Platonides wrote:
> Transom wrote:
> > So---are the permissions wrong? I don't know why MediaWiki
> > would need "execute" permissions for images, but when images
> > are 755 it works, and when set to 640, it doesn't.
>
> I guess it has to do with the read permission for everybody,
> and not with the execute permission. Can the files be read by
> the user under which apache runs?
Thanks for replying.
Apache is running as "nobody". After upload, the read permission is set to "644" (in one place in original message I said "640", but that was a typo). The owner and group are set to the user where MediaWiki is installed.
I would think that 644 was fine; Apache can read the file.
> > ALSO... I have noticed that MediaWiki shows a link on
> > the image page for images uploaded in the past, and the
> > MIME type is displayed as "text/plain".
> The mime type is stored in the database. As those images
> were detected as text/plain in the past, they're still listed
> as such. They would be correctly detected if you reuploaded
> them today.
Hmm. Another responder said to run a maint script to fix the mime types. I'll probably do that, but I don't understand why files uploaded before didn't have the mime type set; they displayed fine, and basically, everything was working as expected before installing phpsuexec.
Another symptom: When I go to the "recent changes" page and limit the namespace to "Image", I get no results even though various images have been uploaded.
I'll run the main script and then I'll be back...
Thanks.
_______________________________________________
Join Excite! - http://www.excite.com
The most personalized portal on the Web!
Hallo,
is there a methode to check the querystring.
I would like to crate a global array in which i place the exceptions!
If part of the query_string is in the exception, the should be a redrect
to another page.
Example:
http://localhost/wiki/index.php?title=Category:Images&action=edit
$wgWIzardExtensions["Category:"]
if the value is part of the url, then redirect
regards
mic
>> Earlier: "...Looks like the method
>> described here is probably what
>> you want:.."
http://shiflett.org/blog/2005/oct/convert-smart-quotes-with-php
> Earlier: "...That looks like exactly
> what I need...I'd never heard them
> called "smart quotes"..."
There are many places Microsoft "reveals*" it's auto-replacement policy
as the user types, using high-order printable typographic characters
replacing standard 256 ASCII characters resulting in "?" and so on for
word-to-web cut-and-paste, look in the following Microsoft Word menus
(note (c) and 1/2 and other replacements):
Tools > Auto Correct Options...
[AutoCorrect]
Show AutoCorrect Options buttons
Correct TWo INitial CApitals
Capitalize first letter of sentences
Capitalize first letter of table cells
Capitalize names of days
Correct accidental usage of cAPS LOCK key
Replace text as you type
[AutoFormat As You Type]
Replace as you type
"Straight quotes" with smart quotes
Ordinals (1st) with superscript
Fractions (1/2) with fraction character ( )
Hyphens (--) with dash (
*Bold* and _italic_ with real formatting
Internet and network paths with hyperlinks
Apply as you type
Automatic bulleted lists
Automatic numbered lists
Border lines
Tables
Built-in Heading styles
Automatically as you type
Format beginning of list item like the one before it
Set left- and first-indent with tabs and backspaces
Define styles based on your formatting
[AutoFormat]
Apply
Built-in Heading styles Automatic bulleted lists
List styles Other paragraph styles
Replace
"Straight quotes" with smart quotes
Ordinals (1st) with superscript
Fractions (1/2) with fraction character ( )
Hyphens (--) with dash (
*Bold* and _italic_ with real formatting
Internet and network paths with hyperlinks
Preserve
Styles
Always AutoFormat
Plain text WordMail documents
==
You're dealing with MS Word as the dynamic input editor for MediaWiki,
and that's similar to the task of re-purposing static MS Word DOCs for
use as MediaWiki pages, so there's much overlap in assignment and tools.
Importing existing documents, as well as cutting and pasting
"live/unsaved/in-process" documents in your case, is a significant need,
and there's much to be learned from document converter/import
experiences that applies equally to cut-and-paste (where "?" often
displays on the web instead of Microsoft's choices for typographical
characters intended for printing).
Note the following characters are anticipated and dealt with by Gunter
Schmidt http://www.beadsoft.de in
http://www.mediawiki.org/wiki/Extension_talk:Word2MediaWikiPlus
' Replace all smart quotes with their dumb equivalents
... and:
"_"
"-"
"+"
"/"
"{"
"}"
"["
"]"
"~"
"^^"
"|"
"'"
"<"
">"
... and adding as needed:
"" 'save some space if several characters in a row
"<nowiki>"
"</nowiki>"
'remove empty paragraphs at begin of document
'add <br> to all manual line breaks
'add <br> to all paragraphs
'Add empty line at document end to prevent error
(address lines beginning with tab or space)
'replace "forced blanks"
'do not allow more than two empty paragraphs in a row
'Clean up Tab Indention
'Header, Footer and Footnote reference
... and much, much more.
==
I do extensive document cleanup (macros) before converting form MS Word
to Wiki, and your users may be able to load the macro at
http://www.mediawiki.org/wiki/Extension_talk:Word2MediaWikiPlus which
will assist them in also cutting-and-pasting tables and images, and
preserving bold, italic and other formatting! I love it.
Good luck. Keep us informed in how it works for you, and please share
your results on http://www.mediawiki.org/
==
* for your entertainment: Microsoft's non-"reveal codes" white paper:
http://www.google.com/search?hl=en&q=microsoft+word+wordperfect+reveal+c
odes+white+paper&btnG=Google+Search - oh how the mighty have fallen, too
bad, WordPerfect and "reveal codes" could have evolved into the perfect
web editor, Macromedia Dreamweaver came close.
Hi,
I've great trouble in saving a new article not using the editpage class.
