Getting "cannot send to channel"? /msg gitinfo .voice | git-slushie: submit your best punch line to https://madeitwor.se/git-jokes
17:12 cgdub joined #git
17:13 Eugene There isn't a spec; its just HTTP Basic Authentication. Password goes in, password comes out.
17:13 jast robnester: right, that'll work, but you have to do it manually in each clone where you need it, and it will still use the original tree structure (and it will, as I said, download everything).
17:13 robnester well shucks :/
17:13 robnester ok.
17:13 robnester thanks.
17:13 WayToDoor joined #git
17:13 Eugene git will respect and use your .netrc file
17:13 Eugene See http://stackoverflow.com/questions/5​343068/is-there-a-way-to-skip-passwo​rd-typing-when-using-https-on-github
17:13 Darcidride joined #git
17:14 chardan ooks like I'm back in business!
17:14 ShalokShalom joined #git
17:14 Eugene Its entirely possible that SourceTree is doing something dumb and ignoring that
17:14 ShalokShalom joined #git
17:14 jast robnester: there are other options instead of submodules, though. you can simply copy the file over (which isn't that bad of a solution, really)... or, particularly for large files, something like git-lfs might be a good option (some other choices don't even need special support from your git server)
17:14 memfrob I like how Eugene says it like it is
17:15 Eugene If computers worked right the first time then none of us would be here
17:15 robnester jast: copying isn't really a good option unfortunately, as this particular file is being handled by another org. :/
17:15 Keverw interesting. I need to go AFK but thanks for the better information. I know SourceTree also uses it's own embeded Git but there's a option to use the system's one too. I'll play with it some more.
17:15 tang^ SourceTree is usually pretty good about things... but then I don't ever use http for my clones
17:15 robnester and I won't know when / if it should be updated.
17:15 robnester but thank you.
17:15 jast you could automate it... but yeah, that's rough
17:15 Es0teric joined #git
17:15 memfrob SourceTree with its own embedded git? Oh boy, no wonder it doesn't work.
17:16 cyan__ joined #git
17:16 pypypypypy joined #git
17:16 memfrob Try using github or something practical then :)
17:16 Eugene All the cool kids are doing it. Nobody likes encountering bundled dependency installers, so just include your favorite out-of-date dependency and stomp on the rest of the system. Its the Windows way
17:17 jast github's client for github is similarly limited
17:17 hwrdprkns joined #git
17:17 Keverw @memfrob Yeah here's the settings screen https://screencast.com/t/doLtCINz37Y
17:17 tang^ It does if you don't already have git installed
17:17 jast of course it's much easier for GitHub, Atlassian and such to tailor their stuff to their own idiosyncrasies
17:17 tang^ no, it's truly embedded
17:18 tang^ doesn't interfere with system git
17:18 tang^ but doesn't get updated as well as system git
17:20 jast my guess would be that the client is querying some additional URIs that Git normally doesn't look at, and gets a response it doesn't expect... but who knows
17:20 Keverw interesting. Looks like git version 2.10.1 (Apple Git-78) is what I've got if I use the builtin one... I had Xcode and command line git installed before I ever installed source tree... but I'm going to tell it to use system and play with it more.
17:20 jast good luck!
17:21 Lauven joined #git
17:22 Keverw oh but Git-flow and git-lfs I don't have... I wonder if using system git but keeping those extensions as embeded is a stupid idea... I should maybe just brew install everything instead. I know I've done that to get the newer bash lol
17:22 jast it *should* work with using the embedded versions of those
17:23 jjasinski joined #git
17:23 jast third-party tools tend to be loosely coupled, and Git's plumbing (low-level) output formats and such are pretty stable
17:23 memfrob A git hosting site should honestly JUST be that. A server. With auth stuff. and a web UI browser as simple as gitweb or cgit
17:24 memfrob Not a server that massages your nipples and gives you a client and a busty secretary etc
17:24 codebam how do I remove the last commit without modifying any files
17:25 memfrob git revert?
17:25 jast there are a bunch of options: !revert
17:25 gitinfo That's a rather ambiguous question... options: a) make a commit that "undoes" the effects of an earlier commit [man git-revert]; b) discard uncommitted changes in the working tree [git reset --hard]; c) undo committing [git reset --soft HEAD^]; d) restore staged versions of files [git checkout -p]; e) move the current branch to a different point(possibly losing commits)[git reset --hard $COMMIT]?
17:26 acetakwas joined #git
17:26 codebam like I want to remove the last commit, add a .gitignore, then commit again without the ignored files and with the .gitignore
17:26 matsaman joined #git
17:27 codebam is it reset --soft?
