IRCloggy #git 2013-07-24

Logs Search ←Prev date Next date→ Channels Documentation

Provider of IRC logs since 2005.
WARNING: As Freenode became unjoinable and lost all warnings in topics, we cannot log channels on Freenode anymore.

2013-07-24

← oliveiraev left00:00
← mklappstuhl left00:01
ozette i'm not too sure on how to prevent a conflict if i have for example 3 patches commited, but before i can check those in i have to pull in changes from upstream00:01
→ schemanic joined00:02
ozette or if i want to merge a topic branch back into master00:02
EugeneKay git doesn't "check in", it pushes.00:02
ozette or vise versa00:02
EugeneKay The point is to fetch/merge BEFORE editing/commiting, so you're never creating a conflict to begin with00:03
The viability of this in a high-traffic repo varies00:03
ozette hmm ok, so saving up is bad practice if its not a personal branch?00:04
EugeneKay Depends upon your project practices00:04
→ enr joined00:04
cjz why dont git submodules pull when you pull the super directory?00:04
EugeneKay I branch off from wherever "unstable"(or "dev") is into a Topic branch, which shares an ID# with our Issue Tracker. I do my thing, then push it. If the code is good, it gets merged back into unstable, resolving any conflicts at that point.00:05
!flow00:05
gitinfo [!gitflow] The description of the gitflow branch workflow model is at http://nvie.com/posts/a-successful-git-branching-model/ while a tool to help implement this workflow is at https://github.com/petervanderdoes/gitflow See http://sethrobertson.github.com/GitBestPractices/#workflow for other workflow suggestions/references00:05
← boombatower_ left00:06
ozette aha00:06
→ tinti_ joined00:06
← johnnyfuchs left00:08
ozette thank you00:08
← lindenle left00:11
→ belak joined00:11
← myoung left00:12
← frgtn left00:12
→ macabre joined00:12
→ joelpittet joined00:12
→ myoung joined00:12
← Marquisk2 left00:12
→ osmosis joined00:13
← enr left00:13
← pitzips left00:14
← iliketurtles left00:14
← Vampire0_ left00:15
→ rchavik joined00:15
→ v0n joined00:16
→ DrGonzo joined00:16
→ SinnerNyx joined00:17
← Gonzo left00:17
→ lavaman joined00:17
→ gua_ould joined00:19
← breck7 left00:20
← osmosis left00:21
← BJfreeman left00:21
← RavenHursT left00:22
→ Cromulent|2 joined00:24
← tpollard left00:25
→ hugo_dc joined00:25
→ awygle joined00:25
← FernandoBasso left00:26
← cbreak left00:26
→ cballenar joined00:26
→ tang^m joined00:26
→ normanrichards joined00:27
← happyg0pher left00:27
→ cballenar1 joined00:28
← cballenar left00:28
← cballenar1 left00:29
→ cbreak joined00:29
← cjz left00:30
juliohm What GNOME has to do with Git? I get an annoying warning about gnome-keyring00:30
← DonRichie left00:30
→ ewnd9 joined00:30
juliohm is running Kubuntu00:31
← AaronMT left00:31
→ DonRichie joined00:31
← zhost left00:31
→ mckendricks joined00:32
→ carlos-huang joined00:32
← sr78ger left00:33
← Noldorin left00:34
← yshh left00:34
→ Noldorin joined00:34
← hyperair left00:34
← Noldorin left00:35
→ Noldorin joined00:35
→ R1KRDO joined00:35
bremner git or git-svn?00:35
→ sr78ger joined00:36
← tang^m left00:37
→ chaw joined00:37
→ Vampire0 joined00:38
→ devoldmx joined00:38
→ cballenar joined00:41
→ JPeterson joined00:42
→ setmeaway joined00:42
→ frgtn joined00:43
→ tgs3_ joined00:43
tgs3_ how to download "00:43
← ozette left00:43
tgs3_ git://git.kernel.org/pub/scm/linux/kernel/git/mmarek/kbuild-2.6.git deterministic-build-v1 on linux exactly?00:43
→ ozette_ joined00:43
→ HenryG joined00:44
← juliohm left00:44
→ byprdct joined00:44
← byprdct left00:45
→ nerdfiles joined00:47
→ byprdct joined00:48
tgs3_ anyone?00:48
← ewnd9 left00:49
→ fairuz joined00:49
← Cykey left00:49
← hahuang65 left00:50
→ hahuang65 joined00:50
← gua_ould left00:50
→ enr joined00:50
← frgtn left00:52
← fairuz left00:52
milki tgs3_: git clone that url you gave00:52
→ weslly joined00:52
→ gmc_ joined00:52
milki tgs3_: then git checkout that branch name you gave00:53
tgs3_: then read !book00:53
gitinfo tgs3_: There are several good books available about git; 'Pro Git' is probably the best: http://git-scm.com/book but also look at !bottomup !cs !gcs !designers !gitt !vcbe and !parable00:53
ivan`_ → ivan`00:53
← hugo_dc left00:53
← gmc_ left00:53
tgs3_ milki: it's not working. Is this git not working? Maybe it was moved00:53
→ LuxuryMode joined00:54
→ gmc_ joined00:54
← enr left00:55
gmc_ → zikemanb00:56
LuxuryMode When I have conflict on a `git pull --rebase` and am told to resolve then run git rebase --continue…i should be adding the resolved files before continuing, right?00:56
← weslly left00:56
← zikemanb left00:56
← harish left00:57
← GT4066 left00:57
→ gua_ould joined00:57
bremner yes00:58
← gua_ould left00:58
← chrob left00:58
→ Marquisk2 joined00:59
→ yshh joined00:59
← ozette_ left01:01
→ bdavenport joined01:01
→ ozette joined01:01
→ Marenz joined01:01
→ zikemanb joined01:01
← zikemanb left01:02
← lifelike left01:02
← tinti_ left01:02
→ lifelike joined01:03
→ hugo_dc joined01:04
→ zikemanb joined01:04
← zikemanb left01:04
← Suprano left01:04
→ hyperair joined01:05
← cdash004 left01:05
→ k_sze[work] joined01:05
← freestyl3r left01:07
← causasui left01:08
← R1KRDO left01:08
← LuxuryMode left01:09
← cballenar left01:09
→ Suprano joined01:10
← joelpittet left01:11
← Marenz left01:15
→ umib0zu joined01:16
→ Goplat joined01:16
← thiago left01:18
→ cballenar joined01:19
→ johnnyfuchs joined01:22
← groglogic left01:23
→ nurey joined01:23
← meLon_ left01:23
→ jalama joined01:24
→ meLon joined01:24
→ jackdempsey joined01:24
→ rylangrayston_ joined01:24
rylangrayston_ .01:25
→ dhruvasagar joined01:26
← towski left01:26
← johnnyfuchs left01:27
rylangrayston_ so im new to git ... and im wanting to track a script inside blender3d is it best to put the script in a txt doc or just track the hole blend file?01:27
→ McEnroe joined01:28
Arokh → Arokh[Off]01:28
jon_r are blend files binary?01:29
← trusktr left01:29
sam113101 → sam113101_afk01:29
sam113101_afk → sam11310101:30
← schemanic left01:31
→ schemanic joined01:31
rylangrayston_ i dont know01:32
.. arnt all files binary ?01:33
→ hetal_ joined01:33
osse yes01:35
but Git doesn't compress files that aren't text very well01:36
but in common lingo there is a seperation between "binary" (JPG, PDF, etc. etc.) and text files01:36
→ whaley joined01:37
→ evjsli joined01:40
rylangrayston_ yes i understand now .. iv been googlein the differnce..01:40
SamB osse: its not so much that git is bad at compressing binary files, as that many binary files are uncooperative :-(01:41
osse SamB: aren't many binary files well compressed already, typically? (E.g. pictures and videos)01:42
SamB yes but this plays bloody hell with the delta compression01:42
osse ahh the delta compression01:43
← belak left01:43
→ luc_ joined01:44
← owen1 left01:45
rylangrayston_ my .blend files are really small .. like 800 kb... in the blend file is a text file(but i dont know if its stored as bianary in that file) im wondering if git will track each like in the script in my blend file01:45
*each line in the script01:46
Or would it be better to put my scripts in text files out side the blend file01:46
← PeterBooker left01:47
← cballenar left01:48
→ weiyang joined01:48
→ Vampire0_ joined01:49
→ iliketurtles joined01:49
← iliketurtles left01:50
← Vampire0 left01:51
EugeneKay Put stuff in plaintext files whenever possible.01:51
Binary files, because of their format, don't diff well in packfiles, so a repo will grow by the size of the file each time you make a revision.01:52
rylangrayston_ ah thankyou thats the answer i was looking for !01:52
→ harish joined01:52
→ boombatower joined01:53
→ Dougie187 joined01:54
rylangrayston_ ok one more question ... i have hired a couple people on odesk to work on 2 scripts that talk to each other over the serial port.. is using git over kill for a small project ? like 3 coders , 500 lines of code ?01:54
milki tgs3_: !doesntwork01:55
gitinfo tgs3_: We're sorry it doesn't work, but "it doesn't work" is about the worst problem description possible. Please give us some details, what exactly are you doing, what are you expecting to and what are you seeing instead?01:55
← jhford_ left01:55
rylangrayston_ sorry just checked make that 1500 lines of code01:56
EugeneKay git isn't overkill for any project; even something as simple as a one-file !jokes repo is deserving of git.01:56
gitinfo A complete listing of the git-jokes used in the /topic can be found at https://madeitwor.se/git-jokes (patches welcome)01:56
EugeneKay Especially when there's >1 person involved01:56
jamesd_ gitinfo: it no work, fix it, make work, make work (in a strong accented voice)01:56
← SirFunk_ left01:57
← Cromulent|2 left01:57
← nurey left01:59
rylangrayston_ ok thanks for your time everyone! ps !dosentwork == itworks #double negative:)01:59
→ daito joined01:59
milki p.O01:59
jast I don't see a double negative in that01:59
← amjam left02:00
→ nurey joined02:01
EugeneKay jast, what size T-shirt do you wear02:01
offby1 don't answer that02:01
jast depends on the maker02:01
→ jhford joined02:02
jast I believe for woot shirts it's M... for most European products is L, sometimes XL02:02
EugeneKay USian types02:02
← hwrd|work left02:02
→ echo-area joined02:02
jast I have very limited experience with US sizes :)02:02
→ SirFunk joined02:03
← rylangrayston_ left02:03
→ eletuchy joined02:04
← alexander__b left02:04
← decotii left02:04
→ alexander__b joined02:05
EugeneKay http://i.imgur.com/1yCZPFJ.png02:05
milki lol02:06
are we gathering jast stats?02:06
EugeneKay I have something far more nefarious in mind02:06
jast eww, non-SI units02:06
on woot's size chart M works for me02:07
EugeneKay They look substantially the same; going by the L measurement anyway.02:08
→ AWinterm_ joined02:08
← bdavenport left02:08
← AWinterman left02:09
→ rabellamy joined02:09
← boombatower left02:09
EugeneKay Anyway..... thanks.02:09
Is the address on jk.gs accurate?02:09
EugeneKay makes notes02:11
→ thiago joined02:12
thiago → Guest4845702:12
EugeneKay Should be there before then02:13
→ bdavenport joined02:13
→ amjam joined02:16
← daito left02:17
→ aartist joined02:17
← ozette left02:18
→ towski joined02:18
→ rickruby joined02:18
← towski left02:20
→ clreed87 joined02:20
← bdavenport left02:21
← tplaner left02:21
→ Kira joined02:23
← arietis left02:23
← UukGoblin left02:24
→ UukGoblin joined02:24
← zizzfizzix left02:25
← DonRichie left02:25
← esc left02:25
← xxtjaxx left02:25
→ elauria_ joined02:25
← k_sze[work] left02:25
→ bdavenport joined02:25
→ esc joined02:25
← hahuang65 left02:25
← nilg` left02:25
→ DonRichie joined02:25
→ xxtjaxx joined02:25
← xxtjaxx left02:25
→ xxtjaxx joined02:25
→ cballenar joined02:26
→ nilg`` joined02:26
→ zizzfizzix joined02:26
→ dsdeiz_ joined02:26
← dsdeiz_ left02:26
← elauria left02:27
→ dsdeiz joined02:27
→ elauria joined02:29
→ daito joined02:29
← elauria_ left02:30
→ tiglog joined02:30
→ Davey joined02:32
← Davey left02:33
→ kerrick joined02:35
→ CheckDavid joined02:36
← jon_r left02:36
← dougsland_ left02:37
← Guest48457 left02:38
→ thiago joined02:38
← jackdempsey left02:38
→ doomi-o joined02:40
doomi-o G u M P | Y ) ( z g d s C Q + M , & 2 \ [ ; C X k U | " b < a b 8 / Z R K U p : ~ \ b { u02:41
> t ! %02:41
? ] : ^ @ K N & V I ? = + ; l . s K j ( V M I N $ } > s y v ^ % ? W a + I A R P * E * I [ / t02:41
v g \ \ P J o k ^ i - " S e A \ - " N , L J - F = : . $ U Q / < Q j l m E , I [ ! g ( 9 [ ^ = \ ` / 602:41
@ A .02:41
B X X02:41
` C s N e H h r y H n } w k ' # h u " " & F S v " V U v o v O = * # o S { ? ! j % ' 6 B ' ; :02:41
A & L K N02:41
e X u L Z } X b h n - f j | , w 0 T i S q L v 0 D ? + ` > / 4 U c X v S Q S K \ - ` X - ~ / d t r02:41
V l # /02:41
= F 4 D Z W s W ; 1 < ] k F 0 a { ] b 3 s ( k w d y { s F o Z r l Q E i b * E j p \ F = G k f } f ~ R02:41
. = n02:41
K a @ h e ` ~ C l K q = x N : X & M f @ l V E U v / E " : w [ ! " l Q U < ? x R W @ l G H j % B L02:41
u c { X 602:41
[ z F x d Y $ ) r < w H o A F ' ; [ f G ] = * D $ ; G , n L 6 ( ' Q Q F l V [ P Z X q , ; W r02:41
U * Z H M02:41
p 0 + B O p X y ~ E * V _ < k a z n g ~ r < M N n - ; U x @ r J Y ( c ^ D { { R " R p N / B S h ! P m02:41
W i \ 902:41
4 > y l A J v X # u Y & O I q = k t h Q * ^ + ( y 1 U } u Y c O Z h _ s o b N h m p q e C _ w02:41
ChanServ set mode: +o02:41
doomi-o 4 q & _02:41
z b t F c V { j \ : I \ ~ y O k } h T I V ' W I E M ? $ & A x } @ m % a E % F P * A V C h \ 4 e D d02:41
x o O x02:41
bremner set mode: +q02:41
mribeirodantas by02:41
:)02:41
bye doomi-o02:41
← doomi-o left02:42
→ IlIIlIIllIlIllI joined02:42
IlIIlIIllIlIllI H / [02:42
$ ^ 3 ] @ C 8 : g i T { A s M \ > ~ g U ; , R G H b v U = o * t \ J : x + K k \ | $ + ` : Y ! k m02:42
Z A 6 D b02:42
O O r E * R p z m } 0 p + q * $ { F ` ) ; i ~ M # = B { V d C : / } Z h Z n = b 7 U E % p x 6 802:42
ChanServ set mode: +o02:42
IlIIlIIllIlIllI \ @ . ) O02:42
thiago set mode: +b02:42
← IlIIlIIllIlIllI left02:42
→ IlIIlIIllIlIllI joined02:42
IlIIlIIllIlIllI Turned into a pollen grain02:42
I mean in ancient times the Egyptians used hard muscle-demanding labor to create large pyramids... and stone age people, the neanderthals, needed to burn 5000-6000 calories a day to survive02:42
bremner kicked IlIIlIIllIlIllI (-)02:42
→ IlIIlIIllIlIllI joined02:43
IlIIlIIllIlIllI But now people at Amazon use fork lifts, or skyscraper builders use heavy machinery02:43
heavy machinery by companies like Hitachi or Kabuto or C.A.T. or whatever requires some object movement by people to make but not that of bodybuilders02:43
But now all weight lifting is good for is police, military, or football/sports players02:43
Police don't really "contribute" and product to society, but it's always good pay and it's sort of needed02:43
Military is a waste of time and money02:43
And football players are like playing video games only in real life02:43
Throwing a ball around doesn't help other people no matter how much someone is paid02:43
Police are mostly worthless02:43
Football is like a virus to the world02:43
What if I throw your balls around?02:43
Police don't deter crime02:43
They just mop up and catch the crooks02:43
thiago set mode: +b02:43
thiago ok, strike three02:43
← IlIIlIIllIlIllI left02:43
thiago goes seek oper help02:43
thiago set mode: -b02:43
← awygle left02:44
→ skmidry joined02:44
offby1 zounds02:44
← nerdfiles left02:44
bremner the first bit looks like a perl program02:45
ChanServ set mode: -o02:45
thiago set mode: -o02:45
→ dsdeiz_ joined02:45
jast I wonder where those people take all the time they sink into stuff like that02:46
← rabellamy left02:46
→ jon_r joined02:46
← konundra left02:47
→ jackdempsey joined02:47
← dsdeiz left02:47
→ anonymuse joined02:47
→ fridim__ joined02:47
← Nicholas_Rishel_ left02:48
offby1 probably best not to think too hard about it02:48
jast works for me02:48
if it happens again, do the op thing and /mode #git +q $~a02:48
we originally introduced that due to this kind of... creativity02:49
← kossy left02:49
→ Nicholas_Rishel joined02:50
← shinnya left02:50
← jfinn left02:51
→ kossy joined02:51
→ enr joined02:51
→ jfinn joined02:51
← jfinn left02:52
→ jfinn joined02:52
← jfinn left02:52
→ jfinn joined02:55
← dreiss left02:56
← enr left02:56
→ kossy_ joined02:58
← amjam left02:59
→ Cromulent joined02:59
← kossy left03:00
← clreed87 left03:01
← jalama left03:02
← eletuchy left03:02
→ eletuchy joined03:02
rohan32|afk → rohan3203:03
→ kpreid joined03:04
← donmichelangelo left03:04
→ donmichelangelo joined03:05
← eletuchy left03:06
← kukks left03:08
← jackdempsey left03:08
→ clreed87 joined03:09
← macabre left03:10
→ parcs_ joined03:11
← aartist left03:12
← alvaro_o left03:13
← McEnroe left03:13
→ _BJFreeman joined03:13
← parcs left03:13
← Marquisk2 left03:13
← daito left03:13
← byprdct left03:13
_BJFreeman → BJfreeman03:14
→ daito joined03:14
→ macabre joined03:16
→ hahuang65 joined03:17
← daito left03:18
← anonymuse left03:20
→ daito joined03:20
→ anonymuse joined03:20
← whaley left03:20
→ towski joined03:21
→ Gooder joined03:22
← dkpi left03:23
→ Destos joined03:23
← CheckDavid left03:23
← towski left03:23
EugeneKay When did that get taken out?03:23
← clreed87 left03:24
→ jackdempsey joined03:24
← anonymuse left03:24
← mribeirodantas left03:25
← dhruvasagar left03:26
← AWinterm_ left03:27
→ aoper joined03:28
← cballenar left03:28
← Dougie187 left03:28
← Milossh left03:29
← SinnerNyx left03:30
→ Marquisk2 joined03:30
→ crundar joined03:31
← chaw left03:31
crundar hey folks. Do I have a doozey (I think).03:31
I've got a repo in an enterprise github, and I've got access to another repo in the public one. I want both of those under a group account in my public one. Hopefully, with their histories.03:32
Is that a laughable request? Trivial?03:32
→ grilix joined03:33
SamB sounds like a #github question03:33
← p3lim left03:33
→ preflex_ joined03:34
→ groglogic joined03:34
← preflex left03:34
→ FreezingAlt joined03:34
→ Milossh joined03:34
preflex_ → preflex03:34
crundar oh, well I'll go try over there03:34
← FreezingCold left03:35
← kerrick left03:36
→ kerrick joined03:36
← evjsli left03:36
→ evjsli joined03:37
← plumpkin left03:38
← macabre left03:39
→ ferene joined03:40
→ sqlnoob joined03:42
← Marquisk2 left03:43
← oojacoboo left03:43
← OnlyMax left03:43
→ amjam joined03:44
→ CheckDavid joined03:44
→ chaw joined03:46
→ belak joined03:46
← nurey left03:49
→ dreiss joined03:49
← kerrick left03:51
← BJfreeman left03:51
← evjsli left03:51
← Destos left03:51
→ enr joined03:52
← mmlac-bv left03:52
→ guest110 joined03:52
← setmeaway left03:54
→ irqq joined03:54
→ setmeaway joined03:54
← setmeaway left03:54
← guest110 left03:55
→ setmeaway joined03:55
← darknyan left03:55
← enr left03:56
← Noldorin left03:56
← chaw left03:57
← groglogic left03:57
← jday_ left03:58
← irqq left03:58
← mitz left03:59
← lavaman left03:59
→ irqq joined03:59
→ _BJFreeman joined03:59
← donmichelangelo left04:00
→ VoidAtValhalla joined04:00
_BJFreeman → BJfreeman04:00
→ mitz joined04:00
→ donmichelangelo joined04:00
← lresende left04:01
→ darknyan joined04:01
← VoidAtValhalla left04:02
← jackdempsey left04:02
