Dev July 2013

dev@lists.roundcube.net

13 participants
19 discussions

Larry: improved attachment preview page?
by A.L.E.C 01 Sep '13

01 Sep '13

While I work on a new plugin for files management. I found that we could improve our current attachment preview page. I'm attaching a screenshot from my plugin. I'd like to make attachment preview page to look the same, i.e. attachment properties on the left and toolbar with buttons (Download for now, but we could add other e.g. Print, Save, Edit). What do you think? An opposite layout would be to make it more looking as message preview page with properties on top and small buttons in right-top corner (and no toolbar). I however, prefer the solution with toolbar. -- Aleksander 'A.L.E.C' Machniak LAN Management System Developer [http://lms.org.pl] Roundcube Webmail Developer [http://roundcube.net] --------------------------------------------------- PGP: 19359DC1 @@ GG: 2275252 @@ WWW: http://alec.pl

3 4

Invalid or Expired Session
by Rodrigo Castillo 31 Jul '13

31 Jul '13

We've had sporadic reports of expired or invalid sessions from users. They'll have a roundcube page open, possibly other tabs, too, then click on compose or on an email message and be presented with a login screen that says the session is invalid or expired. We're thinking it's possibly related to js getting stalled or the keep_alive not being run due to the page being 'busy' for any other reason, but reproducing the behavior, let alone diagnosing it, has been difficult. Our sessions are in mysql and have a length of 10min, keep_alive runs every 60s. Has anyone else had reports like this?

2 1

turning off a plugin from another plugin
by Cor Bosman 30 Jul '13

30 Jul '13

Hey all, anyone know if it's possible to turn off a plugin from another plugin? only certain people are currently allowed to use managesieve in our setup, and right now I have some hacks inside managesieve to do that but it would be cleaner if I could do that from a separate plugin, cor

2 2

Jenkins integration
by Michée Lengronne 29 Jul '13

29 Jul '13

Hello, I try to integrate Roundcube into Jenkins but the ant build seems to have a problem. Is there an official build.xml file ? Do you use Jenkins for continuous integration ? Thanks. -- Signature Michée Lengronne Coppint Michée Lengronne ------------------------------------------------------------------------ Ce message et ses éventuelles pièces jointes sont personnels, confidentiels et à l'usage exclusif de leur destinataire. Si vous n'étes pas la personne à laquelle ce message est destiné, veuillez noter que vous avez reçu ce courriel par erreur et qu'il vous est strictement interdit d'utiliser, de diffuser, de transférer, d'imprimer ou de copier ce message. This e-mail and any attachments hereto are strictly personal, confidential and intended solely for the addressee. If you are not the intended recipient, be advised that you have received this email in error and that any use, dissemination, forwarding, printing, or copying of this message is strictly prohibited. ------------------------------------------------------------------------ Logo CopPInt 30440 Saint Laurent le Minier France *Compagnie pour des Prestations Internet* Téléphone personnel : +41 (0)791 94 71 98 Téléphone professionnel : +33 (0) 4 84 25 23 67 poste 102 Téléphone professionnel : 0805 696 8 9 0 poste 102 France uniquement Courriel : michee.lengronne(a)Coppint.com <mailto:michee.lengronne@coppint.com> ------------------------------------------------------------------------

1 0

Re: [RCD] [RCU] Unexpected session timeouts using memcached session store
by Thomas Bruederli 23 Jul '13