17:27 jast here's the interactive answer guide: !fixup
17:27 gitinfo So you lost or broke something or need to otherwise find, fix, or delete commits? Look at http://sethrobertson.github.com/GitFixUm/ for full instructions, or !fixup_hints for the tl;dr. Warning: changing old commits will require you to !rewrite published history!
17:28 jast reset --soft is one way to do it, you may want to leave out the '--soft' so you don't have to unstage all of the ignored files
17:28 Keverw Oh switching to System Git, deleting the repo locally and restarting it for good measure... Then I checked out and it works! It has my login saved I guess from when I did it with the command line maybe as it never asked but it does say the repo needs repair... I just click cancel and it works as expected :)
17:28 jast this is all assuming you haven't pushed yet
17:28 jast in full: git reset HEAD~
17:29 Keverw So yeah I guess Source Tree works a bit different with the embedded one. Going to system one and it's prefect. Thanks everyone :) I learned a bit more about Git too. I'm guessing BitBucket and Github might be something slightly different maybe.
17:29 codebam what if it was the first commit?
17:29 Cabanossi joined #git
17:29 YuGiOhJCJ joined #git
17:29 jagob joined #git
17:29 codebam do I just redo the whole repo
17:29 codebam since it wont let me reset before the initial commit
17:29 gugah joined #git
17:30 jast another option is to rm .git/index, stage things the way you want them, then 'git commit --amend'
17:30 mohabaks joined #git
17:30 jast that will work for the first commit
17:30 codebam oh okay cool
17:31 jast if it complains about a broken index, that should be easy to fix, too
17:32 GordCaswell joined #git
17:32 memfrob git fix --index (jk)
17:34 nowhereman joined #git
17:34 griffindy joined #git
17:34 d1z joined #git
17:35 GordCaswell Hi folks, I'm using Git Gui, and have just setup gpg signing on an existing repo, but now get an error on committing, and I've had no luck finding an answer online on how to fix the error: commit-tree failed: 'can't read "old_author": no such variable'
17:36 Es0teric joined #git
17:37 Es0teric joined #git
17:37 jagob joined #git
17:38 nebg joined #git
17:39 jast GordCaswell: looks like a bug / incomplete error handling in git-gui. I'm not terribly familiar with git-gui's code, maybe it's worth asking about this on the Git mailing list
17:39 jast !list
17:39 gitinfo [!mailing_list] The mailing list can be reached via git@vger.kernel.org. You don't need to subscribe to the list, you will always be put in cc on reply. Read archives at http://j.mp/gitlist
17:39 GordCaswell Thanks, jast, I'll try there.
17:40 govg joined #git
17:40 jast clearly my knowledge of Tcl is too weak for this... from looking at the code, I don't even see how that message could come into being
17:41 memfrob git grep "can't read" might help narrow it down
17:41 hwrdprkns joined #git
17:41 jast I've found the code, I just don't understand that particular combination
17:41 _ikke_ git grep -r old_author
17:41 jast old_author is conditionally defined a few lines above but it's not even used where the error is generated
17:42 memfrob Send the code I'll take a peek
17:42 memfrob link to it or something
17:42 memfrob I'm a C guy but eh I'll try XD
17:42 jast does anyone remember how this gitlink thing works? ;)
17:43 _ikke_ https://github.com/git/git/bl​ob/master/git-gui/git-gui.sh
17:43 _ikke_ Nope
17:43 jast I wrote it but I never remember the syntax
17:43 nowhereman joined #git
17:43 denete__ joined #git
17:43 jast the code is in a lib subdir
17:43 jast hang on a sec
17:43 mohabaks joined #git
17:43 _ikke_ something like git:git.git:master
17:43 _ikke_ something like git:git:git:master
17:43 vuoto joined #git
17:44 jast I'm re-reading the code, maybe that'll help
17:44 _ikke_ last try: github:git:git:master
17:44 memfrob git-gui/lib/commit.tcl
17:44 jast yeah, that's the one
17:44 jast line 388 is where the error seems to be output
17:45 jast the first deref of old_author seems to be after that in the same error handler...
17:46 _ikke_ what version of git-gui?
17:46 jaafar joined #git
17:46 _ikke_ Not that I suspect this to have changed much though
17:46 Noldorin joined #git
17:46 GordCaswell 0.21.GITGUI, running git 2.11.0.windows.1
17:47 _ikke_ ah
17:47 jast git:github:git/git:git-gui/lib/commit.tcl
17:47 gitinfo Git web link: https://github.com/git/git/co​mmits/git-gui/lib/commit.tcl
17:47 netj joined #git
17:47 jast still wrong
17:47 jast git:github:git/git::git-gui/lib/commit.tcl
17:47 gitinfo Git web link: https://github.com/git/git/blo​b/HEAD/git-gui/lib/commit.tcl
17:48 jast yeah, that's the one
17:48 jast I gotta make this easier to use
17:48 _ikke_ I think it should be tree between the colons
17:48 jast nah, that's the optional ref
17:48 jast the obj type is auto-guessed
17:48 _ikke_ ok
17:48 GordCaswell I do see an if exists mention at line 373, but I can't figure out how to set it
17:49 lindenle joined #git
17:49 DrSlony joined #git
17:49 jast are you amending a commit?
