$wgDBtransactions gets set to true if using InnoDB tables. Is there
an advantage to using InnoDB tables?
The disadvantage is that with MySQL there is a file, ibdata1, that
seems to grow endlessly if InnoDB tables are used. See
http://bugs.mysql.com/bug.php?id=1341
We're wondering if we should just convert everything to MyISAM. Any
thoughts?
=====================================
Jim Hu
Associate Professor
Dept. of Biochemistry and Biophysics
2128 TAMU
Texas A&M Univ.
College Station, TX 77843-2128
979-862-4054
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Hi all,
I've created some custom namespaced on one of my wikis, Botwiki
(previously known as pywikipedia).
I've put these lines in my LocalSettings.php file:
- ---
#Custom namespaces
$wgExtraNamespaces =
array(100 => "Manual",
101 => "Manual talk",
102 => "Python",
103 => "Python talk",
104 => "Php",
105 => "Php talk",
106 => "Perl",
107 => "Perl talk",
108 => "AWB",
109 => "AWB talk",
110 => "IRC",
111 => "IRC talk",
112 => "Other",
113 => "Other talk"
);
$wgContentNamespaces[] = 100;
$wgContentNamespaces[] = 102;
$wgContentNamespaces[] = 104;
$wgContentNamespaces[] = 106;
$wgContentNamespaces[] = 108;
$wgContentNamespaces[] = 110;
$wgContentNamespaces[] = 112;
- ---
However, I have a big problem: when I go to a page in one of these new
namespaces (not the discussion, the main ones), for example
http://botwiki.sno.cc/wiki/Perl:Copyright_Violation_Bot , I found the
red link to the discussion page. It's right, as there is no discussion
page for that article. But if you click on it, it brings you to
http://botwiki.sno.cc/w/index.php?title=Perl_talk:Copyright_Violation_Bot&a…
correct, of course. But have a look of the article and discussion tabs:
they are both red! The first, "article", leads to
http://botwiki.sno.cc/w/index.php?title=Perl_talk:Copyright_Violation_Bot&a…
when it should lead to
http://botwiki.sno.cc/wiki/Perl:Copyright_Violation_Bot and the second,
"discussion", leads to
http://botwiki.sno.cc/w/index.php?title=Talk:Perl_talk:Copyright_Violation_…
, when it should lead to
http://botwiki.sno.cc/w/index.php?title=Perl_talk:Copyright_Violation_Bot&a…
.
It's the first time I deal with custom namespaces :-( but I have some
ideas of what it can be. Can the problem be with the
$wgContentNamespaces settings? So it detects everything as ns0? (don't
think so).
Or can it be the fact that I haven't used an underscore in the
$wgExtraNamespaces definition?
Snowolf
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.7 (MingW32)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org
iD8DBQFGWhk7sdafW5NQMtERAuX+AKDQ7QLNjXv9cu+ZbSLXidMzgi/vNgCaA7VT
+VTgR3iI/BI7FVDqcyRZVJ0=
=a4yP
-----END PGP SIGNATURE-----
Hi,
I have been trying to create a maintainable way of grouping pages
together and allowing readers to page through them in sequence. Most
samples I've seen use templates that require you to supply the
'previous' and 'next' page. This however results in three page edits
to insert a new page between two existing pages and does not guarantee
that your prev-sequence is identical to the next-sequence...
Being a programmer, that is way too much duplicate and error-prone
work for me ;-) There must be a better way to do this and I was hoping
to solve it with a plain MW installation (with ParserFunctions and
DynamicPageList at the moment).
Given an 'index' page that holds a list of all page titles in the
preferred order, isn't it possible to create a template that selects
the correct previous and next page title given the current page title?
I get stuck in getting the correct lines from the index page. DPL can
select based on section name (= page title), but then the contents of
that index section must be the prev/next links themselves:
=First Page=
Prev [[Second Page|Next]]
=Second Page=
[[First Page|Prev]] [[Third Page|Next]]
=Third Page=
.. etc.
This works, except that it is still a lot of duplication of page names
(but the edits are contained in a single page, big plus).
I hoped to simplify the index page by creating a template that writes
the section header and prev/next links, but then DPL no longer
recognizes the sections :-( Apparently DPL 'sees' the page text before
the templates are called ({{Page|Prev page|Page title|Next page}}:
{{Page||First Page|Second Page}}
{{Page|First Page|Second Page|Third Page}}
{{Page|Second Page|Third Page|Fourth Page}}
Basically my questions are:
1. Am I completely off track here?
2. Can DPL be coerced to evaluate templates before looking at the page
3. Can DPL (or another extension) select the text from a section
before/after a matched section?
