<< 09-04-2014 >>
00:28:51*q66 quit (Remote host closed the connection)
00:29:25*q66 joined #nimrod
00:44:26*Skrylar joined #nimrod
00:49:41*psquid joined #nimrod
00:50:41*Guest54139 quit (Ping timeout: 264 seconds)
01:00:40*q66 quit (Quit: Leaving)
01:13:11*DAddYE_ quit (Remote host closed the connection)
01:13:44*DAddYE joined #nimrod
01:17:54*DAddYE quit (Ping timeout: 240 seconds)
01:22:58*nande joined #nimrod
01:27:49reactormonkAraq, so the problem is that cstring is an array f chars?
01:42:22Varriountreactormonk: Have the repo's been purged yet?
01:48:44*xenagi quit (Quit: Leaving)
02:04:56reactormonkVarriount, I didn't do anything...
02:05:43*BitPuffin quit (Ping timeout: 240 seconds)
02:24:56*rta quit (Ping timeout: 265 seconds)
02:38:31Skrylarfor some reason i expected "endsWith" and "startsWith" to be in system.nim
02:53:27*brson quit (Quit: leaving)
02:53:44*brson joined #nimrod
02:59:57*rta joined #nimrod
03:06:31*Qualitative joined #nimrod
03:11:06Qualitativequestion, in aporia.nim source code there are these two lines of code:
03:11:14Qualitative win.oneInstSock.handleAccept =
03:11:14Qualitative proc (s: PAsyncSocket) =
03:11:14QualitativeI'm familiar with = coming after proc but is there an = after win.oneInstSock.handleAccept?
03:13:03runvncthe type of handleaccept is proc
03:13:21runvncwhen the socket is accepted that proc gets called
03:13:49runvncor something like that
03:14:42*rta quit (Ping timeout: 240 seconds)
03:16:11runvncwell, when the connection is accepted
03:17:48Qualitativerunvnc: thanks
03:20:43QualitativeWith the first var is tab automatically defined?
03:20:45Qualitative var tab = win.Tabs[current]
03:20:45Qualitative var endIter: TTextIter
03:25:04runvnctab is declared with var and the type is inferred based on the type of "Tabs"
03:25:11runvncnot sure what you mean
03:25:15runvncdid you read the nimrod manual
03:25:18runvncor tutorial
03:26:03runvnchttp://nimrod-lang.org/manual.html http://nimrod-lang.org/tut1.html
03:27:48Qualitativerunvnc: thanks again
03:28:01runvncsure hope it helps
03:29:14runvncI mean sure, hope it helps
03:29:17*rta joined #nimrod
03:29:24QualitativeThanks for the hard work on Nimrod. It's surely the best programming language even though I found it by accident.
03:29:38runvncI think so too but I can't take credit for any of the hard work
03:29:52runvncI think some of the main people may be sleeping since they are in europe or the uk or something
03:34:14*Qualitative quit (Quit: if you have a good idea keep it good and don't let it spoil. Thank you bbl.)
03:39:13*flaviu quit (Remote host closed the connection)
03:54:08*brson quit (Ping timeout: 246 seconds)
05:00:38*psquid quit (Ping timeout: 240 seconds)
05:01:38*psquid joined #nimrod
06:08:28*betawaffle quit (Excess Flood)
06:10:50*betawaffle joined #nimrod
06:12:31*BitPuffin joined #nimrod
06:56:48NimBotAraq/Nimrod devel c3d04c7 Araq [+0 ±4 -0]: refined and documented regionized pointers
07:31:23*nande quit (Read error: Connection reset by peer)
08:00:30*Varriount claps for Araq
08:06:22Varriountfowl: I *finally* got the allegro 5 library built. I'll test out your wrapper as soon as I wake (going to bed)
08:07:39*psquid_ joined #nimrod
08:09:11*psquid quit (Ping timeout: 250 seconds)
08:25:42BitPuffinVarriount: on windows?
08:47:05*easy_muffin joined #nimrod
08:51:54*eigenlicht quit (Ping timeout: 240 seconds)
09:27:25*psquid joined #nimrod
09:28:29*psquid_ quit (Ping timeout: 250 seconds)
09:32:07*ics quit (Ping timeout: 240 seconds)
09:35:57*ics joined #nimrod
09:50:54*renesac quit (Ping timeout: 240 seconds)
10:38:36*eigenlicht joined #nimrod
11:39:54*faassen joined #nimrod
11:50:41*easy_muffin quit ()
12:08:42*BitPuffin quit (Ping timeout: 240 seconds)
12:17:59*BitPuffin joined #nimrod
12:23:07*BitPuffin quit (Ping timeout: 250 seconds)
12:48:38*BitPuffin joined #nimrod
12:56:12*flaviu joined #nimrod
12:57:44*darkf quit (Quit: Leaving)
13:33:17*nande joined #nimrod
13:37:55*OrionPK joined #nimrod
13:40:30*flaviu quit (Remote host closed the connection)
13:52:11*vendethiel quit (Read error: Connection reset by peer)
13:52:32*vendethiel joined #nimrod
13:53:29*flaviu joined #nimrod
14:02:04*flaviu quit (Remote host closed the connection)
14:02:32*flaviu joined #nimrod
14:30:10*flaviu quit (Read error: Connection reset by peer)
14:48:28*flaviu joined #nimrod
15:39:08dom96hey
15:45:54BitPuffinoi dom96
15:46:03dom96oi m8
16:00:18*Demos joined #nimrod
16:02:14dom96upvotes appreciated: http://www.reddit.com/r/programming/comments/22jwcu/how_i_came_to_write_d/cgo4zur
16:13:30Demosyou got my upvote, not sure if walter is still reading the thread though
16:14:56dom96Demos: thanks
16:15:03dom96He'll likely ignore it anyway
16:15:56dom96I also mentioned Nimrod in other places
16:16:05dom96Upvote that too if you can
16:18:20DemosI like D. I would probably have gone to D after I got fed up with c/c++ bullshit if I had not known of nimrod
16:18:37dom96What makes you prefer Nimrod?
16:19:12*faassen left #nimrod (#nimrod)
16:20:32DemosI like the full on CTFE with slurp and gorge. I like how most compile options are also pragmas (it should be more than most), I like the implicit generic syntax, I like the departure from the OOP idea of having functions "in" your data, and I think araq has good taste
16:22:36dom96good list
16:23:44Demosand modules. We have them. They are a good idea
16:26:03dom96D doesn't?
16:28:13Demosno D has em
16:28:29Demospretty much only C and C++ lack them
16:32:01*mac01021_ joined #nimrod
16:34:11mac01021_forgive my ignorance: I would like to ask (out of curiosity) why the standard library wants to discourage the use of unsigned integers.