23 Jul '13

Hi David Thanks a lot for the analysis and the proposed solution! Memcache support is still considered experimental and not tested by a broad audience. But your observations and the proposed fix seem reasonable and just applied the changes to git. Best regards, Thomas David Carter wrote: > We are currently running a pilot Roundcube service and are expecting to go > live a a couple of weeks time. Something that my users have picked up on is > that idle sessions appear to time out in unpredictable ways despite the > keep_alive requests fired off by the Web browser. > > A bit of digging over the weekend revealed a potential race condition in > program/lib/Roundcube/rcube_session.php : mc_write(). (Roundcube 0.9.2). > > I'm happy to write the following up as a ticket on trac.roundcube.net. I > wanted to make sure my analysis and proposed solution was correct first. > > Given the 10 minute (600 second) default session_lifetime, keep_alive > requests are fired off every 300 seconds. However these may not arrive > immediately, particularly if the Web browser, network or server is busy. > > Sessions stored in memcached expire after precisely 600 seconds unless they > are refreshed. > > mc_write() only refreshes session entries if the following is true: > > ($newvars !== $oldvars || $ts - $this->changed > $this->lifetime / 2) > > Most of the time with a idle login session: > ($this->lifetime / 2) will be 300 > ($ts - $this->changed) will be a tiny fraction more than 300. > > However if I add some logging it becomes clear that ($ts - $this->changed) > is occasionally less than 300. I imagine this happens when the previous > keep_alive was stuck in transit for a fraction of a second. It is an > asynchronous request and the timer for the next keep_alive doesn't wait. > > The mc_write() refresh on the session data is skipped when this happens. > > The following keep_alive is scheduled to arrive precisely the same second > that the session is due to expire in memcached. If you are unlucky the > keep_alive arrives just after the session has been expired, which leads a > "Session is invalid or has expired" error page. > > This is using memcached for session storage: the problem appears to go away > if I switch to using a database backend. I believe that this is because > session entries stored in memcached expire after precisely 600 seconds. > With a database backend sessions are expired by PHP garbage collection > later on. This will add a few seconds delay, at least on a inactive system. > > The attached patch appears to fix the problem, using two separate approaches: > > 1) A lower threshold for deciding when to refresh the session entry. (I am > tempted to remove the guard condition altogether and always update. > There isn't any I/O cost associated with a memcached back end). > > 2) A slightly longer expiry time for entries stored in memcached. > > However I may be missing some subtlety about the refresh process. > > -- > David Carter Email: David.Carter(a)ucs.cam.ac.uk > University Computing Service, Phone: (01223) 334502 > New Museums Site, Pembroke Street, Fax: (01223) 334679 > Cambridge UK. CB2 3QH. > > --- program/lib/Roundcube/rcube_session.php-DIST 2013-07-21 > 12:11:49.000000000 +0100 > +++ program/lib/Roundcube/rcube_session.php 2013-07-22 > 10:00:40.000000000 +0100 > @@ -310,9 +310,9 @@ > > $newvars = $oldvars !== null ? $this->_fixvars($vars, $oldvars) : > $vars; > > - if ($newvars !== $oldvars || $ts - $this->changed > > $this->lifetime / 2) { > + if ($newvars !== $oldvars || $ts - $this->changed > > $this->lifetime / 3) { // DPC > return $this->memcache->set($key, serialize(array('changed' => > time(), 'ip' => $this->ip, 'vars' => $newvars)), > - MEMCACHE_COMPRESSED, $this->lifetime); > + MEMCACHE_COMPRESSED, $this->lifetime + 60); // DPC > } > > return true; > _______________________________________________ > Roundcube Users mailing list > users(a)lists.roundcube.net > http://lists.roundcube.net/mailman/listinfo/users

1 0

Configuration changes in git master
by Thomas Bruederli 22 Jul '13

22 Jul '13

Hello devs With today's commit [1] we took the first step towards a new and simplified configuration for Roundcube. We merged the two files main.inc.php and db.inc.php into one single config.inc.php file and a defaults.inc.php file respectively. The config.inc.php is meant to be the local config file and only containing those options which differ from the defaults. The defaults.inc.php will be part of the distribution is will be updated with every software update and thus should not be modified. The current git master will remain using the old config files as a fallback but we strongly recommend to upgrade your configs if you're running a bleeding edge git master version somewhere. You can do this by running bin/update.sh and letting the script migrate your config. The manual way of doing this is to copy config.inc.php.sample file and start from scratch. There's more to come related to configs [2] but this was the major change that just found its way back into git master. Kind regards, Thomas [1] https://github.com/roundcube/roundcubemail/commit/0703bfecd0 [2] http://trac.roundcube.net/ticket/1487311

3 8

RC in iframe: print, download, show source and open in new window button don't work properly.
by Michée Lengronne 22 Jul '13

