Hi all,
CPAN’s Protocol::DBus now includes a connector for Mojolicious,
Protocol::DBus::Client::Mojo.
Hope it’s useful!
cheers,
-Felipe Gasper (FELIPE)
--
You received this message because you are subscribed to the Google Groups
"Mojolicious" group.
To unsubscribe from this grou
Hi all,
I’ve got a library called Net::Curl::Promiser that does the “dirty-work” of
interfacing between different event loops and Net::Curl::Multi. I recently
added a Mojo::IOLoop backend, so it should now be possible to integrate
libcurl seamlessly into Mojolicious via Net::Curl::Promiser::Moj
.
-FG
> Le 10 déc. 2019 à 13:19, Stefan Adams a écrit :
>
> Hi, Felipe! Thanks for sharing this! What advantages does integrating with
> libcurl have over using Mojo::UserAgent?
>
> On Tue, Dec 10, 2019, 8:41 AM Felipe Gasper wrote:
> Hi all,
>
> I’ve got a library
> Le 10 déc. 2019 à 13:57, sri a écrit :
>
> The disadvantage is that you’ll need to use Net::Curl::Easy, which
> isn’t a very Perlish interface. HTTP::AnyUA’s backend for same demonstrates
> how to “tame the beast”, though, and Net::Curl’s own documentation is quite
> helpful (IMO)
> Le 10 déc. 2019 à 14:08, Sebastian Riedel a écrit :
>
>>> Another big disadvantage is that it doesn't use Mojo::Promise. I don't
>>> think it will attract too many Mojo users without.
>>
>> The latest version does, actually.
>
> Ah, nice. Then you can simplify your example too.
>
>my
I’ve spent very little time with Mojo as a WS client, but I found the example
in Mojo::UserAgent’s documentation to be enough to get me where I needed to go.
=
$ua->websocket('ws+unix://%2Ftmp%2Fmyapp.sock/echo.json' => sub {
my ($ua, $tx) = @_;
say 'WebSocket handshake failed!' and retur
to use the new XS
code as an implementation when available.
I don’t know Mojo well enough to be at ease with doing the work, but if
anyone’s interested, it might make a nice improvement in Mojo to have XS-backed
promises when available.
-Felipe Gasper
Mississauga, Ontario
--
You
I’m just encountering Mojo::AsyncAwait and notice that it lists Coro as a
dependency.
Isn’t Coro problematic in newer Perls? I thought Marc had basically written off
support for 5.22 onward.
-F
> On Jan 10, 2020, at 8:49 AM, 'Tobias Oetiker' via Mojolicious
> wrote:
>
> I just read Joel's
Hello,
I see t/mojolicious/websocket_lite_app.t in the Mojolicious
distribution, but this seems to test Mojo as a WS server. Are there any
examples of using Test::Mojo--or any other tool--to test Mojo-powered WS
clients?
Thank you!
-FG
--
You received this message because yo
-Dan
>
> On Wed, Jan 22, 2020 at 10:38 AM Dan Book wrote:
> https://github.com/mojolicious/mojo/blob/master/t/mojo/websocket.t
>
> -Dan
>
> On Wed, Jan 22, 2020 at 10:34 AM Felipe Gasper
> wrote:
> Hello,
>
> I see t/mojolicious/websocket_lite_app.
> On Jan 22, 2020, at 12:25 PM, Felipe Gasper wrote:
>
> Hi Dan,
>
> Thank you for your response!
>
> So, maybe what I need isn’t Test::Mojo. I have:
>
> -
> package t::MockApp;
>
> use Mojo::Base -strict;
>
> use Mojolicious::Li
Hi Joel,
I’ve got a proof-of-concept branch of Net::ACME2 that uses
Net::Curl::Promiser rather than HTTP::Tiny; to use Mojo would likely be a
pretty similar change. If anyone’s interested in trying it out, let me know.
-FG
> On Feb 19, 2020, at 10:28 PM, Joel Berger wrote:
>
> As I'
https://developer.mozilla.org/en-US/docs/Web/API/FormData
^^ Also, if you’re not sure whether the problem is in Perl or in the browser,
you can use this to see what data the browser intends to send prior to sending.
-FG
> On Mar 2, 2020, at 10:51 AM, Dan Book wrote:
>
> That is correct, it s
https://github.com/FGasper/p5-Net-ACME2/tree/async
Supports Mojo via Net::Curl::Promiser.