16:34:46dom96https://github.com/Araq/Nimrod/wiki/Unofficial-FAQ#why-are-unsigned-types-discouraged
16:35:23dom96There was talk of putting the unsigned ints into the system module IIRC
16:37:05Demosmac01021_: it is because unless you are dealing with the unsigneds as bits or you actually want mod2 behavior unsigned types are not a good idea
16:37:28*easy_muffin joined #nimrod
16:38:23Demosor maybe if you have like a huge amount of data (think framebuffers and meshes)
16:38:28mac01021_Demos: why are they not a good idea?
16:40:11Demosthey are a low level construct and it is easy to forget that they wrap at zero
16:40:46Demosthey are not satan himself, but a lot of programmers think that they should use unsigned when they have a value that is always positive, they should not
16:40:47flaviuBasically, think of them as a sequence of bits, not as a number
16:41:10Demosactually nimrod has a typeclass (Natural I think) that represents positive numbers
16:41:19flaviuYes, natural
16:41:49Demosdoes the constraint get checked at runtime? How does one make a natural of a specific type?
16:42:58flaviuIIRC, you have to do --checks:on
16:43:41mac01021_ok. Thanks guys.
16:43:52Demossomewhat interesting that typeclasses can be compile time or runtime properties, almost dependent
16:44:44flaviumac01021_: The stdlib defines natural as `type Natural* = range[0..high(int)]`
16:45:14Demoswowah that is no typeclass
16:45:46Demoswould it not be better to have somethink like type Natural = generic x x >= 0 ?
16:45:57Demosalthough I guess that may not actually work
16:46:42mac01021_So basically the reason is that, when a programmer error causes some value to be lower than the acceptable range, the presence of a negative number there makes it easier to diagnose the problem than would the presence of a positive overflow result?
16:48:12Demosyes
16:48:21Demosusually the programmer "expects" a negitive
16:48:36Demoscomparisons all go "the correct" way
16:48:36EXetoCflaviu: yep. it's on by default though
16:49:11*DAddYE joined #nimrod
16:49:57mac01021_Demos: that makes sense. Thanks again.
16:51:10EXetoCDemos: those programmers should use unsigned if the positive range of signed is too small, but that can't be common
16:51:35Demosyeah it is not common, and usually you can just move to a bigger signed type
16:51:51Demosso it does come up when you have a whole lot of data, but not much otherwise
17:29:58*Matthias247 joined #nimrod
17:37:24*taapa joined #nimrod
17:39:52*[1]Endy joined #nimrod
17:42:30EXetoCwill the command syntax support specifying parameters by name?
17:48:40flaviuEXetoC: What new command syntax?
17:49:40EXetoCflaviu: the alternative way of calling a proc. it's not new
17:50:40flaviuOk, I read too fast
17:54:55*Demos quit (Read error: Connection reset by peer)
17:59:01VarriountBitPuffin: Yes
18:07:47*BitPuffin quit (Ping timeout: 246 seconds)
18:09:53*BitPuffin joined #nimrod
18:10:48*q66 joined #nimrod
18:15:38Araqhi taapa welcome
18:17:54*BitPuffin quit (Ping timeout: 240 seconds)
18:18:18*BitPuffin joined #nimrod
18:24:20*q66 quit (Ping timeout: 252 seconds)
18:24:24AraqVarriount: so ... purge?
18:26:00*q66 joined #nimrod
18:26:00*q66 quit (Changing host)
18:26:00*q66 joined #nimrod
18:28:25*BitPuffin quit (Ping timeout: 250 seconds)
18:35:16*Matthias247 quit (Quit: Matthias247)
18:35:20*easy_muffin quit ()
18:35:33*Matthias247 joined #nimrod
18:37:49*renesac joined #nimrod
18:37:49taapaAraq: thks
18:39:23*Mat3 joined #nimrod
18:39:31Mat3good afternoon
18:41:57*BitPuffin joined #nimrod
18:42:11*BitPuffin quit (Client Quit)
18:42:19EXetoClo
18:42:28*BitPuffin joined #nimrod
18:42:31Mat3hi EXetoC
18:43:07Mat3what's the status of the new, planned version ?
18:45:43*Mat3 quit (Quit: Verlassend)
18:48:40VarriountAraq: I'm ready when you are.
18:48:59*nande quit (Read error: Connection reset by peer)
18:54:42*BitPuffin quit (Ping timeout: 240 seconds)
18:56:56taapatrying nimrod using brew on osx. Any plan to update a version with symlink?
18:57:36taapaCurrently “The compiler will currently fail to find system.nim if called through a
18:57:37taapasymlink. To compile nim files, specify the full path to the compiler”
19:03:12Araqtaapa: no idea whether we fixed that in the devel version. I remember doing *something* about it. :P
19:04:45AraqVarriount: dom96 wants a "Plan". Do you happen to have a gist or something?
19:07:18*brson joined #nimrod
19:09:05VarriountPlan? Of What?
19:09:14Araqthe purge
19:09:16EXetoCworld domination
19:10:19VarriountPurge the repo -> Post thread on forum concerning the purge -> Send emails and notifications to forkers concerning the purge -> ???
19:10:41flaviuVarriount: You forgot make a backup
19:10:54flaviualthough each fork is sort of like a backup
19:13:33*BitPuffin joined #nimrod
19:16:43VarriountAraq, dom96: https://gist.github.com/Varriount/10304645
19:17:26AraqVarriount: I'm waitin for dom96 to give his OK.
19:17:49Araqhe's worried so let's see if he has some idea why
19:28:39*BitPuffin quit (Ping timeout: 250 seconds)
19:30:36Varriountdom96: Hurry up.
19:35:07*Mat3 joined #nimrod
19:38:03dom96Varriount: I would like to see how you're going to purge the repo, i.e. what commands you'll be executing.
19:39:33VarriountI thought reactormonk was doing that.
19:39:56VarriountI /never/ said I was going to purge the repo - only send out the messages.
19:41:20dom96Ahh ok.
19:41:30dom96In that case your plan looks good.
19:42:50Varriountdom96: You'd only find me purging the repo if you gave me three weeks notice, 50 dollars, and time to sign up for a life insurance policy.
19:43:18VarriountI'd want to leave something behind in case git ate me, or something.
19:43:43dom96JS backend why you flooring my float?!
19:45:04Araqoh yeah dom96 should push his JS fixes before we purge
19:45:08Araqbrb
19:45:44Skrylari don't have any compiler patches to rebase :B
19:45:54*Skrylar is one of those derps who's only writing libraries
19:47:39*Mat3 wonders what the heck is going on here
19:48:01flaviuThe git repo is too big, so its being trimmed down
19:48:11VarriountMat3: Araq and dom96 are putting off purging parts of the main Nimrod repositories history.
19:48:37flaviuWhich parts?