← BJfreeman left04:04
→ sbbrtn joined04:04
→ macabre joined04:04
← m0viefreak left04:05
→ beautifulmind joined04:05
← CheckDavid left04:06
← beautifulmind left04:07
→ beautifulmind joined04:09
→ kerrick joined04:09
→ evjsli joined04:09
← mitz left04:12
← macabre left04:12
→ eletuchy joined04:13
← beautifulmind left04:13
→ mitz joined04:13
→ BashCode_ joined04:14
→ beautifulmind joined04:16
→ scientes joined04:16
← scientes left04:16
→ scientes joined04:16
← BashCode_ left04:17
← beautifulmind left04:17
← eletuchy left04:18
→ BashCoder joined04:18
→ alexO joined04:20
anildigital_away → anildigital04:22
→ max_ joined04:22
← huoxito left04:25
anildigital → anildigital_away04:27
← grilix left04:27
anildigital_away → anildigital04:27
→ johnnyfuchs joined04:29
→ beautifulmind joined04:29
anildigital → anildigital_away04:30
anildigital_away → anildigital04:30
→ kcj joined04:32
← scientes left04:33
→ scientes joined04:34
← scientes left04:34
→ scientes joined04:34
← kcj left04:34
← mitz left04:34
← myoung left04:35
→ myoung joined04:36
→ AWinterman joined04:37
← vedant left04:38
→ scientes_ joined04:39
← scientes left04:39
→ hoschi joined04:39
rohan32 → rohan32|afk04:40
← scientes_ left04:40
→ macabre joined04:41
← akurilin left04:41
← the_cybersphinx left04:42
→ tkil joined04:42
← AWinterman left04:42
← undecim left04:43
tkil i'm did a branch against linux-next from 2013-07-05, and it seems that greg kh took a few of my patches. what's the easiest way to ask git to rebase my changes onto the latest linux-next tree?04:43
← daito left04:44
tkil i think that i can extract patches, switch to the latest edition of linux-next, and apply the patches manually / scripted in order.04:44
but i would assume there's an easier way to do it?04:44
jon_r you have a branch with just your commits04:44
based off another branch?04:44
and you wish to switch the base branhc?04:45
*branch04:45
tkil jon_r -- not entirely sure, but i think the answer is yes; i created a branch off the 07-05 linux-next tree.04:45
jon_r did you commit to it?04:45
your patches?04:45
tkil er... yes?04:45
→ gtback joined04:45
→ acharlieh joined04:45
tkil is not sure what else one would do with a branch.04:45
→ Tommy[D] joined04:46
jon_r so rebase it?04:46
(was just making sure you didn't make a mess or something ;))04:46
if you do `git rebase --interactive linux-next-latest` (where latest is whatever) it'll walk you through (re)applying your changes ontop of the latest from linux-next04:47
tkil this is what "git log" shows for my branch, note 985b84a0d18aaad943103c75f82accc94f475a15 at the bottom, that's where i branched off linux-next04:47
http://fpaste.org/27383/64121213/ er.04:47
i almost certainly made a mess, just trying to extricate myself with minimal further damage. :)04:47
← Tommy[D]_ left04:48
tkil mmm... i did a few different rebase attempts, one failed instantly, the other took a while and then complained about how my rename limit wasn't high enough.04:48
let me stare at what you suggested.04:48
→ byprdct joined04:48
← revyd left04:48
→ jaequery joined04:49
tkil i suspect i can just check out the latest -next tree, and then cherry pick the ones that greg didn't take.04:50
← sgambino left04:50
← acharlieh left04:50
tkil (i believe he took the first three there.)04:50
→ lavaman joined04:51
→ mitz joined04:51
← mckendricks left04:51
jon_r if you have only a few commits you should be able to rebase04:51
unless it's the size of the repo making problems04:51
← madewokherd left04:52
→ enr joined04:52
tkil is this a problem? http://www.fpaste.org/27384/13746415/04:52
and while linux-next is obviously pretty large, this is a pretty beefy system, so i wouldn't expect to be limited by hardware...04:54
← enr left04:54
→ cads joined04:54
← umib0zu left04:55
jon_r do a git fetch04:57
to make sure you're uptodate04:57
and make sure you've pulled any local tracked branches04:57
it's not a problem but will help if you're uptodate04:57
tkil no idea what the latter are, so i'll assume i don't have any.04:58
and i just did a fetch, let me make sure.04:58
yes, git fetch returned instantly.04:58
→ ntus1017 joined04:58
tkil reading this now: http://stackoverflow.com/questions/2452226/master-branch-and-origin-master-have-diverged-how-to-undiverge-branches04:59
→ rurufufuss joined04:59
← alfplayer left04:59
← hahuang65 left04:59
→ PCChris joined04:59
tkil git rebase origin/master -> failed with a conflict after just 3-4 files.05:00
← ferene left05:00
← dreiss left05:00
tkil hm. git pull --rebase seems to have executed without complaint.05:01
jon_r wait you have local commits in master?05:02
o_O05:02
tkil .... not that i know of.05:02
jon_r well that's why they'd differ… unless remote master was force pushed05:03
tkil [subtitle: i continue to not understand git, and therefore might be spouting completely meaningless gibberish.]05:03
linux-next is rebased all the time, IIRC... would that cause this sort of issue?05:03
← Dreadnaught left05:03
jon_r yes05:04
but it's normally done on a branch05:04
not on master?05:04
tkil is the question "which master" a reasonable one, here?05:04
← Nicholas_Rishel left05:04
tkil because i didn't do anything odd when i made these changes; i cloned linux-next, made a branch, made fixes, committed to that branch.05:04
→ ferene joined05:05
→ private joined05:05
tkil maybe the fact that upstream is rebased means that my branch is now disconnected. so just dealing with them textually as patches is probably the easiest solution for now.05:06
jon_r oh I see05:07
you're attempting to rebase your branch off master05:07
is linux-next a branch or a repo?05:08
tkil ... repo, i think, is the correct answer?05:08
← iateadonut left05:08
tkil git remote -v => origin git://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git05:08
(for both fetch and push)05:08
jon_r ah ok, for some reason I thought you were talking about a branch05:09
← xxcagedwisdom left05:10
jon_r git pull --rebase should have given you a branch that contains all of master plus your commits05:10
you couldn't have done git pull origin master into your branch because you diverged it from master, thats normal05:10
← sbbrtn left05:11
→ pretty_function joined05:11
jon_r you could have done git co master; git pull origin master; git co $branch_name; git rebase master05:11
tkil "git pull --rebase" gives me "no tracking information for the current branch."05:11
jon_r it's essentially the same in my understanding05:11
→ sajan joined05:11
jon_r yeah you need to specify origin/master05:11
→ Vampire0 joined05:11
jon_r cause your local branch isn't tracking any remote branches05:11
tkil fatal: 'origin/master' does not appear to be a git repository05:12
joy05:12
← sr78ger left05:12
jon_r try origin:master05:12
I get my : and my / mixed up05:12
→ hahuang65 joined05:12
tkil ssh: Could not resolve hostname origin: Name or service not known05:12
git pull --rebase origin master seems to be doing something...05:13
→ dreiss joined05:13
tkil ah-ha. that was it. thanks!05:13
→ oojacoboo joined05:13
→ fluxbox_ joined05:13
← Vampire0_ left05:14
tkil although it's still showing the three commits that greg took. let me try that same thing against greg's tree...05:14
← Kira left05:14
→ iateadonut joined05:15
jon_r you'll need to replace origin with greys remote05:16
*gregs05:16
or master with his branch if its a branch05:16
etc05:16
tkil right. and it might still 'splode since i was gbased off linux-next, and greg drives staging-next...05:16
← natechan left05:17
← ntus1017 left05:17
→ jackdempsey joined05:17
← fridim__ left05:18
→ dhruvasagar joined05:19
ChanServ set mode: +o05:19
thiago set mode: -b05:19
thiago set mode: -o05:19
← YoungFrog left05:21
← cigarshark left05:21
← Cromulent left05:22
→ shruggar joined05:22
← jaequery left05:24
← weiyang left05:25
→ macabre^ joined05:25
← PaulCapestany left05:25
→ jaequery joined05:26
→ breck7 joined05:26
→ stliu joined05:27
← macabre left05:27
→ mechanicalduck joined05:27
← maslowbeer left05:28
→ daito joined05:29
← nicedice left05:30
→ Guest__ joined05:32
← darknyan left05:33
→ bodom joined05:35
→ ffio joined05:36
→ ThomasLocke joined05:36
→ kcj joined05:36
← smw left05:37
← MXBrandon2 left05:37
← dhruvasagar left05:37
← v0n left05:37
→ dhruvasagar joined05:38
← mechanicalduck left05:38
→ MXBrandon2 joined05:39
← jaequery left05:40
← FreezingAlt left05:41
← Guest__ left05:43
→ sr78ger joined05:44
→ Guest__ joined05:44
← skmidry left05:44
→ skmidry joined05:45
→ mechanicalduck joined05:46
← jackdempsey left05:46
← sr78ger left05:49
← dhruvasagar left05:49
→ Sigma joined05:49
← alunduil left05:49
→ dhruvasagar joined05:50
→ alunduil joined05:50
← vipzrx left05:51
← shruggar left05:51
→ vipzrx joined05:51
→ rath joined05:53
→ enr joined05:53
← BashCoder left05:56
→ ayaz joined05:56
→ Gooder` joined05:57
→ KernelPanicMK3 joined05:57
← enr left05:57
→ Spyro_ joined05:57
← skmidry left05:58
← joki left05:59
→ ntus1017 joined05:59
→ adnap_ joined05:59
← hetal_ left06:00
← bodom left06:00
→ ejnahc_ joined06:01
→ rapha_ joined06:02
→ ryanjh joined06:02
→ k_sze[work] joined06:02
→ arvind_khadri joined06:02
→ joki joined06:03
← Gooder left06:04
← KernelPanicMK2 left06:04
← Spyro left06:04
← adnap left06:04
← ryanhirsch left06:04
← beawesomeinstead left06:04
← ejnahc left06:04
← iateadonut left06:04
← rapha left06:04
← kcj left06:04
← alexander__b left06:05
→ alexander__b joined06:05
← hahuang65 left06:05
→ mamtina joined06:06
→ mutbuerger joined06:06
→ kcj joined06:06
← leandrosansilva_ left06:07
→ leandrosansilva joined06:07
→ tatsuya_o joined06:07
→ nilg``` joined06:08
← nilg`` left06:08
→ pretty_f_ joined06:09
→ rudak joined06:10
→ cbreak-work joined06:11
→ lb1a joined06:11
← pretty_f_ left06:12
→ pretty_f_ joined06:12
← pretty_function left06:12
← tatsuya_o left06:13
← zbcm left06:13
→ juanmabc joined06:13
← ferene left06:13
← mechanicalduck left06:14
← macabre^ left06:16
→ iateadonut joined06:17
← juanmabc left06:17
→ juanmabc joined06:18
→ mechanicalduck joined06:18
← normanrichards left06:21
← cads2 left06:22
→ Caritade joined06:23
← crundar left06:25
← mitz left06:25
← kerrick left06:27
→ jayfk joined06:28
← johnnyfuchs left06:28
← mechanicalduck left06:29
← lepine left06:29
→ kerrick joined06:29
→ lepine joined06:30
→ marthinal joined06:30
→ soee joined06:31
→ Adlai joined06:31
→ hahuang65 joined06:32
→ mechanicalduck joined06:32
→ rednaks joined06:33
← breck7 left06:34
→ mitz joined06:35
→ fRAiLtY- joined06:35
← bvi left06:36
→ GeertV joined06:37
← aoper left06:37
← max_ left06:38
← mechanicalduck left06:38
→ quasimodo joined06:38
→ a-l-e joined06:38
← rchavik left06:38
← hahuang65 left06:40
→ undecim joined06:41
→ netiul joined06:41
← tiglog left06:42
→ rchavik joined06:42
← hugo_dc left06:42
→ fairuz1 joined06:43
← fairuz1 left06:43
→ JohnBat26 joined06:44
→ mechanicalduck joined06:45
→ gua_ould joined06:48
→ bvi joined06:48
→ FawnOnyx joined06:49
gitinfo set mode: +v06:50
← mechanicalduck left06:50
lb1a .info 1+106:51
gitinfo [calc] 1 + 1 = 206:51
← rednaks left06:52
→ shwouchk joined06:53
shwouchk Hello06:53
lb1a !hi06:53
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.06:53
→ enr joined06:53
← gua_ould left06:54
shwouchk Is there a way to have git treat a second repository that I put inside a first repository as "just files"?06:54
← lavaman left06:54
← danicampa9011 left06:54
← pretty_f_ left06:54
→ eijk joined06:54
iveqy shwouchk: no06:55
PjotrOrial Yves,06:55
lb1a shwouchk, since git isnt aware of 'files' no06:55
→ pretty_function joined06:55
lb1a shwouchk, if you want to nest repositories, checkout !subrepo06:55
gitinfo shwouchk: [!subprojects] So, you want to add git repositories inside of other git repositories? Well, you have four main options. First is to just do it, add the repo to the outer project's .gitignore, and treat them entirely separately. Best if they are entirely separate. Otherwise your best options are "!submodule", "!gitslave", and "!subtree". Try those commands in this channel, or in a PM to avoid flooding.06:55
→ pduin joined06:56
iveqy shwouchk: you could always add .git to .gitignore, it might help somewhat (I think) but that wouldn't give you tracking of the repo06:56
SamB shwouchk: I've seen that for dulwich's test data, I think06:56
iveqy shwouchk: you really should follow lb1as advice06:56
lb1a iveqy, thanks, it's lb1a btw ;)06:56
SamB but it's not a normal thing to do at all06:56
→ mishok13 joined06:57
→ mklappstuhl joined06:57
lb1a SamB, it is 'normal' usecase for example shared libary and such06:57
iveqy lb1a: don't you add an s for possessive things? ;)06:58
← enr left06:58
cbreak-work shared libraries have nothing to do with git06:58
shwouchk SamB: can you please elaborate?06:58
← amjam left06:58
→ enr joined06:58
lb1a iveqy, ^^06:58
→ pantsman joined06:59
← pantsman left06:59
→ pantsman joined06:59
SamB shwouchk: dulwich implements git stuff, so it has/had one or more git repositories checked into its git repository for use in testing06:59
shwouchk iveqy, lb1a: I would like the .git files to be versioned as well rather than ignored (forgive me if this seems senseless to you, but I need it).06:59
lb1a cbreak-work, correct, but putting the git repo of the libaray inside your own project to track its progress ( using the available sub repo options ) is a normal usecase for me at least06:59
shwouchk, stick with !subprojects07:00
gitinfo shwouchk: So, you want to add git repositories inside of other git repositories? Well, you have four main options. First is to just do it, add the repo to the outer project's .gitignore, and treat them entirely separately. Best if they are entirely separate. Otherwise your best options are "!submodule", "!gitslave", and "!subtree". Try those commands in this channel, or in a PM to avoid flooding.07:00
cbreak-work lb1a: then it's not shared :)07:00
iveqy shwouchk: would you mind telling why you need it? You know that a .git is mostly binary files and you will have a hard time doing conflict resolution with them.07:00
→ koda joined07:00
cbreak-work sub repositories are by their very definition strongly coupled to the parent, they are directly owned by it07:00
lb1a cbreak-work, i mean in the sense, that this libary is shared amongst projects, not like a shared lib in the C way :P07:00
shwouchk !subtree07:00
gitinfo The subtree merge method is great for incorporating a subsidiary git repo into your current one with "unified" history. Read http://git-scm.com/book/en/Git-Tools-Subtree-Merging for more info, or try one of the !subtree_alternatives07:01
→ benoit joined07:01
SamB lb1a: actually checking in the git repository is senseless in that sort of situation though07:01
→ barneybook|2 joined07:01
lb1a that's why i refered to alternatives07:02
← irqq left07:03
← JohnBat26 left07:03
← barneybook left07:04
shwouchk SamB: but necessary nevertheless07:04
→ tausigma joined07:04
→ tsdh joined07:05
iveqy shwouchk: what's your situation and why do you think you need that solution?07:05
→ YoungFrog joined07:06
← iateadonut left07:06
← UukGoblin left07:06
→ UukGoblin joined07:06
→ gebi joined07:07
→ psoo joined07:07
SamB shwouchk: I really don't think it is, unless you also need test data for an implementation of git07:07
shwouchk iveqy: I want to use a single git repository the way I use dropbox (aside from needing to commit).07:07
SamB: you are mistaken. I want it, hence I need it. There needs not be a further reason.07:08
iveqy shwouchk: then you best chance is to add some pre-commit and post-checkout hooks that will zip our inside git repos before commiting07:09
shwouchk iveqy: I see. Thanks for the tip!07:09
→ vedant joined07:09
iveqy however git is not meant for this use07:09
← byprdct left07:09
iveqy you would probably be better of with a file centric approach like svn or cvs07:09
lb1a !dropbox07:10
gitinfo Storing git repositories on DropBox is a Very Bad Idea™. Use a free hosting service (such as Bitbucket or GitHub), or read !dropbox_why if you insist upon having your repo corrupted.07:10
→ tatsuya_o joined07:10
shwouchk iveqy: I understand. Perhaps I won't use it that way - but perhaps I will nevertheless.07:10
iveqy an other alternative would be to use sshfs and maybe rsync-backup (if you need incremental backups)07:10
→ Cromulent joined07:10
→ mechanicalduck joined07:11
shwouchk iveqy: sshfs is not local07:11
→ danicampa901 joined07:11
iveqy shwouchk: since git doesn't work on files but on whe whole repo you scale poorly with git. In the "dropbox usage way" files usually doesn't have a relationship with eachother with makes the git way useless07:11
shwouchk: not local?07:12
→ buibex joined07:12
shwouchk iveqy: sshfs just mounts a remote fs, it does not make it local07:12
iveqy shwouchk: yes it does. Depending on your needs that's an alternative07:12
or are you only going to use git to versioning your home tree locally?07:13
→ TomyLobo joined07:13
← bvi left07:13
shwouchk iveqy: I repeat, and there's no arguing: the sshfs files are not local. If you lose internet, you have nothing.07:13
→ harobed_ joined07:13
shwouchk anyway, I need to go. Thanks for the tips everyone!07:14
iveqy shwouchk: depends on what you mount and what's your usecase is. You can use sshfs locally (but why would you?). However it's not clear if the local thing is a requirement for you or not07:14
← dsdeiz_ left07:15
→ dsdeiz joined07:15
← Caritade left07:15
← vedant left07:17
← gebi left07:17
→ gebi joined07:17
← Matrixiumn left07:17
→ iateadonut joined07:17
shwouchk iveqy: ah, yes it is.07:17
a requirement that is07:17
← buibex left07:17
→ _Qman joined07:17
→ Matrixiumn joined07:18
YoungFrog IIUC, the only problem with a normal git is that it won't add files in a directory named .git, right ?07:18
→ Amandil joined07:18
← fluxbox_ left07:18
YoungFrog if that's the case, I suggest something like mv .git .tig; git add .tig; ...; mv .tig .git07:19
← evjsli left07:19
YoungFrog (or use some hardlink magic to avoid the constant renaming)07:19
← danicampa901 left07:20
← ffio left07:20
→ LimeBlast joined07:23
← gebi left07:24
← tausigma left07:25