17:50 flavius joined #git
17:50 DrSlony Hey again. What is the best way to detect whether HEAD is currently on a tag? The purpose is to check whether what is being compiled is a release or just a development version.
17:50 GordCaswell Nope, trying to add a new one. The only other change is that I added gpg signing prior to this commit.
17:51 jast DrSlony: check that 'git describe --exact-match HEAD' exits with code 0
17:51 jast and suppress its output :)
17:52 DrSlony thanks jast
17:52 jast or git name-rev --tags --no-undefined HEAD
17:52 memfrob echo $?
17:52 cheater joined #git
17:52 cheater hi
17:53 memfrob hi
17:53 _ikke_ hi
17:53 jast hi ;)
17:53 cheater is there a way to refer to the current state of the tree (with unstaged and uncommitted changes) as a rev or some other identifier when doing git diff?
17:53 Derperpe1 joined #git
17:53 gitinfo hi
17:54 cheater hi
17:54 jast cheater: if you give only a single rev, it defaults to diffing against the working tree
17:54 cheater yea but i want to diff the other way
17:54 ^7heo hi
17:54 timdiels left #git
17:55 jast as in, you want to see the reverse of the difference?
17:55 gitinfo hi
17:55 ^7heo hi
17:55 memfrob hi
17:55 ^7heo (git diff -R)
17:55 cheater jast: yes
17:55 ^7heo cheater: ^
17:55 ^7heo cheater: hi
17:55 cheater hi ^7heo
17:55 cheater hi
17:56 ^7heo hi
17:57 gitinfo http://g.jk.gs/Eh.png
17:58 jast silly bot
17:59 tyreld joined #git
18:01 ^7heo https://makeameme.org/meme/hi-317kmv
18:01 bilb_ono joined #git
18:01 memfrob Ugh my parents keep cranking the freezer so cold that the popsicles always stick to my tongue
18:01 ^7heo or even https://media.makeameme.org/created/hi-317kmv.jpg
18:01 memfrob i can't even eat them!
18:02 ^7heo memfrob: /nick memfrost
18:02 cheater lol
18:02 memfrob XDD
18:02 ^7heo also, #firstworldproblems...
18:02 cheater #git isn't always welcoming, but when it's welcoming it's very welcoming
18:02 nickabbey joined #git
18:02 ^7heo that's what he said.
18:03 cheater STOP GENDERING ME
18:05 jaafar joined #git
18:05 memfrob earth is the 3rd planet from the sun. any problem is a third world problem.
18:05 _ikke_ jast: who's the cheather here :P
18:06 Gsham joined #git
18:06 m0viefreak joined #git
18:07 denete_ joined #git
18:07 elsevero joined #git
18:08 jast _ikke_: I have no idea what you're talking about :P
18:08 sea-gull joined #git
18:08 mischat joined #git
18:08 KV joined #git
18:09 ikelso joined #git
18:09 anddam left #git
18:11 grayjoc joined #git
18:13 zeroed joined #git
18:13 zeroed joined #git
18:13 anuxivm left #git
18:13 Balliad joined #git
18:13 _ikke_ @gitinfo │ hi
18:14 doener joined #git
18:15 jast _ikke_: so what if gitinfo likes to be friendly sometimes
18:15 spacelord_ joined #git
18:16 mojjo joined #git
18:17 wizardedit joined #git
18:17 wizardedit joined #git
18:18 Snugglebash joined #git
18:19 _ikke_ yeah, whati f
18:19 _ikke_ what if
18:20 leftriver41 Hey all. Not sure if this is the right place for this, but hoping to get some help. Or even if someone can point me to a good resource.
18:21 _ikke_ !just_ask
18:21 gitinfo You can just ask your question. If anybody knows the answer, they will answer soon (most of the time)
18:21 leftriver41 I work in a repo, where we have to use different code for different OSes (RHEL, SLES, Debian, Ubunti, etc) and we have different versions of the product (1.1, 1.2, 2.1, 2.2, 2.3, 3.0, etc)
18:21 leftriver41 We currently have a new branch for each OS and Version combination, and it's getting to be very complicated.
18:21 stickperson joined #git
18:21 leftriver41 Does anyone deal with code like this? Or is there a good git branch/flow model to follow?