19:48:57VarriountParts of the history, as tracked by git, are really large (size-wise). I think it's the parts with the DLL's and csources
19:52:21flaviuThat sounds straightforward. `git filter-branch -f --tree-filter 'rm -rf ./csources/ ./anythingelse/' --prune-empty -- --all`
19:53:32Varriountreactormonk: ping
19:55:07*rta quit (Ping timeout: 240 seconds)
19:55:45*Demos joined #nimrod
19:56:28dom96Yeah, it should be straightforward.
19:56:56dom96But I worry that we will run into many problems.
19:57:54flaviudom96: We can test localy, so if there are problems, they can just not be pushed
19:58:26*io2 joined #nimrod
19:58:53Mat3try it out part for part ?
19:59:26dom96True. But what if the problems start occurring when we push and others start attempting to pull the changes?
19:59:35dom96I guess everyone will just have to reclone?
19:59:47flaviuA reclone isn't going to be optional anyway
19:59:56flaviurebasing changes the hashes
20:00:05dom96good.
20:00:09Araqhi Mat3 as you can see, we're purging the repo, devel will be merged into master and then we will have a relaxed release
20:05:19flaviuThe purge is going to take a while
20:06:21Demosusing filter-branch?
20:06:25flaviuYes
20:06:30fowlAraq, how can i run a nakefile with the vm
20:08:34Skrylarhm.. i might have to fiddle with my event module
20:08:45Skrylari saw 'on' is now an alias to 'true'; i had On(...) as call syntax
20:09:13flaviuWell, not that bad I guess. About 8 commits per second, it should take ~16 miniutes
20:09:15AraqSkrylar: no, 'on' has been an alias for years now
20:09:32Skrylari'm surprised the compiler didn't get confused then oO
20:09:43Araqwell who DOES the purge?
20:09:50Araqno me, I'm a git noob
20:09:56*rta joined #nimrod
20:10:18Skrylargit filter-branch does the delete, though i seem to remember github gets mad if you try to wipe out master and replace it
20:11:07dom96we don't need to wipe out the master do we?
20:11:24flaviudom96: Not doing so would defeat the point
20:11:39dom96how so?
20:11:54dom96We are going to be merging devel into master anyway
20:12:01flaviuGit stores files in a giant hashmap, if anything refrences them they will be kept
20:12:02dom96We can remove devel instead.
20:12:21Araqfowl: the VM still is not ready for nake, sorry
20:13:17DemosAraq: the VM is getting FFI though right?
20:13:37flaviudom96: Actually, it looks like you can do things the way you want with --all
20:13:39fowlAraq, can it be used for scripting
20:14:05Araqnot anytime soon, I'm now against -d:useFFI
20:14:42VarriountWhy?
20:14:54AraqFFI is a just a never ending story where constantly things *almost but not quite* work
20:15:05Araqand we have enough of that in nimrod already
20:15:17Demosleave it for after 1.0
20:15:19flaviuAraq: You should see Scala
20:15:49flaviu'Feature X, but not if combined with features A or B'
20:16:02Araqfowl: I designed a nice way to provide extensions for the VM though which should be ready soon
20:16:22Araqthe VM will support lots of stuff from OS and math and then that's it
20:16:27Demosalthough FFI in the VM would be cool. Would have to make sure implicit static eval did not go and call FFI stuff
20:16:43Araqif you need more compile time voodoo, use gorge instead
20:16:55Demosyeah. I suppose
20:17:01AraqDemos: that's easy to do, we already have permission flags
20:17:10VarriountWe do?
20:17:16Araqsure thing
20:17:34AraqallowEndlessLoops, allowFFI, allowCast iirc
20:18:06Araqthe compiler disables any of these for its compiletime evaluation
20:18:43VarriountNeeds documentation...
20:19:36*BitPuffin joined #nimrod
20:20:16flaviuAraq: I think I have a command that works
20:20:27flaviusudo git filter-branch -f --tag-name-filter cat --index-filter 'git rm --cached --ignore-unmatch -rf ./build/' --prune-empty -- --all
20:21:32flaviusorry, get rid of the sudo
20:21:46flaviuI was testing on a ramdisk
20:22:05Araqwell I suppose I have to do it
20:22:25Araqnobody else has what it takes, apparently
20:23:03Araqbtw nobody commented on the new regionized pointers...
20:23:11VarriountI did!
20:23:36flaviuI would, but I have no idea what they do
20:23:50Varriount[04:00;] * Varriount claps for Araq
20:24:08NimBotnimrod-code/packages master addfa81 Dominik Picheta [+0 ±1 -0]: Get rid of trailing comma.
20:24:10Demosis there a forum post>
20:24:25fowlAraq, i thought reactormonk was taking care of the purge
20:24:43Araqfowl: yeah indeed but he is away
20:25:42AraqDemos: did you report that regression that keeps devel from building your stuff?
20:25:42fowldom96, why u hate on the trailing comma
20:25:53AraqI guess I need to fix that before release
20:25:56dom96fowl: I don't. The JSON spec does.
20:26:01DemosAraq: what regression?
20:26:07DemosI have been building on devel for a while
20:26:17AraqDemos: you tell me. It was template related.
20:26:25Araqmaybe it was somebody else?
20:27:13DemosI have been having continueing issues with static[T], I also had a problem with a the scope of a proc call from a template. But I could not construct a minimal case
20:28:37fowldom96, thats valid json
20:29:26Varriountfowl: No..I've had many errors with JSON configurations concerning trailing commas.
20:30:20fowlthose libraries are in error
20:32:59*BitPuffin quit (Quit: WeeChat 0.4.3)
20:33:14*BitPuffin joined #nimrod
20:33:15dom96http://stackoverflow.com/a/201856/492186
20:35:26*dom96 doesn't feel like reading the spec to verify this
20:35:40dom96But in any case, JSON.parse in Firefox fails with a trailing comma.
20:37:53reactormonkAraq, gotta say, I like the regionized pointers.
20:38:22Demoswhere is the info on regionized pointers
20:38:32fowldom96, o
20:38:44fowlin that case i hate json
20:38:54reactormonkVarriount, commands: get http://rtyley.github.io/bfg-repo-cleaner/ , java -jar bfg.jar --delete-folders build path/to/nimrod/repo
20:39:09reactormonkDemos, one of the recent commits
20:39:19reactormonkVarriount, rebasing the PRs is a bit more complicated :-/
20:39:25Araqah reactormonk is here :-)
20:39:38reactormonkAraq, aaaand he's gone. Gotta do some homework :-(
20:39:49Araqnoooo!
20:39:58Araqwhen will you do the purge then?
20:43:36flaviuAraq: The command I gave you will delete these files: https://gist.github.com/10312838
20:43:40DemosI like the regionized pointers quite a bit, presumably they are not going into 0.9.4 though
20:44:03fowlDemos, link?