4. Is it possible to determine the section sequence number given the
section name (so 'Second Page' results in 2, allowing me to use DPL to
retrieve the name of section 2 -/- 1 and 2 + 1 to create the prev/next
links?
Apologies for the long post, hopefully someone can point me to some
good resources (I've been to Meta, Wikibooks, Medawiki.org but could
very well have overlooked something there as the amount of info is a
bit overwhelming and it is difficult to judge how up to date it is).
--
Regards,
Jean-Marc
Hi all, hi Bernard... probably, my response is a bit late, but I spent
time reading (hopefully!) carefully all the answers...
Thus, my 0,02$:
I think that probably the correct question is: "Please convince me to
make my business model" :-)
I agree with others that probably, the SaaS business models require
closed source. Thus, probably, no choice...
IMHO the open source software implies a different business model,
something similar to "partnership".
Good luck, honestly,
Giuseppe Briotti.
--
Giuseppe Briotti
g.briotti(a)gmail.com
"Alme Sol, curru nitido diem qui
promis et celas aliusque et idem
nasceris, possis nihil urbe Roma
visere maius."
(Orazio)
Hello everyone,
We want to implement a MW-based WikiFamily with a reserved "pool" instance,
very similarly to Wikipedia, but we're encountering some problems with the media sharing.
We followed this process:
1) we started from
- three language instances
- containing about 1000 images altogether
- configured with Wikipedia-style URL scheme: http://<lang>.domain.com/wiki/<Node>
2) we installed and set up one "pool" wiki instance, according to "Scenario 5"
of http://www.mediawiki.org/wiki/Manual:Wiki_family
3) we modified maintenance/importImages.php to import automatically from any
language instance to the "pool" all images along with metadata (description, author etc)
The pool contains all medias of all language instances, and the "Upload file" link in the
toolbox of each wiki points to the pool's Special:Upload correctly.
The relevant configuration for the language wikis is as follows:
===
$wgUploadNavigationUrl = "http://pool.domain.com/wiki/Special:Upload";
$wgUseSharedUploads = true;
$wgSharedUploadPath = 'http://pool.domain.com/wiki/File:';
$wgSharedUploadDirectory = '/var/www/wikis/wiki-pool/images/';
$wgHashedSharedUploadDirectory = true;
$wgEnableUploads = false;
===
However, trying to enforce the "pool fallback lookup" documented
http://www.mediawiki.org/wiki/Manual:$wgUseSharedUploads
we tried moving away the instance's own images/ directory. Doing this, all images
result into a broken link.
We suspect the problem is in a misconfiguration of $wgSharedUploadPath , which
isn't quite clear from the WikiFamily example or its own documentation, but cannot
track down the problem more specifically.
Notably, we observed that none of the includes/filerepo/* files is loaded.
Can anyone help out with nailing down this problem?
Thanks!
Hello,
We've recently installed a wiki for IT dept usage but came to the conclusion
that it would be very useful to let end users see some of the articles. The
thing is we don't want them to see all the content mainly because it would
confuse them. For instance, if they search for "Accounting" we want to
provide definitions and troubleshooting on the Accounting system that might
help them, but not every server list, SQL code and tech spec that we have.
We've been looking around and found a couple of options based on namespaces,
which we're still not sure will work the way we need (for one thing,
limiting search results for end users) and seem kind of hard to maintain.
The simplest option seems to be keeping two separate wikis, but I don't want
to disconnect the end users documents from our wiki because it's valuable
info for us too. What we're thinking about is having both wikis but
maintaining them in sync with some batch process that could export and
import all articles that belong to a certain category or even backup and
restore the entire database and afterwards delete non-end user articles.
That way, the end user wiki would be read-only and get updates every day. In
both cases we'd have to know how to deal with images (seems easy with the
backup/restore option but no so much with the export / import).
Do you think we're on the right path with any of these? Any other
suggestions? Here is our version info:
Product Version
MediaWiki 1.15.1
PHP 5.2.10 (isapi)
MySQL 5.0.22
Thank you in advance!!
> From: Ron Laufer <gonzoron(a)hotmail.com>
>
> I've had a couple images uploaded to my wiki that are so big, it
> seems they make my thumbnail program crash.
Are you using ImageMagick?
We've had no problems using the biggest files a 12 megapixel camera
can produce.