18:23 roygbiv joined #git
18:23 hahuang65 joined #git
18:23 LBV_User joined #git
18:24 Disgeae joined #git
18:24 XenophonF joined #git
18:26 memfrob How is it complicated? Just have a branch named in a way like Mac_3.0 so you know that that branch is v3.0 for Mac
18:26 memfrob then git branch -a to pick your branch
18:27 memfrob Deb8_v1.2 for example
18:28 leftriver41 memfrob: What happens is if we have to support 7 distros, and 5 version, we now have 35 branches that are active at once. And the merge strategy isn't clear either. So people are just cherry-picking commits everywhere.
18:28 memfrob It shouldn't be difficult at all to figure out what version of OS / distro + product version if you name your branches right
18:28 leftriver41 memfrob: We are currently doing what you are suggesting; I'm just wondering if there is a better way. But i'm not sure.
18:28 _geoff joined #git
18:29 memfrob Well as opposed to cherry picking everything, at least maybe have a branch semi common (if possible) that you can just pull from other branches?
18:29 leftriver41 Yeah. Knowing the branches isn't too tough. I just hate that we don't have a good merge strategy. And maybe that's the problem I should be looking into more.
18:29 mda1 joined #git
18:29 memfrob like if you know you're going to be adding something that will be used everywhere, put all those commits (if any) all in one branch
18:29 leftriver41 memfrob: That's actually making a bit of sense. Having an intermediate branch, where we can merge.
18:29 memfrob yep!
18:30 leftriver41 I think your suggestion is helping me realize the problem. The program managers don't always have a clear idea of what features will be supported. So they just say "chery-pick commit XXX to here and here".
18:30 leftriver41 Ok. I'm open to other suggestions, but thanks!
18:30 memfrob np doin my best
18:31 ecuanaso joined #git
18:31 Squarism is "git-flow" what everyone uses or is just some scrum-prophet inventing air?
18:32 GodGinrai Squarism: prophet giving out koolaid that many people like to drink
18:33 reynierpm left #git
18:33 leftriver41 Squarism: I think in an ideal world, people try to use git-flow. But for any number of reason, people don't.
18:34 iron74 joined #git
18:34 jccn joined #git
18:34 memfrob What I've been doing personally is checking out a branch that has lots of things I want, but as opposed to cherry picking ALL of them, I check out the latest branch of the patch set I want, running git diff HEAD~X number of commits where it first started &> my_clobbered_patch.patch and applying that to my tree so it saves all the time copypasta commits and such
18:34 memfrob git format-patch also does that but it depends on the desired effect.
18:35 memfrob I personally use git in it's many ways, some methods less approved of of others but I do what works best for me and my own workflow, not what git decides it should be.
18:37 memfrob The way I think and process is very different from other people. I'm not normal by any means so I get my work done in the most efficient way for myself and how I'm wired.
18:38 Disgeae joined #git
18:38 thiago joined #git
18:39 memfrob I mean, anyone here will tell you I'm crazy :P It's the same with martial arts. No single style works for every human, for every fight, in every environment, under all the same circumstances. same philosophy applies to work ethic. at least IMO.
18:40 memfrob err.. *under all circumstances
18:44 adino joined #git
18:45 reynierpm joined #git
18:45 Doginal left #git
18:45 reynierpm I have a repository with two branches: master and httpd, I want to get rid of master and make httpd the master branch, it's possible? how?
18:47 miczac joined #git
18:48 tabakhase reynierpm "nothing is stoping you from that"
18:48 tabakhase git branch -D master and its gone
18:49 memfrob rename branch httpd to master?
18:49 reynierpm tabakhase, that will erase branch from remote and from local but how do I make httpd the master?
18:49 tabakhase reynierpm do you want to keep calling it https?
18:49 tabakhase *d
18:49 reynierpm tabakhase, no
18:50 reynierpm it doesn't matter to me
18:50 reynierpm could be renamed into master if that's the solution
18:50 memfrob well if master is gone, then httpd will became the master automatically
18:50 memfrob but to rename httpd to master idk
18:50 TomyLobo joined #git
18:50 tabakhase then go to master and merge in your branch (with the additional changes it has i guess)
18:51 reynierpm memfrob, so if I remove master leaving only httpd this last will become master, that's what you said? I don't need to run any other command?
18:51 tabakhase you could also just delete the master, checkout httpd, and recreate master there with a "git checkout -b master" --- but the merge is more likely what youre actually looking for
18:51 memfrob or to remove all of master's history to make it look like httpd has always been master, to REALLY make it master from scratch, git reset back to the initial commit and apply all changes from httpd into master
18:52 reynierpm tabakhase, I don't want to merge httpd in master they are complete different and at the end I will end with stuff from httpd in master and viceversa
18:52 jnavila joined #git
18:52 tabakhase maybe cherry picking / interactive rebase is then your way to go?