20:44:07Araqthey are in 0.9.4 but I only have 2 tests for them
20:44:15Demoshttp://build.nimrod-lang.org/docs/manual.html#memory-regions
20:44:29AraqI'd like to patch the python and lua wrappers to make use of them
20:45:38Araqgah typo in the docs
20:45:46fowlhm
20:45:57Araqit has to be ptr[A, T] <: ptr[B, T] obviously
20:46:35Demoshow does it work with the pointer type?
20:47:58Araqwhat do you mean?
20:48:09fowli was just thinking about a temporary annotation for ptr
20:48:46Mat3ciao
20:48:48fowli could use regions for that
20:48:49*Mat3 quit (Quit: Verlassend)
20:48:52fowlcya mat3
20:50:41AraqI wonder if these things really help type safety in a kernel or whether it's just an academic idea of mine
20:53:30*BitPuffin quit (Ping timeout: 240 seconds)
20:54:29reactormonkAraq, the commands above do the purge for every branch. Do you want to be able to merge port PRs too?
20:55:03Araq"to merge port"?
20:56:09DemosAraq: well kernels are not going to be written in nimrod anyway :D also why do we have both #! and {.someOption: on.} for module level options? Is it just controlling parseing vs controlling codegen?
20:56:20NimBotAraq/Nimrod devel 8e697ee flaviut [+0 ±1 -0]: Inline docs for sequence appending
20:56:20NimBotAraq/Nimrod devel aa3f165 flaviut [+0 ±1 -0]: Code docs for `contains`, `in`, `is`, and `of`
20:56:20NimBotAraq/Nimrod devel 97b5380 flaviut [+0 ±1 -0]: Code examples for string concatination
20:56:20NimBotAraq/Nimrod devel 4edcab6 flaviut [+0 ±1 -0]: Document inplace string appending
20:56:20NimBot4 more commits.
20:57:05AraqDemos: yeah; {.strongSpaces: on.} is technically harder to do
20:57:06flaviuDemos: Are you sure? https://github.com/dom96/nimkernel
20:57:44fowlAraq, should i wait for the purge to do a PR making add/len nil safe?
20:57:55DemosAraq: because you have to parse the pragma to figure out how to parse the rest of the source?
20:57:56Araqfowl: hmm yeah
20:58:23fowlAraq, also, what to do about safeLen? its only used in the compiler afaict
20:58:26Demosfowl: how is that going to work?
20:58:28AraqDemos: yup.
20:58:33Araqfowl: deprecate it
20:59:19Araqreactormonk: no PRs are left to bother with
20:59:35fowlDemos, for string/seq add/len will create a new thing if the thing is nil
21:00:17fowlDemos, the old version will be addx, lenx if you want to skip "if obj.isnil: newseq obj"
21:01:19Araqfowl: I thought it's xadd and xlen
21:01:29fowlok
21:01:41fowli was going to ask lol
21:04:31*OrionPK quit (Remote host closed the connection)
21:07:22runvncHello so you guys did the purge already? I copied the file I changed to a backup just in case, and was assuming I would just need to reclone
21:08:41dom96not yet
21:09:45reactormonkAraq, purging will break all and every repo. Recommended action: git cherry-pick <your commits>. an easy way to specify <your commits> is old_devel..your_branch_head
21:10:57Araqso what are you telling me?
21:11:18Araq"will break all and every repo". well yes, we know that... don't we?
21:12:03NimBotAraq/Nimrod purged_devel aab0885 Simon Hafner [+1 ±1 -0]: Added `$` for seq
21:12:03NimBotAraq/Nimrod purged_devel bece17e Simon Hafner [+0 ±2 -0]: explicit firstElement for `$` in collections
21:12:03NimBotAraq/Nimrod purged_devel 57de0db Simon Hafner [+0 ±1 -0]: removed `$` for openarray, was commented out
21:12:03NimBotAraq/Nimrod purged_devel f443a07 Simon Hafner [+0 ±1 -0]: use `.f` instead of `.16e` for floats... 3 more lines
21:12:03NimBot27 more commits.
21:12:24reactormonkhmmm
21:12:32reactormonknot sure if correct.
21:12:38reactormonklet me try again.
21:15:40dom96hey, what about my JS changes?
21:16:04dom96I don't think you should be doing the purge in a separate branch
21:16:39fowlthis is too stressful
21:16:43fowllet me know when the purge is done
21:16:44*fowl left #nimrod ("Derogatory terms for gynecomastia can include moobs (for male boobs) and bitch tits.[34]")
21:16:55flaviugive me a few seconds to finish cloneing
21:17:54*[1]Endy quit (Ping timeout: 240 seconds)
21:22:13flaviuok
21:25:39reactormonkdom96, the new branch is the purged one.
21:26:17reactormonkAraq, yup, there we go. purged_devel is the devel without the build/ history.
21:26:33reactormonkdom96, got your JS changes in a commit?
21:26:48reactormonkgit pull origin purged_devel && git cherry-pick <your JS commit>
21:29:28dom96not yet
21:29:40dom96As long as they aren't committed it should be fine
21:32:51VarriountHas the purge been applied?
21:35:59flaviuNot on master and devel, no
21:36:13Araqreactormonk: go ahead and do the real thing. enough of this separate branch testing
21:36:41VarriountPush the big red button! Doo iiitt!
21:39:11VarriountAlso, if possible, tell me the new size of the repository.
21:40:27*nande joined #nimrod
21:43:15flaviuVarriount: I'm doing the same thing on my own for fun, and I'm getting 27MB
21:43:26reactormonkgot 33MB here
21:43:35reactormonkAraq, ok, big red button inc.
21:43:48reactormonkdone.
21:43:53flaviuI was a bit more agressive, I cut everything below 1M
21:44:03reactormonkflaviu, I only cut build/*
21:45:10reactormonkjust saying - this did not yet cut the commits from master.
21:45:18reactormonkshould I bring the big sledgehammer there as well?
21:46:13reactormonksledgehammer inc. give me a bit.
21:48:01Araqdo it
21:52:57reactormonkAraq, I hope you have the old repo somewhere safe.
21:53:16Araqon 3 computers
21:53:21reactormonkI do too, but maybe my SSD fries this moment.
21:53:43NimBotAraq/Nimrod asyncmacro 411683c Dominik Picheta [+2 ±1 -0]: First implementation of the async/await-like system with macros.
21:53:43NimBotAraq/Nimrod asyncmacro 91648ee Dominik Picheta [+0 ±1 -0]: Merge branch 'master' into asyncmacro
21:53:43NimBotAraq/Nimrod asyncmacro 0ab77a6 Dominik Picheta [+0 ±2 -0]: Better param passing. Implemented send, readLine and support for... 1 more lines
21:53:43NimBotAraq/Nimrod asyncmacro 608272c Dominik Picheta [+1 ±2 -0]: Calling an async proc with invalid arguments will now fail at compile-time.
