From mboxrd@z Thu Jan 1 00:00:00 1970 From: Tatiana Sholokhova Subject: Re: Web interface review Date: Sun, 22 Jul 2018 16:57:18 +0200 Message-ID: References: <87sh4encn4.fsf@lassieur.org> <87k1pq5k87.fsf@lassieur.org> <87h8ksmyn4.fsf@lassieur.org> Mime-Version: 1.0 Content-Type: multipart/alternative; boundary="0000000000007008fe057197bf4e" Return-path: Received: from eggs.gnu.org ([2001:4830:134:3::10]:55984) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1fhFnf-0007Q6-Sg for guix-devel@gnu.org; Sun, 22 Jul 2018 10:57:33 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1fhFne-00007K-Pm for guix-devel@gnu.org; Sun, 22 Jul 2018 10:57:31 -0400 Received: from mail-oi0-x22e.google.com ([2607:f8b0:4003:c06::22e]:36408) by eggs.gnu.org with esmtps (TLS1.0:RSA_AES_128_CBC_SHA1:16) (Exim 4.71) (envelope-from ) id 1fhFne-000078-J2 for guix-devel@gnu.org; Sun, 22 Jul 2018 10:57:30 -0400 Received: by mail-oi0-x22e.google.com with SMTP id n21-v6so13598953oig.3 for ; Sun, 22 Jul 2018 07:57:30 -0700 (PDT) In-Reply-To: <87h8ksmyn4.fsf@lassieur.org> List-Id: "Development of GNU Guix and the GNU System distribution." List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: guix-devel-bounces+gcggd-guix-devel=m.gmane.org@gnu.org Sender: "Guix-devel" To: =?UTF-8?Q?Cl=C3=A9ment_Lassieur?= Cc: guix-devel --0000000000007008fe057197bf4e Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable Hi Clement, Thank you for your advice! I resolved the 'with-critical-section' issue. To do so, I had to remove 'with-critical-section' call from 'handle-build-request' function and wrap critical section around each call of this function. Today I am trying to fix the pagination and I will let you know about my results. P. S. Now I send this changes as separate small commits, we can rebase the history again when I finish with pagination. Best regards, Tatiana =D1=81=D0=B1, 21 =D0=B8=D1=8E=D0=BB. 2018 =D0=B3. =D0=B2 16:50, Cl=C3=A9men= t Lassieur : > Hello Tatiana! > > Tatiana Sholokhova writes: > > > Hello Cl=C3=A9ment! > > > > Thank you for your review! > > > > I fixed most of the problems you noticed and rebased commits as you > advised > > . > > > > I couldn't fix the problem with several calling of > (with-critical-section). > > As I wrote to IRC channel, I tried to put '(with-critical-section > > db-channel (db)' around '(let* ...)' and I received an error: > > ``` > > In web/server.scm: > > 279:25 0 (_) > > Throw to key `vm-error' with args `(vm-run "Too few values returned to > > continuation" ())'. > > ``` > > It's because 'respond-html' returns several values. I think you could > do: > > (respond-html > (with-critical-section ... > (let* ...))) > > > Could you give a status about the pagination? > >> > > Pagination works correctly with evaluations, but it doesn't work > correctly > > with builds. In some cases, we have builds missing. It happens due to > equal > > timestamp values, so we need to filter build by (timestamp, id) tuple > key. > > > > What else do we need to do before the merge? > > Once we have something consistent, we can push. And we can add stuff > afterwards of course. Do you think it would be feasible to fix the > pagination before the merge? > > I won't have time to look at your update before tomorrow night, I'll let > you know then! > > Thanks, > Cl=C3=A9ment > --0000000000007008fe057197bf4e Content-Type: text/html; charset="UTF-8" Content-Transfer-Encoding: quoted-printable
Hi Clement,

Thank you for your advice! = I resolved the 'with-critical-section'=C2=A0issue. To do s= o, I had to remove 'with-critical-section' call from 'handle-bu= ild-request' function and wrap critical section around each call of thi= s function.
=C2=A0
Today I am trying=C2=A0to fix the pa= gination and I will let you know about my results.

P. S.
Now I send this changes as separate small commits, we can = rebase the history again when I finish with pagination.

Best regards,
Tatiana


=D1=81=D0=B1, 21 =D0=B8=D1=8E=D0=BB. 2= 018 =D0=B3. =D0=B2 16:50, Cl=C3=A9ment Lassieur <clement@lassieur.org>:
Hello Tatiana!

Tatiana Sholokhova <tanja201396@gmail.com> writes:

> Hello Cl=C3=A9ment!
>
> Thank you for your review!
>
> I fixed most of the problems you noticed and rebased commits as you ad= vised
> .
>
> I couldn't fix the problem with several calling of (with-critical-= section).
> As I wrote to IRC channel, I tried to put '(with-critical-section<= br> > db-channel (db)' around '(let* ...)' and I received an err= or:
> ```
>=C2=A0 =C2=A0 In web/server.scm:
>=C2=A0 =C2=A0 279:25=C2=A0 0 (_)
> Throw to key `vm-error' with args `(vm-run "Too few values re= turned to
> continuation" ())'.
> ```

It's because 'respond-html' returns several values.=C2=A0 I thi= nk you could
do:

(respond-html
=C2=A0 (with-critical-section ...
=C2=A0 =C2=A0 (let* ...)))

> Could you give a status about the pagination?
>>
> Pagination works correctly with evaluations, but it doesn't work c= orrectly
> with builds. In some cases, we have builds missing. It happens due to = equal
> timestamp values, so we need to filter build by (timestamp, id) tuple = key.
>
> What else do we need to do before the merge?

Once we have something consistent, we can push.=C2=A0 And we can add stuff<= br> afterwards of course.=C2=A0 Do you think it would be feasible to fix the pagination before the merge?

I won't have time to look at your update before tomorrow night, I'l= l let
you know then!

Thanks,
Cl=C3=A9ment
--0000000000007008fe057197bf4e--