19:05:01 #startmeeting Fedora DotNet (2017-02-07) 19:05:01 Meeting started Tue Feb 7 19:05:01 2017 UTC. The chair is Rhea. Information about MeetBot at http://wiki.debian.org/MeetBot. 19:05:01 Useful Commands: #action #agreed #halp #info #idea #link #topic. 19:05:01 The meeting name has been set to 'fedora_dotnet_(2017-02-07)' 19:05:03 #meetingname dotnet 19:05:03 The meeting name has been set to 'dotnet' 19:05:07 #nick dotnet 19:05:56 #chair bt0 rmartinelli nmilosev omajid tmds 19:05:56 Current chairs: Rhea bt0 nmilosev omajid rmartinelli tmds 19:06:07 hi all! 19:06:15 hello everyone! 19:06:16 hey tmds 19:06:17 hi o/ 19:06:24 hey all 19:07:03 .hello ttorling 19:07:05 moto-timo: ttorling 'Tim Orling' 19:07:19 * jflory7 watches from the back 19:07:23 #topic Agenda 19:07:32 #link https://fedoraproject.org/wiki/Meeting:DotNet_2017-02-07 19:07:34 #info (1) Roll Call 19:07:37 #info (2) Announcements 19:07:40 #info (3) Action items and Tickets 19:07:42 #info (4) Packaging progress 19:07:45 #info (5) Open Floor 19:07:49 #topic Roll Call 19:07:52 #info Name; Timezone; Sub-projects/Interest Areas 19:07:56 #action dotnet New members, make sure you introduce yourself on the DotNet mailing list [ https://fedoraproject.org/wiki/DotNet ] 19:08:06 If this is your first time at a DotNet meeting, feel free to introduce yourself to everyone and say hello! If anyone has any questions before we get started with the rest of the agenda, now is also a good time to ask. 19:08:19 #info Radka Janek; UTC+1; CommOps, Diversity, DotNet,... 19:08:31 * Rhea waves at jflory7 :D 19:08:38 #info Alberto Rodriguez; UTC-6; Commops, DotNet, Infra... 19:08:51 .fas nmilosev 19:08:51 nmilosev: nmilosev 'Nemanja Milosevic' 19:08:53 #info Justin W. Flory; UTC+1; CommOps, Magazine, Marketing, Diversity, and moreā€¦ [ just observing for this meeting! ] 19:09:00 .fas rimolive 19:09:01 rmartinelli: rimolive '' 19:09:14 .fas roddiekieley 19:09:14 rkieley: roddiekieley 'Roddie Kieley' 19:09:27 .hello tmds 19:09:28 tmds: tmds 'Tom Deseyn' 19:09:49 .hello omajid 19:09:50 omajid: omajid 'Omair Majid' 19:10:06 .fas bt0dotninja 19:10:08 bt0: bt0dotninja 'Alberto Rodriguez Sanchez' 19:11:06 #topic Announcements 19:11:11 #info === "Update for ASP.NET Core 1.1" === 19:11:14 #link https://blogs.msdn.microsoft.com/dotnet/2017/01/30/january-2017-update-for-asp-net-core-1-1 19:11:19 #info Security update - if you're using asp.net core you might want to update :P 19:12:13 * nmilosev should update the 1.1 and clean RPMs 19:12:26 * nmilosev should also obsolete 1.0 repo 19:12:27 Now my question on topic, omajid/tmds do we need to update the F25 package? I have to admit that i wasn't following our internal discussions about this... 19:12:44 nmilosev: obsolete yes definitely 19:12:45 nmilosev: afaik rpms don't include any asp.net stuff. so should be good 19:12:52 nmilosev++ 19:12:57 Yeah so its all in the nuget package for aspnet right 19:13:00 Oh it's just a project.json modification? 19:13:07 1.0.0-preview2-1-003175 << yes, it needs update 19:13:20 Yes, for the user-dev they have to bump a tiny number in project file 19:13:25 :D 19:13:28 rmartinelli, that's just tooling version :) 19:13:36 oh 19:13:39 Yeah you could go with 1.1 now tho 19:13:46 Yeah :D 19:13:46 Anyway moving on then 19:13:56 sorry then 19:13:58 #info === "Announcing .NET Core, .NET Native and NuGet Updates in VS 2017 RC" === 19:14:00 #link https://blogs.msdn.microsoft.com/dotnet/2017/01/30/announcing-net-core-net-native-and-nuget-updates-in-vs-2017-rc 19:14:05 ...no extra info... updated stuff 19:14:11 :D 19:14:29 Go read if you're curious, i just breezed through and didn't notice anything worth mentioning 19:15:12 #info === ".NET Core Image Processing" === 19:15:13 #link https://blogs.msdn.microsoft.com/dotnet/2017/01/19/net-core-image-processing 19:15:19 This however... this... finally 19:15:23 \o/ 19:15:30 nmilosev: you could hash-tag action those items :) 19:15:48 #action nmilosev obsolete the old 1.0 copr packages 19:15:52 hua hua 19:16:01 * moto-timo shivers 19:16:02 now you HAVE TO do it 19:16:06 Our fearless reader should do that 19:16:09 There you go! :D 19:16:12 :D 19:16:19 fearless rheader 19:16:23 :D 19:16:32 :P 19:16:33 one more 19:16:37 #info === "The .NET Language Strategy" === 19:16:40 #link https://blogs.msdn.microsoft.com/dotnet/2017/02/01/the-net-language-strategy 19:16:44 #info A little bit about different languages in dotnet 19:16:58 And thats it, anyone got anything to mention here? 19:17:25 Can I go slightly offtopic? 