21:53:43NimBot14 more commits.
21:53:43NimBotAraq/Nimrod master 3407eed Reimer Behrends [+0 ±1 -0]: Fixes various issues with shallow copying.... 11 more lines
21:53:43NimBotAraq/Nimrod master f27d8ba Andreas Rumpf [+0 ±1 -0]: Merge pull request #691 from rbehrends/fix-shallow-pragma... 2 more lines
21:53:43NimBotAraq/Nimrod standalone 06e2f61 Simon Hafner [+0 ±2 -0]: added InSystemModule for hti.nim
21:53:43NimBotAraq/Nimrod standalone 30e5d24 Simon Hafner [+0 ±1 -0]: PSTR for arduino literals
21:53:43NimBotAraq/Nimrod standalone 8616c32 Araq [+0 ±1 -0]: bugfix: pragmas allowed for implicit generics
21:53:43NimBotAraq/Nimrod standalone 4f4f66d Simon Hafner [+24 ±94 -1]: Merge remote-tracking branch 'upstream/master' into standalone
21:53:43NimBot1 more commits.
21:54:48dom96Yep, you just killed NimBuild.
21:55:06reactormonkdom96, figured.
21:55:24Araqreactormonk: you should really join #nimbuild
21:55:54reactormonkAraq, kk
21:55:56reactormonkmuh
21:56:41dom96But actually, it's good.
21:56:41reactormonkAraq, github has its own references for the pull requests and I can't really delete them :-/
21:56:50dom96Otherwise this channel would have been spammed to hell.
21:57:01reactormonkbut the repo size is now 50MB.
21:57:16reactormonkI assume that's because master and devel now may have split histories.
21:57:52dom96reactormonk: Delete all the branches except devel and master
21:57:56dom96and merge devel into master
21:58:07dom96Try purging more perhaps after.
21:58:13reactormonkdom96, let's see.
21:58:15dom9650MB is still kind of a lot
21:59:17reactormonkdom96, hmm, apparently nope. everything's still intact.
21:59:19reactormonkdom96, ok
21:59:55reactormonkdom96, a few merge conflits, compiler/ropes.nim and compiler/ccgexprs.nim
22:00:22reactormonkAraq, ^ could you check out the new repo and merge those two? I don't really trust my ability to judge in those parts of the code.
22:01:52Araqok let's see
22:02:44*fowl joined #nimrod
22:05:04*Matthias247_ joined #nimrod
22:07:04*BitPuffin joined #nimrod
22:07:30*Matthias247 quit (Ping timeout: 240 seconds)
22:10:09VarriountShould I send out the notifications?
22:10:59reactormonkVarriount, yup
22:11:06reactormonkVarriount, we're currently messing with the repo.
22:11:10VarriountAnd the repo size?
22:13:12reactormonkdown to 50 atm
22:13:16VarriountOk, 35 emails sent. Now for the issues.
22:13:20reactormonktrying to half it again.
22:19:42*Styled joined #nimrod
22:20:44StyledWhen I followed installalation instructions sh build.sh I received this error.. how to remedy?
22:20:46Styledbuild/2_2/semfold.o: In function `levalueconv_214015':
22:20:46Styledsemfold.c:(.text+0x1273): undefined reference to `lrint'
22:21:21dom96Try the latest from github.
22:21:22reactormonkStyled, let's see
22:21:27reactormonkoh right.
22:21:52Varriountdom96: That might be a bad idea, considering that reactormonk and Araq are in the middle of rewriting history.
22:23:34VarriountStyled: What OS are you on?
22:23:45reactormonkVarriount, nah, build sources are different from the main repo
22:23:46Styledubuntu
22:23:50fowlthis is the 0.9.2 bug for ubuntu
22:23:50reactormonkVarriount, guess what we just stripped
22:24:06*darkf joined #nimrod
22:24:12Varriount25 mb?
22:24:32AraqI got 33 MB for a clone
22:25:00Araqer ... no
22:25:02fowlStyled, this bug is fixed in the latest changes, you should wait until they are done working on the repository to get it (or wait for 0.9.4 release which will be real soon)
22:25:42AraqI only cloned the 'deel' branch
22:25:46Araq*devel
22:25:54Araqreactormonk: what's up with that?
22:25:57dom96Well if Araq can clone it then the repo is in a useable state
22:26:06Stylednot sure if it is a good idea to have a bug like this in the main zip download though. It could turn people away.
22:26:33fowlStyled, release is long overdue, yep
22:26:41reactormonkAraq, that sould be fine
22:27:00reactormonkStyled, release is actually due after we finished messing with the git repo, aka this week
22:27:11AraqStyled: lrint is part of Ansi C; strictly speaking it is not our bug, but ubuntu's
22:28:00reactormonkAraq, default branch is devel, so it will only clone that one IIRC
22:28:21reactormonkgit checkout -b master origin/master
22:28:39Styledubuntu is not nice
22:29:31Araqwith the recent OpenSSL bug everybody knows open source C development simply doesn't work anyway. Took people a few decades to notice.
22:30:05dom96reactormonk: I wonder if the fact that your local version is 50MB is because you still have some old cached files in there. I think you can run git gc or something to further get rid of useless files.
22:30:15reactormonkdom96, nope, that's a fresh clone
22:30:30dom96ahh
22:30:39reactormonkmight be that I clone everything over here
22:30:48Araqwell I now have 45MB too
22:30:49renesac[17:59:35] <fowl> Demos, for string/seq add/len will create a new thing if the thing is nil <-- why len would call newseq?
22:31:04reactormonkAraq, which is about 30MB over the wire
22:31:08fowlrenesac, sorry, len will just return 0 if the thing is nil
22:31:09reactormonkI think that's good enough for now.
22:31:14StyledAdding a note to for Ubuntu users, Irint won't be recognized please download latest from Github maybe needs to be added on linux download section?
22:31:17reactormonkunless you want to be a bit more aggressive
22:31:19renesacfowl, ah, ok
22:31:45reactormonkStyled, nah, we will update the csources in short notice
22:31:47reactormonk*on
22:31:51renesacreactormonk, 'for now' you mean forever, right?
22:32:02renesacI don't think we will do another purge
22:32:16reactormonkrenesac, I don't think so
22:32:23fowlheh
22:32:26reactormonkapparently there's still some bigger parts in there
22:32:31fowljust wait until stdlib becomes a separate repository
22:32:44Araqno way, 45MB is good enough
22:32:49Araqwe should never do it again
22:32:58dom96Come on
22:33:03Styledbtw what linux distro works best with nimrod?
22:33:10dom96Keep trying.
22:33:12renesacif there is things that should be trimmed, it should be now
22:33:17dom96We're doing this once, may as well do it right.
