I develop random projects to experiment and for fun.


Last active 5 months ago


Last active 2 years ago
View more

Recent activity

[PATCH gamja v3] Linkify channel names 3 months ago

From Tom Lebreux to ~emersion/public-inbox

Changes from v2:
- Rebased on master

Changes from v1:
- Fixed issues mentionned

As said in the v2 patch, I chose the TODO route for the channel name
regex, I could try experimenting with (\w|-|_) though if you prefer.
 components/app.js           | 17 ++++++++++++++-
 components/buffer-header.js |  2 +-
 components/buffer.js        |  9 +++++---
 lib/linkify.js              | 42 ++++++++++++++++++++++++++++++++++---
 4 files changed, 62 insertions(+), 8 deletions(-)
[message trimmed]

[PATCH gamja v2] Linkify channel names 3 months ago

From Tom Lebreux to ~emersion/public-inbox

Fixed issues from v1.

For channel name regex, I added a TODO. A possible option would be to
match on ([\w-_]+).

Curious, why using function(){} instead of const .. = () => {}
 components/app.js           | 17 ++++++++++++++-
 components/buffer-header.js |  2 +-
 components/buffer.js        | 14 +++++++++----
 lib/linkify.js              | 42 ++++++++++++++++++++++++++++++++++---
 4 files changed, 66 insertions(+), 9 deletions(-)
[message trimmed]

[PATCH gamja] Linkify channel names 3 months ago

From Tom Lebreux to ~emersion/public-inbox

As discussed, this creates links for channel names and on click it will
automatically join the channel and switch to the buffer.

I am not sure if there is a better way to detect channel names, and I
also wasn't sure if we wanted to detect all channel (# and &).
 components/app.js           | 17 ++++++++++++++-
 components/buffer-header.js |  6 +++++-
 components/buffer.js        | 19 +++++++++++++----
 lib/linkify.js              | 42 ++++++++++++++++++++++++++++++++++---
 4 files changed, 75 insertions(+), 9 deletions(-)

diff --git a/components/app.js b/components/app.js
index 38d3cb46cc9b..5c482c3e0775 100644
[message trimmed]

Re: [builds.sr.ht] Make bash default shell on debian 4 months ago

From Tom Lebreux to ~sircmpwn/sr.ht-discuss

On 2021-05-26 6:18 p.m., Askar Safin wrote:
>> If you’re writing a portable script (which you should be for build manifests)
>> you shouldn’t be assuming that the shell is bash in any case, regardless of
>> the distribution. If you need to use bash, run /bin/bash explicitly.
> My original mail was not about shell used to execute tasks. My mail is about shell I connect to when I type "ssh -t builds@azusa.runners.sr.ht connect ...". I want this shell to be bash, because this is convenient. And because bash is default login shell in debian (i. e. default shell for accounts created using "adduser")

Wouldn't it be simpler if the default shell when you `ssh -t 
builds@azusa.runners.sr.ht connect ...` was the same as the shell being 
used by the manifest though?

I would find it confusing if I would try commands in a debug ssh session 
and those would behave differently in the manifest.

[PATCH] testing/navidrome: new aport 5 months ago

From Tom Lebreux to ~sircmpwn/alpine-aports

Modern Music Server and Streamer compatible with Subsonic/Airsonic
 testing/navidrome/APKBUILD              |  64 +++++++++++++++
 testing/navidrome/navidrome.confd       |   4 +
 testing/navidrome/navidrome.initd       |  20 +++++
 testing/navidrome/navidrome.pre-install |   6 ++
 testing/navidrome/navidrome.toml        | 102 ++++++++++++++++++++++++
 5 files changed, 196 insertions(+)
 create mode 100644 testing/navidrome/APKBUILD
 create mode 100644 testing/navidrome/navidrome.confd
 create mode 100644 testing/navidrome/navidrome.initd
 create mode 100644 testing/navidrome/navidrome.pre-install
 create mode 100644 testing/navidrome/navidrome.toml
[message trimmed]

[PATCH] testing/openrc-exporter: upgrade to 0.0.2 and fix command 5 months ago

From Tom Lebreux to ~sircmpwn/alpine-aports

 testing/openrc-exporter/APKBUILD              | 6 +++---
 testing/openrc-exporter/openrc-exporter.initd | 2 +-
 2 files changed, 4 insertions(+), 4 deletions(-)

diff --git a/testing/openrc-exporter/APKBUILD b/testing/openrc-exporter/APKBUILD
index e7338283f9fd..3e64a7f1ee82 100644
--- a/testing/openrc-exporter/APKBUILD
+++ b/testing/openrc-exporter/APKBUILD
@@ -1,7 +1,7 @@
# Contributor: Tom Lebreux <me@tomlebreux.com>
# Maintainer: Tom Lebreux <me@tomlebreux.com>
[message trimmed]

Re: Opt out of FLoC 5 months ago

From Tom Lebreux to ~sircmpwn/sr.ht-discuss

On 2021-04-16 5:38 a.m., Simon Ser wrote:
> Should already be done.
>      > curl -v -X HEAD https://meta.sr.ht
>      […]
>      < permissions-policy: interest-cohort=()

Would it make sense to add the header to pages hosted on page.sr.ht?


	> curl -v -X HEAD https://pages.sr.ht

[PATCH] testing/openrc-exporter: new aport 5 months ago

From Tom Lebreux to ~sircmpwn/alpine-aports

Prometheus exporter for OpenRC
First package here so hopefully I got everything correct.

I doubt creating the user openrc-exporter is required, but I didn't want
to run the exporter as root. Maybe `nobody` should be used instead?

The port number has been allocated officially with Prometheus through
this page: https://github.com/prometheus/prometheus/wiki/Default-port-allocations
 testing/openrc-exporter/APKBUILD              | 44 +++++++++++++++++++
 testing/openrc-exporter/openrc-exporter.confd |  3 ++
 testing/openrc-exporter/openrc-exporter.initd | 13 ++++++
 .../openrc-exporter.pre-install               |  6 +++
[message trimmed]

[PATCH] Document SELinux error 5 months ago

From Tom Lebreux to ~kennylevinsen/greetd-devel

 index.md | 10 ++++++++++
 1 file changed, 10 insertions(+)

diff --git a/index.md b/index.md
index c540c319e5ce..90dc2b73c31b 100644
--- a/index.md
+++ b/index.md
@@ -210,6 +210,16 @@ The default greetd configuration uses `agreety` (which looks just like `agetty`

You can change the configuration as mentioned in the steps above to make `agreety` start another login environment of your choice (e.g. `sway`), or use a different greeter altogether.

## greetd fails with `unable to exec: Sys(EACCES)`

[message trimmed]

Re: Unable to list HEAD with graphql 8 months ago

From Tom Lebreux to ~sircmpwn/sr.ht-dev

On 2021-01-25 10:12 a.m., Drew DeVault wrote:
> Turns out that this was a GraphQL validation error. You need to select
> some sub-fields from the HEAD object; it's a Reference type. I've
> written a patch which displays GraphQL validation errors on the explorer
> UI instead of a generic error message.

It's much better now with the validation error in the explorer UI. Thanks.

> (re-sent due to some lists.sr.ht issues, bear with me)

Maybe related: Is this why I don't see your response on lists.sr.ht[0]?