19:17:30 #info === "devconf talks: nmilosev++ dotnet in Fedora" === 19:17:39 do you9 have links to youtube? 19:17:51 i have it somewhere... 19:17:52 if you are an Eclipse IDE fan 19:17:53 * Rhea digging 19:18:00 #link https://github.com/mickaelistria/aCute 19:18:06 nmilosev++ I lost your presentation because I needed to help a friend with his talk . 19:18:21 #info C# editor plugin for Eclipse IDE 19:18:36 nmilosev++ :) 19:19:04 dont link it here please 19:19:05 I have a link to share too. It it the right moment to do that? =D 19:19:09 its mid-something 19:19:10 #undo 19:19:11 Removing item from minutes: INFO by moto-timo at 19:18:21 : C# editor plugin for Eclipse IDE 19:19:13 #undo 19:19:13 Removing item from minutes: 19:19:14 #undo 19:19:14 Removing item from minutes: INFO by Rhea at 19:17:30 : === "devconf talks: nmilosev++ dotnet in Fedora" === 19:19:16 sorry 19:19:30 #info === "C# editor plugin for Eclipse IDE" === 19:19:34 #link https://github.com/mickaelistria/aCute 19:19:49 moto-timo++ for finding this, looks good :) 19:20:03 yeah, it uses the new Language Services Protocol so awesome 19:20:29 unless I'm confusing it with some llvm/clang stuff I also saw today 19:20:30 #info === "devconf.cz talks: tmds++" === 19:20:37 #link https://www.youtube.com/watch?v=TADd4HnnoeI 19:20:56 tmds++ 19:21:00 tmds++ 19:21:01 tmds++ 19:21:05 tmds, a youtube celebrity, who would have thought... 19:21:07 * jflory7 has to run for a bus, but will catch up later or read the minutes :) 19:21:12 tmds++ on the talk! 19:21:12 jflory7: Karma for tmds changed to 7 (for the f25 release cycle): https://badges.fedoraproject.org/tags/cookie/any 19:21:29 tmds++ for devconf talk 19:21:39 karma roulette 19:21:40 thx guys! 19:21:44 #info === "devconf.cz talks: nmilosev++" === 19:21:47 #ink https://www.youtube.com/watch?v=MD-2ITGN2Wg 19:21:55 there found it haha 19:22:18 nmilosev++ 19:22:25 great talk! 19:22:31 Rhea, s/ink/link/ 19:22:32 Devconf was awesome and i'm really happy that i had the chance to meet the two of you +1... kinda shame that i didnt have a bit more time to spare >_< 19:22:40 #link https://www.youtube.com/watch?v=MD-2ITGN2Wg 19:22:43 :D 19:22:47 thanks 19:22:51 * rmartinelli wondering if there is Rhea talk too 19:23:01 Same, awesome to meet Rhea and tmds :) devconf was awesome 19:23:01 nope, there will be at open source summit 19:23:36 I'm giving a talk at ELC if you want to meet me ;) 19:23:49 I'll start working on it in two months though, cause i'm working on some code that if i succeed i'd like to include in the talk 19:24:08 moto-timo: where is that? 19:24:19 Portand, Oregon, USA 19:24:25 Feb 21-24 19:24:38 oh yeah i cant get passport anytime soon i'm european union bound :D 19:24:49 "you're welcome" 19:24:57 i never had one, my citizenship is in different country - so much trouble 19:25:17 my government is a reality tv show 19:25:24 xD 19:25:29 #topic Action items and Tickets 19:25:32 #link https://meetbot.fedoraproject.org/teams/dotnet/dotnet.2017-01-17-19.04.html 19:25:35 #link https://pagure.io/fedora-dotnet/issues?tags=meeting 19:25:39 #info How This Works: We look at past #action items from the last meeting for quick follow-up. If a task is completed, we move on to the next one. If it isn't, we get an update and re-action it if needed. If no status, we'll try to get a quick update and move forward. 