22:33:24renesacStyled, ubuntu works fine
22:33:26Araqok ok, do what dom96 says
22:33:26*Matthias247_ quit (Read error: Connection reset by peer)
22:33:33renesacI'm using it
22:33:42renesacbut not with the 0.9.2 zipped relase
22:33:45Styleddoes debian have this Irint issue?
22:34:09AraqI built it with Linux Mint Debian edition, so I don't think so
22:34:31renesacStyled, the problem is really the ancient 0.9.2 release
22:35:11renesaceverybody is using the latest devel now, that will turn in 0.9.4 this week (sooner than later)
22:35:34Styledlooks good.
22:35:46reactormonkAraq, let's see if I can find something else that's big too
22:35:54renesacjust wait a bit while they fiddle with the git history
22:36:23reactormonkAraq, some old versions where rod/nimrod was checked in too
22:36:24VarriountOk, I sent the emails, and I sent the issue notifications (and will close the issues in about a weeks time.)
22:36:47dom96reactormonk: Have you tried this? http://stackoverflow.com/a/20609719/492186
22:37:31reactormonkdom96, yup
22:37:35reactormonkon it
22:37:38dom96reactormonk: what's the output?
22:37:45reactormonka bunch of hashes
22:37:57reactormonkah, that's better.
22:38:05dom96can you gist it?
22:38:30reactormonkdom96, wait, I'll rerun it.
22:38:46reactormonkanyway, if you also bfg away files called `nimrod', the repo size shrinks to 34MB
22:39:13reactormonkdom96, nah, that one mostly bugs out.
22:39:13fowlwhats that, the binary?
22:39:34reactormonkfowl, yup
22:39:51reactormonkor actually I could just throw bin/
22:39:56*OrionPK joined #nimrod
22:40:24*Araq prays it still bootstraps
22:40:57*Araq is bootstrapping
22:41:20reactormonkAraq, got another iteration where I kicked bin/ from the repo history.
22:41:23Araqdie Spannung steigert sich ins Unerträgliche...
22:41:41dom96Cool. The graph no longer shows ~6 million additions/deletions heh https://github.com/Araq/Nimrod/graphs/contributors?from=2008-06-22&to=2014-04-06&type=a
22:41:53reactormonk... reduces total repo size to 34MB on disk
22:42:10Araqer ... does that mean I have to clone again, reactormonk ?
22:42:18flaviureactormonk: Have you gotten rid of `rod/nimrod`?
22:42:54reactormonkAraq, yup :-/
22:42:57reactormonkflaviu, nope
22:42:59AraqVarriount: now where is your forum announcement
22:43:02flaviuand `lib/oldwrappers/*`
22:43:08Araqreactormonk: ok, never mind
22:43:25Araqyeah remove lib/oldwrappers too then
22:43:44reactormonkI can only work on directory name, so I kicked oldwrappers...
22:43:45flaviuYeah, the rod files are a few 3-4mBs
22:44:16Araqgood news. bootstrapping still works.
22:44:38reactormonkflaviu, doesn't help too much, but I'll still kick them.
22:44:54renesacgithub don't inform the repository size anywhere right?
22:45:10flaviureactormonk: Also get rid of tests/tester
22:45:22reactormonkflaviu, kk
22:45:24dom96I hope we don't lose something important in the history
22:45:56reactormonkdom96, the sources should still be there
22:45:56flaviudom96: Do you have backups?
22:46:01reactormonkflaviu, Araq got a few
22:46:13dom96flaviu: Yeah.
22:46:18renesacI will make my own here too
22:46:18reactormonkkk, down to 17M
22:46:21dom96Araq: Put them somwhere safe.
22:46:30renesacwow
22:46:34reactormonkI deleted the full directory `oldwrappers'
22:46:38dom96reactormonk: well done :)
22:46:41reactormonkpurged from the annuals of time
22:46:54flaviutry making a new repo for the backups
22:47:02reactormonkAraq, can I push again? Will overwrite everything yet again.
22:47:10fowli dont see lib/oldwrappers or rod/ o.o
22:47:19dom96I wonder what those 100k addition spikes are in the contribution graph
22:47:20reactormonkfowl, it's in the histroy
22:47:22flaviufowl: They are in the history
22:47:24Araqsure but I think you can remove more
22:47:28fowloh
22:47:32reactormonkAraq, tell me
22:47:42reactormonkI mean we're down to 17MB
22:47:47dom96reactormonk: MORE
22:47:55renesaclol
22:47:58Araqthere used to be 'nim' and 'rod', 'nim' containing pascal sources
22:48:12Araqnot sure if that ever made it into git though
22:48:16dom96I think we should leave that
22:48:28dom96For its historic value
22:48:30Araqit's more than 1 MB iirc
22:48:31reactormonkAraq, I want to keep sources.
22:48:41Araqit's old pascal stuff
22:48:47reactormonkwe're only purging generated / binary
22:48:52dom96reactormonk: Search for dlls maybe
22:48:52reactormonkeverything else has its place in history.
22:49:00reactormonkdom96, ok
22:49:01StyledAm I looking at the wrong location? https://github.com/Araq/Nimrod/tree/devel/build there is only an empty.txt file in that directory
22:49:04flaviuAlso get rid of the binary tools/nimweb and misc/
22:49:12flaviuStyled: Git keeps history
22:49:29flaviuEach file that has been commited ever gets put into giant hash table
22:49:51flaviuAnd its kept there forever, so it takes a while to clone
22:49:53reactormonkdom96, deleted three *.dll
22:49:54renesacwe were hopping for a factor of 10 reduction, and we may get a 20+ reduction
22:50:18reactormonkStyled, yup. https://github.com/nimrod-code/csources
22:50:22OrionPKello
22:50:39dom96Awesome.
22:50:42reactormonkdom96, killed a few old *.dll, nothing new.
22:50:55Araqoh la la a french guy. ello OrionPK
22:51:02reactormonkkilled nimweb
22:51:16reactormonkanything else?
22:51:20flaviuWhat is myprofile2.txt?
22:51:37Araqsome profiling info, get rid of it
22:51:40Styledokay I'm on my way
22:51:51flaviureactormonk: Misc too?
22:52:07reactormonkflaviu, not too sure about that
22:52:23Araquh oh
22:52:42reactormonkok, I think we're good now.
22:52:43*Araq remembers his misc directory. Was full of treasures.
22:52:46dom96well that's a pity. Github can't compare really old commits.
22:52:59reactormonkAraq, 16MB, good?
22:53:12dom96what are the top 3 biggest files now?
22:53:14Araqdunno, still doesn't fit on a floppy
22:53:17reactormonkno idea
22:53:21dom96can you check?
