profile
viewpoint
If you are wondering where the data of this site comes from, please visit https://api.github.com/users/jaor/events. GitMemory does not store any data, but only uses NGINX to cache data for a period of time. The idea behind GitMemory is simply to give users a better reading experience.

jaor/geiser 275

emacs and scheme talk to each other

bigmlcom/bigml-php 8

BigML.io Php Bindings

jaor/elibs 7

Elisp libraries

antoniomachine/bigml-ruby 3

A Ruby wrapper for the BigML REST API

jaor/iwlib 2

A binding to the iw library for getting info about the current WiFi connection.

bigmlcom/fipp 1

Fast Idiomatic Pretty Printer for Clojure

jaor/clojure-mode 1

Emacs support for the Clojure language

jaor/congomongo 1

Clojure wrapper for the mongo-db java api

jaor/dcpl 1

Code inspired by 'Design Concepts in Programming Languages'

jaor/io 1

Libraries and examples for the BigML API

issue commentjaor/xmobar

Dhall Configuration?

On Wed, Sep 15 2021, Guy Gastineau wrote:

Well, I don't care about trays :stuck_out_tongue_winking_eye:

i don't do either anymore, solved that problem in a better way :)

those are however quite a bit of work, and won't provide any new feature, so it's a bit of a grunt job that you might well not feel like undertaking at all! :)

This kind of work feels quite honest to me. It is likely challenging, tedious, and frustrating.

agreed on all accounts.

I am worried enough about all my X tools and the future that I have extra incentive :laughing:

and on this one too :)

I wonder if it will be simpler to migrate to pango before migrating to cairo?

i haven't looked at this in detail, but my impression is that moving to cairo would make trivial using pango with it, but not the other way round. so, if it were me, i think i'd investigate directly the cairo possibility (there's already a bug about pango, i think).

I am afraid of neither C nor allocaBytes and the rest of the Haskell Foreign stuff, and I would feel proud of my work if I migrate xmobar. I'll make a new issue about it I guess.

there are some haskell cairo bindings in hackage that will possibly help (https://hackage.haskell.org/package/gi-cairo, even though they're auto-generated, it seems... i think there were some hand-crafted ones too.)

guygastineau

comment created time in 5 days

issue commentjaor/xmobar

Dhall Configuration?

On Wed, Sep 15 2021, Guy Gastineau wrote:

I agree, and I expect my time could be better devoted elsewhere. FWIW, dhall is an impressive and serious project with dedicated developers. They are devoted to a safe yet featureful, turing-incomplete config language based on system f.

sounds interesting, i'll have to take a look one of these days :)

I have been looking through some of the issue log to see what there is to do. I might open some PRs with an attempt at closing some existing issues over the next week.

many thanks in advance for that, truly appreciated!

If there is anything that you think is important or would like to see, @jaor, please let me know.

a thing that's been in my todo list like forever is moving away from Xfts and use pango instead; the former is very, very poorly documented and, in my experience, terribly buggy (at least in the way we use it in xmobar), and it's always been a headache to, simply, align things in the bar!

relatedly, i'd be ecstatic if we could get rid of direct Xlib calls and use something higher level like cairo instead, which might even pave the way to a wayland future... at the end of the day, we just display a rectangle and write text to it, how hard could that be :))

those are however quite a bit of work, and won't provide any new feature, so it's a bit of a grunt job that you might well not feel like undertaking at all! :)

other things i've wanted in the past are some form of support for tray icons, perhaps using XEMBED (#299), although that's linked to Xlib.

there are a few other bugs marked with "feature request", but i'm noticing that, after so many years, i've implemented most of the little things i felt missing and there's no big feature i crave for by now :))

possibly that's the best way for you too: just find some functionality you really need, and go for it!

I do hope to get a sense from other users about their preferences. I don't expect responses to be too different, but I'll keep this issue open for discussion for a while if that is fine with you.

yes, by all means.

guygastineau

comment created time in 5 days

issue commentjaor/xmobar

Dhall Configuration?

On Wed, Sep 15 2021, Guy Gastineau wrote:

  1. Recompilation is not especially fast.

not a big deal in my experience, and i got the impression that i recompile my xmobar instances more often than average. but this is just a datapoint.

  1. I install GHC on everything I own (well not phones), but some users might not be haskellers. Does anyone use xmobar without xmonad?

fwiw, that's a false dichotomy: i don't use xmonad and compile my own xmobars. so i guess the question is more like: do non-haskellers use xmobar and are they unhappy with the config syntax? my impression is that there are a non-insignificant number of non-haskellers, but over the years we haven't received any complaints about the configuration file syntax...

[...]