19:25:42 #info === [COMPLETE] Rhea Finalise the fas-group and pagure setup. === 19:25:46 #info Kinda done there, except that we can't reaaally finish all that until we get nmilosev into the group, which would be done only when we submit package, so it's kinda weird. 19:26:17 I have to get sponsored as a packager first, I think. 19:26:20 Not much to add to that. The fas group has now packager requirement, and will be used to cover our packages, we need to sort you out 19:26:21 I gotta go, but if we want a badge I already have a draft and can file a ticket with #fedora-design 19:26:27 I'll read the logs 19:26:51 nmilosev: basically you get sponsored when we submit the package, you will be listed as one of the co-maintainers, and stuff around it, and that will get you sponsored 19:27:03 By the way, I've talked to Martin from MS (at DevConf) and they added me to the new source-build repository so I can test it out. 19:27:14 I'll have a chat about details with certain someone when fosdem / gsoc chaos settles down. 19:27:16 nmilosev: cool 19:27:31 amazing 19:27:33 Yeah, it's great, I will tell you how it goes on the next meeting 19:28:07 Yeah i still didnt have time to sort out my github - ill get to that some time next week i think heh 19:28:26 #info === GSoC === 19:28:41 #idea Do we want to mentor a Google Summer of Code project? 19:29:23 We've got just about 24 hours to get something together if so, i'd be heading that innitiative and draft the proposal, but i need ideas... Is there anything we could do? 19:29:33 Is there something we want a student to contribute to? 19:30:00 Fedora and/or RedHat? omajid tmds what do you think about RH side? :d 19:30:34 ping jflory7 gsoc topic 19:30:43 if you have anything to add 19:30:44 * nmilosev thinks jflory7 is afk 19:30:54 * omajid has had his head down in the builds. sorry no ideas atm. 19:30:56 nop he's ... he's actually walking and on phone i think so yeah 19:31:10 :D 19:31:26 Rhea, any idea what tasks for GSoC should look like? 19:31:30 I could use a student to work on MY project haha 19:31:44 Rhea: what's your project? ;) 19:31:48 Anything really, any project in dotnet, or around packaging it 19:32:10 We've got packaging covered i think, so i'm thinking about some dotnet based tools for Fedora 19:32:50 like Eclipse integration? 19:32:58 omajid: two empty repositories right now, cause i'm in the process of splitting the project up, fixing up structure, i've rewritten about 30k lines of code in last two days... 19:33:00 or vscode? 19:33:05 an idea just came to mind, a .net tool that packages a .net app as an rpm? 19:33:13 yup 19:33:18 tmds, that's actually a very good idea 19:33:19 something like that is what i was thinking tmds 19:33:19 tmds: that would be great! 19:33:31 I'm just not sure about... details heh 19:33:38 what would the product look like really 19:33:45 we would probably want that packaged as well? 19:33:58 Or have it as a simple script to be downloaded 19:34:03 that works too 19:34:18 like the .net tools in project.json which perform db migrations 19:34:27 mm 19:34:37 Or package NuGet -> RPM 19:34:41 yeah 19:34:43 lets do that 19:34:45 like Python does 19:34:56 i mean 19:35:14 But that's a separate idea, than building a tool to package user's application 19:35:25 yup thats what i mean haha 19:36:01 for example: https://github.com/aspnet/MusicStore/blob/rel/1.1.0/samples/MusicStore/project.json#L66 19:36:12 that line is a .NET tool that 19:36:20 Yup we could go with this.. lets wait for jflory7 - or i'll poke him later about details, he could have some insight 19:36:47 i'm not entirely sure that this would be large enough project 19:36:56 yeah, maybe both :D 19:37:17 if not then we can combine all of the above into one app/script/thing 19:37:40 i would suggest writing it all in dotnet just to be more evil to the student (as opposed to bash or other scripting) 19:38:53 Okay, who's gonna be around for a few more hours to explore further on this after hte meeting? 