~euantorano

England

https://www.euantorano.co.uk

Recent activity

Re: builds.sr.ht CI doesn't merge PR's against repo HEAD, unlike other CI services, causing issues a month ago

From Euan Torano to ~sircmpwn/sr.ht-discuss

Yeah, I think that's a good call. There are a __lot__ of tests being ran, we can certainly disable some of them.

On Mon, 22 Feb 2021, at 14:02, Drew DeVault wrote:
> On Sun Feb 21, 2021 at 10:42 PM EST, Timothee Cour wrote:
> > I also wasn't aware of resource usage issues; we are now using 2
> > instead of 3 openbsd jobs, hopefully this help; openbsd CI is now the
> > bottleneck in our CI jobs, and reducing it to 1 job would double our
> > CI running time; we are tracking in
> > https://github.com/nim-lang/Nim/issues/17107 for ways to improve CI on
> > openbsd with various approaches
> > (https://github.com/nim-lang/Nim/issues/17107#issuecomment-782232960)
> > so we can bring it down to just 1 job on openbsd.
> 
> Looks like your new merged builds are timing out, which isn't good

Re: builds.sr.ht CI doesn't merge PR's against repo HEAD, unlike other CI services, causing issues 2 months ago

From Euan Torano to ~sircmpwn/sr.ht-discuss

Good point RE bash-isms, will fix - thanks!

Fair enough. I did originally use a single job but things got changed. I've reported this to the project and will be opening a PR shortly to fix this - the last thing I intended was to cause problems for you or other users! Issue ref: https://github.com/nim-lang/Nim/issues/17107

If there are any other concerns with Nim's usage or anything, please do feel free to contact me. I usually lurk in IRC too if you need a more prompt response.

On Fri, 19 Feb 2021, at 14:50, Drew DeVault wrote:
> On Fri Feb 19, 2021 at 9:47 AM EST, Euan Torano wrote:
> > I __think__ I've solved it like this:
> > https://github.com/nim-lang/Nim/pull/17104
> 
> LGTM, but note that your bashisms (`if [[`) may stop working without
> notice as we plan on updating each image to use its native shell at some
> point in the future (for the BSDs, this would be a more POSIXy shell,

Re: builds.sr.ht CI doesn't merge PR's against repo HEAD, unlike other CI services, causing issues 2 months ago

From Euan Torano to ~sircmpwn/sr.ht-discuss

I __think__ I've solved it like this: https://github.com/nim-lang/Nim/pull/17104

I wasn't aware of any complaints RE OpenBSD utilisation issues, but I know that the original build I wrote got split into 3 separate builds because the build was taking a long time on OpenBSD. Any chance you can summarise the issues and I'll report them to Nim and hopefully find some time to look into them?

On Fri, 19 Feb 2021, at 13:57, Drew DeVault wrote:
> I think you should be able to achieve this behavior yourself with an
> extra task step. Test if you're in a pull request (the environment
> variables will help here), and if so, fetch upstream and check out a
> branch which looks the way you want. If a merge conflict occurs, fail
> the build.
> 
> I might be more open to helping you with this if you guys fixed the
> OpenBSD utilization issues I've been complaining about for the last few
> weeks :)

Re: builds.sr.ht CI doesn't merge PR's against repo HEAD, unlike other CI services, causing issues 2 months ago

From Euan Torano to ~sircmpwn/sr.ht-discuss

I think that if nothing else the documentation could use a call out warning users coming from other services of the difference in behaviour. That may help save some confusion at least. 

I’m going to look at updating the build scripts in question to do the merge into the main branch before running the CI. 

On Fri, 19 Feb 2021, at 07:40, Matthias Beyer wrote:
> On 19-02-2021 07:20:53, Euan Torano wrote:
> > the current behaviour is unexpected
> 
> I disagree. It is not unexpected, it is just not doing anything *implicitely*,
> that's a huge difference IMHO.
> 
> Right now, from my limited perspective as a mere user of builds, it does _very_
> little (install stuff, set ENV, clone the repos and run the script), and I like
> that, because it still gives the user power to do more explicitely if they want

Re: builds.sr.ht CI doesn't merge PR's against repo HEAD, unlike other CI services, causing issues 2 months ago

From Euan Torano to ~sircmpwn/sr.ht-discuss

It probably is, but the current behaviour is unexpected should one have pre-conceptions based upon other offerings. I set up the CI scripts in question here and didn’t realise this was how builds.sr.ht behaved at all. 

I assume it functions differently for CI for patches (e.g. it behaves the same way as Travis/Azure Pipelines/etc.)?

On Fri, 19 Feb 2021, at 07:14, Matthias Beyer wrote:
> On 18-02-2021 18:05:44, Timothee Cour wrote:
> > Can this be fixed in builds.sr.ht (following behavior of other CI's,
> > ie merge PR's against target branch HEAD on each push and on
> > closing/reopening) ?
> 
> I wonder: Isn't that possible via the CI script one provides in their
> repository?
> 
> -- 

Re: Undocumented services outage 3 months ago

From Euan Torano to ~sircmpwn/sr.ht-discuss

You may also create an empty commit and push that, rather than ammending:

    git commit --allow-empty -m "Trigger build"

On Fri, 8 Jan 2021, at 10:14, Nguyễn Gia Phong wrote:
> Thank you for the prompt response, Simon.
>

Re: [Discussion]: Note taking service 1 year, 5 months ago

From Euan Torano to ~sircmpwn/sr.ht-discuss

I'm a bit late to the discussion here, but ideally any note taking service would support the CalDAV protocol[1]. That way existing systems could automatically integrate with the service (for example, both iOS and macOS support CalDAV out of the box for the notes apps).

[1]: https://en.wikipedia.org/wiki/CalDAV

[PATCH man.sr.ht] Spelling and structural changes for Git documentation. 2 years ago

From to ~sircmpwn/sr.ht-dev

From: Euan Torano <euan@torano.co.uk>

Fixed some minor spelling mistakes in the index of the Git documentation, and fixed what appears to be an incorrect heading level.
 
---
 git.sr.ht/index.md | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/git.sr.ht/index.md b/git.sr.ht/index.md
index 422abcc..786c197 100644
--- a/git.sr.ht/index.md
+++ b/git.sr.ht/index.md
@@ -32,7 +32,7 @@ The git.sr.ht SSH host key is

[message trimmed]