----------------
An American is a person who demonstrates against a new power plant,
then goes home and flips on all the lights, turns up the air
conditioner, puts a tape in the stereo, opens the refrigerator door,
plugs in the coffee maker and sits down to see if the television
cameras caught him protesting. -- Wendell Trogdon
:::: Jan Steinman, EcoReality Co-op ::::
Hi,
I went through the sample steps in the API
example<http://www.mediawiki.org/wiki/API:Edit_-_Create%2526Edit_pages#Example>
on using the api to edit a page:
1. login your bot: ( I just used my admin login) - it succeeds
2. get a token - it suceeds
3. post to api.php - this is the form I
use<http://pathboston.com/forms/mw/append21.html>- after I get the
token I replace the +\ with %2B%5C
Alas, I get this error:
<?xml version="1.0"?>
<api>
<error code="badtoken" info="Invalid token" xml:space="preserve">
my form code is:
<form id="form1" name="Update" method="post"
action="../../hum310/api.php?action=edit">
<label>title: <input type="text" name="title" id="textfield"
value="Talk:Main_Page" /></label><br />
<label>section: <input type="text" name="section" id="textfield"
value="new"/></label><br />
<label>summary: <input type="text" name="summary" id="textfield"
value="Hello%20Worldwide"/></label><br />
<label>text: <input type="text" name="text" id="textfield"
value="Hello%20everyone!"/></label><br />
<label>watch: <input type="text" name="watch" id="textfield" /></label><br />
<label>basetimestamp: <input type="text" name="basetimestamp"
id="textfield" value="2008-03-20T17:26:39Z"/></label><br />
<label>token: <input type="text" name="token" id="textfield"
value="96a50d37e6c48e6d0def3dd64dc9450d%2B%5C" size=50/></label><br />
<input name="" type="submit" value="send" />
</form>
<a href="http://pathboston.com/hum310/api.php?action=query&prop=info|revisions&intok…">get
token</a>
Thanks,
Timothy S. McKenna
Humanities teacher
Parkway Academy of Technology and Health, a Boston Public High School
1205 VFW Parkway, Boston, MA 02132 rm404
tim(a)sitebuilt.net, http://sitebuilt.net
exp. course news on twitter, id: mrtmckenna
class wiki sites:
http://pathboston.com/hum Humanities 3
http://pathboston.com/poets class159
(857)498-2574 (mobile), (617)524-0938 (home)
Hello!
I have Mediawiki 1.15 installed with Apache 2.2.
I have created some virtual namespace (similar to categories) called Rubric.
The articles can belong to one ore more rubrics. I want to use this namespace
instead of Categories by some reasons.
And I need to show these Rubrics' content. For example:
http://mysite.com/wiki/Rubric:Economy/
should open a list of articles under Economy rubric.
I've created a special page which does that. So, now I can open
http://mysite.com/wiki/Special:Rubric/Economy
and view Economy's content. It works, no problem.
But I want to get rid of Special namespace in the URL! Can I do that?
I've tried several RewriteRules in the apache's .htacces, but failed so far...
The last try is:
=============================================================================
RewriteEngine On
RewriteRule wiki/Rubric:(.*)$ wiki/Special:Rubric/$1
RewriteRule wiki/(.*)$ w/index.php?title=$1 [L]
=============================================================================
It does not work. When I'm opening
http://mysite.com/wiki/Rubric:Economy/
this rule still opens "Rubric:Economy" instead of "Special:Rubric/Economy" and
says "there is no text on the page blah-blah-blah..."
What do I do wrong?
NOTE: I need an internal redirect. The external one works, but that's not what I
want. I mean, this one:
RewriteRule wiki/Rubric:(.*)$ wiki/Special:Rubric/$1 [R,L]
works great of course, but it results in that the Special namespace presents in
the URL.
Best regards.
Michael Wozowoy
e-mail: strannik(a)cnt.ru
Bernard,
On 28 January 2010 14:58, Bernard(a)bernardHulsman.nl
<bernard(a)bernardhulsman.nl> wrote:
> Jean-Marc, you have to use
> http://handleiding.wikiation.nl/MediaWiki:Common.css . That is where it is
> defined. It is no standard MW feature. There is no LocalSettings involved.
The Common.css is only the part where you define the background color
depending on the category you use. I.e. you specify that
.cat-Timecode-future results in a background: #fff7e7
That part is entirely clear to me, I use Common.css extensively.
But in my mediawiki installation (1.13.3, and neither on mediawiki.org
which uses 1.16alpha), there are no cat-[category name] classes being
output. I see a ns-xxx and a page-yyy part in the body class, but no
cat-zzz part. That is what confuses me.
--
Regards,
Jean-Marc
--
. ___
. @@ // \\ "De Chelonian Mobile"
. (_,\/ \_/ \ TortoiseSVN
. \ \_/_\_/> The coolest Interface to (Sub)Version Control
. /_/ \_\ http://tortoisesvn.net