19:39:14 ME 19:39:26 Rhea, do you think it's too soon for this? :( 19:39:31 other idea, add support for .net core to mono-sharp 19:39:42 err, gtk-sharp I mean 19:39:46 nmilosev: what do you mean too soon 19:40:01 I mean for GSoC offer 19:40:08 ideas should be in the gsoc application with deadline in 43 hours 19:40:32 so we need to put some draft together to help our org applicationa s well 19:40:57 afaik, they broke the tools api between project.json and csproj migration. i don't know what the path foward is. seems a bit too unstable change to ask a student to work on it. 19:41:02 Hence why i'm trying to get it out of the way rather today than tomorrow 19:41:30 I really liked the project.json :( 19:41:41 Rhea, I meant too soon to ask someone to work on it 19:41:47 Since .NET is changing all the time for now 19:41:54 not really 19:42:08 not much will change in regards to packaging an app 19:42:29 i mean... some code -> dotnet publish -> rpm -> done? 19:42:39 with some parameters as to specify target, etc 19:43:28 we write proposel for what we've got now, project.json, and if it changes before summer, then the student will work on xproj if not then they stick to .json and we can easily fix it up later 19:43:41 Or we keep the student and have him work on it in winter as well 19:43:48 hua hua hua 19:44:06 sounds ok, but we would have to specify somewhere that the student has to be familiar with RPM packaging and .NET Core 19:44:21 yup there are requirements and stuff 19:44:43 i wouldnt say that they have to be that familiar with it, its school thing for a reason - they will learn 19:45:05 There is a designated mentor for this? 19:45:10 yes, me 19:45:20 Oh, I'm gonna apply then ;D 19:45:21 and whoever else out of our lovely dotnet team here would go 19:45:29 :D 19:45:29 But from the other side 19:45:39 nope, i dont know what do you mean by other side 19:45:46 student side 19:45:49 :D 19:46:09 It's just our idea, then students browse them, pick, discuss with us, make an application to google, they get accepted, we work together and teach the student while creating something cool 19:46:34 sounds good :) 19:46:38 (+ frankly its all new for us so we learn as well, a lot) 19:47:10 We've got maybe slight advantage to a student that we've been around the industry whether as developers or as packagers for a while and got an idea how does it work... 19:47:45 nmilosev and myself dev side and a bit of packaging, omajid tmds the other side of things - experienced packagers 19:47:51 we've got it covered 19:47:54 :D 19:48:48 what topic are we? 19:48:48 You can count on me to help :) 19:48:53 GSOC 19:48:55 still 19:49:18 I'd like to continue via email/etherpad a bit later, in about an hour ill send you some info 19:49:28 nmilosev <3 19:49:43 Sure, you can ping me on telegram as well 19:50:06 tru 19:50:11 Okay lets wrap it up 19:50:16 #topic Packaging progress 19:50:19 Anything here? 19:50:24 I have a couple of things 19:50:25 Doubt it with all the events over last two weeks 19:50:27 oh 19:50:29 okay 19:50:31 unexpected 19:50:39 This maybe interesting: https://blog.rraghur.in/2017/02/building-efficient-aspnet-core-docker.html 19:50:47 (for origin/openshift) 19:51:02 tmds omajid ^ 19:51:22 thanks, saw it last night 19:51:32 Also: https://pagure.io/fedora-dotnet/issue/8 there is an issue with dotnet-clean package 19:51:47 System.Data.SqlClient DLL is suspiciously missing 19:51:49 :( 19:51:54 Any ideas? 19:52:09 hmm 19:52:36 Pulling System.Data.SqlClient from Nuget doesn't work. What is very weird: application gets compiled, crashes at runtime 19:53:03 did you try on f24? 19:53:11 i mean with official msft package 19:53:16 No, I only tested on F25 19:53:24 But with nmilosev/dotnet1.1 everything works 19:53:26 Same project 19:53:36 Hmm. 19:53:43 I will look into it, just wanted to let you guys know. 19:53:54 thanks! 19:54:06 Also, a question for omajid 19:54:23 Do you know why rover tools produces much bigger binaries than they are supposed to be 19:54:44 MS package is around 50MB, ours 90MB 19:54:46 do you have any examples? 