22:53:24fowllol
22:53:26reactormonklet's see
22:53:40flaviureactormonk: misc is safe
22:53:56flaviuhttps://gist.github.com/flaviut/0a775ea645fd873a2dd8
22:54:06Araq16 floppies, that's more than Sam and Max
22:55:57renesacAraq, a compiler whose repository weights 300+mb is much more impressive than one that weights messily 16MB
22:55:58renesacXD
22:56:18flaviuAraq: What is a patricia? https://gist.github.com/flaviut/10327411#file--L13
22:57:00OrionPKoui oui araq
22:57:07reactormonk16MB. Araq, big red button?
22:57:14dom96reactormonk: well? top 3 biggest?
22:57:16dom96what are they?
22:57:22reactormonklet me see
22:57:22OrionPKI can finally retire my old windows server box, just got everything running on my cubox / nimrod
22:57:46flaviureactormonk: get rid of misc/tre-0.7.5/ at least
22:58:02flaviua copy of GNU regular expressions source
22:58:12reactormonkdom96, http://pastie.org/9067306
22:58:12Araqreactormonk: get rid of misc for good
22:58:20reactormonkAraq, ok
22:58:32OrionPKnow I can go back to working on irc familiar in my spare time... maybe :)
22:58:59dom96reactormonk: thanks. Looks good.
22:59:27Araqreactormonk: can you *restore* the lazarus example?
22:59:41Araqreactormonk: get rid of myprofile1.txt
22:59:57dom96Araq: The lazarus example is still there.
23:00:02reactormonkhttp://pastie.org/9067366
23:00:06reactormonkAraq, restore?
23:00:12fowlgtk could be removed
23:00:14reactormonkAraq, already did
23:00:25dom96no, don't remove gtk2.nim
23:00:35Araqand build.sh?
23:00:42reactormonkAraq, not relevant.
23:00:50dom96build.sh is part of csources
23:00:52dom96You should remove it
23:00:56Araqyeah
23:01:03fowlOrionPK, oo cubox is a cutie
23:01:07reactormonkthat's true
23:01:07flaviureactormonk: Change line 14 in the script so it ends with head -n 50
23:01:08OrionPKya
23:01:10dom96what about lib/base/libcmt-0.1.0/configure ?
23:01:13OrionPKfowl its gorgeous
23:01:14reactormonkanything else?
23:01:39OrionPKfowl a lot more powerful than the pi as well
23:01:43fowlOrionPK, what distro do you have on it
23:01:48OrionPKdebian
23:01:49Araqdom96: I have no lazarus example in examples/ anymore since quite some time
23:01:53flaviuthat way you get more deletion candidates
23:02:09reactormonkhttp://pastie.org/9067438
23:02:20dom96Araq: This exists: https://github.com/Araq/Nimrod/tree/devel/examples/cross_calculator/lazarus
23:02:26dom96Unless you're talking about some older example?
23:02:37Araqyeah
23:02:40reactormonkI think we're good.
23:02:52Araqoh I see
23:02:56Araqgot moved to there
23:02:59dom96What about pcre.dll
23:03:03flaviureactormonk: 1 sec
23:03:03dom96and tools/niminst
23:03:05reactormonkdom96, still active somewhere
23:03:08dom96and libcurl.dll
23:03:21dom96IMO we should remove them
23:03:32flaviureactormonk: get rid of lib/base/libcmt-0.1.0/
23:03:34reactormonkdom96, bfg told me they're still referenced somewhere
23:03:41dom96by remove I mean.
23:03:48dom96Remove them with a commit
23:03:53dom96and then purge them from the history
23:04:01flaviureactormonk: You have to get the top commit not to refrence them
23:04:03flaviuthen you can remove
23:04:10Araq lib/ucmaps/
23:04:11dom96or however it should be done
23:04:12fowlOrionPK, wow these are cool, im looking at the cubox-i
23:04:22OrionPKyeah, definitely recommend em
23:04:28fowldo they have wifi?
23:04:31OrionPKyep
23:04:39dom96reactormonk: you should grep for .dll
23:04:41OrionPKand bluetooth and ir
23:04:44reactormonkdom96, they are not even 1MB
23:04:47Araqwe have #nimrod-offtopic btw
23:04:54dom96also remove tests/vm/toverflowopcmulint
23:05:00Araqand all the cool kids are there already
23:05:10dom96reactormonk: It all adds up though
23:05:15reactormonkdom96, I think we should stop here, otherwise there's some foot-shooting going on
23:05:39AraqI agree with dom96
23:05:40OrionPK:p
23:05:45reactormonkso be it
23:05:47OrionPKthis is #nimrod-offtopic
23:05:56flaviuI think that we shoudn't get rid of previous sources
23:06:15dom96I think all binaries should be gone from the repo.
23:06:35Araqyeah plus all the mapping.txt, these are generated too
23:06:47flaviuthats fine then
23:06:52reactormonkso the whole dist/ directory should be gone?
23:07:03dom96yeah, pretty much.
23:07:06reactormonkok
23:07:11dom96But be careful. There are still currently files inside there.
23:07:14Araqplus lib/ucmaps/*
23:07:16dom96But you know that already.
23:07:27flaviureactormonk: You've gotten rid of rod/* right?
23:07:33reactormonkflaviu, nope
23:07:46reactormonknow I did
23:07:47flaviuIIRC thats the compiler's caching
23:07:50dom96I think rod contained the pascal sources at one point
23:08:03reactormonk... fuck
23:08:11dom96.. yeah
23:08:13reactormonknow let's start over.
23:08:16Araqnote: lib/pure/unidecode/unidecode.dat is NOT to be deleted
23:08:20Araqno wait
23:08:38Araqrod contained generated nimrod code, rod is fine to be deleted
23:08:44reactormonkdamn
23:08:46Araq'nim' contained the pascal sources
23:08:49reactormonkalready started over
23:08:53Araqlol
23:08:54reactormonkoh well
23:09:00dom96lol sorry
23:09:01reactormonkzsh history ftw
23:09:24flaviuoh, rod had a bunch of nimrod sources too
23:09:37flaviuhttps://gist.github.com/10328344
23:10:01flaviuI love having gist as a command
23:10:06dom96yeah, let's not remove it.
23:10:15fowlthats just the old compiler
23:10:21reactormonkstill
23:10:30dom96sources are sources
23:10:56flaviusorry, I thought it was just compiler caches
23:11:52flaviureally sorry, the *.rod files are caches
23:12:23Araqyup
23:15:19reactormonkso delete all *.rod files?
23:15:45flaviureactormonk: There aren't any
23:15:49reactormonkok
23:15:50flaviugit rev-list --all --objects | grep '.*\.rod'
23:16:13flaviuBut delete all mapping.txt
23:16:56reactormonkand all *.dll?
23:16:56dom96what's nimrod.aip?
23:17:23dom96That's Araq's call
23:18:26reactormonkAraq, got the gun ready, wanna kick all *.dll?