18:52 reynierpm memfrob, ummm I get lost
18:52 reynierpm :-(
18:53 canuk joined #git
18:53 memfrob yeah if you dont want to merge you're gunna either have to do a clean swipe of master or delete the master branch and create a branch called master off of httpd
18:53 memfrob clean swipe meaning removal of all commits from master branch
18:54 e14 joined #git
18:54 tabakhase depends a bit on wheres your starting situation and what problem you wanne solve ;-) - the branchname in the end is just a nametag on a treebranch, and you can swap them arround as you please
18:55 reynierpm tabakhase, doesn't matter to me either the history from master, I just want to make httpd the master because I started from master but at the end I finish doing all the work in httpd and merging them now will became a mess
18:55 frostythesnowman joined #git
18:55 tabakhase so did you create httpd branch?
18:56 reynierpm tabakhase, yes
18:56 tabakhase and you did that before youve made changes in master - that you "would like to have as well"?
18:57 reynierpm tabakhase, I did made different changes in both that's why I don't want to go with merging
18:57 reynierpm tabakhase, at some point they became complete different each other
18:58 leftriver41 left #git
18:58 memfrob git checkout httpd && git branch -D master && git checkout -b master
18:58 tabakhase ever played around with git cherry-pick? sounds like your fix... - selectivly pull singular commits "from one branch to another"
18:59 thiago joined #git
18:59 reynierpm tabakhase, I don't played never with cherry-pick
19:00 memfrob or..
19:00 nickabbey joined #git
19:00 ikelso joined #git
19:00 jstimm joined #git
19:01 tabakhase next quest would be what you expect to happen from "changing the master"
19:02 memfrob git checkout httpd && git format-patch HEAD~XYZ first commit blah blah blah && git checkout master && git reset --hard <very first commit of master> && git am <patch file of all commits in httpd> && git push -f
19:02 memfrob BAM!
19:02 Es0teric joined #git
19:03 tabakhase if youre ending on a "tested smth in my local master, then made httpd and finished all there" you might be just fine as is ;-)
19:04 memfrob eh im close, something like that
19:04 btree joined #git
19:04 tabakhase maybe you actually want to drop master and relabel, - maybe you want to merge (after droping some stuff from master) - or maybe you just want to rebase ontop of it... not sure from what ya said so far...
19:04 seni joined #git
19:04 memfrob mind you, my way is a bit dangerous, this is why people advise to not do what i do
19:05 memfrob i think he wants to go fuck all based on everything he's said so far
19:05 memfrob he wants to burn master to the ground and have httpd take over like its nobody's business
19:05 titankiller joined #git
19:05 reynierpm tabakhase, memfrob ok then which one is the way to go?
19:05 memfrob my way kills master to the point of no return and makes httpd the new master
19:06 tabakhase "solution depends on your problem" - give us your problem, not your suggested solution :P
19:06 memfrob (and being called master)
19:06 memfrob a full reset and a git am will do the trick if thats what you really want
19:06 reynierpm tabakhase, problem: I have two branches master and httpd, I want httpd be the master but without merging
19:07 reynierpm memfrob, you wrote to examples, which one to go?
19:07 buzzLightBeer joined #git
19:07 memfrob the first one is much safer and cleaner
19:07 tabakhase git checkout master; git checkout -b old_master; git branch -D master; git checkout httpd; git checkout -b master;
19:07 memfrob ^that
19:08 memfrob do that
19:08 memfrob git aming an entire branch history and purging it of everything is not something you wanna do on your first try :P
19:08 b100s joined #git
19:09 tabakhase this will go to master, put down a sign calling it old_master, rip down the master sign, swap over to httpd and pull up a new master sign there
19:09 memfrob yeah but dont forget to run -D old_master unless you want to keep what was the master from before
19:10 tabakhase or disable GC Kappa
19:10 memfrob GC Kappa?
19:10 hasc joined #git
19:10 dreiss joined #git
19:11 tabakhase garbagecollection + a twitch emote - habbits...
19:11 tabakhase sarcasm warning basically ;-)
19:11 jsolano joined #git
19:11 memfrob XD
19:12 reynierpm tabakhase, memfrob "hint: Updates were rejected because the tip of your current branch is behind
19:12 reynierpm hint: its remote counterpart. Integrate the remote changes (e.g.
19:12 reynierpm hint: 'git pull ...') before pushing again.
19:12 reynierpm hint: See the 'Note about fast-forwards' in 'git push --help' for details."
19:12 tabakhase reynierpm hint: See the 'Note about fast-forwards' in 'git push --help' for details."
19:12 jast did anyone mention that all clones have to be manually fixed or re-cloned when you change master to something else?