→ bvi joined07:25
→ gebi joined07:25
→ threesome joined07:25
→ Balliad joined07:26
← gebi left07:27
← harobed_ left07:29
→ gebi joined07:30
→ Kad-- joined07:30
← Kad-- left07:30
→ Kad-- joined07:30
→ sr78ger joined07:30
cbreak-work YoungFrog: sounds idiotic07:31
don't track git internals with git07:31
all you'll get is a constantly exploding bloated mess of a repository07:31
→ bondar joined07:31
← bondar left07:32
← daito left07:32
→ harobed_ joined07:33
→ OOPMan joined07:33
← beautifulmind left07:33
← jayfk left07:33
← pantsman left07:35
→ diegoviola joined07:36
→ gua_ould joined07:36
→ hahuang65 joined07:36
→ maxmanders joined07:37
← ejnahc_ left07:39
→ ejnahc joined07:39
→ dfanjul joined07:41
← mechanicalduck left07:41
← hahuang65 left07:42
← gebi left07:42
→ gebi joined07:42
← undecim left07:42
→ danicampa901 joined07:42
→ mechanicalduck joined07:43
→ vedant joined07:44
→ breck7 joined07:44
← gebi left07:46
← scorphus left07:46
← harish left07:47
→ gebi joined07:47
← Cromulent left07:48
← breck7 left07:49
← oojacoboo left07:50
→ JohnBat26 joined07:51
← JohnBat26 left07:51
→ daito joined07:51
→ raf joined07:52
← alexwhitman left07:53
← jon_r left07:54
→ jon_r joined07:54
→ arietis joined07:55
→ alexwhitman joined07:56
← simesy left07:56
← danicampa901 left07:56
← Guest__ left07:56
→ beawesomeinstead joined07:56
→ Guest__ joined07:57
← mklappstuhl left08:00
mamtina git is grit \o/08:00
→ barneybook joined08:01
→ ffio joined08:01
← juanmabc left08:03
← barneybook|2 left08:04
→ barneybook|2 joined08:04
→ buibex joined08:05
← barneybook left08:07
→ marhaban joined08:07
→ Sonderblade joined08:07
← benoit left08:07
← nilg``` left08:07
← dreiss left08:07
→ nilg``` joined08:08
← thiago left08:09
← barneybook|2 left08:10
→ kesha_ joined08:11
← iateadonut left08:11
← ffio left08:11
→ trusktr joined08:12
← dfanjul left08:12
→ danicampa901 joined08:12
← mechanicalduck left08:12
→ benoit joined08:12
← Goplat left08:12
→ antoine__ joined08:13
→ mechanicalduck joined08:13
Fissure_logbot → Fissure08:15
→ dfanjul joined08:15
→ irqq joined08:15
→ pdx joined08:15
← carlos-huang left08:16
→ towski joined08:17
→ iateadonut joined08:17
← jon_r left08:18
← danicampa901 left08:19
← gebi left08:19
→ Plummer joined08:19
← jayrobot left08:19
→ gebi joined08:19
← benoit left08:19
← towski left08:20
← mechanicalduck left08:21
→ jayrobot joined08:21
→ carlos-huang joined08:22
→ mechanicalduck joined08:23
→ dreiss joined08:23
← prettyrobots left08:26
→ ilogger2_ joined08:30
→ a-l-e joined08:30
→ hagabaka joined08:30
→ mitsuhiko_ joined08:30
→ zz_michael_mbp joined08:30
→ shruggar joined08:30
→ AndreSomers joined08:30
→ jhoepken joined08:30
→ xSmurf joined08:30
→ gtback joined08:30
→ DolphinDream joined08:30
→ Azis joined08:30
→ AAA_awright joined08:30
→ alexander__b joined08:30
zz_michael_mbp → michael_mbp08:30
→ jokajak joined08:30
→ cmn joined08:32
→ Guest39808 joined08:32
→ tchan joined08:32
← alexander__b left08:32
→ alexander__b joined08:32
→ jon_r joined08:33
→ weiyang joined08:34
→ Guest__ joined08:34
→ kesha_ joined08:35
Guest39808 → Matrixiumn08:36
← Matrixiumn left08:36
→ Matrixiumn joined08:36
→ tazle_ joined08:36
→ danicampa901 joined08:36
→ daanavitch joined08:37
→ Etheco joined08:37
→ fuzzybunny69y joined08:37
→ hahuang65 joined08:37
→ jargon joined08:38
→ benoit joined08:38
→ carlos-huang joined08:39
→ Kulrak joined08:39
→ gebi joined08:39
← alexander__b left08:41
→ alexander__b joined08:42
← hahuang65 left08:42
← alexander__b left08:43
→ alexander__b joined08:43
→ FrobinRobin joined08:43
→ dsdeiz_ joined08:43
→ rednaks joined08:45
← gebi left08:46
→ gebi joined08:46
→ diegoviola joined08:47
→ mklappstuhl joined08:47
← Guest__ left08:48
← alexander__b left08:49
→ Guest__ joined08:49
← kesha_ left08:51
→ howlymowly joined08:51
→ kesha_ joined08:51
→ mikecmpbll joined08:52
→ dsdeiz joined08:52
← kesha_ left08:52
← dsdeiz_ left08:52
→ |Freeman| joined08:53
← jargon left08:55
→ tvw joined08:56
YoungFrog cbreak-work: IIUC, that was shwouchk's requirement : track a git repo using a surrounding git repo08:56
→ simesy joined08:56
cbreak-work idiotic requirement then.08:57
→ stamina joined08:57
→ dsdeiz_ joined08:57
→ jargon joined08:57
cbreak-work it'll result in infinitely growing repository size, impossible merges and worthless history graphs08:57
← rednaks left08:58
← dsdeiz left08:58
→ Arafangion joined08:59
jon_r surely i'd result in just n*2 repo size08:59
cbreak-work jon_r: no.09:00
jon_r: n*m09:00
jon_r git doesn't grow exponentially...09:00
→ mechanicalduck joined09:00
YoungFrog I suggest we don't waste time analyzing bizarre requests09:00
jon_r each new object would just gain an entry in the outer repository09:00
→ matteo joined09:01
cbreak-work jon_r: every time the inner repository gets repacked, the outer has to store a complete copy of the whole inner repository09:01
← simesy left09:01
cbreak-work jon_r: so potentially every commit in the inner repository leads to a complete copy of the inner repository in the outer repository09:01
Arafangion blinks.09:01
Arafangion What's this?09:01
cbreak-work idiocy of high purity.09:02
jon_r well it still wouldn't be n*m09:02
it'd me somewhere between 2 and m depending on the number of repacks09:02
which depends on repository size09:03
still a stupid idea though09:03
Arafangion cbreak-work: I'll try to ignore it, then.09:03
→ alexander__b joined09:06
→ jkremser joined09:06
← jkremser left09:06
→ jkremser joined09:06
← gebi left09:08
→ gebi joined09:08
← weiyang left09:10
→ iokill joined09:11
→ mastro joined09:12
← jargon left09:12
→ paul424 joined09:14
→ harish joined09:15
→ Cromulent joined09:16
← gebi left09:16
mastro Hi, I'm experiencing a weird behavior of git describe. I've a submodule, with git submodule status I see my submodule positioned at: +6903774653de52d0206e0e6026ca6914def8a333 submodule (1.0-2-g6903774)09:16
→ gebi joined09:16
mastro But entering the submodule I see: git describe --> 1.0-2-g6903774 | git describe --tags 1.0-rc109:17
sorry... git describe --tags ---> 1.0-rc109:17
anyway the submodule is positioned on the 1.0-rc1 tag09:17
why the git describe command show me I'm in 1.0-2-g6903774 ?09:17
→ achadwick joined09:18
← gebi left09:18
→ enr joined09:19
cbreak-work mastro: because normal git describe only considers annotated tags09:21
→ angelsl joined09:21
→ sharpshooter joined09:21
→ gebi joined09:21
pradeepc hello all, in one of my bare repo, i found RENAMED-REF file09:22
mastro cbreak-work, uh... why is that?09:22
pradeepc can someone explain me the meaning of that file09:22
cbreak-work mastro: because annotated tags are usually used to mark releases, version relevant things09:22
normal tags are often used to just tag random things, even just locally09:22
mastro cbreak-work, I see... I should start using annotated tags then09:22
→ weiyang joined09:23
→ cenuij joined09:23
← cenuij left09:23
→ cenuij joined09:23
→ ryanhirsch joined09:23
mastro cbreak-work, I'd probably used the reverse: -a as default and a flag to do a non-annotated tag09:24
cbreak-work, but you surely discussed it on list before choosing to make non-annotated the default09:24
cbreak-work I didn't09:24
but it's kind of obvious09:24
common thing of normal tag: easy, default09:25
special thing that requires further user input, such as a tag message: special flag09:25
→ SicoAnimal joined09:28
← ryanhirsch left09:28
mastro cbreak-work, hum... from the user perspective the tag, specially if coming from other VCS, is something used to tag a release so the "default" behavior should be tagging a release09:29
cbreak-work no...09:29
normal tags are lightweight. annotated tags are heavy09:29
that's how it is.09:29
← gebi left09:30
→ yacks joined09:30
mastro cbreak-work, I'm not objecting that, I'm objecting what the user expect vs default -- anyway I can't find a way to list annotated tags only (or vice-versa)09:32
→ OwiecPL1986 joined09:33
mastro cbreak-work, I'm saying this is an area causing (unnecessary) confusion in the end user09:33
cbreak-work why?09:33
it's clearly documented09:33
→ ryanhirsch joined09:33
→ dsdeiz joined09:34
← dsdeiz_ left09:35
← weiyang left09:35
mastro cbreak-work, not really, it's clearly documented if you already know there are lightweight tags and annotated tag / if you know what a lightweight tag means vs annotated one and what different porpoise they have09:35
cbreak-work ... which is all clearly documented09:35
→ _br_ joined09:36
OwiecPL1986 I have "hub" on bitbucket and "prime" on production server. On production server i have special accounts for develop branch. How can i auto-update to the newest develop version on this accounts? When i push to hub develop branch, i want, server pull the newest version to the develop account.09:36
→ zmike123 joined09:36
→ buibex joined09:36
cbreak-work OwiecPL1986: 1. Git doesn't have accounts. 2. Automated pulling is idiotic, pick something else, like automated fetching or resetting.09:36
mastro cbreak-work, nowhere in git help tag I see the doc saying annotated tags are meant for release and lightweight for tagging random thing....09:37
cbreak-work OwiecPL1986: also, you might not be allowed to install any hooks on bitbucket09:37
so you can't make automation happen that way09:37
OwiecPL1986 but only for develop version, i want on serwer always the newest veriosn for testing09:37
cbreak-work mastro: they aren't meant for that. They are used for that.09:37
mastro cbreak-work, also you make a distinction between the two but "git tag" show both of them and there is no option to show only annotated tag or lightweight tags09:37
cbreak-work mastro: you can use them how ever you want09:37
mastro: you have to read man git describe to find out what describe does... duh... :)09:38
gitinfo mastro: the git-describe manpage is available at http://jk.gs/git-describe.html09:38
← enr left09:38
mastro cbreak-work, partially true, cause commands like git describe assume that annotated tags are the ones that matters - git submodule status use git describe without options making it, de-facto, an imposition on how you should use those tags09:39
→ enr joined09:39
cbreak-work wrong.09:39
OwiecPL1986 so.... mayby use cron to check the newest version for develop?09:39
cbreak-work you can use them how ever you want09:39
mastro cbreak-work, ok, mind elaborate a little on that? how can I?09:39
→ nicholasruunu joined09:39
cbreak-work mastro: by using them how ever you want...09:40
mastro: what's stopping you?09:40
mastro cbreak-work, that's not something stopping me, it's something either not well designed for the end user either not well documented.09:41
cbreak-work mastro: what's not well documented?09:41
mastro cbreak-work, suppose you are a new git user, you don't know anything about git and you are learning it. You want to create a tag.09:41
cbreak-work, you read git help tag and find out there are annotated tags and lightweight tag. You may choose (like I did) to use -a for some tag and don't use it for others because it seems -a is just useful when you have something to say about that tag09:42
← yacks left09:42
→ rednaks joined09:43
mastro cbreak-work, Then you use git submodule and find out that git submodule status sometimes show the commit and sometimes show a previous commit09:43
cbreak-work mastro: so?09:43
← enr left09:43
→ eletuchy joined09:43
mastro cbreak-work, at THAT point you read the git submodule help and find out status use describe, so you read git describe to find out it only show annotated tags09:44
cbreak-work, since time has passed you do not remember which tags you annotated and which you didn't and there's no option in git tag to show annotated tags only09:44
← Cromulent left09:44
cbreak-work sounds reasonable.09:44
mastro cbreak-work, assuming you look at each tag in sequence to discover which tags is not annotate: you can't convert lightweight tag in annotated tag without doing a public announcement requiring all others users to delete their local tags09:46
cbreak-work, (I may be wrong in this last sentence)09:46
cbreak-work you can create new tags09:46
just change the name09:46
you can also overwrite tags when pushing if you change them09:46
→ dsdeiz_ joined09:46
cbreak-work this is not friendly obviously09:46
mastro this all confusion could be have avoided by adding in the git tag documentation something that say: lightweight tags (non-annotated) are not meant for release and other commands ignore them by defaults09:47
cbreak-work I use them for releases09:47
works fine.09:47
mastro cbreak-work, sure, but that's your informed decision ;)09:47
cbreak-work of course it is...09:47
← dsdeiz left09:48
← iokill left09:48
mastro cbreak-work, I mean that a new user do not know the annotated tags only are used by other commands by default and the git tag help should mark it in some way09:48
cbreak-work, also, can you give me a way to list annotated tags only?09:48
cbreak-work complain to the mailing list09:48
→ mechanicalduck_ joined09:48
→ john2x joined09:49
cbreak-work I don't know an easy way to do that09:49
john2x how do I amend a commit that's right before HEAD?09:49
mastro cbreak-work, that's not actually complaining :) I want to help actually, now that I know it nothing change for me if they document or not09:49
cbreak-work john2x: git rebase -i HEAD~209:49
mastro john2x, git rebase -i HEAD~2 (be careful, DON'T if you already pushed it)09:50
cbreak-work john2x: then mark the commit you want for edit, and amend it, then rebase --continue09:50
john2x: obviously you should not do that if you pushed that commit already09:50
mastro xD09:50
john2x yup haven't pushed :) thanks!09:50
OwiecPL1986 so... do you think auto-update by cron on server from hub for develop branch is ok or not?09:50
cbreak-work mastro: if you want to get documentation changed, you'll have to tell the devs or do it yourself09:50
OwiecPL1986 i dont see any reason to be not ok?09:50
cbreak-work OwiecPL1986: sounds dumb.09:51
← mklappstuhl left09:51
cbreak-work OwiecPL1986: why do you need auto update?09:51
OwiecPL1986 cbreak-work why?09:51
mastro cbreak-work, I asked cause maybe this things have already been discussed, I don't want to bring up an argument already ruled out09:51
OwiecPL1986 why loose my time to do that everytime?09:51
cbreak-work mastro: I wouldn't know, I don't read the mailing list09:51
← mechanicalduck left09:52
cbreak-work OwiecPL1986: updating when ever you need it is more efficient and faster than fetching automatically09:52
OwiecPL1986: you'll have to merge manyally in any case anyway09:52
OwiecPL1986 cbreak-work but i need update always when new develop verion appear without expection09:52
cbreak-work OwiecPL1986: then push directly, or via some hook09:53
→ jatt joined09:53
OwiecPL1986 i cant use hook, because i use bitbucket as hub09:53
→ wasanzy joined09:54
wasanzy hi09:54
I want to know the different between gitosis, git, github. which one do I go for?09:55
OwiecPL1986 wasanzy public or private repository?09:55
→ iokill joined09:55
← iokill left09:55
cbreak-work wasanzy: go for git09:56
it's the best09:56
and you'll need it either way09:56
wasanzy OwiecPL1986: private but will be access and used by a team of people09:56
cbreak-work forget about gitosis, it's ancient and unmaintained09:56
→ iokill joined09:56
→ simesy joined09:57
OwiecPL1986 wasanzy bitbucket.org + git + GUI (if you want) sourcetree.09:57
bitbucket.org to 5 users in team is free09:57
wasanzy I want to setup and maintain my own repo so I will go for git09:58
jast git is a system for versioning projects. gitosis is a tool for hosting git repositories yourself (but it's no longer maintained and there's a replacement called gitolite). github is a site that hosts repositories for you, for free if they're open source09:58
gitosis and github are basically services related to git09:59
← buibex left10:00
wasanzy git has GUI?10:00
cbreak-work wasanzy: man git gui10:00
gitinfo wasanzy: the git-gui manpage is available at http://jk.gs/git-gui.html10:00
cbreak-work it's very ugly10:00
wasanzy cbreak-work: hehehe, how can it be made beautiful then?10:01
cbreak-work use something else10:01
like sourcetree10:01
← simesy left10:02
wasanzy sourcetree can sit on git?10:02
→ mechanicalduck joined10:03
iveqy there should be css for tcl, so that you just could apply a css file and make git-gui handsome10:04
john2x what does `remote = .` mean in .git/config?10:04
wasanzy iveqy: ok10:04
iveqy wasanzy: there isn't such thing...10:04
wasanzy: and most people just use git from the command line10:05
wasanzy what of gitweb?10:05
iveqy wasanzy: and afaik sourcetree isn't available for ubuntu10:05
gitweb (and cgit, etc.) is a web interface for viewing a git repository10:05
you cant work with gitweb10:05
wasanzy ok10:06
← mechanicalduck_ left10:06
→ Zeelot joined10:06
→ decci joined10:07
decci We have Stash installed and working with multiple projects, and multiple repositories; We need to be able to upon commit to have automatically push to either development web server, or production server, or both... in each instance (project) we need to be able to configure to which server information is pushed post-commit.10:07
wasanzy now how do I setup git? any good document?10:08
TMO wasanzy: http://git-scm.com/book10:08
→ Kicer86 joined10:08
→ iateadonut joined10:08
→ sqlnoob joined10:08
→ enr joined10:09
→ _iron joined10:10
cbreak-work john2x: man git remote10:11
gitinfo john2x: the git-remote manpage is available at http://jk.gs/git-remote.html10:11
→ Cromulent joined10:11
iveqy wasanzy: what TMO said, otherwise git init and you're done, however to grasph what you're doing, you need to read the book10:12
→ rob_w joined10:13
iveqy decci: what's Stash? You could use a post-commit hook for this. But the whole solution you're asking for sounds ugly. Why would you force a push? Are you trying to use git as centralized system? Why are you using git at all?10:13
→ beautifulmind joined10:14
mastro cbreak-work, https://groups.google.com/forum/?fromgroups#!topic/git-users/jtSoQ6Lv3eM10:14