19:54:55 do you filter out .dbg files? 19:55:17 * nmilosev smacks himself on the head 19:55:21 thanks :D 19:55:30 hehe, been there, done that :) 19:55:42 (me is just now debugging the same problem with openjdk9 copr :) ) 19:56:04 that's probably it, didn't want to open issue until I ask 19:56:15 Rhea, you can action that aswell :D 19:56:34 huh 19:56:38 you can 19:56:40 I have some more links to share :) 19:56:41 i dnot know what 19:56:48 give all links 19:56:50 * Rhea eats them 19:56:55 interesting reads on project.json -> msbuild: http://www.natemcmaster.com/blog/2017/01/19/project-json-to-csproj/ http://www.natemcmaster.com/blog/2017/02/01/project-json-to-csproj-part2/ 19:57:07 ubuntu arm is supported by .net core: https://github.com/dotnet/core-setup/issues/725#issuecomment-275252640 19:57:20 and so is avalonia: https://github.com/AvaloniaUI/Avalonia/pull/875 19:57:29 arm support is in netcore to begin with, so i wouldnt give ubuntu TOO much credit.. 19:57:39 #action nmilosev cleanup the dotnet-clean rpm (remove debug symbols) 19:58:03 I am not giving credit, I am bringing the message 19:58:11 I've tried ARM binaries from that thread on Fedora as well 19:58:35 and? 19:58:53 But these ones: https://github.com/dotnet/core-setup/issues/725#issuecomment-273745685 19:59:19 Works like a charm (with symbolic linking correct libicu) 19:59:21 did they work on fedora? 19:59:25 cool 19:59:27 (RPi2, F25) 19:59:28 on rasberry pi? 19:59:30 k 19:59:52 hehe 20:00:09 ubuntu-arm is in the rid tree (https://github.com/dotnet/corefx/blob/master/pkg/Microsoft.NETCore.Platforms/runtime.json) 20:00:21 huh 20:00:27 Lets push Fedora-arm 20:00:29 uahaaa 20:00:52 Can we do that? 20:00:55 I mean... why not 20:01:30 *cough* f25 *cough* 20:01:40 :D 20:01:42 nmilosev so what's stopping us from packaging dotnet for fedora on raspberry 20:02:21 I could package it (the ugly way), but it would be much better to wait for source builds :) 20:02:24 In my opinion 20:02:34 Should I create an issue to add F25 rid btw? 20:02:47 Yup, so we can do taht one day 20:03:23 tmds_ what is their policy about this, do they require their own builds before they create rid or what 20:04:00 * omajid can't make head or tails of their policy. 20:04:04 if you send a PR to add your RID, they will accept it 20:04:18 Okay i'll get onto that then maybe 20:04:25 o.o 20:04:49 We've got more-less packages almost there, so might as well have this in dotnet for future updates soon TM 20:04:50 don't know what the plans are, but this RID stuff has been proven quite inflexible 20:05:05 I hope we can work with linux-x64 20:05:28 Yeah but given that they also have all the explicit stuff, we should probably maintain these as well... 20:05:33 in my opinion you should be able to specify the rid as a build parameter and a list of parents, and that should just get baked into the output 20:05:49 hmm 20:07:05 if you want to know more, see https://github.com/dotnet/coreclr/issues/9210 20:07:20 Yup i've been following that 20:07:37 kind of weird in their ecosystem, though. for example, if we create our own f25 rid, no one else can target it. 20:08:43 I've found out recently that you can change RID with an env variable 20:08:52 I didn't see that documented anywhere 20:09:01 omajid: the apps of your distro can target it 20:09:01 Or just didn't look deep enough 20:09:14 and anyone else, would probably target linux-x64 20:09:16 tmds_: yup, better than nothing. 20:09:35 hmm 20:10:06 #action Rhea create an issue on pagure to keep rid info and discussion together... 20:10:12 Lets call it for today? 20:10:17 good idea :) 20:10:19 #topic Open Floor 20:10:23 Please talking too you all :) 20:10:27 to* 20:10:34 too you to nmilosev 20:10:37 :P 20:10:53 https://blogs.apache.org/activemq/entry/using-net-libraries-with-activemq 20:11:26 The Artemis team published a tutorial to use dotnet core producing/consuming messages in Artemis. 20:11:30 It uses AMQP 20:13:02 interesting 20:15:22 #endmeeting