19:13 hasc_ joined #git
19:13 memfrob jast, well where you 10 minutes ago? XD
19:13 reynierpm tabakhase, memfrob https://paste.fedoraproject.org/537192/
19:13 Cabanossi joined #git
19:13 jast memfrob: busy :P
19:14 reynierpm now I can't push to Github and I am still seeing both branches there
19:14 memfrob try -f
19:14 memfrob git push -f
19:14 darkSeid_ joined #git
19:14 ertes joined #git
19:15 Sasazuka joined #git
19:15 reynierpm memfrob, that worked
19:15 reynierpm thanks!!! guys
19:15 memfrob sweeeeeeeeeeeeeeet
19:15 iron74 joined #git
19:16 * memfrob wins MVP award
19:16 Darren_ joined #git
19:16 * memfrob kisses MVP trophy and receives applause
19:16 * memfrob becomes full of himself
19:16 tyreld joined #git
19:16 Vinnie_win joined #git
19:16 jast most valued platypus?
19:17 darkSeid_ hi all
19:17 gitinfo darkSeid_: 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.
19:17 memfrob platypi look like ducks with long beeks
19:17 alex1a joined #git
19:17 cjohnson gitinfo: you have beautiful eyes
19:17 memfrob gitinfo: have sex with me please
19:17 tabakhase ;D that welcome message... " 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."
19:17 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
19:17 tabakhase mem 2 minutes ago "try -f"
19:17 jast strange ducks, those
19:18 gitinfo cjohnson: well I do have beautiful contacts
19:18 cjohnson lol
19:19 gitinfo memfrob: EINVALIDPORT
19:19 memfrob XDD Best response ever
19:20 GodGinrai EIN VALID PORT <.< >.>
19:21 jast DIE BART DIE
19:23 hwrdprkns joined #git
19:24 d1z joined #git
19:26 re1 joined #git
19:28 darkSeid_ hi, so im quite new to git and im getting E325: ATTENTION saying there’s a swap file, after i tried git pull
19:28 darkSeid_ literally have no clue whats going on lol
19:29 jast doesn't look like a message from Git itself, but maybe the full message would be helpful
19:29 Balliad joined #git
19:29 qqx That's a message from vim
19:30 darkSeid_ ah ok
19:30 qqx It happens when another instance is already editing the same file, or was killed while editing that file and not able to clean up.
19:31 darkSeid_ i think i accidentally pressed some things i shouldn’t have while trying to update a forked repo
19:31 ShapeShifter499 joined #git
19:31 darkSeid_ yeah i _think_ i accidentally edited it? any way to undo this mess?
19:31 darkSeid_ i know this isn’t actually a git issue so apologies :/
19:32 qqx At least for me it gives additional information listing the file that it's trying to open, the swap file it found, and the PID that wrote that.
19:32 qqx It also prompts about how to handle the situation.
19:34 darkSeid_ ok i recovered the file, i’ll just cross my fingers that i didn’t destroy something in the process
19:34 darkSeid_ thanks, btw :)
19:36 svm_invictvs joined #git
19:36 jnavila joined #git
19:37 matoro joined #git
19:38 Phlunk3 joined #git
19:38 nickabbey joined #git
19:43 matoro joined #git
19:45 kyan joined #git
19:45 alipoor90 joined #git
19:46 netj joined #git
19:46 MattMaker joined #git
19:47 cjohnson I have the following code as part of a deploy process (I would love to have a better dpeloy process but this is what I have for hte moment)
19:48 cjohnson git reset --hard HEAD && git checkout -- . && git clean -df && git pull && git submodule update --init --recursive
19:48 cjohnson a dev pushed a change which removed a submodule, and now it's getting hung up on git checkout -- . saying "the following untracked the files would be deleted by checkout"
19:48 cjohnson can I just move the git clean -df before the checkout to solve this?
19:48 alipoor90 Hi, I have a repo with commit history like this: `A (initial) - B - C - D - E - F - G - H`, I want to squash A and B and C and D and E into one commit (i.e. remove unneeded old history which takes disk space), how can I do this?
19:51 ToxicFrog alipoor90: git rebase --interactive and then set B, C, D, and E to "squash" or "fixup".
19:51 ToxicFrog That said, unless you have alarmingly large files in there, the history really doesn't take much space at all; it's all deduplicated and compressed
19:52 ToxicFrog And if you *do*, you will additional work to clean it all out.