decci iveqy: Atlassian Stash is the Git repository management solution for enterprise teams. It allows everyone in your organisation to easily collaborate on your Git repositories.10:14
→ cworth joined10:15
← beautifulmind left10:15
cbreak-work man git stash10:15
gitinfo the git-stash manpage is available at http://jk.gs/git-stash.html10:15
mastro cbreak-work, ouch! that's not maint for bug reports10:15
→ dkpi joined10:16
→ Latro joined10:16
iveqy decci: oh, so it's like a in-house github/bitbucket thing. It doesn't really concern how to use git (as in git.git, the program)10:16
← diegoviola left10:17
iveqy decci: so to my other question, why do you need a push after each commit?10:17
decci iveqy: We need to be able to upon commit to have automatically push to either development web server, or production server, or both... in each instance (project) we need to be able to configure to which server information is pushed post-commit.10:17
← enr left10:18
iveqy decci: why do you need that?10:18
← sqlnoob left10:19
Latro Hi to all. I'm begining to use GIT in my company and I hope than you can help me when need it.10:19
→ beautifulmind joined10:20
→ dhruvasagar joined10:20
Latro someone from spain, please?10:20
iveqy Latro: Hi, welcome to git. We try to offer you support but we have no guarantees. If you want professional git support that always will be availiable, you need to buy it from someone.10:20
OwiecPL1986 is it possible to push directrly form my computer to server? When i want finish new develop version i want push that to bitbucket and also to develop server. But is it possible to push directly to working directory on server /hoem/user/www/public by sftp?10:20
Latro someone that speak Spanish?10:20
iveqy Latro: !anyone10:20
gitinfo Latro: Usually, it does not help to ask for someone specific to help you. Without knowing your specific problem, nobody knows if they can be of assistance. Please ask your questions and wait until somebody speaks up.10:20
Latro thanks, Iveqy10:21
iveqy OwiecPL1986: !deploy10:21
gitinfo OwiecPL1986: Git is not a deployment tool, but you can build one around it(in simple environments) or use it as an object store(for complex ones). Here are some options/ideas to get you started: http://gitolite.com/the-list-and-irc/deploy.html10:21
Latro Iveqy: Correct. At the moment, I am reading the tutorial of this page http://git-scm.com/book/es10:22
iveqy OwiecPL1986: short answer; no. a little longer answer; yes but you will eventually really screw up. The correct answer; it works perfectly fine if you use the right technics and read the docs I pointed you to10:22
OwiecPL1986 iveqy but this is for situation when hub is on the same server as production?10:23
iveqy decci: I would appriciate if you try to explain to me. Because to me it sounds really stupid (however, much things seems really stupid when you don't have the whole picture)10:23
Latro: okay, good luck. You will be easier off if you're able to use english, most documentation and all commands is in english10:24
OwiecPL1986: what's hub?10:24
OwiecPL1986: and you're still talking about a push, doesn't matter if it's to a local server or a remote server, it will still be threated as a remote git repository10:25
Latro but I will have questions later (sure)10:25
iveqy Latro: you're welcome to ask. If you don't get any help here, try [email@hidden.address]10:25
→ deuterium joined10:26
← mechanicalduck left10:27
Latro Thanks you very much, Iveqy10:27
for now, all is well10:28
→ rapha joined10:28
← dhruvasagar left10:28
TMO for automatically deployment from git-repositor to test/Development environment. you might want to look at Jenkins.10:28
→ user_ joined10:29
→ dhruvasagar joined10:29
OwiecPL1986 iveqy i have develop account on server, i want put here newest version branch develop. I am using bitbucket.org. So i see only this ways to update develop code: 1) checking in crone the newest version and pull then 2) send by sftp the newest version directrly to /home/user/www and run some scripts after 3) doing this manually (this one i dont want)10:29
→ ismlages joined10:30
OwiecPL1986 TMO its to me?10:30
TMO could be.. you never said why you wanted to do the thing you asked for :)10:31
but now youve elaborated a bit more.10:31
OwiecPL1986 i dont want lose my time every time when new develop version out and want people to test newest version of develop10:31
iveqy OwiecPL1986: use hooks to do this automatic10:31
← decci left10:31
→ decci joined10:32
OwiecPL1986 iveqy how? i am using bitbucket.org10:32
← alexander__b left10:32
→ alexander__b joined10:32
iveqy OwiecPL1986: doesn't bitbucket has support for hooks?10:32
OwiecPL1986 iveqy somebody said on this channel no, but i didnt check this personaly10:32
→ sr78ger joined10:33
→ ntus1017 joined10:33
iveqy OwiecPL1986: in that case, stop using bitbucket... or your server needs to poll your bitbucket repo for new changes10:33
OwiecPL1986 even if have hooks i cant run update on server?10:33
iveqy OwiecPL1986: why not?10:33
OwiecPL1986 iveqy how server will know to start update? In that case i have to create repository in server and after update it will run update in /home/user/www ?10:34
→ pdurbin joined10:35
OwiecPL1986 too long way, no sense10:35
or its something what i dont know?10:35
iveqy OwiecPL1986: suppose you've hooks on bitbucket, you could just sftp the changed files to your server10:35
← Cromulent left10:36
→ Chryyz joined10:36
iveqy and it's not stupid to have a repo on the server, that will minimize traffic because you can just send the changed deltas to the server with help from git instead of the whole files10:36
Arafangion Although do be wary of who you give access to, as always.10:37
→ stephCoue joined10:37
→ the_cybersphinx joined10:37
→ daito joined10:37
OwiecPL1986 iveqy mmm so how should look url by sftp to ~/www folder? add remote ssh://git@server/accountname/www ?10:39
→ Guest81169 joined10:39
Guest81169 → jan1337z10:39
iveqy OwiecPL1986: look at how git-ftp does this.10:40
https://github.com/git-ftp/git-ftp10:40
OwiecPL1986 but cant do this with clear git?10:41
iveqy OwiecPL1986: no, as I said: !deploy10:41
gitinfo OwiecPL1986: Git is not a deployment tool, but you can build one around it(in simple environments) or use it as an object store(for complex ones). Here are some options/ideas to get you started: http://gitolite.com/the-list-and-irc/deploy.html10:41
← dhruvasagar left10:42
OwiecPL1986 iveqy ok so the final question is: jenikins could be tool what solve my problems?10:42
→ dhruvasagar joined10:42
iveqy OwiecPL1986: it would be like buying a tank to get to school. It would work, but it would be able to do a lot more than you need it to do.10:42
← daito left10:43
→ sqlnoob joined10:43
→ RonScagz joined10:43
OwiecPL1986 if it can testing etc. code it will be also needed in the close future10:44
i will read about that10:44
thx for help10:44
iveqy OwiecPL1986: then it would be an investment. Good luck!10:44
→ tinti_ joined10:45
→ decci_ joined10:45
TMO jenkins will do automatic testing as well.10:45
← sharpshooter left10:46
→ daito joined10:47
TMO dunno with bitbucket.. but with regular git-repository. you can have a post-receive hook in git telling jenkins to start a new build/test deployment10:48
← decci left10:48
→ keller joined10:49
→ shuaib joined10:51
→ babilen joined10:51
shuaib Any ideas on how to setup a local git mirror of a remote svn repo, so that I can work with my git repo, and at the end sync my changes into svn repo?10:51
← mikecmpbll left10:52
→ BashCoder joined10:52
shuaib I would want all the setup to be local for most part, as I don't want to get on the shady nerves of the svn system's admin10:52
← paul424 left10:53
iveqy shuaib why not just use git-svn?10:53
TMO git-svn clone -s http://example.com/my_subversion_repo local_dir10:53
shuaib iveqy: Doesn't git-svn do some magic everytime you push to svn repo (rebase?)10:54
magic to my local repo that is10:54
← dhruvasagar left10:54
→ dhruvasagar joined10:54
Arafangion shuaib: It does a lot of magic.10:56
shuaib Arafangion: and I want to avoid that. As I would like to keep my local repo, in pure git form, i.e., later be able to sync with another git repo10:56
→ enr joined11:00
← pdurbin left11:00
iveqy shuaib: the magic it all on your side and it's done with you "pull" and commit, not when you push11:01
however do have an "usual" git repo to sync with a svn repo is almost impossible (at least to do in a good way), svn has a linear history while git has a DAG11:01
← dhruvasagar left11:01
TMO dcommit11:01
iveqy you need to convert the DAG to a LINE11:02
→ dhruvasagar joined11:02
→ diegoviola joined11:03
→ mikaelj joined11:04
shuaib Lets see. I might just as well use git-svn11:04
→ spaceonline joined11:04
shuaib Does anyone know of a way to make git svn translate .gitignore to something svn specfic?11:04
so my ignored files keep being ignored on svn too11:04
anyone using svn to access repo that is11:05
← enr left11:06
→ enr joined11:06
→ lov joined11:07
→ emergion joined11:07
→ Octalot joined11:09
→ enr_ joined11:10
← enr left11:10
→ alorence joined11:11
→ chaw joined11:12
iveqy shuaib: your files won't be on svn if you're ignoring them in git11:14
shuaib: otherwise just keep a .svnignore too (or whatever esvn uses)11:15
shuaib iveqy: yea but if someone else clones that repo, reproduces those files on his side while using svn, he is going to push them in11:15
iveqy shuaib: yes. That's onw of the downsides with using git and a svn server. I don't think a conversion of .gitignore can be made in a secure way11:16
shuaib iveqy: yea well, seems like I will have to handle some mess manually11:16
How do I push dvcs down the management's throat? :/11:17
→ rshetty joined11:17
iveqy shuaib: I don't know. dvcs is scarry and not always the right thing to use11:17
shuaib iveqy: not as scary as centralized vcs11:18
→ ToBeFree joined11:18
iveqy shuaib: IMHO dvcs is much scarrier because you can do more. Centralized repos are easy to understand, however they are hard to work with because they have limited features11:18
shuaib: I guess you just have to prove that code quality and development speed increases with git11:19
shuaib iveqy: yea, so centralized vcs is easy to learn, hard to collaborate easily on. DVCS hard to learn, but heavens to use11:19
anyway...11:19
shuaib back to work11:19
← stamina left11:20
← daito left11:21
← Guest__ left11:21
→ CheckDavid joined11:21
→ daito joined11:21
mitsuhiko_ → mitsuhiko11:22
← mitsuhiko left11:22
→ mitsuhiko joined11:22
→ Guest__ joined11:22
→ h0bbit joined11:24
→ sgambino joined11:24
← ryanhirsch left11:25
← daito left11:26
← ntus1017 left11:26
→ UnacceptableUse joined11:28
→ dsdeiz joined11:28
← dsdeiz_ left11:28
→ sajan joined11:29
→ plof joined11:29
→ FernandoBasso joined11:29
plof hi , i am trying to develop a website with my freind who is remotly distant from me , i wonder if we can host the files in git repos and develop in a smooth way ? (am totally beginner to git concept )11:30
→ dfanjul joined11:31
← sqlnoob left11:32
→ marhaban joined11:33
→ buibex joined11:35
john2x is it possible to reset just one file?11:38
→ grilix joined11:38
← beautifulmind left11:39
dfanjul john2x: git reset -- <file>11:39
john2x: or git checkout -- <file>, depending on what 'reset' means for you11:40
→ Somelauw joined11:40
→ blami joined11:40
→ Cromulent joined11:40
john2x dfanjul: it's already commited, so `reset` right?11:40
← fuzzybunny69y left11:40
← Somelauw left11:40
→ nicklasbo joined11:41
john2x wait, that doesn't sound right11:41
dfanjul john2x: then, do you want to create another commit restoring the old file? or do you want to amend the original commit?11:41
john2x dfanjul: I want another commit restoring the old file..11:41
dfanjul so, then: git checkout <original commit> -- file11:42
john2x thanks!11:42
→ deed02392 joined11:42
← sajan left11:42
← eletuchy left11:43
→ eletuchy joined11:43
← rob_w left11:47
← eletuchy left11:48
← rednaks left11:48
→ tmcmahon joined11:48
→ saurabh joined11:49
Perdu plof: Of course, that's what git is for11:49
plof Perdu , thx :)11:50
→ godgodgodgo joined11:52
→ da5id joined11:53
gitinfo set mode: +v11:53
← cenuij left11:53
→ frgtn joined11:53
→ anonymuse joined11:55
← Cromulent left11:56
→ simesy joined11:58
→ p1ng joined11:58
→ beautifulmind joined11:58
→ ffio joined11:59
← angelsl left12:00
→ guest171 joined12:01
→ mikecmpbll joined12:01
→ alphanyx joined12:01
← alphanyx left12:01
→ IrishPrime joined12:02
← simesy left12:02
→ amjam joined12:03
→ eletuchy joined12:05
Sp4rKy ␄/W 3612:06
← dsdeiz left12:07
shuaib bangs his head on the table as he figures that this new org he joined has one svn repo, having all their projects as subdirectories12:07
Arafangion shuaib: Nothing wrong with that.12:08
shuaib shoots himself on hearing Arafangion12:08
Arafangion shuaib: It simplifies maintenance and backups, and it's a reasonable fit for SVN.12:08
alexander__b what do Windows users need to set in order to get line endings?12:08
i.e. I want the Windows guy to get the same as me. I'm on GNU+Linux.12:08
Arafangion alexander__b: I suggest you just set up your .gitattributes and forget about it.12:09
alexander__b: Make them text=auto.12:09
alexander__b Arafangion: I'm not changing anything. I want him to change whatever so that he doesn't send me files without \ns.12:09
Arafangion alexander__b: They always have \n's.12:09
alexander__b no they always send me commits where they the last line of my file and add a new end line without \n.12:10
*delete the last12:10
Arafangion Oh, that. That's a text editor thingy.12:10
alexander__b yes but I seem to recall git being able to "fix" it for windowsers12:11
→ m0viefreak joined12:11
Arafangion alexander__b: That's not a windows issue. git allows conversion of line endings, from crlf to and from simple lf's.12:11
muep at least on GNU/Linux git will happily accept files without a newline as their last character12:11
Arafangion alexander__b: But I don't think it will arbituarily *add* or *remove* line endings.12:11
muep: Likewise on windows.12:12
cbreak-work there's a pre-commit hook that can check for whitespace errors12:12
alexander__b Arafangion: so is there nothing I can tell him to do to not send me that crap? except... "fix ur editor"?12:12
cbreak-work: I thought maybe it was builtin.12:12
cbreak-work alexander__b: you have to make the sample pre-commit hook executable and give it the correct name12:12
alexander__b i.e. that he could git config --stop-annoying-alexander and then I could be a happy camper.12:13
cbreak-work: right.12:13
Arafangion alexander__b: Indeed, or you could take cbreak-work's suggestion, or use .gitattributes and setup filters.12:13
→ freeed joined12:13
→ normanrichards joined12:13
→ ntus1017 joined12:13
→ Wh1plash|m joined12:14
→ paul424 joined12:14
← amjam left12:14
alexander__b Arafangion: thanks.12:15
Arafangion It's amazing how crap windows editors are, though.12:15
← UnacceptableUse left12:16
← saurabh left12:17
EugeneKay Notepad++ works fine for me.12:17
jast I use vim12:17
← decci_ left12:18
Arafangion EugeneKay: That's one of the better ones.12:18
OwiecPL1986 Arafangion use sublime text12:19
Arafangion Worser ones are, eg, thelikes of Borland's IDEs, Notepad.exe...12:19
alexander__b he uses sublime, which looks like a proprietary emacs without any of the... features...12:20
← Latro left12:20
alexander__b he saw my vim plugins for git and found a bunch for sublime though, so for git stuff it works nicely. except for line endings at the end of the file apparently.12:21
→ yshh joined12:21
← tmcmahon left12:23
jast sublime does have a few interesting features, e.g. being able to edit multiple lines simultaneously (and visually)12:23
nothing that I need to switch away from vim for, though12:23
← dhruvasagar left12:23
Arafangion alexander__b: Strictly speaking, newlines at the end of the file shouldn't be needed.12:24
alexander__b it may be advanced and feature rich. he doesn't appear to be a poweruser of it, so my impressions are based on his knowledge of the editor. I read that it was proprietary and left interest.12:24
Arafangion: it's just convention, so it's annoying that the windows crowd insist on filling every commit with --nothing ++nothing.12:25
→ netsroht joined12:25
→ mmitchell joined12:25
→ yacks joined12:25
← shuaib left12:25
EugeneKay No we don't.12:25
Arafangion yes, that's annoying. Get him to remove unneccessary changes.12:25
→ shwaiil joined12:25
shwaiil hi12:25
EugeneKay Hate the user, not the tools.12:25
alexander__b EugeneKay: sorry, a moderation: the ones I work with do.12:25
EugeneKay Except when PHP is involved. Then hate the tools.12:26
Arafangion EugeneKay: PHP is nicer than windows.12:26
← mikecmpbll left12:26
Arafangion But only barely.12:26
bremner plenty of hate to go around.12:26
alexander__b then again they also have a weird habit of doing git add -A && git commit -m "lol this fixes stuff".12:26
dfanjul j/win 1512:26
sorry, :(12:26
iveqy shwaiil: !welcome12:26
gitinfo shwaiil: 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.12:26
← ffio left12:27
→ ffio_ joined12:27
Arafangion alexander__b: reject it, then ask when he's going to fix a specific issue.12:27
alexander__b: And then claim you can't find the commit.12:27
shwaiil how to delete a commit from remote repository ? I've made some changes and did git add -u instead of just add the changes I wanted then pushed git push origin branchX. I undo locally but I need to do the same in the repo, is there a way to delete that commit ?!12:27
cbreak-work force push12:27
alexander__b Arafangion: sadly I do not have enough time because it's a super short project, and interns and blah. but in srs projects I of course just reject that crap.12:27
iveqy alexander__b: git has some tools for helping with lineending conversion (as you've already beeing told) however I find it really creepy to have a SCM altering my files. That's what I've an editor for.12:28
alexander__b !force12:28