22 Jul '13

Hello, I have roundcube embbeded in an iframe and I would like to have all the functionnalities working properly. Currently, the "print", "download", "show source" and "open in new window" buttons don't work properly in the iframe. They work properly in direct access. How can I adjust that ? Thanks. -- Signature Michée Lengronne Coppint Michée Lengronne ------------------------------------------------------------------------ Ce message et ses éventuelles pièces jointes sont personnels, confidentiels et à l'usage exclusif de leur destinataire. Si vous n'étes pas la personne à laquelle ce message est destiné, veuillez noter que vous avez reçu ce courriel par erreur et qu'il vous est strictement interdit d'utiliser, de diffuser, de transférer, d'imprimer ou de copier ce message. This e-mail and any attachments hereto are strictly personal, confidential and intended solely for the addressee. If you are not the intended recipient, be advised that you have received this email in error and that any use, dissemination, forwarding, printing, or copying of this message is strictly prohibited. ------------------------------------------------------------------------ Logo CopPInt 30440 Saint Laurent le Minier France *Compagnie pour des Prestations Internet* Téléphone personnel : +41 (0)791 94 71 98 Téléphone professionnel : +33 (0) 4 84 25 23 67 poste 102 Téléphone professionnel : 0805 696 8 9 0 poste 102 France uniquement Courriel : michee.lengronne(a)Coppint.com <mailto:michee.lengronne@coppint.com> ------------------------------------------------------------------------

2 1

Google Spellchecker
by Victor Benincasa 17 Jul '13

17 Jul '13

Hello Folks, The Google apparently turned off its spell check API since Jul 10, so the default Roundcube spell checker stopped working: https://www.google.com/tbproxy/spell Anybody have any additional/official information about it? Cheers, -- *Victor Benincasa* # *Mobile*: +55 (11) 99222-7227 | *Nextel*: 55*11*51775 | *Skype*: vbenincasa # *CTO / CIO* :: Networks, Applications, Systems, Security, Advanced Support

1 0

trac problem
by Michael Heydekamp 14 Jul '13

14 Jul '13

Just wanted to post a comment to http://trac.marketingxp.com/ticket/1488553, but when sending it, I get the following: ---------------------------------------------------------------------- > Trac wertet Ihre Übertragung als Spam. Um das Gegenteil zu beweisen, > geben Sie bitte Antwort auf folgendes. > > Input error: Invalid referer ---------------------------------------------------------------------- So I can't send the comment. What does "Invalid referer" mean and how can I avoid it? This spam warning is annoying anyway. Can't it be disabled at all?? I mean, I'm logged in, why does trac then think it might be spam...? Cheers, -- Michael Heydekamp Co-Admin freexp.de Düsseldorf/Germany

3 3

Re: [RCD] [Svn] [roundcube/roundcubemail] 73ad4f: Finally: make message list header stay on top when scrolling (#1295420)
by Michael Heydekamp 14 Jul '13

14 Jul '13

See commit quoted below: Larry only? :-( ...Seconds later...: Ah, the message LIST header, not the message header...? Ideally, both (message list header as well as message header) should stay on top upon scrolling - and of course in both skins, IMO. I'd still love to kill the folks who invented Larry... ;-) Seriously: It's unfortunate that the Classic skin doesn't get the same attention by the devs anymore as the Larry skin. Although the Classic skin is still the better usable one, as being less fancy. <sigh> Cheers, -- Michael Heydekamp Co-Admin freexp.de Düsseldorf/Germany Am 09.07.2013 23:42, schrieb GitHub: > Branch: refs/heads/master > Home: https://github.com/roundcube/roundcubemail > Commit: 73ad4f1bfdf104055104907c11f97315d6fb2ebe > > https://github.com/roundcube/roundcubemail/commit/73ad4f1bfdf104055104907c1… > Author: Thomas Bruederli <thomas(a)roundcube.net> > Date: 2013-07-09 (Tue, 09 Jul 2013) > > Changed paths: > M program/js/app.js > M program/js/list.js > M skins/larry/mail.css > M skins/larry/templates/mail.html > M skins/larry/ui.js > > Log Message: > ----------- > Finally: make message list header stay on top when scrolling (#1295420) > > > > > _______________________________________________ > Roundcube SVN commits mailing list > svn(a)lists.roundcube.net > http://lists.roundcube.net/mailman/listinfo/svn

2 2

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

Dev July 2013