From minus to ~sircmpwn/sr.ht-dev
On 01/12/2021 09.51, Drew DeVault wrote: > Not in love with the "direct" TLS nomenclature, but I don't have any > better ideas at hand. It's usually called implicit TLS
From minus to ~sircmpwn/sr.ht-discuss
> This is most likely due to using a non-standard SSH port. What could I do to use a non-standard SSH port without exposing it in the build manifest and the job log?
There's no point in hiding the port. Anyone with a port scanner can
trivially find it within minutes. The only benefit of running on a
non-standard port is less log spam.
From minus to ~sircmpwn/sr.ht-dev
On 02/04/2021 04.33, Drew DeVault wrote: > NACK to all of these patches. We don't support this installation method. > Don't waste your time. Then why bother with having an install_requires section at all if it's incomplete and obviously not necessary?
From minus to ~sircmpwn/sr.ht-dev
On 12/03/2021 14.33, Drew DeVault wrote:
> This doesn't make sense. Foreign keys are always indexed.
PostgreSQL does not automatically create indices for foreign keys, it
only does so on unique constraints and primary keys.
From minus to ~sircmpwn/sr.ht-dev
On 30/12/2018 19.43, Drew DeVault wrote: > I'm allergic to polling. One thing builds.sr.ht has over the competetion > is the boot-up speed - you submit a job and it's running by the time you > take your next breath. I want to get that even faster still, not slower. I see. I suggested polling because it's simpler and works reasonably fast with a polling frequency of a couple of seconds. In that case, I'd still prefer a custom subscription system over HTTP (WebSockets or SSE) or TCP, where you connect, send your desired subscription/advertise your capabilities and then wait for new build requests on that connection. This allows for complex filtering and that the coordinator directly sees which worker is responsible for the build. I'm not sure how this would work with AMQP with fixed queues.
From minus to ~sircmpwn/sr.ht-dev
Hi there, I'm not using builds.sr.ht, but I can offer some feedback, partly based on experience with similar projects. > The first change will be to switch from Celery to RabbitMQ for build > distribution. Workers polling an API may be a simpler alternative offering more control over distribution (since you don't push jobs into a queue but have them pulled by the workers). That also avoids having to maintain and know RabbitMQ (or Celery for that matter). At least that's my experience from a cancelled AMQP project.
From minus to ~sircmpwn/sr.ht-dev
--- installation.md | 20 ++++++++++++++++---- 1 file changed, 16 insertions(+), 4 deletions(-) diff --git a/installation.md b/installation.md index b601dba..d0202a8 100644 --- a/installation.md +++ b/installation.md @@ -47,9 +47,9 @@ sr.ht a consistent look and feel. The core package is listed as a dependency of the official `*.sr.ht` packages, so installing it explicitly is not necessary if you are using our package repositories. If you are not using our packages, obtain the [source code](https://git.sr.ht/~sircmpwn/srht) and install it like a typical Python package (`./setup.py install --prefix=/usr`). You will need to install its[message trimmed]
From minus to ~sircmpwn/public-inbox
--- listssrht/templates/archive.html | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/listssrht/templates/archive.html b/listssrht/templates/archive.html index 7a662a7..e2d9fcc 100644 --- a/listssrht/templates/archive.html +++ b/listssrht/templates/archive.html @@ -66,7 +66,7 @@ <a href="mailto:{{ post_address(ml) }}" class="btn btn-default btn-block" >New pos {{icon("caret-right")}}t</a> >New post {{icon("caret-right")}}</a>[message trimmed]
From minus to ~sircmpwn/public-inbox
On 2018-07-08 21:57, Drew DeVault wrote: > On 2018-07-08 9:51 PM, minus wrote: >> diff --git a/README.md b/README.md >> index ddcd6b3..fe33d06 100644 >> --- a/README.md >> +++ b/README.md >> @@ -103,10 +103,10 @@ You can become an admin like so: >> - >>> u.approvalDate = datetime.now() >> + >>> u.approvalDate = datetime.now(timezone.utc) > > Prefer datetime.utcnow(), here and throughout the patchutcnow() doesn't work, as the resulting datetime object is not timezone-aware. I tried that first :) > >> diff --git a/alembic/versions/07477ca58bf4_timestamps_with_time_zone.py b/alembic/versions/07477ca58bf4_timestamps_with_time_zone.py