19:53 hwrdprkns joined #git
19:54 alipoor90 ToxicFrog: Thanks
19:55 mischat joined #git
19:56 kpease joined #git
19:56 jccn joined #git
19:58 penguinoneman joined #git
19:58 nd joined #git
19:58 notebox joined #git
19:58 Cabanossi joined #git
19:59 e14 joined #git
20:00 ecuanaso joined #git
20:01 Derperpe1 joined #git
20:01 cdown joined #git
20:02 Balliad joined #git
20:04 cyan__ joined #git
20:05 cyan__ joined #git
20:05 M-nated left #git
20:06 cyan__ joined #git
20:08 Exagone313 joined #git
20:10 crayon joined #git
20:14 swift1 joined #git
20:15 bilb_ono joined #git
20:15 apotry joined #git
20:21 hashpuppy joined #git
20:22 Exagone313 joined #git
20:24 inflames joined #git
20:24 stickperson joined #git
20:25 vuoto joined #git
20:25 aidalgol joined #git
20:25 MacWinner joined #git
20:26 ccolorado joined #git
20:26 jnavila joined #git
20:28 pur3eval joined #git
20:28 ccolorado Hi there, so I am using git-svn an I am having troubles undertanding why on my linux box I get multiple timeouts when trying to clone the svn repo. Howver there is this mac mini that has no issue doing the clone, it does it faster and in the first try.
20:28 Cabanossi joined #git
20:28 ccolorado *understanding
20:33 askb joined #git
20:33 jimi_ joined #git
20:34 davimore joined #git
20:35 Es0teric joined #git
20:36 KV joined #git
20:37 Exagone313 joined #git
20:40 rivarun joined #git
20:41 grayjoc joined #git
20:42 shampine joined #git
20:43 Synthbread joined #git
20:45 joelsmith joined #git
20:45 Toordog joined #git
20:48 XenophonF joined #git
20:50 Atm0spher1c joined #git
20:51 adino joined #git
20:53 Exagone313 joined #git
20:53 Xenophon1 joined #git
20:55 mehola joined #git
20:56 oskarkv joined #git
20:56 e14 joined #git
20:56 compeman left #git
20:56 svm_invictvs joined #git
20:58 Domhack joined #git
21:00 b100s joined #git
21:04 Tobbi joined #git
21:07 svm_invictvs joined #git
21:07 crose joined #git
21:11 thebinary joined #git
21:14 Cabanossi joined #git
21:15 dreiss joined #git
21:16 Wja joined #git
21:17 rivarun joined #git
21:17 cdg joined #git
21:23 cdg joined #git
21:23 MTecknology left #git
21:23 WifiBrain joined #git
21:23 MattMaker joined #git
21:25 jast ccolorado: different versions maybe?
21:27 d10n-work joined #git
21:27 jeffreylevesque joined #git
21:33 ismaris joined #git
21:35 Noodlewitt joined #git
21:36 manuelschneid3r joined #git
21:37 inflames joined #git
21:37 matsaman joined #git
21:38 codemann joined #git
21:43 chipotle joined #git
21:44 BackEndCoder joined #git
21:44 KV joined #git
21:46 jimmyar joined #git
21:46 jimmyar hello
21:46 gitinfo jimmyar: 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.
21:46 jimmyar can i know the git repo size before hand? before downloading the repo?
21:46 jimmyar more like git size up that repo?
21:46 jimmyar GingerGeek: ?
21:47 nowhereman joined #git
21:47 jimmyar anyone is here? to answer me ?
21:47 jimmyar nowhereman: hi
21:49 Anja joined #git
21:51 jimmyar beatdown: BlessJah da5id reactormonk Renter anyone
21:51 jimmyar ThumbBone: ToBeFree
21:54 govg joined #git
21:55 cjohnson When I switch branches, old branch has a submodule at a given path, and the new branch replaces that submodule with tracked files, I get: "error: The following untracked working tree files would be overwritten by checkout:"
21:55 cjohnson How can I switch branches?
21:57 cjohnson git checkout -f boom
21:57 jast jimmyar: Git doesn't know beforehand, sorry
21:58 nowhere_man jimmyar: don't tag random people, please
21:59 jast yeah, no waking up everyone, they need their beauty sleep
21:59 jast so do I, actually
21:59 kulelu88 joined #git
22:00 BlueProtoman joined #git
22:00 dhollinger joined #git
22:01 BlueProtoman Some files in my Git repo got damaged (including files in .git).  How can I restore my repo to a good state?
22:01 jast by restoring from backups
22:02 jast manual recovery is not very straightforward, and you still need to get the original objects from somewhere
22:03 BlueProtoman Damn.
22:03 dviola joined #git
22:04 jast if only ref files are affected it might be not too hard
22:05 Singmyr joined #git
22:06 BlueProtoman jast: I just recloned the repo anyway
22:06 Hudu joined #git
22:06 BlueProtoman Still weird, though.  What could have damaged my filesystem?
