Pike-devel November 2011

pike-devel@lists.lysator.liu.se

5 participants
6 discussions

Sql.pgsql behavior in case of errors
by aldem-pikedev＠aldem.net 21 Aug '14

21 Aug '14

Hi everyone, I am preparing to use Postgres in my app, found the following strange behavior: ---snip--- Sql.Sql db = Sql.Sql("pgsql://localhost", "test", "test", "test"); db->query("BEGIN"); db->query("INSERT INTO requests (SvTrId) VALUES ('lala')"); catch(db->query("INSERT INTO requests (SvTrId) VALUES ('lala')")); db->query("INSERT INTO requests (SvTrId) VALUES ('juju')"); db->query("ROLLBACK"); ---snip--- Database is opened, row with 'lala' is inserted, second query (obviously) fails as SvTrId is the primary key, but row with 'juju' is inserted *and* appears in the database - i.e. the transaction is *committed*! Studying source code show, that once error is encounter, session is reset, i.e. running transaction is rolled back, everything is closed, temporary tables are deleted and session variables are reset as well, and after that, the next statement(s) will be committed immediately, as there is no active transaction. In my opinion, this is not exactly correct behavior when statement fails, as session still may be needed (especially temporary tables, open cursors etc). Im some cases, this is perfectly normal when SQL statement fails, especially in SERIALIZED transaction isolation level, and it would be more practical to leave the decision as of what to do to the application. Yes, this is true that it is impossible to continue executing statements in failed transaction, but temporary tables etc. may still be needed. Also, silent reconnection is not a good idea too - when there are many transactional statements are executed - it would be nice to have an option to modify this behavior. Best regards, /Alexander

5 6

kqueue based file event notification and the backend
by Bill Welliver 02 Dec '11

02 Dec '11

I've taken a cue from Arne's inotify module and have started working on one for kqueue based systems. The code basically works but suffers from a (imho) major drawback: it's not integrated with the Pike backend. The inotify code slides nicely into the backend, primarily because inotify presents itself as a standard fd with a read data event to indicate events. kqueue, however, uses different event types monitored by kqueue. Right now, a secondary kqueue loop can be run in a thread to wait for events, but that seems pretty ugly to me. I'd like to find a better approach. An alternate approach might be to hook a second, immediate-timeout call to kqueue() into the backend, similar to the GTK or ObjectiveC module. That's a little better, as at least the end user sees a more consistent approach to asynchronous events. Internally, I think it probably at least less efficient, with other possible drawbacks. I think the nicest approach would be to modify the kqueue backend so that it can register VNODE, PROC and SIGNAL filters and shunt them off to an appropriate handler. That, however, means that the backend would be split again (creating a KqueueBackend or similar), with the new implementation having different features. If anyone's interested in looking at the code, it's available here: https://bitbucket.org/hww3/public_system_kqueue/ I should note that I've commited some changes today away from my development box, so it's likely that the last few revisions have errors in them. Does anyone else have thoughts about these approaches (or have an alternate one?) Perhaps I've missed something obvious... it wouldn't be the first time! Best, Bill

3 3

Filesystem.Monitor and inotify
by H. William Welliver III 28 Nov '11

28 Nov '11

Did I hear correctly that someone had modified Filesystem.Monitor to use methods other than polling? Can anyone fill me in on what's been done in that department (including code)? I have 7.9.5 and it doesn't appear to be present… Best, Bill

2 1

pike.ida.liu.se offline?
by Ryan 18 Nov '11

18 Nov '11

I am unable to get to the pike homepage to download the latest bits and read the documents. Anyone have any information?

5 11

Re: [patch] allow multiple parts of a multipart form with the same name
by Erik Dahl 10 Nov '11

10 Nov '11

Maybe you should add the mime type as well? Something like (not tested): if(variables[part->disp_params->name+".mimetype"] && arrayp(variables[part->disp_params->name+".mimetype"])) variables[part->disp_params->name+".mimetype"] += ({part->disp_params->mimetype}); /erik This is a response to: http://lists.lysator.liu.se/pipermail/pike-devel/2011-October/014499.html

1 0

More of a Roxen question
by Coen Schalkwijk 10 Nov '11

10 Nov '11

Hi there, I have a Roxen specific question: we had a server breakdown and had to install Roxen on a new (virtual) server. All seemed well untill we found out we missed a lot of Roxen user accounts. In the mean time we recreated one that was absolutely neccesary, but there are more in need of a restore. Now the question is: can I safely import the(/a) user export? Will the recreated user be safely overwritten or will there occur issues based on user ids. The recreated user is used, by ourselves, based on user name. Thank you for your time & effort! -- Met vriendelijke groet / With kind regards / Mit besten Grüßen, Coen Schalkwijk Software Engineer coen.schalkwijk(a)rtl.nl <mailto:coen.schalkwijk@rtl.nl> coen(a)rtlinteractief.nl <mailto:coen@rtlinteractief.nl> +31 (0)35 671 8915 MSN - rtl(a)navelpluis.nl Werkt niet op vrijdag

1 0

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

2003

2002

Pike-devel November 2011