gitinfo [!force_push] If you need to overwrite the history of a remote git repository (very bad idea, see !rewrite), you can do so with `git push -f`. Note the remote server may reject this. See man git-config and search for receive.denyNonFastForwards. Best practice is for upstream servers to denyNonFastForwards.12:28
iveqy shwaiil: !public12:28
gitinfo shwaiil: [!rewriting_public_history] Rewriting public history is usually bad. Everyone who has pulled the old history have to do work (and you'll have to tell them to). If you must, you can use `git push -f` to force (and the remote may reject that, anyway). See http://goo.gl/waqum12:28
alexander__b iveqy: well these people do git add -A && git commit -m "lulz!!!!", so I doubt they'd care... or even notice...12:29
bremner that's what managers are for.12:29
Arafangion alexander__b: Bah.12:29
← diegoviola left12:29
alexander__b Arafangion: it's a prototype so thankfully this repo will never see daylight. the code... the horror...12:29
iveqy alexander__b: yeah I know, people are stupid (or rather, they don't understand why to use SCM). My suggestion is that they should just stick to subversion and be out of my way if possible12:30
Arafangion alexander__b: Lets pray you're right.12:30
alexander__b Arafangion: maybe I should remove myself from the copyright headers in case it ever gets used. :-P12:30
Arafangion iveqy: That reminds me.12:30
→ ffio joined12:30
→ mikecmpbll joined12:30
Arafangion How does one set text attributes such that the line endings are left as-is.12:30
Arafangion runs and puts a '?' at the end of that.12:31
alexander__b though TBF the code I have committed is almost exclusively "Remove stupid redundant crap" or "Use a decorator instead of copypasting the same damn thing 103104302413241 things".12:31
iveqy Arafangion: I don't know. I don't use windows12:31
Arafangion iveqy: Irrelevant.12:31
← Guest__ left12:31
→ devoldmx3 joined12:32
← danicampa901 left12:32
→ scampbell joined12:33
← guest171 left12:35
→ troker joined12:35
← mmitchell left12:36
→ whaley joined12:37
iveqy Arafangion: AFAIK git doesn't care about your line endings by default on non-windows systems12:38
← wasanzy left12:38
→ TeckniX joined12:39
cbreak-work by default it doesn't care at all, even on windows12:40
→ TheBrainWork joined12:40
cbreak-work you have to enable autocrlf/core.eol/safecrlf/... stuff12:40
or use attributes12:40
← RonScagz left12:41
→ diegoviola joined12:42
Arafangion iveqy: BUt it does on windows, by defaults.12:43
→ rnsanchez joined12:43
Arafangion cbreak-work: The team has that stuff enabled inconsistently, and we dont' use attributes.12:43
cbreak-work: So how do you do it using attributes?12:43
jast Arafangion: -text12:43
cbreak-work Arafangion: git config -l12:43
read it12:43
Arafangion jast: That prevents diff'ing, and all that, doesn't it?12:43
cbreak-work does it contain crlf or eol related things?12:44
jast not according to the docs12:44
that's what the 'diff' attr is for12:44
Arafangion jast: I should read it again, then.12:44
← devoldmx3 left12:45
← BashCoder left12:46
→ kaldrenon joined12:46
→ hellopat joined12:46
→ krz joined12:48
← ffio left12:48
← ntus1017 left12:48
krz i accidently did git merge github/foo into the current branch. how would i undo this?12:48
→ nicknisi joined12:49
← jon_r left12:49
→ radiochatr joined12:50
radiochatr hi12:50
← normanrichards left12:50
gitinfo radiochatr: 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.12:50
→ wasanzy joined12:50
radiochatr anybody knows why after: git remote add drupal_facetapi http://git.drupal.org/project/facetapi.git12:51
when I do: git fetch drupal_facetapi12:52
the command hangs indefinitely ?12:52
→ zhost joined12:52
→ angelsl joined12:52
→ ShaPOC joined12:53
radiochatr but git clone http://git.drupal.org/project/facetapi.git works perfectly12:53
EugeneKay !repro12:55
gitinfo Please paste (using https://gist.github.com/ or similar) a transcript (https://gist.github.com/2415442) of your terminal session -- or, even better for complex issues, design a minimal case in which your problem can be reproduced, and share it with us. This will help immensely with troubleshooting.12:55
→ rfoust joined12:55
← FernandoBasso left12:56
→ Cromulent joined12:56
← chaw left12:56
→ helpD joined12:56
→ ryanjh joined12:57
radiochatr why is git fetch -vv <remote_branch> showing12:57
→ konundra joined12:57
lb1a radiochatr, http://pastebin.mozilla.org/270001712:57
radiochatr POST git-upload-pack messages indefinitely?12:57
→ nerdfiles joined12:57
→ simesy joined12:58
→ macabre joined12:58
CheckDavid Hello.13:00
gitinfo CheckDavid: hi! I'd like to automatically welcome you to #git, a place full of helpful gits. Got a question? Just ask it — chances are someone will answer fairly soon. The topic has links with more information about git and this channel. NB. it can't hurt to do a backup (type !backup for help) before trying things out, especially if they involve dangerous keywords such as --hard, clean, --force/-f, rm and so on.13:00
→ hyperboreean joined13:00
CheckDavid OH well. I'm new to this and not very techie. So here it goes.13:00
→ geekbri joined13:01
CheckDavid I would like to take a course and do some tasks. But apparently I need a linux console to do it, based on a reply I got from someone who helped me, which was: "no problem....all u need is to use ur linux terminal to push code to github and your heroku account...that is the only reason a console is needed and to edit files...if your main operating system is ubuntu thats even better"13:01
→ devoldmx joined13:01
CheckDavid But I am having an very hard time running linux distros on my pc13:01
Or even virtual box13:01
← buibex left13:02
← paul424 left13:02
→ breakingthings joined13:02
→ p0wn3d joined13:02
jordanm CheckDavid, this channel is not for linux distro support. try #linux or #ubuntu13:02
← simesy left13:03
jordanm windows can use git and push to github though13:03
muep CheckDavid: git can also be used just fine from operating systems that are not linux based13:03
CheckDavid I am not talking about linux distro support =_13:03
jordanm CheckDavid, you were making it sound like you were requesting help getting linux on your PC13:03
CheckDavid Nope13:04
#ubuntu, ##linux... etc etc...13:04
if I wanted that.13:04
lb1a CheckDavid, git is available for windows too13:04
← devoldmx left13:04
→ rednaks joined13:05
CheckDavid I just wanted to maybe get some help interpreting that suggestion.13:05
I got from a fellow coleague13:05
lb1a CheckDavid, the whole git thing is very linux centric and on linux it just works. on windows you'll have to do some extra steps, fiddeling with ssh for remote conections and stuff. but it is possible of course13:06
→ huoxito joined13:07
muep in my experience, it does not much differ on windows either13:07
TMO cygwin+git ?13:07
muep though it may seem a bit weird for someone who is only used to windows13:07
← ffio_ left13:07
EugeneKay msysgit works just fine with the included openssh tools13:08
You /can/ fiddle with the GIT_SSH env var to make it use plink.exe+Pageant, but there's no need to.13:09
← plof left13:09
→ chaw joined13:09
→ hahuang65 joined13:09
→ jerius joined13:09
EugeneKay I hate the stereotype that "Windows is worse!", as perpetuated by FOSS zealots who've never sat down to learn it. The street goes both ways, y'know.13:10
→ natechan joined13:10
lov ^^13:10
→ seigmann joined13:10
lov there are different focuses, of course. linux tends to be better for development, windows tends to be better for booting to a usable OS without having to edit your xorg.conf, etc.13:11
← nerdfiles left13:11
→ trungqtran joined13:11
← chaw left13:11
lov strangely, linux continues to be unpopular outside of developer crowds.13:11
EugeneKay "linux" as in traditional distributions perhaps; my mother loves her Chromebook.13:12
jordanm i haven't had to manually edit a xorg.conf is 5 years13:12
shadok EugeneKay: how could two things be worse than the other one? That makes them equal! ^^13:12
→ danicampa901 joined13:12
→ mike_af joined13:12
← shruggar left13:12
→ mr_mustard joined13:13
→ sqlnoob joined13:13
→ hyperair joined13:13
EugeneKay shadok - imaginary numbers13:13
← hahuang65 left13:13
← rshetty left13:14
← dfanjul left13:14
osse they would have to be complex, otherwise it wouldn't make a difference13:15
← rednaks left13:15
osse and be non-trivial. in principle all numbers are complex numbers13:17
→ mechanicalduck joined13:18
→ snearch joined13:19
→ normanrichards joined13:19
→ OnlyMax joined13:19
→ gebi joined13:20
→ rednaks joined13:20
← anonymuse left13:21
→ decotii joined13:21
→ mechanicalduck_ joined13:22
→ mdhopkins joined13:22
CheckDavid Well, I would like to do it with a Linux console13:22
for the learning purposes I guess13:23
since it's an educational program that I want to do13:23
EugeneKay My advice? Rent a VPS.13:23
CheckDavid But I just can't get one.13:23
Yeah, maybe I should do that.13:23
EugeneKay A Linode is $20/mo, and gives good performance, great stability, and it vomits unicorns.13:23
→ BashCoder joined13:23
→ AaronMT joined13:23
← gebi left13:23
→ mechanicalduck__ joined13:24
→ hetal_ joined13:24
← mdhopkins left13:24
← BashCoder left13:25
← trungqtran left13:25
→ BashCoder joined13:25
← wasanzy left13:25
→ gebi joined13:25
← mechanicalduck left13:26
canton7 CheckDavid, or fire up an EC2 instance for a couple of hours a day, if you want13:26
→ Sigma[Mtp] joined13:26
← mechanicalduck_ left13:26
→ tplaner joined13:26
→ BashCode_ joined13:27
→ saml joined13:27
CheckDavid I was trying to do it without charging anything from my bank account13:27
EugeneKay You can get a free EC2 micro for 12 months13:27
But the prices are exorbitant if you exceed the allowances13:28
→ trungqtran joined13:28
← beautifulmind left13:28
← BashCode_ left13:28
→ Uzk joined13:28
Uzk Olan13:28
← gebi left13:28
canton7 micro instances are like $0.016/hr on the spot market13:29
EugeneKay I'd rather not have my development environment spontaneously combust13:29
canton7 keep that on a ebs volume13:29
s/a/sn13:30
EugeneKay Still annoying ;-)13:30
canton7 slightly, but it's cheap!13:30
EugeneKay Free is cheaper13:30
→ gebi joined13:31
Arafangion Eh? I've learnt windows, and linux. Still prefer linux, despite the chaos.13:31
canton7 I was assuming a lack of free tier. Of course that's preferable if it's available13:31
Arafangion It, at least, comes with a text editor.13:31
EugeneKay <EugeneKay> You can get a free EC2 micro for 12 months13:31
← OwiecPL1986 left13:32
→ Christi123 joined13:32
EugeneKay https://aws.amazon.com/free/13:32
You can even boot a Windows box, too13:32
CheckDavid Yeah,, well that's another problem13:32
Arafangion What does he want, anyway?13:32
CheckDavid apparently I made an account at Amazon long ago13:32
And I didn't even use it13:33
→ mdhopkins joined13:33
alexander__b does anyone know of a vim plugin that lets me add a line/hunk?13:34
Arafangion alexander__b: I'd use interactive git add.13:34
← ryanjh left13:35
→ Adlai joined13:35
alexander__b Arafangion: I usually do too but sometimes I can't split it enough, and it'd be nice to just have vim deal with that13:35
Arafangion alexander__b: git gui's not so bad at that.13:35
alexander__b because line numbers etc gah.13:35
Arafangion: gui? I use git add -p.13:35
← gebi left13:35
alexander__b but anyway this is very beside my question.13:35
Arafangion alexander__b: You haven't tried git add -i?13:35
alexander__b maybe I should take it to #vim though.13:35
I don't like -i very much13:36
Arafangion alexander__b: I don't like vim very much. :)13:36
→ gebi joined13:36
→ dwmw2_gone joined13:37
→ Spyro joined13:37
← rapha left13:37
→ devoldmx joined13:37
→ ryanjh joined13:38
← BashCoder left13:38
→ wasanzy joined13:38
→ segher joined13:38
← CheckDavid left13:39
← sr78ger left13:40
← krz left13:40
→ dfanjul joined13:40
→ ojacobson joined13:40
→ eugenejen joined13:41
→ Sam_OTnosc joined13:42
→ dsdeiz joined13:42
← john2x left13:42
← radiochatr left13:43
← Cromulent left13:43
→ BashCoder joined13:43
← babilen left13:46
← gebi left13:46
← helpD left13:47
→ gebi joined13:47
→ kesha joined13:47
→ _BJFreeman joined13:48
← kesha left13:48
→ vedant joined13:48
_BJFreeman → BJfreeman13:48
→ CodeShark joined13:49
→ valgrind joined13:49
← gebi left13:49
→ devoldmx3 joined13:50
→ gebi joined13:52
← tvw left13:52
→ buibex joined13:52
→ helpD joined13:53
→ nurey joined13:53
← CodeShark left13:53
← normanrichards left13:54
← carlos-huang left13:54
← devoldmx left13:55
← nicknisi left13:55
→ tdm00 joined13:56
→ helpD_ joined13:57
← gebi left13:58
← hetal_ left13:58
→ mmitchell joined13:58
→ simesy joined13:58
→ gebi joined13:59
→ jaequery joined13:59
← jaequery left14:00
→ donmichelangelo joined14:00
← helpD left14:00
← BashCoder left14:01
→ dougsland joined14:01
← dougsland left14:01
→ dougsland joined14:01
← gebi left14:01
← troker left14:01
Sigma[Mtp] → Sigma14:02
← simesy left14:03
→ gebi joined14:04
→ johnnyfuchs joined14:05
→ stamina joined14:05
→ groglogic joined14:05
→ BashCoder joined14:05
→ aksnot joined14:06
← sqlnoob left14:07
→ v0n joined14:08
→ freeed_ joined14:08
→ salamanderrake joined14:09
← freeed left14:09
→ anonymuse joined14:09
→ scientes joined14:09
← scientes left14:09
→ scientes joined14:09
→ hahuang65 joined14:09
→ acharlieh joined14:10
wcwrel I found a bunch of commits using --grep= , how do i inspect each commit to see what files changed14:10
→ _BJFreeman joined14:10
iveqy wcwrel: git log14:10
→ OOPMan joined14:10
→ ClamShell joined14:11
→ drizzd joined14:11
← scientes left14:11
BJfreeman → Guest2479414:11
→ orafu joined14:11
_BJFreeman → BJfreeman14:11
wcwrel i used git log, but it doesnt show me the diffs14:11
just the commit message14:12
→ cannonball joined14:12
← Guest24794 left14:12
← gebi left14:12
→ rpg joined14:12
wcwrel I did: git log --all --grep=<ticket#>14:12
→ DonRichie joined14:12
→ gebi joined14:13
← hahuang65 left14:13
→ scientes joined14:13
← scientes left14:13
→ scientes joined14:13
→ cballenar joined14:14
iveqy wcwrel: then you might look for git show instead14:14
→ hetal_ joined14:14
→ ldh joined14:14
→ Etheco- joined14:14
wcwrel do i pass a specific hash to git show?14:15
iveqy wcwrel: yes, see man git show14:15
gitinfo wcwrel: the git-show manpage is available at http://jk.gs/git-show.html14:15
← Etheco left14:16
← emergion left14:16
wasanzy am having some issue here with pushing projects to my git server14:16
← nurey left14:16
wcwrel fatal: bad object 0083358fe2df12c07495bbb8612346287f4dfadd14:17
→ Etheco joined14:18
← tdm00 left14:18
osse wcwrel: git log -p14:18
← vedant left14:18
← hetal_ left14:18
wasanzy I created SSH public keys and in my gitadmin home, I did create .ssh directory and copied the key to authorized_keys but when I tried to push to the server, it is still asking for password14:18
wcwrel osse: what is -p?14:19
← grilix left14:19
jast show patches14:19
osse wcwrel: try it and see14:19
jast all explained in the manpage :P14:19
← gebi left14:19
osse git log -p --grep=foo14:19
→ normanrichards joined14:20
→ gebi joined14:20
← Etheco- left14:20
→ shamess joined14:20
← BJfreeman left14:21
← devoldmx3 left14:21
→ simesy joined14:21
→ oliveiraev joined14:21
← mr_mustard left14:21
← dsdeiz left14:22
→ devoldmx joined14:22
← ldh left14:24
← gebi left14:24
→ ldh1 joined14:24
→ gebi joined14:25
michael_mbp → zz_michael_mbp14:25
bluj i've clone'd a repo for the first time and an tracking branch X... is there a way to go-back-in-time a little before they had merged branch Y and introduced problems? with repo's ive had around longer, i would use 'reflog' to checkout a certain state, but with a fresh clone my reflog is empty. do i just checkout HEAD~20 , for e.g, to go back?14:26
osse bluj: yes14:26
you can checkout any commiy you want. no need for reflog14:26
bluj ok i'll try that14:26
← iokill left14:27
→ Fandekasp joined14:27
← snearch left14:28
wcwrel is there a way to do git log -p --grep=blah on github?14:29
→ vedant joined14:29
→ dhruvasagar joined14:29
iveqy wcwrel: I think you can get a better answer for that in #github. AFAIK you've to clone the repo first14:30
→ dsdeiz joined14:30
← mikecmpbll left14:30
← scientes left14:30
→ aldodelgado joined14:31
← buibex left14:31
→ jerius_ joined14:31
← gebi left14:32
→ gebi joined14:33
→ vedant_ joined14:33
← jerius left14:34
← vedant left14:34
← gebi left14:34
Arafangion wcwrel: github's search is crap, from what little I've seen. Much easier to just clone the repo first. :)14:35
wcwrel yeah they really needa better search14:36
← helpD_ left14:36
→ FernandoBasso joined14:36
→ gebi joined14:38
→ grilix joined14:38
→ Dougie187 joined14:40
→ bddy joined14:40
gitinfo set mode: +v14:40
→ corni joined14:41
bddy Hi there. Are there any articles/books about how to code with git? E.g., when to commit, when to merge and etc?14:41
→ Guest____ joined14:42
j416 bddy: there's always !book, but perhaps you need to think about what you need git for?14:42
gitinfo bddy: There are several good books available about git; 'Pro Git' is probably the best: http://git-scm.com/book but also look at !bottomup !cs !gcs !designers !gitt !vcbe and !parable14:42
← benoit left14:42
YoungFrog it's the more general question : what's a (D)VCS good for ?14:42
→ shuaib joined14:43
→ rabellamy joined14:44
ojacobson tracking, publishing, and discussing changes to a project over time, if the project is largely text14:44
(there are very few distributed *asset-oriented* tools)14:44
→ lresende joined14:45
bddy For example, in my projects I chaotically jump from one piece of code to another and add some functionality, and I don't know how could git help here.14:45
← gebi left14:45
→ jcromartie joined14:46
→ gebi joined14:46
osse bddy: commit before each jump14:46
bddy: or you can keep the stuff seperated so they don't interfer with eachother14:46
bddy And I use git instead of rsync :(14:46
← shuaib left14:47
→ shuaib joined14:47
← dougsland left14:48
→ dougsland joined14:48
→ kermit joined14:50
→ rudak joined14:50
← gebi left14:50
← shuaib left14:51
→ voila joined14:52
voila hello14:52
← bddy left14:52
voila helo14:52