Given both the pros and cons of adding dhall config integration to xmobar I hope to get some idea from the community. Is this something anyone wants? Will this add too much complexity/will it get rejected on such grounds? I like xmobar, and as such I would like to continue contributing. I think this would be fun to implement, but if no one wants it, then my time would be better spent implementing an outstanding feature request or bug-fix.

i won't go as far as saying that i'd reject it, but i honestly feel it'd be a bit of a waste of your time, which could be put in improving xmobar in other ways. and some of the drawbacks you mention might perhaps be addressed via helper utilities written in haskell, to be used from haskell.

it would also make things more complex implementation-wise: right now, it's already a bit of a chore having to implement monitors keeping in mind that they should be also configurable via the config language (i have several that i use through haskell and haven't put here because of that). having a second configuration language would just add to that problem, possibly make things a bit more confusing to newcomers, and potentially also to contributors.

finally, i for one had never heard of dhall, and wonder how much of fad it might be (compared with haskell syntax).

guygastineau

comment created time in 5 days

issue closedjaor/xmobar

HandleReader only ever says `Updating...`

https://github.com/jaor/xmobar/blob/3a5004e39e6dc42f2691012b4513a210202311b2/src/Xmobar/Plugins/HandleReader.hs#L43

closed time in 5 days

guygastineau

issue commentjaor/xmobar

HandleReader only ever says `Updating...`

closing...

guygastineau

comment created time in 5 days

PR merged jaor/xmobar

Add the QueueReader plugin.
  • A queue reader for xmobar using TQueue a from stm. This is a flexible and performat solution for sharing data between arbitrary haskell and xmobar.
  • I am not sure if I did the haddocks correctly.

Should there be an safe and unsafe version? If stripActions were exposed, then users could make their own safe version trivially. Let me know.

I ran hlint. No errors in the whole codebase. Awesome maintenance job you all!


I didn't make any tests. Tests seem hard and unnecessary for this. The plugin is very simple, so it practically feels tautological, but it feels arrogant to say this.

Let me know if I should make any improvements.

+97 -0

10 comments

4 changed files

guygastineau

pr closed time in 6 days

push eventjaor/xmobar

Guy Gastineau

commit sha ec96ec469e9dff06bc503366ff7e3022c595d8f7

Add the QueueReader plugin. * A queue reader for xmobar using `TQueue a` from `stm`. This is a flexible and performat solution for sharing data between arbitrary haskell and xmobar. * I am not sure if I did the haddocks correctly.

view details

Guy Gastineau

commit sha 38adaf0ded1ca40bd7366af82f8e5bae737233ae

Add documentation for QueueReader. I added `QueueReader` as a section under `Software Transational Memory`, but it is all alone. Some other plugins actually use STM, and maybe they should actually go in that section too. I didn't make a safe versions since the plugin is not readable. It still would be nice to have, but I think exposing `stripActions` is the appropriate solution. That should be handled in a new PR I guess.

view details

jao

commit sha 6525d5102ef3e1649bc71d29734bbae93cc8c1b7

Changelog and whitespace

view details

push time in 6 days

PullRequestReviewEvent

Pull request review commentjaor/xmobar

Add the QueueReader plugin.

 choice.      hPutStr writeHandle "Hello World"    #+end_src +** Software Transactional Memory++When invoking xmobar from other Haskell code it can be easier and+more performant to use shared memory.  The following plugins leverage+=Control.Concurrent.STM= to realize these gains for xmobar.++*** =QueueReader (TQueue a) (a -> String) String=++ - Display data from a Haskell =TQueue a=.++ - This plugin is only usefule if you are running xmobar from another

yes, i think we're fine right now... and i usually prefer ff, yes, although it's a big deal... i'll let the CI run and merge!

guygastineau

comment created time in 6 days

issue commentjaor/xmobar

Pipewire and Pulseaudio monitors?

i think it'd be nice, yes.

jsoo1

comment created time in 6 days

PullRequestReviewEvent

pull request commentjaor/xmobar

Add the QueueReader plugin.

oh, forget about my rant on the placement of the doc section. it's already where i was asking!!!

guygastineau

comment created time in 6 days

Pull request review commentjaor/xmobar

Add the QueueReader plugin.

 choice.      hPutStr writeHandle "Hello World"    #+end_src +** Software Transactional Memory++When invoking xmobar from other Haskell code it can be easier and+more performant to use shared memory.  The following plugins leverage+=Control.Concurrent.STM= to realize these gains for xmobar.++*** =QueueReader (TQueue a) (a -> String) String=++ - Display data from a Haskell =TQueue a=.++ - This plugin is only usefule if you are running xmobar from another

typo: useful

guygastineau

comment created time in 6 days

PullRequestReviewEvent

pull request commentjaor/xmobar

Add the QueueReader plugin.

Thanks Guy, this looks good... i think exposing in the library stripActions is a good idea; having also a safe version would be useful for people not using xmobar as a library, but, iiuc, they won't be using this module at all! would it be possible to have some docs for it in doc/plugins.org (somewhere near https://github.com/jaor/xmobar/blob/master/doc/plugins.org#L1418)? Thanks again!

guygastineau

comment created time in 6 days

PullRequestReviewEvent

pull request commentjaor/xmobar

Added link to a list of station ids

merged, thanks!