Still experimental; file bugs as you find them. Enjoy!
-FG
--
You received this message because you are subscribed to the Google Groups
"Mojolicious" group.
To unsubscribe from this group and stop receiv
FWIW, you can use libcurl with Mojo via Net::Curl::Promiser::Mojo. That may be
your easiest solution?
-FG
> On May 14, 2020, at 07:38, 'Michael Lackhoff' via Mojolicious
> wrote:
>
>
>> Am 14.05.2020 um 05:08 schrieb Stefan Adams:
>> > I'm certain you'll be able to accomplish this with M::U
looks great but if it has some severe shortcomings I want to know before
> doing lots of mission critical stuff with it.
>
> But thanks for your suggestion. It is always good to know further options.
> -Michael
>
>
> Am Donnerstag, 14. Mai 2020 13:47:41 UTC+2 schrieb Feli
> On May 18, 2020, at 7:45 PM, Miguel Manso wrote:
>
> Hi there,
>
> Ive a Perl client that fetched some data every 3 seconds and post it to an
> API. Both client and server are written using mojo. Client uses
> Mojo::UserAgent to post JSON and server is a Mojolicious app being served
> wit
That seems an odd error from Chrome, if that’s the fix .. was the browser
sending a subprotocol in the handshake?
-F
> On Sep 14, 2020, at 07:37, Mikhail wrote:
>
> Thanks! Forever ♥ Mojolicious!
>
> понедельник, 14 сентября 2020 г. в 13:41:46 UTC+5, Sebastian Riedel:
>> Then you need to ne
RFC 6455/1.9 states:
The client can request that the server use a specific subprotocol by
including the |Sec-WebSocket-Protocol| field in its handshake. If it
is specified, the server needs to include the same field and one of
the selected subprotocol values in its response for the conne
> On Sep 15, 2020, at 9:30 AM, Sebastian Riedel wrote:
>
> Given this, should Mojo even send a response? Or if it does, should it at
> least warn “failed to negotiate subprotocol; sending handshake response
> anyway”?
>
> So you want Mojolicious to cause a TCP level connection error instead
> On Sep 15, 2020, at 11:20 AM, Sebastian Riedel wrote:
>
> An HTTP-level error response--maybe 426--would probably be most sensible.
>
> What error would common browsers give in that case?
They’d give whatever error as with any other HTTP-level failure during a WS
handshake.
Chrome shows
Hi all,
Does Mojo::UA offer a control to cancel an in-flight HTTP request? I
don’t see one in the docs but wonder if I’m just missing it.
(The promise, of course, can be pre-resolved/rejected, but that doesn’t
appear to affect the underlying HTTP request.)
Thank you!
c
=
… then strace that script, the request continues even after the call to
completed().
Am I just “holding it wrong”?
cheers,
-FG
> On Oct 23, 2020, at 10:13 AM, Dmitry L. wrote:
>
> https://metacpan.org/pod/Mojo::Transaction#completed
>
> On Fri, 23 Oct 2020 at 17:07, Felipe Gasper
> On Oct 23, 2020, at 10:42 AM, Dmitry L. wrote:
>
>> my $p = $ua->get_p('http://www.google.com');
>>
> https://metacpan.org/pod/Mojo::UserAgent#get_p returns a Mojo::Promise object
>
>> Mojo::Promise->timer(0.01)->then( sub { print "canceling\n"; $p->completed()
>> } );
>>
> https://metacp
> On Oct 23, 2020, at 10:42 AM, Dmitry L. wrote:
>
>> my $p = $ua->get_p('http://www.google.com');
>>
> https://metacpan.org/pod/Mojo::UserAgent#get_p returns a Mojo::Promise object
>
>> Mojo::Promise->timer(0.01)->then( sub { print "canceling\n"; $p->completed()
>> } );
>>
> https://metacp
I think you could do:
$tx = $tx;
… instead. (It might be worth updating the FAQ to do that?)
-F
> On Nov 16, 2020, at 8:08 AM, Edward Baudrez wrote:
>
> Hello list
>
> There is this hint in the FAQ about keeping a strong reference to the
> transaction object $tx around, so that the
Hello,
In a PR thread yesterday Sebastian mentioned that Mojo allows multiple
concurrent event loops.
When and why would someone want such a configuration?
Thank you!
cheers,
-Felipe Gasper
--
You received this message because you are subscribed to the Google Groups
27 matches
Mail list logo