ToxicFrog bddy: commit before jumping; if you forget to do this, use selective add (git add -p, git gui) to split your changes into logically separate commits/branches afterwards.14:52
→ tdm00 joined14:52
ToxicFrog I tend to write 3-4 logically distinct things at a time without remembering to commit, then split them up afterwards.14:53
→ hahuang65 joined14:53
→ bddy joined14:53
→ micah joined14:53
← bddy left14:53
iveqy voila: !welcome14:53
gitinfo voila: 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.14:53
→ mikecmpbll joined14:53
ojacobson in order to make sensible use of an SCM tool -- any SCM tool, not just git -- you need to care about making the "trajectory" of a project comprehensible, even if your working habits aren't14:54
micah i've got three stashes, and I'm trying to figure out which one is which. So I looked in gitk --all, but I only see one of the stashes listed14:54
ojacobson micah: git stash list14:54
micah ojacobson: yeah, i was looking at that, but I wanted more information, so that is why I ran gitk --all14:55
osse micah: git stash show -p <stash>14:55
← aksnot left14:55
micah i know i know... my question was about gitk only showing one stash14:55
i'm familiar with the 'git stash' commands14:56
osse oh14:56
→ thiago joined14:56
→ gebi joined14:56
osse all but the first stash is kept in the reflog. maybe that's why gitk won't list them14:56
micah i see, thanks14:56
does that mean only the first stash is shown in gitk?14:57
YoungFrog I guess you can force gitk to show other stashes by giving their sha on the command line14:57
→ jerius joined14:57
← jerius_ left14:57
← hahuang65 left14:57
→ shuaib joined14:58
← shuaib left14:58
← rednaks left14:58
ojacobson micah: gitk doesn't show reflogs (since that would be incredibly noisy for most refs)14:58
← Uzk left14:58
micah stash{0} is the earliest stash?14:58
ojacobson the stash *relies* on reflogs to store past stashes -- in fact, the {N} syntax is nothing more than reflog access, not specific to stash at all14:58
latest stash14:58
micah confused about what he was doing a month ago now :P14:59
osse micah: it's the latest stash14:59
← devoldmx left14:59
← Guest____ left15:00
lahwran if I git clone a local repo, then it will use hard links for the git objects, correct?15:00
← rudak left15:00
→ jaequery joined15:00
osse lahwran: by default yes, unless you use a "file://" URL thingy15:00
lahwran if I then pull a force-pushed history revision in the original repo, what happens? do the git objects get altered?15:01
or do they get unlinked from the original repo and then replaced?15:01
or what15:01
ojacobson lahwran: ref changes only ever add objects15:01
osse they probably stay in the remote repo15:01
and get gc'd in time15:01
ojacobson even force-pushing a branch and removing objects from history doesn't delete or modify those objects (they're probably in pack files anyways)15:01
← lresende left15:01
ojacobson they might get gc'd eventually15:01
note that pull is "fetch, then merge", though, which rarely goes well in the presence of force pushes15:02
lahwran so I have three imaginary repos here; [email@hidden.address] ~/repo (with a remote to the previous), and ~/repo2 (with a remote to ~/repo)15:02
← diegoviola left15:02
ojacobson you can happily pretend objects' physical representations are never deleted, and shared if possible15:02
← konundra left15:03
→ Guest__ joined15:03
ojacobson and if you touch them or rely on that assumption *directly* rather than through a git command, you get to keep both halves when your system breaks15:03
← daanavitch left15:03
→ devoldmx joined15:03
ojacobson (if you're really deeply concerned about storage requirements for multiple copies of the "same" repo on the same system, look at the alternates mechanism)15:03
← Guest__ left15:04
← gebi left15:04
→ nerdfiles joined15:04
lahwran I'm expecting a huge force-push of commit deletion soon in a very large repo15:04
← anonymuse left15:05
lahwran I would like to keep a local backup, but my main local copy of that repo should be in sync15:05
→ anonymuse joined15:05
→ gebi joined15:05
ojacobson make a local branch pointing to the current state you'd like to keep15:06
then when the remote branch is rewritten to point to different history, you'll still have a local name for the original branch15:06
lahwran hm, I guess that is what branches do, isn't it15:06
for some reason it feels more appropriate to throw it in a different repo, I'm not sure what about this scenario is making me pattern match that15:07
→ ozette joined15:07
← howlymowly left15:07
← enr_ left15:07
→ enr joined15:08
← cballenar left15:08
→ anonymus_ joined15:08
← dwmw2_gone left15:09
→ dwmw2_gone joined15:09
→ jackdempsey joined15:09
→ secondjeff joined15:09
← anonymuse left15:09
→ axl_ joined15:10
voila here somebody direct me for some book .. git for computer science .. or something like that .. Any one knows what was the exact name15:11
:)15:11
→ ManiaArmyYurt joined15:11
ojacobson !cs15:11
gitinfo "Git for Computer Scientists" is a quick introduction to git internals for people who are not scared by phrases like Directed Acyclic Graph. http://eagain.net/articles/git-for-computer-scientists/ See also !concepts !bottomup15:11
lahwran oooohh15:11
cool15:11
→ Rix joined15:11
→ ldh joined15:12
← enr left15:12
→ mklappstuhl joined15:12
← Spyro left15:12
→ pitzips joined15:12
← gebi left15:12
→ Spyro joined15:13
← simesy left15:13
Spyro → Guest6663015:13
← ldh1 left15:14
→ cballenar joined15:14
→ ValDuare joined15:14
← nicklasbo left15:14
ValDuare how do you pin a file that you don't want to over write in git pull15:15
→ lavaman joined15:15
→ Marquisk2 joined15:15
jast git pull doesn't overwrite files15:16
that is, if they have uncommitted changes15:16
→ Destos joined15:16
jast if all content you care about has been committed, you can easily get it back at any point15:17
voila whenever we try to push commit on remote , it throw us error .. do pull first ... my question is , how git come to know that our local master does not incorporate latest merge ?? is it through any file or what ?15:17
ValDuare i.e. on my web server I just logged in and cloned my repo for a site I'm working on and made access restrictions to .git folder15:17
→ eletuchy_ joined15:18
jast voila: the git server and client communicate and figure out what history (i.e. which commits) the two sides have in common15:18
ValDuare so i'd like to not override the config file for the production db server15:18
jast !configfiles15:18
gitinfo It is recommended to store local configuration data in a file which is not tracked by git, but certain deployment scenarios(such as Heroku) may require otherwise. See https://gist.github.com/1423106 for some ideas15:18
→ mechanicalduck joined15:18
ValDuare how do I untracked a file15:18
→ joevano joined15:18
jast !untrack15:19
gitinfo To remove a file from git's tracking, without deleting it from your working tree, `git rm --cached <file>`. Note that any repo which pulls this change will delete their local copy of that file. You can "bring it back" using `git checkout HEAD^ file`15:19
ValDuare whats the --cached option do?15:19
voila jast, So to be more precise .. Can I say ... Git want latest commit on remote in history of my master before pushing anything on server ?15:20
→ kenperkins joined15:20
ToxicFrog ValDuare: deletes it from the index. man git-rm15:20
gitinfo ValDuare: the git-rm manpage is available at http://jk.gs/git-rm.html15:20
voila this gitinfo is really nice15:20
ValDuare ok ty15:21
EugeneKay !botsnack15:21
gitinfo Om nom nom15:21
→ ldh1 joined15:21
EugeneKay pets the bot15:21
EugeneKay Good boy!15:21
← eletuchy left15:21
voila So to be more precise .. Can I say ... Git want latest commit on remote in history of my master before pushing anything on server ?15:21
← ldh left15:22
← mechanicalduck__ left15:22
cmn by default your local git will refuse to change history on the remote15:23
← devoldmx left15:23
cmn which at the graph level means that what's on a remote branch must be an ancestor of what you want to set it to15:23
jast voila: yeah, that's correct. any remote history that you haven't merged yet will prevent a normal push.15:23
voila okies15:24
← |Freeman| left15:24
→ devoldmx joined15:24
→ breck7 joined15:25
→ causasui joined15:26
cmn but remember this has nothing to do with branch names15:26
→ Noldorin joined15:26
→ shruggar joined15:27
→ gusnan joined15:27
→ Guest__ joined15:27
→ bklane joined15:28
GarethAdams How can I display the git log for a file that doesn't exist any more?15:29
EugeneKay Same way you do for any file. `git log -- path/to.file`15:29
→ r04r joined15:29
← devoldmx left15:30
← jatt left15:31
→ PaulCapestany joined15:31
→ cods joined15:32
← bklane left15:32
← joevano left15:32
voila My branch is diverge by 1 commit on either side ... then i merge it .. then when i did git status .. it is saying ... 'Your branch is 2 commits' ahead of origin/experiment ... I know how it is ahead .. 1 for my changes , 2 for merge commit .... At this point HEAD points to merge commit and remote/experiment points to its latest commit .. But problem is I am now able to figure it out when i draw a DAG ..15:32
EugeneKay !lol15:33
gitinfo A nifty view of branches, tags, and other refs: git log --oneline --graph --decorate --all15:33
EugeneKay Let git do the work for you15:33
voila let me try :)15:33
→ Davey joined15:33
← mklappstuhl left15:33
← marhaban left15:34
→ marhaban1 joined15:34
→ devoldmx joined15:35
→ sferry joined15:35
← sferry left15:35
→ sferry joined15:35
→ howlymowly joined15:35
← ryanjh left15:35
→ Wh1plash_ joined15:36
→ ryanjh joined15:37
→ rudak joined15:38
← groglogic left15:38
← marhaban1 left15:38
← Wh1plash|m left15:38
← blami left15:38
→ babilen joined15:38
chernjie feels like bragging15:38
lahwran oh.15:38
→ enr joined15:38
milki ah.15:39
jast eh.15:39
lahwran goofballs15:39
chernjie :p15:39
→ aksnot joined15:39
milki spaceballs15:39
← kermit left15:39
→ mckendricks joined15:40
lov Silly question; I've modified a file and changed several lines. However, I realized that a good number of the changes that I've made are completely pointless, and want to rollback. One way to do this would be to commit the changes I do want, checkout the file, then do a git reset, but that seems a little odd. Is there a smarter way to do this?15:41
→ joevano joined15:41
← nerdfiles left15:41
← DonRichie left15:41
chernjie git revert ?15:41
ToxicFrog Why do you need the git reset in there?15:41
cmn reset -p or checkout -p15:41
ToxicFrog git add -p; git commit; git checkout -- file15:41
cmn depending on what you want to change15:41
chernjie ooo... have never tried "checkout -p" before15:42
have always used add/reset with -p :D15:42
lahwran lov: I'm going to vouch for cmn's answer15:42
→ TomyLobo joined15:43
→ endiruna joined15:44
← mckendricks left15:44
← v0n left15:44
→ v0n joined15:45
→ paolooo joined15:45
← devoldmx left15:45
→ marhaban joined15:45
← deuterium left15:46
→ devoldmx joined15:46
← FernandoBasso left15:47
← enr left15:47
← jackdempsey left15:48
lov oh neat, I didn't know that reset and checkout supported -p.15:48
← micah left15:49
← trungqtran left15:50
→ schemanic joined15:50
→ kermit joined15:51
← Christi123 left15:51
→ api joined15:51
→ LuxuryMode joined15:51
← marhaban left15:51
→ Christi123 joined15:52
→ McEnroe joined15:52
← paolooo left15:52
← aksnot left15:52
→ mklappstuhl joined15:53
LuxuryMode Accidentally did git add && commit in the middle of a rebase with conflict. then, obviously, when i did rebase --continue git says "No changes - did you forget to use 'git add'?" etc.. What should I do now? I'm on HEAD (no branch)15:53
← dhruvasagar left15:54
→ paula_ joined15:54
← thiago left15:54
→ yoshx joined15:55
LuxuryMode Anyone?15:55
← paula_ left15:55
→ padhah joined15:55
Dougie187 LuxuryMode: try `git rebase --abort`15:55
voila There are many files in my staging area . I want to remove them .. I know I can do git rm --cached <file> .. but point is there are huge .. is there any other efficient way15:56
← padhah left15:56
← kenperkins left15:56
Dougie187 voila: are they tracked?15:56
voila Dougie187, yup ..15:56
LuxuryMode Dougie187, that will just put me back at where i was prior to the rebase, no?15:57
Dougie187 voila: do you want to undo all of them?15:57
LuxuryMode: yeah.15:57
LuxuryMode: you'd have to start over with the rebase, but you'd know where you were at least.15:57
→ Cromulent joined15:57
LuxuryMode Dougie187, right15:57
ok15:57
→ sniffells joined15:57
voila Dougie187, possibley ... ok if i do undo then can i add whatever i needed to tracked right ?15:58
possibly*15:58
Dougie187 maybe I don't quite understand your situation.15:58
→ oaoao joined15:58
gitinfo set mode: +v15:58
Dougie187 Basically, you do something like `git add .` when you didn't want a lot of the files staged?15:58
→ Anders_J joined15:59
voila ok .. there are many config files of eclipse and codeigniter .. I donot want them to be versioned .. I want to remove them from repository .. only from repository not from filesystem15:59
← devoldmx left15:59
EugeneKay !untrack15:59
gitinfo To remove a file from git's tracking, without deleting it from your working tree, `git rm --cached <file>`. Note that any repo which pulls this change will delete their local copy of that file. You can "bring it back" using `git checkout HEAD^ file`15:59
→ ldh joined16:00
← tdm00 left16:00
oaoao Hi. I'm using msysgit on windows and I don't want to enter password each time. I've tried all possible variants, but ssh-add always tells me that "Could not open a connection to your auth agent"16:00
← Guest__ left16:00
ojacobson oaoao: is ssh-agent or pageant running?16:00
TMO oaoao.. run pageant16:00
→ mckendricks joined16:00
TMO load your key..16:00
← ldh left16:01
oaoao TMO: There is no ssh-pageant there...16:01
→ mmlac-bv joined16:01
oaoao There is ssh-agent.exe when I ps | grep ssh-agent16:01
← tplaner left16:02
← mklappstuhl left16:02
← endiruna left16:02
TMO download pageant.. from same place as putty16:02
oaoao I've tried this script also https://help.github.com/articles/working-with-ssh-key-passphrases but it hangs for30-35 seconds and then says the same error.16:02
ojacobson oaoao: ssh-agent and ssh-add rely on environment variables to find one another, sounds like ssh-agent's env vars aren't set in your current session16:03
← yoshx left16:03
← ldh1 left16:03
← alorence left16:03
oaoao TMO: But pageant works with putty only?16:03
→ kesha joined16:04
oaoao I can't see how to add id_rsa there.16:04
TMO right click icon in systray16:04
→ dreiss_ joined16:04
oaoao Yeah, it proposes to add a key.16:04
→ Mode-M joined16:05
TMO and what is id_rsa16:05
oaoao That's what I uploaded to github, id_rsa.pub16:05
→ maslowbeer joined16:05
oaoao And created it with ssh-keygen16:05
TMO pub yes..16:05
→ vipzrx joined16:06
TMO add private to pageant16:06
→ Gonzo joined16:06
← Adlai left16:06
oaoao TMO: It definitely doesn't like it: Couldn't load this key (OpenSSH SSH-2 private key)16:06
→ kenperkins joined16:07
← kesha left16:07
→ melee joined16:08
← kermit left16:08
oaoao I also have a line "ForwardAgent yes" in ssh config, maybe it matters?16:08
→ Protux joined16:09
→ catcher joined16:09
← dreiss_ left16:09
← breck7 left16:10
→ tpollard joined16:10
oaoao SSH_AUTH_SOCK variable is set.16:10
catcher I intended to create a new branch from master, but created it from another feature branch instead. What's the best way to recover after several commits have been made? Cherry pick them into master?16:10
→ threesome joined16:12
→ devoldmx joined16:12
dfanjul catcher: git rebase --onto <the corret feature branch> <the incorrect feature branch, or the commit where you branched> <your new branch>16:13
→ simesy joined16:14
→ yoshx joined16:14
← yoshx left16:15
← valgrind left16:15
→ yoshx joined16:15
catcher so you're telling me I can't go through life without ever learning about rebase?16:15
← Arafangion left16:15
← devoldmx left16:16
dfanjul catcher: no, you can go through with cherry pick, but you asked for the best way16:17
→ valgrind joined16:17
← ShaPOC left16:18
→ deuterium joined16:18
← simesy left16:18
→ saintcajetan joined16:19
← mckendricks left16:19
← angelsl left16:20
→ knowj joined16:20
catcher cool, ty16:20
→ happyg0pher joined16:20
→ mklappstuhl joined16:20
→ nerdfiles joined16:21
← nerdfiles left16:21
→ hahuang65 joined16:21
→ dreiss_ joined16:21
knowj git pull (no options) what does it exactly do (yes I've read the man page)? From what I can see it updates your local repo branches with their remotes?16:21
← ValDuare left16:22
→ mckendricks joined16:23
knowj i.e. git fetch/git merge on each local branch?16:23
→ tplaner joined16:23
cmn no16:24
PerlJam knowj: did you also read the man page on "git fetch" and "git merge"?16:24
ojacobson on the current branch, from its tracked upstream branch & remote16:24
cmn it fetches with whatever args you passed and merges either that or the upstream branch16:24
ojacobson or on the current branch, from the passed branch16:24
← p1ng left16:24
knowj cmn: What if you have no arguments?16:24
→ dhruvasagar joined16:25
cmn the second part of the "or"16:25
→ shinnya joined16:25
← Christi123 left16:25
→ ewnd9 joined16:25
cmn knowj: see also the very first sentence in the manpage16:26
though I notice the rest isn't accurate anymore16:26
man git pull16:26
gitinfo the git-pull manpage is available at http://jk.gs/git-pull.html16:26
knowj cmn: Did read the man page but obviously very badly :)16:27
→ Christi123 joined16:27
cmn oh wait, it is accurate, just too specific16:27
→ ntus1017 joined16:27
→ p1ng joined16:27
voila When i create a new branch and push it to remote ... where should its first commit lie ?? Is it at master ??16:28
→ paolooo joined16:29
cmn pushing isn't related to where a branch starts16:29
← donmichelangelo left16:30
→ rednaks joined16:30
cmn you branch off of whatever is the right place for the particular branch, the tip of the main branch, the latest stable release, something else...16:30
→ donmichelangelo joined16:31
← valgrind left16:31