23:18:38flaviudom96: git show 4d1439db406ee1c409991a040eb1ca607a288e0c
23:19:10flaviuIt isn't important, looks like the installer for a pascal program
23:19:18flaviuwait nm
23:19:37flaviuIts an installer for nimrod from back when it was still partially pascal
23:23:38*runvnc quit (Ping timeout: 240 seconds)
23:23:52Araqreactormonk: yes
23:24:15Araqdelete all .dll
23:24:30Araqno idea what nimrod.aip is, remove it
23:24:50reactormonkkk, down to 12MB
23:24:59reactormonkAraq, good enough?
23:25:09flaviuWow, thats amazing
23:25:21Araqyeah, push it, I need to sleep
23:26:25reactormonkdone
23:26:29EXetoCyou sleep like every day
23:26:39NimBotAraq/Nimrod asyncmacro 75893d1 Dominik Picheta [+2 ±1 -0]: First implementation of the async/await-like system with macros.
23:26:39NimBotAraq/Nimrod asyncmacro 84ea209 Dominik Picheta [+0 ±1 -0]: Merge branch 'master' into asyncmacro
23:26:39NimBotAraq/Nimrod asyncmacro d15e67e Dominik Picheta [+0 ±2 -0]: Better param passing. Implemented send, readLine and support for... 1 more lines
23:26:39NimBotAraq/Nimrod asyncmacro a1a8925 Dominik Picheta [+1 ±2 -0]: Calling an async proc with invalid arguments will now fail at compile-time.
23:26:39NimBot14 more commits.
23:26:40NimBotAraq/Nimrod master 68b7779 Reimer Behrends [+0 ±1 -0]: Fixes various issues with shallow copying.... 11 more lines
23:26:40NimBotAraq/Nimrod master e2d02d5 Andreas Rumpf [+0 ±1 -0]: Merge pull request #691 from rbehrends/fix-shallow-pragma... 2 more lines
23:26:40NimBotAraq/Nimrod master a86c774 Simon Hafner [+0 ±0 -0]: deleted & purged dlls
23:27:07AraqEXetoC: everybody needs a hobby
23:27:15reactormonkAraq, where we go
23:28:52reactormonkslimmed & trimmed.
23:29:18Araqdamn downloading 12MB still takes an eternity
23:29:40reactormonkbetter than the initial 200MB?
23:29:43flaviuIts actually 25MB
23:29:51flaviuNo idea why
23:30:06reactormonkbetter than the previous occasion.
23:30:27flaviuDefinitely
23:31:26dom96reactormonk: very nice. Thanks for your hard work!
23:31:27renesacdo some of those deleted files need a entry in .gitignore to not be added again to the repository?
23:31:31dom96and for everyone else that helped :)
23:32:24VarriountOk so... how do I get my local repo working?
23:32:35flaviureactormonk: Can you post your recent command history? So everything can be added to .gitignore
23:32:51flaviuVarriount: rm -rf ./Nimrod; git clone https://github.com/Araq/Nimrod.git
23:32:52Araqliar, it's not 12MB
23:33:01reactormonkflaviu, http://pastie.org/9068442
23:33:11reactormonkAraq, was on my end :-/
23:33:18reactormonkAraq, well, the bare repo
23:33:21reactormonkwithout the checked out files.
23:33:30flaviureactormonk: How did you measure it?
23:35:05dom96Try this: git gc --aggressive --prune=now
23:35:23dom96It reduced the repo size from 40MB to 25MB.
23:35:28dom96For me at least.
23:35:49dom96But I guess that's just a local thing.
23:35:49*runvnc joined #nimrod
23:35:51flaviuIt reduced my .git/pack from 27MB to 11MB
23:36:02reactormonkVarriount, you can pull the new branches via git checkout -b new-devel origin/devel and then cherry-pick over
23:36:32dom96Hopefully merging devel into master won't be difficult now...
23:37:09*io2 quit (Quit: ...take irc away, what are you? genius, billionaire, playboy, philanthropist)
23:37:30Styledis the nimrod compiler written in nimrod or c?
23:37:36dom96nimrod
23:38:00flaviuStyled: Everything but perl regexes are in nimrod
23:38:27Araqdom96: it's not, I already did
23:38:46dom96Araq: good
23:39:18*xenagi joined #nimrod
23:39:35VarriountStyled: But it translates to C before being compiled.
23:40:51dom96ok, so is that the purge done then?
23:41:58reactormonkdom96, yup
23:42:35dom96Great
23:43:25Varriountreactormonk: How do we prevent people with unpurged history sending PR's that revert things?
23:43:51flaviuI
23:44:09flaviu'm fairly certain that most people who do PRs know about the purge
23:44:31Varriountflaviu: But I don't know whether I incorperated the purge correctly. :/
23:44:48Varriountreactormonk: Also, I was under the impression that the original devel and master branches were purged, not the new-devel and new-master branches
23:45:04flaviuThe entire repo is purged
23:45:19flaviuGit's hash table of files has the purged files completly removed
23:45:23dom96Yeah, what was the point of the purged_devel and purged_master branches?
23:48:39Araqer I pushed but nimbot is silent
23:48:59VarriountOk, so I update my devel branch by force pulling Araq's repo's devel branch?
23:48:59dom96yeah, it crashed
23:49:09flaviuWhat I've done is just delete my fork completly
23:49:16flaviuand fork again
23:49:27VarriountYes, but I have local branches and whatnot
23:49:48VarriountBah, I guess I'll just save a copy of my changes, and move stuff over manually.
23:51:12Araqdom96: well please fix it
23:51:24AraqI merged devel into master
23:51:26dom96Araq: Go into your web hooks and grab me a copy of the JSON it sent
23:54:10dom96https://github.com/Araq/Nimrod/settings/hooks
23:54:30Araqyeah
23:54:33AraqI'm there
23:54:38Araqno idea how to get the json
23:54:48Araqshould I simply press the button?
23:54:50dom96click the web hook
23:55:06dom96you'll see a list of recent deliveries
23:55:14Araqwhich one?
23:55:15dom96click the ... beside the top one
23:55:21Araqgithub2 or github?
23:55:29dom96uhh
23:55:31dom96Why do you have two?
23:55:49dom96ahh, one must be old.
23:55:53dom96picheta.me/github
23:55:56dom96that one
23:56:27dom96craete a comment with the JSON and other info here: https://github.com/nimrod-code/nimbuild/issues/15
23:56:30dom96*create
23:56:55dom96put the headers there too
23:57:55Araqthe JSON is 22695 lines
23:58:03dom96gist it then
23:58:11dom96and create a comment with the link
23:58:16Araqyeah
23:59:08Varriountdom96: I've re-downloaded the repository copies the testers use. All thats left is for me to update the installer repo copies, and my own personal copy
23:59:08Araqthat's github2 though
23:59:26Araqare you sure it's "github"?
23:59:30dom96Araq: picheta.me/github2 ?
23:59:50dom96what's the date on it?