NebuPookins

comment created time in 18 days

push eventjaor/xmobar

NebuPookins

commit sha 3a5004e39e6dc42f2691012b4513a210202311b2

Added link to list of station ids. (#565)

view details

push time in 18 days

PR merged jaor/xmobar

Added link to a list of station ids
+1 -1

0 comment

1 changed file

NebuPookins

pr closed time in 18 days

issue commentjaor/xmobar

Enabling extensions when using xmobar as a library?

this is my xmobar-config repo, which uses xmobar as a library with cabal to create three different xmobar executables:

https://codeberg.org/jao/xmobar-config

i have together with it a checkout of xmobar, and xmobar-config uses that and compiles it with --all-extensions.

ssbothwell

comment created time in 19 days

issue commentjaor/xmobar

Wireless plugin signal/quality not displaying

On Tue, Aug 31 2021, Justin Dove wrote:

It looks to me like the debian build for xmobar uses iwlib: https://sources.debian.org/src/xmobar/0.36-2/debian/rules/.

possibly not a big influence, but that's a few versions behind the latest release of xmobar.

I'm still curious how other people handle this though. Or is it unusual that I can't usefully cat /proc/net/wireless or otherwise get link signal/quality without sudo? Google suggests that there are at least a few others in that boat.

i'm on debian unstable, amd64, and /proc/net/wireless is readable by all users, no need to use sudo:

$ ls -l /proc/net/wireless 
-r--r--r-- 1 root root 0 Sep  1 19:36 /proc/net/wireless
$ cat /proc/net/wireless 
Inter-| sta-|   Quality        |   Discarded packets               | Missed | WE
 face | tus | link level noise |  nwid  crypt   frag  retry   misc | beacon | 22
wlp1s0: 0000   70.  -39.  -256        0      0      0      0     20        0
dovej

comment created time in 19 days

push eventbigmlcom/bigml-php

Mercè Martín Prats

commit sha fb61b7e8adca223dc853ff818239c99bd359a170

PHP version bump to 8.0 and other upgrades

view details

Mercè Martín Prats

commit sha 9ffe98467cf8dae588907dcf19519f3069240faa

Phpunit version bump

view details

Mercè Martín Prats

commit sha 18082809c104d20de59bb3ee481c25822ea1d992

Changin PHP tests version

view details

Mercè Martín Prats

commit sha aad27f592bce6676c642218d6c3d17ae25615a0f

Fixing actions .yml

view details

Mercè Martín Prats

commit sha 1bd839819e2730783f980b8b0e930ed2d4c829ca

Fixing actions files

view details

Mercè Martín Prats

commit sha 2014c4a99ba4e442344b4a1ccc0a9fa99dfb2128

Trace to check phpunit version

view details

Mercè Martín Prats

commit sha 00e10cac1fc384b81048663e484ef65bb8eeab2c

Changing phpunit version

view details

Mercè Martín Prats

commit sha b28eb4bcb1d5ec0778f6d9482ba69f676603287b

Changing test names to comply with Phpunit 9.3

view details

Mercè Martín Prats

commit sha 7f052a7b096f9081bb90563ea79819ee13d3b5e4

Fixing typo in fields class

view details

Mercè Martín Prats

commit sha c9bbce535e390bedc45186b97f12b5c45ac793cc

Fixing using multivote for probabilities and fields class

view details

Mercè Martín Prats

commit sha 9b95329f78914cea9599ddeb1762ec573b4091e2

Fixing dinamic property syntax

view details

Mercè Martín Prats

commit sha 09cc4fd3a939fad00c9663d558e96d520d27bb89

Fixing multimodel batch predictions

view details

push time in a month

created tagbigmlcom/bigml-php

tagv4.0.0

BigML.io Php Bindings

created time in a month

PR merged bigmlcom/bigml-php

PHP version bump to 8.0 and other upgrades
+156 -237

0 comment

68 changed files

mmerce

pr closed time in a month

PullRequestReviewEvent

push eventjaor/xmobar

jao

commit sha 339c07746487dc80735fbcf38689b07ffa2a1ccc

Typo

view details

push time in a month

release jaor/xmobar

0.39

released time in a month

created tagjaor/xmobar

tag0.39

A minimalistic status bar

created time in a month

push eventjaor/xmobar

jao

commit sha b8bcc61a632f49533ff476b3483cb7299866d5dc

readme and changelog

view details

push time in a month

PR merged jaor/xmobar

Fix kbd crashes

Fixes #313 and some related corner cases too. Ran into this myself yesterday, thought I could fix this while at it.

The main point is avoiding partial functions as to not crash/freeze; the rest is mostly fluff, although I did fix what I believe to be a memory leak in Xmobar.System.Kbd. Also, vmods, indicators and groups were marshalled incorrectly, so I also fixed that.

I am somewhat unsure about applying map (map toLower . take 2) to group names. Perhaps first checking for matches with the full group name in KbdOpts and preferring that is a good idea? Let me know what you think.

+58 -34

1 comment

2 changed files

lierdakil

pr closed time in a month