voila cmn, actually I just create branch with git branch solr .. then git checkout solr .. git push origin solr ... I saw that git rev-parse solr .... points to where master is pointing ..16:32
→ m8 joined16:32
← mikecmpbll left16:32
← Christi123 left16:32
cmn just as you asked, yes16:33
but the pushing is irrelevant16:33
→ Christi123 joined16:33
oaoao Okay, "ssh -T [email@hidden.address] says hello to me, but git push requires username and password.16:33
← ewnd9 left16:33
oaoao What could be the reason?16:33
cmn oaoao: you're using http16:34
voila cmn, yes i got it thanks :)16:34
→ goshawk joined16:34
← Christi123 left16:34
← sferry left16:35
→ Shusshu joined16:35
← acharlieh left16:35
→ Christi123 joined16:35
→ belak joined16:36
← freeed_ left16:36
← ryanjh left16:36
→ ryanjh joined16:37
→ _BJFreeman joined16:37
oaoao cmn: Thanks, now I don't need that ssh-agent.16:37
← rabellamy left16:37
→ ology joined16:37
cmn now?16:37
→ exoon joined16:37
_BJFreeman → BJfreeman16:38
→ ewnd9 joined16:38
oaoao I was trying to stop github from asking me password and ssh-agent didn't work, but there was no need to use it.16:38
jkremser → jkremser|bbl16:38
← Shusshu left16:38
oaoao Changing origin to ssh fixed the problem.16:38
← rednaks left16:38
SamB yeah, ssh-agent certainly isn't going to be much use with an http[s] transport16:39
TMO oaoao.. using a key without password?16:39
oaoao TMO: yeah16:39
← happyg0pher left16:39
knowj One of the guys on our repo has commit a file rename changing the casing of the filename. This resulted in the file disappearing from the repo. He fixed it on a new commit and pushed the fix into the repo. - I've now pulled the changes and it's merged fine apart from deleting the file he's fixed.16:40
→ bvi joined16:41
→ AWinterman joined16:41
→ Milossh joined16:41
knowj This is the result git pull origin env/dev -> git status -> # deleted: path/to/file/Importstatus.php16:42
Before the pull I had nothing staged16:42
Now this delete is staged and git reset isn't having any affect16:42
→ enr joined16:43
← enr left16:45
← oaoao left16:45
→ freeed joined16:46
← dreiss_ left16:46
← ozette left16:46
→ rabellamy joined16:47
← yoshx left16:48
→ yoshx joined16:49
→ DrGonzo joined16:49
→ Amandil joined16:50
← mklappstuhl left16:50
→ nfroidure joined16:50
← Gonzo left16:52
→ acharlieh joined16:52
← dfanjul left16:53
← rabellamy left16:53
← DrGonzo left16:53
→ Gonzo joined16:53
← m8 left16:54
← yoshx left16:55
→ dc5ala joined16:56
→ devoldmx joined16:56
→ troker joined16:57
← v0n left16:58
← exoon left16:59
→ helpD joined16:59
← Cromulent left16:59
← normanrichards left16:59
← shruggar left17:00
→ yoshx joined17:00
→ mribeirodantas joined17:00
giraffeslacks i have a branch that was create several commits back from master. i want to merge the changes in master to my branch. should i rebase to do that or checkout the other branch then merge master to the branch?17:01
→ rabellamy joined17:01
← kaldrenon left17:02
→ trungqtran joined17:02
PerlJam giraffeslacks: the latter sounds sane.17:02
giraffeslacks PerlJam: makes more sense to me too.17:03
→ breck7 joined17:03
← mdhopkins left17:04
← bvi left17:05
← cballenar left17:05
→ amjam joined17:06
← kenperkins left17:07
← mechanicalduck left17:08
← lavaman left17:09
→ mdhopkins joined17:09
→ Thra11 joined17:09
→ Balliad joined17:09
← Christi123 left17:09
← alexander__b left17:10
→ Christi123 joined17:10
→ lavaman joined17:10
← Christi123 left17:10
→ Christi123 joined17:10
→ bvi joined17:11
→ mechanicalduck joined17:11
→ v0n joined17:12
← Balliad left17:12
→ kaldrenon joined17:13
→ simesy joined17:14
← rabellamy left17:14
→ kenperkins joined17:16
← achadwick left17:16
← yoshx left17:17
← Sam_OTnosc left17:17
← zmike123 left17:17
→ zerorez joined17:18
→ ffualo joined17:18
→ uidzer0 joined17:18
← simesy left17:18
← seigmann left17:19
← donmichelangelo left17:20
→ tallest_red joined17:21
→ donmichelangelo joined17:21
→ endiruna joined17:21
→ btree joined17:22
→ maxmanders joined17:24
← trungqtran left17:24
← rudak left17:26
← Christi123 left17:26
→ kermit joined17:26
→ Christi123 joined17:27
→ covi joined17:27
covi Hi guys. I have a local branch A. I pushed it to remote at some point. Then, I made two commits in my local machine, but I regret adding them now (at this point I haven't pushed A again). Is there a way to somehow abandon these two commits on my local machine?17:28
→ evjsli joined17:29
→ boombatower joined17:31
→ f8l joined17:31
← maxmanders left17:31
→ zmike123 joined17:31
bret covi yes17:32
covi bret: git reset --hard origin/MYBRANCH?17:32
← devoldmx left17:32
bret find the SHA1 of the commit you want to roll back to, then do git reset --hard [the sha1 of the commit you want to roll back to]17:32
assuming you are on the branch you want to roll back17:33
covi bret: done. works like a charm17:33
← amjam left17:33
covi so the rule of thumb is: any unpublished history can be safely modified?17:33
← v0n left17:33
← gusnan left17:34
ojacobson covi: for values of "safely" that cover what you probably care about, yeah17:34
← wasanzy left17:34
→ mishok13 joined17:34
← hyperair left17:35
→ boombatower_ joined17:35
← dsdeiz left17:36
bret you can unpublish safely as long someone else doesn't get mad at you17:36
← btree left17:36
ojacobson indeed17:36
→ Vampire0 joined17:36
ojacobson scm tools are not a replacement for communication17:36
← OOPMan left17:36
← boombatower left17:36
bret but if others are working with you, its generally not ideal to mess with master's history17:36
→ v0n joined17:36
osse if car companies can recall models then godammit I can push --force17:37
lov o_O17:37
→ hwrd|work joined17:37
bret ^_^17:37
covi bret: how do you define 'working with you'? what if they only fetch my branches and look at them but will not make any commits?17:38
← acharlieh left17:38
bret thats fine17:38
← Mode-M left17:39
← boombatower_ left17:39
→ boombatower joined17:39
bret especially in your situation since you didnt even push your bad changes17:39
→ cballenar joined17:40
→ dreiss_ joined17:40
bret covi, it sounds like what you did deserved its own branch off of master17:40
→ umib0zu joined17:40
bret that way you dont have to mess with history17:40
← Davey left17:40
→ hugo_dc joined17:40
→ Davey joined17:41
bret covi, even if you end up just fastforwarding master into your feature branch, its worth the convinence of not having to deal with messing with resets (opinion)17:42
← ewnd9 left17:43
← ntus1017 left17:43
← mechanicalduck left17:44
← endiruna left17:44
← AWinterman left17:44
→ Cromulent joined17:45
→ mechanicalduck joined17:45
→ devoldmx joined17:46
→ trungqtran joined17:46
→ sheeny joined17:49
sheeny I just cloned from github and now im trying to push and im getting this..17:50
ERROR: Permission to craigsheen/paper_trail.git denied to craigsheen/date_output.17:50
← shwaiil left17:50
→ tdm00 joined17:50
sheeny why is it getting confused?17:50
← yshh left17:50
→ paul424 joined17:52
→ johnbenz_ joined17:53
← bandroidx left17:53
← oso96_2000 left17:54
← Octalot left17:54
→ yshh joined17:55
grawity you added your SSH key to the wrong location, it seems17:55
← bvi left17:56
grawity Github has two places to configure them – normally, you'd add your key to your own account, and it would allow access to all repos of yours17:56
instead, you seem to have added your key to a repo directly, which allows access *only* to that repo17:56
sheeny ah17:56
grawity remove it from https://github.com/craigsheen/date_output/settings/keys and add it to https://github.com/settings/ssh17:57
I'm assuming that's what you meant to do in the first place17:57
→ bandroidx joined17:57
← bandroidx left17:57
→ bandroidx joined17:57
← Cromulent left17:58
grawity but of course apply common sense17:58
sheeny I did indeed! thank you17:58
i didnt realise you could just add to a repo17:58
makes perfect sense now17:58
grawity The per-repo keys are called "Deploy keys" in the website, and that describes their intended use17:59
→ wong1275 joined17:59
→ oso96_2000 joined17:59
sheeny thank you :)18:00
much appreciated18:00
← johnbenz_ left18:00
→ johnbenz_ joined18:01
→ acharlieh joined18:02
→ paulweb515_ joined18:03
gitinfo set mode: +v18:04
← sheeny left18:05
← mribeirodantas left18:05
← stephCoue left18:07
← McEnroe left18:08
→ undecim joined18:08
← h0bbit left18:09
→ guest96 joined18:10
→ ffio joined18:12
→ hyperair joined18:12
→ variable joined18:13
← uidzer0 left18:13
→ Bluec joined18:14
← Gonzo left18:14
→ AnomalusUndrdog joined18:14
AnomalusUndrdog anyone here proficient in git subtrees? what if I rename my subtree's prefix folder? will that corrupt my repo?18:15
← v0n left18:15
→ Gonzo joined18:15
→ scorphus joined18:17
← mechanicalduck left18:19
→ uidzer0 joined18:19
← evjsli left18:19
→ mfulz joined18:19
→ mechanicalduck joined18:20
← mfulz left18:21
← johnbenz_ left18:22
→ bddy joined18:22
→ maxmanders joined18:22
→ tatsuya_o joined18:23
← ffio left18:23
← umib0zu left18:23
→ sferry joined18:23
← sferry left18:23
→ sferry joined18:23
gkatsev is there a way undo a rebase? Or should I just create a new branch and cherry-pick the commit range?18:24
→ AWinterman joined18:24
grawity you can always `git reset --hard` to the last-good commit18:25
(which you can find in `git reflog` and/or `git reflog <branchname>`)18:25
old commits that aren't part of history still remain available locally for a few weeks.18:26
→ johnbenz_ joined18:26
gkatsev so, basically reset --hard to where I last branched/rebased and then re-rebase?18:27
→ Guest__ joined18:28
→ Ngt joined18:28
grawity you asked how to undo a rebase; the answer is `git reset --hard` to the commit that you want to return to18:29
← AWinterman left18:29
← mechanicalduck left18:30
← danicampa901 left18:30
→ ozette joined18:31
← maxmanders left18:31
→ kerrick joined18:32
→ kesha joined18:33
← mdhopkins left18:34
→ rabellamy joined18:34
→ umib0zu joined18:35
→ ErKa joined18:35
→ SebanK joined18:37
AnomalusUndrdog I can't seem to make git subtree pushes anymore. it says "Updates were rejected because the tip of your current branch is behind its remote counterpart. Merge the remote changes (e.g. 'git pull') before pushing again."18:37
I don't get that in the context of subtrees18:37
is it because I moved my subtree's path within my repo?18:38
→ daanavitch joined18:38
← cballenar left18:38
← guest96 left18:39
SamB that seems to have nothing to do with subtrees and everything to do with some entity having pushed something to the same branch18:39
→ Dreamer3 joined18:40
← melee left18:40
← troker left18:41
→ melee joined18:41
← kerrick left18:41
→ cballenar joined18:41
→ troker joined18:42
← troker left18:43
→ kerrick joined18:43
→ sq-one joined18:43
→ maxmanders joined18:43
→ mfulz joined18:44
← maxmanders left18:45
← Sigma left18:46
← LuxuryMode left18:47
→ alexander__b joined18:49
→ creese joined18:49
← alexander__b left18:50
→ alexander__b joined18:50
→ SHyx0rmZ joined18:51
→ evjsli joined18:51
creese I have a merge conflict I don't understand... "CONFLICT (rename/delete): some_file delete in master and renamed in HEAD. Version HEAD of some__file left in tree." What does this mean? The file moved?18:52
← PaulCapestany left18:52
→ v0n joined18:52
→ Sam_OTnosc joined18:52
→ CodeShark joined18:53
→ troker joined18:53
→ gusnan joined18:54
→ AWinterman joined18:54
→ PaulCapestany joined18:54
← kerrick left18:54
zz_michael_mbp → michael_mbp18:54
→ AWinterm_ joined18:57
← AWinterman left18:57
→ normanrichards joined18:58
← tatsuya_o left18:58
→ RonScagz joined18:58
cbreak creese: that means the branch deleted the file, you moved the file18:58
creese: and git doesn't know who's changes to use18:58
SebanK creese: My guess: It got deleted in master, but renamed/moved in HEAD (your current branch). Therefore it conflicts and you have to choose what to do with it18:59
→ tatsuya_o joined18:59
→ Sigma joined18:59
SebanK damn i'm slow18:59
→ LuxuryMode joined18:59
← evjsli left18:59
← sniffells left18:59
← troker left19:00
← matteo left19:00
← spaceonline left19:00
← SHyx0rmZ left19:00
→ eightyeight joined19:00
→ SHyx0rmZ joined19:01
→ chuckharmston joined19:01
→ Irish` joined19:01
← AWinterm_ left19:02
→ nbouscal joined19:02
→ russt joined19:03
→ spaceonline joined19:04
→ ldh joined19:04
← mishok13 left19:04
creese we resolved with a git checkout --ours <wildcard match>19:05
sound right?19:05
← IrishPrime left19:05
→ ffio joined19:05
← RonScagz left19:05
→ mattcen joined19:07
← TomyLobo left19:07
cbreak no19:09
sounds dumb :)19:09
why do you merge if you throw away everything?19:09
← mmlac-bv left19:10
← ozette left19:11
← Christi123 left19:11
→ Christi123 joined19:11
→ evjsli joined19:12
← aldodelgado left19:12
dwmw2_gone → dwmw219:12
michael_mbp → zz_michael_mbp19:13
creese the wildcard matches all the files in a certain directory19:13
→ kesha_ joined19:13
← Bluec left19:13
creese there were about 50 of them, we could add them one by one ...19:13
← kesha left19:13
← devoldmx left19:14
cbreak creese: use git add, not git checkout19:15
→ simesy joined19:15
→ CheckDavid joined19:15
→ Xiti joined19:16
ojacobson conflict resolution by throwing away one side of a merge entirely is pretty unusual19:16
← Christi123 left19:16
ojacobson if you really did want to throw away the incoming changes for that part of a merge, then yes, git checkout --ours will do that, though keep in mind that globs are expanded by your shell, not by git19:16
→ enr joined19:17
→ cenuij joined19:19
← cenuij left19:19
→ cenuij joined19:19
← voila left19:19
→ joebobjoe joined19:19
← simesy left19:20
← secondjeff left19:20
← tdm00 left19:22
→ pandeiro joined19:22
← ffio left19:22
→ kerrick joined19:22
← paul424 left19:23
← DolphinDream left19:23
→ DolphinDream joined19:23
→ petter joined19:23
← Guest__ left19:23
→ voila joined19:24
← maslowbeer left19:24
→ DolpheenDream joined19:24
← uidzer0 left19:24
← DolphinDream left19:25
← DolpheenDream left19:25
→ dijonyummy joined19:25
→ DolphinDream joined19:25
creese cbreak: I used git checkout, then git add19:25
cbreak forget about checkout19:26
Ngt do you usually put a drupal database dump when you push your on git? As a lot of information is stored in the Drupal database.19:26
your work*19:26
cbreak Ngt: no. Databases are not part of source code19:26
creese how would you do the equivalent of git add --ours?19:26
cbreak they are completely worthless for a program19:26
← ErKa left19:26
cbreak since it must be able to create such things on its own19:27
Ngt: see !config19:27
gitinfo Ngt: [!configfiles] It is recommended to store local configuration data in a file which is not tracked by git, but certain deployment scenarios(such as Heroku) may require otherwise. See https://gist.github.com/1423106 for some ideas19:27
Ngt cbreak: so you use module like Features to export your work?19:27
cbreak creese: you don't need --ours19:27
Ngt: what?19:27
→ irqq joined19:27
cbreak I commit my work of course19:27
→ mdhopkins joined19:28
← cballenar left19:28
cbreak Ngt: would you commit images in the repository of photoshop? Or mp3 in the repository for VLC?19:28
→ tdm00 joined19:28
→ konundra joined19:29
Ngt cbreak: when you use Drupal pretty much everything's is stored in your local database. There are no real source code like when you write your own PHP/HTML files to make a website19:29
→ private joined19:29
cbreak doesn't sound like you could store that in git19:29
← petter left19:29
→ Christi123 joined19:29
Ngt cbreak: so if you want to share your work with your co-workers then you have to make it into files to be able to share them19:30
cbreak git is not for sharing19:30
git is for history management19:30
→ mecool joined19:30
cbreak that only works with human edited files on text basis well19:30
→ MacGyver joined19:30
← MacGyver left19:30
→ MacGyver joined19:30
cbreak everything else is very hard to merge19:30
probably impossible19:30
and the diffs are worthless in most cases when computer generation is involved19:31
Ngt cbreak: yeah but it allows other people to work with you as well right?19:31
creese cbreak: so I should just add or rm the file that is removed19:31
→ AWinterman joined19:31
→ devoldmx joined19:31
cbreak add it if you want to keep the rename, rm it if you want to keep the delete19:32
creese thanks19:32
← Sigma left19:32
→ koda joined19:32
→ RonScagz joined19:34
→ rickruby joined19:35
← ryanjh left19:35
← creese left19:35
→ ryanjh joined19:36
→ secondjeff joined19:38
← daanavitch left19:38
← harish left19:38
→ harish joined19:38
← devoldmx left19:39
← rabellamy left19:39
→ gmac joined19:40
→ s00pcan joined19:40
← paolooo left19:41
→ mmlac-bv joined19:42
← rickruby left19:43
→ devoldmx joined19:43
→ whoami joined19:44
→ enr_ joined19:44
→ rabellamy joined19:45
→ stephCoue joined19:45
→ ddv joined19:46
← stephCoue left19:48
← dhruvasagar left19:49
← RonScagz left19:49
← enr_ left19:49
← axl_ left19:51
→ dustinm joined19:51
← rabellamy left19:52
→ axl_ joined19:52
voila I have many files in stage .. some are modified and some are new , some are deleted .. How can I stage them all together?19:52
→ mklappstuhl joined19:53
← Irish` left19:53
← Christi123 left19:53
→ uidzer0 joined19:54
→ Christi123 joined19:54
→ CatYak joined19:54
→ groglogic joined19:55
gkatsev can I cherry-pick a range of commits? i.e. would `git cherry-pick foo..bar` pick all the commits from foo to bar?19:56
← joebobjoe left19:57
← kerrick left19:58
→ alvaro_o joined19:58
← evjsli left19:58
← tplaner left19:59
→ evjsli joined20:00
← rpg left20:00
← tdm00 left20:01
→ ffio joined20:01
← mklappstuhl left20:01
← acharlieh left20:02
→ oojacoboo joined20:04
→ pantsman joined20:04
← a-l-e left20:04
← pandeiro left20:04
← johnbenz_ left20:04
← evjsli left20:04
← lavaman left20:04
← zmike123 left20:05
← breck7 left20:05
← normanrichards left20:05
→ normanrichards joined20:05
→ mklappstuhl joined20:07
→ lavaman joined20:07
→ troulouliou_dev joined20:07
→ chrob joined20:07
→ insejn_ joined20:08
troulouliou_dev hi i would like to maintain a fork of an active project ; make some changes on the fork while keeping it up to date with the upstream/master20:08