22:07 ac3takwas joined #git
22:08 rwb joined #git
22:08 alex1a joined #git
22:08 canuk joined #git
22:11 FreakySneaky joined #git
22:12 Es0teric joined #git
22:13 re1 joined #git
22:14 Cabanossi joined #git
22:14 astrofog joined #git
22:14 senaps joined #git
22:16 dsdeiz joined #git
22:16 dsdeiz joined #git
22:18 btree left #git
22:21 e14 joined #git
22:23 pur3eval joined #git
22:24 overlord_tm joined #git
22:26 chachasmooth joined #git
22:27 e14 joined #git
22:27 jast BlueProtoman: sudden power failure, faulty RAM, hardware/driver issue
22:27 astrofog joined #git
22:28 sveinse left #git
22:28 hasc joined #git
22:30 Kobaz so i have this: * 2.0                     45abd3e [ahead 4] Merge branch '2.0'
22:31 Kobaz but then: git diff origin/2.0..2.0
22:31 Kobaz is no diffs
22:31 Kobaz so it's ahead by 4, but there's no actual diffs
22:31 Kobaz how do i 'clean'
22:32 phroa 2.0 might be tracking a different branch than origin/2.0?
22:32 Kobaz it's the same
22:33 Kobaz i have Merge stuff in the logs
22:33 Kobaz so it's like extra entries of merges, but no actual changes
22:34 phroa oh, merge commits.  just push 2.0 to origin to get origin in sync
22:34 Kobaz what if i don't want the not useful extra merges in there
22:34 Kobaz any way to just drop them locally
22:34 phroa sure, check out 2.0 locally then `git reset origin/2.0`
22:35 Kobaz ah that works
22:35 Kobaz yay
22:36 Kobaz diff is completely nothing, so there's no point in pushing fluff
22:36 phroa if it helps, see man git cherry if you want a diff of commit logs rather than diff of file changes
22:36 gitinfo the git-cherry manpage is available at http://jk.gs/git-cherry.html
22:37 Kobaz oh yeah that would be nice
22:37 Kobaz show the log diff
22:37 e14 joined #git
22:37 hwrdprkns joined #git
22:37 ToBeCloud joined #git
22:40 e14_ joined #git
22:44 Gsham joined #git
22:44 hecliunyx joined #git
22:45 roelmonnens joined #git
22:45 KV joined #git
22:45 mikecmpbll joined #git
22:47 crayon joined #git
22:49 danimal2 joined #git
22:50 mellernoia joined #git
22:52 przemoc joined #git
22:52 przemoc joined #git
22:56 nidr0x joined #git
23:01 fmeerkoetter joined #git
23:01 afuentes joined #git
23:03 duderonomy joined #git
23:06 hwrdprkns joined #git
23:06 ShadeS joined #git
23:06 ShadeS (use "git pull" to merge the remote branch into yours)
23:06 ShadeS 4 and 7 changes
23:06 dirtyroshi joined #git
23:06 Vinnie_win_f joined #git
23:09 Dougie187 left #git
23:09 ShadeS oh whoops i had a memory issue if ixed i'm ahead of master locally derp
23:11 cfoch joined #git
23:11 cfoch hi
23:11 cfoch I am using interactive rebase to edit a commit
23:11 cfoch can I use "git add --patch" there? because I get "No changes"
23:11 cfoch ?
23:11 re1 joined #git
23:12 phroa I *believe* when you enter edit mode all the changes are staged by default.  (git status to see, of course.)  try `git reset --patch` instead, to unstage the changes you don't want
23:18 courrier joined #git
23:20 cdg joined #git
23:20 Hudu joined #git
23:20 cfoch that doesn't work
23:21 mostlybadfly joined #git
23:25 idlemind joined #git
23:28 msonntag joined #git
23:29 Node_874 joined #git
23:29 fshp joined #git
23:30 justanotheruser joined #git
23:31 alphor joined #git
23:31 fshp Hi. I can change commit and author date over env variable GIT_AUTHOR_DATE and GIT_COMMITTER_DATE. How to change gpg sign date?
23:31 rkazak joined #git
23:31 fshp Only faketime?
23:34 Dougie187 joined #git
23:36 Dougie187 left #git
23:38 lindenle_ joined #git
23:40 roelmonnens joined #git
23:45 KV_ joined #git
23:50 cfoch I am using git add -p
23:50 cfoch I have this message
23:50 cfoch Your edited hunk does not apply. Edit again (saying "no" discards!) [y/n]
23:50 cfoch what does Git means with "discards"?
23:50 cfoch it discards the hunk?
23:50 cfoch or it discards the edition?
23:53 a_thakur joined #git
23:55 FilipNortic__ joined #git
23:55 jeffspeff joined #git
23:58 Gsham joined #git
23:58 ecuanaso joined #git
23:59 jave joined #git
23:59 cfoch I found other way, but I am still curious :)

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