all i have to do is to merge/resolve conflict the upstream/master ?20:08
→ swair joined20:10
← LuxuryMode left20:10
insejn_ → insejn20:10
gkatsev yeah20:10
→ tplaner joined20:11
← Ngt left20:12
→ Ngt joined20:12
→ cjz joined20:13
← ffio left20:14
→ paul424 joined20:14
→ nj3ma joined20:15
← rfoust left20:15
→ simesy joined20:15
← Thra11 left20:16
← ryanjh left20:16
← jordanm left20:16
→ benbloom joined20:17
← CheckDavid left20:18
→ CheckDavid joined20:18
← mklappstuhl left20:19
Marquisk2 If you're using git checkout other-branch filename, does it remove it from the other branch or share the file between them?20:19
← simesy left20:20
→ mklappstuhl joined20:20
→ drasko joined20:21
ojacobson neither20:22
it stages the file, as it exists in other-branch, on the current branch, and writes it to the work tree20:22
troulouliou_dev gkatsev, ok20:22
ojacobson effectively copying it from the other branch to (a) the work tree (b) the staging area20:22
drasko hi all, I have pulled with --rebase and have conflits on files I removed. How to resolve this?20:22
← dc5ala left20:23
→ Etheco- joined20:24
→ epicowl joined20:24
← nj3ma left20:24
→ devoldmx3 joined20:24
epicowl hello, I am wondering how to get all remote branches from a repo Im working on into my local machine? the remote is set up but all I see is master branch and running git pull origin master (or git fetch, git fetch --all) doesn't show the remote branches...20:25
← oojacoboo left20:25
→ rpg joined20:25
tango_ epicowl: git branch -a will also show remote branches20:26
← sferry left20:26
tango_ to check them out, git checkout localname origin/remotename20:26
ehm git checkout -b20:26
git checkout -b localname origin/remotename20:26
← Christi123 left20:26
→ mklappst_ joined20:27
epicowl tango: ah thank you!20:27
← mklappstuhl left20:27
← devoldmx left20:28
→ oojacoboo joined20:28
← Etheco left20:28
→ SinnerNyx joined20:28
← ojacobson left20:29
← sq-one left20:30
→ p3lim joined20:32
→ breck7 joined20:32
← freeed left20:32
← swair left20:34
→ Jim_Lahey joined20:36
← gusnan left20:36
Jim_Lahey → Guest4935720:36
← Guest49357 left20:37
← geekbri left20:38
← mdhopkins left20:38
→ mechanicalduck joined20:39
→ mdhopkins joined20:39
→ acharlieh joined20:40
← CatYak left20:40
→ malkomalko joined20:40
→ rfoust joined20:41
← rpg left20:41
← zerorez left20:42
← irqq left20:44
→ Cassiel joined20:44
→ valgrind joined20:45
→ happyg0pher joined20:45
→ enr_ joined20:45
← TeckniX left20:45
wcwrel can i set a title or description for a branch so i can always see it when i do "git branch" ?20:46
→ maslowbeer joined20:46
→ eijk joined20:46
← helpD left20:46
← TheBrainWork left20:47
→ byprdct joined20:49
← byprdct left20:49
→ byprdct joined20:49
← enr_ left20:49
→ joelpittet joined20:50
← Davey left20:50
Guest66630 → Spyro20:51
← Spyro left20:51
→ Spyro joined20:51
→ psoo joined20:51
→ nathancahill joined20:51
→ cigarshark joined20:52
TMO git branch --edit-description ?20:53
→ gldn joined20:56
← p0wn3d left20:56
← spaceonline left20:56
→ sniffells joined20:57
→ kerrick joined20:57
← trungqtran left20:59
→ kcj joined20:59
→ SneakySnake joined20:59
← Ngt left21:00
→ trungqtran joined21:00
← zhost left21:00
← voila left21:01
← nicholasruunu left21:01
← api left21:01
→ blami joined21:01
→ jordanm joined21:01
← DolphinDream left21:01
← breakingthings left21:02
← schemanic left21:02
← kerrick left21:03
← ClamShell left21:04
→ goddard joined21:04
goddard hey if i have a file set in my .gitignore i can do a git add folder/* and it wont add that file if its in that folder correct?21:04
m0viefreak yes21:05
no wait21:05
← AaronMT left21:06
m0viefreak if you do "git add folder/*" , the * is expanded by your shell, so you are passing the ignored filename literally to git add21:07
in this case git will warn you21:07
if you do "git add folder", it will just add the not-ignored ones21:07
goddard m0viefreak: hmm interesting21:08
← kesha_ left21:09
→ devoldmx joined21:09
← devoldmx3 left21:09
← kaldrenon left21:11
→ shuaib joined21:12
→ lresende joined21:12
← konundra left21:14
← devoldmx left21:14
→ kit joined21:15
← happyg0pher left21:17
← jkremser|bbl left21:17
→ johnbenz_ joined21:19
← johnbenz_ left21:19
→ paulweb515_1 joined21:19
← paulweb515_ left21:20
→ devoldmx joined21:20
← howlymowly left21:20
← gldn left21:21
← ldh left21:21
← SneakySnake left21:22
← Anders_J left21:22
→ mishok13 joined21:23
← groglogic left21:23
← mechanicalduck left21:24
→ kerrick joined21:25
← scampbell left21:27
← donmichelangelo left21:30
← eugenejen left21:30
→ donmichelangelo joined21:31
← r04r left21:33
goddard m0viefreak: thanks21:33
m0viefreak: i like movies as well21:33
→ amjam joined21:34
← troulouliou_dev left21:35
← dougsland left21:37
← Dougie187 left21:37
← amjam left21:38
← macabre left21:39
→ fridim_ joined21:40
zachrab how can I check which branch I branched off of?21:41
→ nicedice joined21:42
← chuckharmston left21:43
→ ryanhirsch joined21:43
offby1 zachrab: run "gitk --all" and eyeball it21:43
← mike_af left21:43
cbreak you can run git merge-base between your branch and all others21:44
and just take the one with the closest base21:44
← tplaner left21:44
→ danicampa901 joined21:44
→ FernandoBasso joined21:44
offby1 heh21:45
cbreak: that's the job-interview-question answer :)21:45
← drasko left21:45
→ ClamShell joined21:45
← ClamShell left21:45
→ ClamShell joined21:45
→ enr_ joined21:45
← p1ng left21:47
→ setmeaway joined21:47
→ tplaner joined21:47
← enr_ left21:48
← valgrind left21:48
→ r04r|away joined21:49
← epicowl left21:50
← lresende left21:51
← bddy left21:51
← nicedice left21:52
→ nicedice joined21:52
→ lresende joined21:52
← mishok13 left21:52
← lresende left21:53
← mmitchell left21:55
→ endiruna joined21:58
← tplaner left21:59
→ lresende joined21:59
r04r|away → r04r22:00
→ simon_weber joined22:01
simon_weber I'm forced into using git to deploy an app binary -- I know it's a bad idea. I'd like to not keep revisions of the binary around. Anyone have suggestions?22:02
I'm thinking of something like filter-branch in a post-commit hook22:03
offby1 yow22:05
might22:05
might be easier to just keep the revisions around :-|22:05
← lavaman left22:05
← jerius left22:05
cbreak simon_weber: use a separate repository22:06
← shuaib left22:06
cbreak a submodule22:06
simon_weber cbreak: what advantages does that have?22:06
← lresende left22:07
← axl_ left22:07
cbreak the bloat is separate22:07
← kit left22:07
simon_weber hah, fair enough22:07
← natechan left22:07
simon_weber I suppose if it's a real pain I'll just manually prune it when/if things get ridiculous22:08
cbreak you know about git annex?22:08
simon_weber oh, yes; good idea22:08
← dijonyummy left22:09
osse wouldn't filter-branch in a hook make the repo worthless?22:09
← acharlieh left22:10
paulweb515_1 → paulweb515_22:10
← paulweb515_ left22:10
→ paulweb515_ joined22:10
simon_weber osse: I'd do something similar to this: http://stackoverflow.com/questions/930612/git-remove-oldest-revisions-of-a-file22:10
cbreak you could forget about history22:10
→ acharlieh joined22:10
cbreak and commit new orphan branches every time :)22:10
← normanrichards left22:10
simon_weber hah22:10
← Destos left22:11
simon_weber I don't suppose there's something like "track this file without saving revisions", is there?22:11
sonOfRa Who is forcing you anyways?22:11
← mfulz left22:11
sonOfRa And why is it not an option to do it otherwise?22:11
cbreak simon_weber: you could just commit --amend every time22:11
← Protux left22:11
← FrobinRobin left22:12
→ Samcio joined22:16
← Sam_OTnosc left22:16
← pantsman left22:16
← paul424 left22:18
← v0n left22:18
wcwrel I'm getting a fatal: Unable to create '/..../.git/index.lock': File exists. error, any ideas?22:20
→ mfulz joined22:20
← psoo left22:21
← trungqtran left22:21
← mfulz left22:21
→ mfulz joined22:21
→ Balliad joined22:21
→ thiago joined22:21
← devoldmx left22:22
← Samcio left22:23
→ dijonyummy joined22:23
→ zerorez joined22:23
→ normanrichards joined22:24
→ madewokherd joined22:24
→ trungqtran joined22:24
→ tplaner joined22:24
← tinti_ left22:24
← dijonyummy left22:25
← tplaner left22:25
offby1 From a script, what's the best way to see if a working tree has untracked files? "git ls-files --other"? "git status --porcelain"? Something else?22:26
← trungqtran left22:26
← kermit left22:26
← secondjeff left22:27
← rnsanchez left22:28
→ Rarrikins joined22:32
→ lavaman joined22:34
← private left22:34
→ noname001 joined22:35
← grilix left22:36
← whaley left22:36
→ devoldmx joined22:37
→ LuxuryMode joined22:39
← acharlieh left22:40
← mastro left22:41
← Balliad left22:42
← drizzd left22:43
← umib0zu left22:43
→ Wearos joined22:44
→ tjbiddle joined22:44
→ dijonyummy joined22:44
→ v0n joined22:45
← godgodgodgo left22:46
← dijonyummy left22:47
→ codemac joined22:48
→ mtk joined22:49
→ groglogic joined22:49
← hwrd|work left22:49
codemac Anyone know how to move a commit before an empty merge, and keep the merge there? like, commitA -> merge commit -> commitB to something like commitA -> commitB -> merge commit? git resolve man page warns me a bunch about doing git rebase -i and -p at the same time, so I'm probably thinking about this incorrectly.22:50
← saintcajetan left22:51
← nathancahill left22:51
→ hwrd|work joined22:52
→ rudak joined22:53
→ chuckharmston joined22:53
codemac The ending idea would be to squash commitA and commitB together.22:54
← eijk left22:54
→ zbcm joined22:54
← noname001 left22:56
→ jon_r joined22:57
← mfulz left22:57
← mckendricks left22:58
← chuckharmston left22:58
← _iron left22:59
← hwrd|work left22:59
→ hwrd|work joined22:59
← decotii left23:00
← anonymus_ left23:02
→ buu joined23:03
→ mmitchell joined23:03
← devoldmx left23:04
JakeSays if i have branches A and B, and i merge A in to B, can i still commit to A?23:04
→ dijonyummy joined23:04
← mdhopkins left23:05
→ alainus joined23:05
→ saintcajetan joined23:05
alainus is there a way in git to leave comments for other developpers in the code? (long shot here)23:05
→ kermit joined23:05
JakeSays alainus: not sure how git would help there - if you want comments, just leave comments23:06
← SHyx0rmZ left23:06
osse JakeSays: yes23:06
JakeSays osse: if that is the case, how would merging A to B be different than cherry picking all the commits in A to B?23:08
frogonwheels alainus: what kind of comments are you talking about?23:08
← kermit left23:08
alainus frogonwheels, for example: # Is this class the one that's required here?23:08
← mmitchell left23:08
alainus I realize that if I use that kind of comments, no one will ever see them23:08
so adding them to the code like that is not optimal23:08
I was hoping some kind of way that actually lets devs know about the comments23:09
osse JakeSays: cherry picking all commits in A to B is what rebase does in a way23:09
→ Liir joined23:09
osse JakeSays: a merge creates a seperate commit that has two parents: the tips of A and B23:09
→ mckendricks joined23:09
osse JakeSays: so the history will show that two lines of development combined into one23:09
→ R1KRDO joined23:09
← wong1275 left23:10
alainus The reason I'm asking this in #git23:10
→ kermit joined23:10
alainus is that it would be great if upon a merge, git could output these comments to the dev23:10
OMGOMG no23:10
figure it out before committing23:10
← jaequery left23:10
OMGOMG and be less likely to get punched in the dick by the other devs23:10
JakeSays osse: up until this very moment i thought i understood git :(23:10
← dustinm left23:11
JakeSays alainus: those sorts of issues should be raised in a design review23:13
osse JakeSays: what do you not understand?23:13
JakeSays osse: lets start with rebase23:13
osse JakeSays: http://sprunge.us/gGSd23:15
JakeSays osse: actually let me ask you this. if i cherry pick A in to B and get 10 conflicts, would the number of conflicts be different if i did a merge instead23:15
← cannonball left23:15
→ simesy joined23:16
osse JakeSays: I don't know. In principle they could be different but I think in most cases they wouldn't.23:16
JakeSays osse: yeah thats what i was thinking.23:16
← sgambino left23:17
JakeSays i have a dev who is complaining because he had 300 conflicts after doing a cherry pick. he hates cherry picking.23:17
→ simesy_ joined23:17
← simesy left23:17
← LuxuryMode left23:17
osse after doing *one* cherry pick ?23:17
JakeSays lol yes23:17
osse *one* commit ?23:17
JakeSays it was a squash merge of two branches23:18
osse then you should merge between the branches more often23:18
JakeSays indeed23:18
osse keep B up to date on what's going on in A23:18
JakeSays i've been preaching that23:18
osse but since it was a squash merge I assume he won't like that23:18
JakeSays well, it was a release branch. the idea was branch for release, test, merge back to production if needed, then cherry pick any changes back to master23:19
← enr left23:19
JakeSays the idea being you only do very minor tweaks if needed23:20
preaction don't make changes in the release branch. cherry-pick from master to release23:20
that's what we do, at least23:20
JakeSays but we've been on that release branch for six weeks now23:20
← R1KRDO left23:20
JakeSays preaction: that would be my preference23:20
osse I thought the custom was to tag releases :S23:21
preaction master should be what everyone develops on, cherry-picking commits to the release branch if necessary. otherwise you get problems23:21
JakeSays preaction: i totally agree.23:21
i cant seem to sell it though23:21
preaction sure, you tag individual releases, but sometimes you have to have a "stable" release branch23:21
gkatsev why cherry-pick as opposed to just merge?23:21
← koda left23:21
preaction master contains both new features and bugfixes. only bugfixes go to the release branch23:22
JakeSays gkatsev: to only get what you want to release23:22
osse wouldn't it be better to do the bugfix on the release branch and then cherry-pick it from there if it is still relevant?23:22
→ griotspeak joined23:22
preaction JakeSays: i would think it sells itself, considering this problem23:22
JakeSays preaction: yup23:23
gkatsev osse: well, that's for hotfixes only, no?23:23
osse or: what do you do if you have a bug that isn't present in the current code anymore, but only in the release branch?23:23
JakeSays osse: well, that would work too, if you have the right devs.23:23
osse gkatsev: yes23:23
preaction if you have a bug that isn't present in master, you don't need to cherry-pick it anywhere23:23
← threesome left23:23
← joelpittet left23:23
JakeSays osse: in that case i'd patch the release branch and not pick it to master23:23
osse preaction: but you will have to commit directly on the release branch then. That seems to go against what you're saying23:24
JakeSays: sure you need the right devs. There is no such thing as git-nanny :P23:24
preaction osse: i'm not writing laws in stone here23:24
JakeSays osse: yeah :( i'm to the point of suggesting we move to TFS.23:24
osse JakeSays: you could quit23:25
preaction i've had more problems with doing changes on the release branch and cherry-picking to master than i have doing them on master and cherry-picking to release. i realize that the plural of anecdote is not data23:25
osse I see23:25
JakeSays osse: that is an option, but i'd rather replace the other devs first23:25
osse It just seems reasonable to me that if you're doing something that is at all relevant to a release branch you should do it directly there.23:25
gkatsev well, usually when we're releasing we want whatever is on master, so we just merge. I guess it makes sense to cherry pick if we only want a specific item23:25
preaction honestly, i hate cherry-pick, and i hate the necessity for it. it does bad things23:26
→ grilix joined23:26
← stamina left23:26
preaction i've spent way too much time cleaning up after cherry-picks + merges23:26
osse I don't see how a cherry-picked commit could do any damage to a subsequent merge23:27
← griotspeak left23:27
← s00pcan left23:27
preaction when the cherry-pick conflicts, the subsequent merge will conflict. multiply by dozens of cherry-picks and create a messy messy repository23:28
which is why i had to make the guideline: always go from master to release, never the reverse23:29
JakeSays so the idea of a rebase is to bring a feature branch up to date with master? (or a child branch up to date with its parent)23:30
← rudak left23:30
→ cannonball joined23:30
← groglogic left23:32
osse preaction: if hadn't cherry-picked wouldn't you get a merge conflict anyway?23:32
JakeSays: yes23:32
JakeSays: you're basically moving it to the front, so to speak23:33
JakeSays so thats how you would keep your feature branch up to date then23:33
rebase daily, etc23:33
osse many people work that way, yep23:33
← lavaman left23:33
JakeSays that makes sense23:34
← normanrichards left23:36
← uidzer0 left23:37
← pitzips left23:38
← catcher left23:40
← saintcajetan left23:40
→ Kvothe joined23:42
→ troker joined23:42
← Kvothe left23:42
→ Cromulent joined23:42
→ Kvothe joined23:43
← SebanK left23:43
→ s00pcan joined23:44
→ relipse joined23:44
→ KernelPanicMK2 joined23:45
← Gonzo left23:46
→ VoidAtValhalla joined23:47
→ epicowl joined23:47
→ enr joined23:47
← chrob left23:50
→ spaceonline joined23:50
→ uidzer0 joined23:50
buu Help! I did a git pull, got a merge conflict, stupidly did git rm $file; and now I'd like the file back. git checkout $file; says "$file is not known to git"23:51
← Etheco- left23:51
← enr left23:52
buu Any suggestions?23:52
I haven't yet committed the merge23:52
→ emergion joined23:52
→ w0ng joined23:53
→ Etheco joined23:54
← cjz left23:54
offby1 buu: it might still be in the index.23:55
→ stliu joined23:55
offby1 there's some weird syntax for pulling the various bits of a conflicted merge out of the index. I don't remember what it is, but I'm sure the man pages do ...23:56
→ jalama joined23:56
← epicowl left23:57
← Liir left23:57
← BJfreeman left23:57
← hahuang65 left23:58
← Milossh left23:59
← zerorez left23:59

Logs Search ←Prev date Next date→ Channels Documentation