I thought I would be a nice idea to take the attempsave function from
Editpage.php to save the new article. And it seems to work.
The article is saved and stored in the wiki and i can view the page but
the articleID is 0.
So, I don't know, whats going wrong
regards
mic
Is this even possible in the current version of Mediawiki? Looking to
basically be able to block users from being able to even read the wiki
unless they log in.
-Dennis
I just upgraded to 1.11.0 over the weekend and have a problem with the
Short URLs. When I click the "edit" link for a page, the editor tries to
edit a page named "index.php". Basically, it's impossible to edit any
page because the edit URL will not work with Short URLs enabled.
The site URL structure is *http://www.domain.com/Page_Name*. When I
click the edit link on a page, the URL is
*http://www.domain.com/index.php?title=Page_Name&action=edit*
However, the edit box is blank and and the title of the page being
edited is wrong; here is what the page shows: /Editing Index.php. You've
followed a link to a page that doesn't exist yet. To create the page,
start typing in the box below.
/
I deactivated all plugins and reverted to the default skin; no change.
I added the following to the *LocalSettings.php* file to activate the
Short URLs:
$wgArticlePath = "/$1";
_If I remove the above entry_ from the *LocalSettings.php* file and load
the same edit URL listed above, it picks the correct page and _edits
work perfectly_.
Any ideas?
*Some additional info about the site:*
All of the mediawiki files are installed in the root directory (not in a
sub-directory). PHP 5.2.x, MySQL 5.0.37. Here is my .htaccess file:
Options FollowSymLinks RewriteEngine On php_value register_globals 0
RewriteCond %{REQUEST_FILENAME} !-f RewriteCond %{REQUEST_FILENAME} !-d
RewriteRule ^(.+)$ /index.php?title=$1 [L,QSA]
I'm having a problem where I'm trying to hack a system message 'affirmation'
into a form (specifically, the upload form in SpecialUpload.php), following
a new checkbox (the checkbox is required to upload). I'm currently using
this:
...
<tr>
<td></td>
<td align='left'>
<input tabindex='9' type='checkbox' name=\"wpUploadAffirm\" value=\"1\"
id=\"wpUploadAffirm\" onchange=\"change_form();\" onclick=\"change_form();\"
/>
");
$wgOut->addWikiText( wfMsgForContent( 'affirmation' ) );
$wgOut->addHtml( "
</td>
...
However, when I do this, there's a linefeed output in front of the
'affirmation' message, so the message does not appear on the same line as
the checkbox, but the next one. I looked at the Mediawiki:Affirmation page
and there is no extra linefeed at the start.
Is there another method to correctly parse a message like this that doesn't
include the extra linefeed?
-Jeff
--
View this message in context: http://www.nabble.com/System-message-without-initial-linefeed--tf4518139.ht…
Sent from the WikiMedia General mailing list archive at Nabble.com.
Anyone know if there is a way to include the text of a particular
wiki page in a regular HTML page? I just want the text of the page
(but converted to HTML, not wiki formatted), not the whole wiki
skin. Thanks.
--
Daniel M. Israel
dmi1(a)cornell.edu
> RE: Debugging MediaWiki crash in PHP
> Earlier: "...[how] to determine
> the cause of a MediaWiki crash
> "PHP has encountered an Access
> Violation at 77FCC2C0"
Peter Blaise responds:
I just did some reading at Google:
http://www.google.com/search?hl=en&q=PHP+has+encountered+an+Access+Viola
tion+at+77FCC2C0&btnG=Google+Search
http://www.google.com/search?hl=en&q=77FCC2C0&btnG=Google+Search
Sounds like an illegal instruction trying to do something in a memory
address that can't be done - perhaps a corrupt executable or dll dynamic
link library.
You can try rebooting clean and unloading any non-needed programs (even
ant virus) and see if the problem goes away.
Also, reboot in safe mode and clean out all cache and temp files, and
scandisk the drive, reboot, and try again.
You may have to replace corrupt programming wherever it is - hard to
find. Operating system? Drivers? Utilities? Web server? PHP?
Extensions?
There are many memory scanners (including debug) that may allow you to
browse and identify what's sometimes (emphasis on "sometimes") located
at 77FCC2C0, but I suspect that what's there changes whimsically, and
it's a meaningless error message - most are.
You could also have a real bad memory chip that you can swap or remove,
or messed up CMOS, which you can update. I'd r&r remove and reinstall
all computer parts, clean connections, leave the cabinet open and add
big fan to keep it cool to test. Everything in the computer may be
capable, but perhaps the power supply is under-powering the memory
chips, or other accessory is draining too much power - remove unused
accessories (modem, sound card).
Threshold problems like this are not processional to troubleshoot. I've
had to work on everything, one step at a time, and sometimes moving to
an entirely new computer turns out easier.
Is it time to migrate to a new computer, rebuild the current server, and
implement it as a backup anyway? Take advantage of the crash to buy
new.
Let us know.
--
> Re: MediaWiki Error on Windows
> Earlier: :...[help with] nasty
> systems log error...in Windows:
9/24/2007 12:40:07 PM WAM Error None 204 N/A
WWW-STL-52 The HTTP server encountered an unhandled exception while
processing the ISAPI Application '
wam!WAM_REQ_CORE::GetSz(unsigned long)const + 0x51
php5isapi + 0x1172
wam!DllCanUnloadNow + 0x636
wam!DllCanUnloadNow + 0x20C
w3svc!HTTP_HEADERS::FindValue(char const *,unsigned long *) + 0xE2
w3svc!STR::Copy(char const *,unsigned long) + 0xC71
w3svc!STR::Copy(char const *,unsigned long) + 0xB49
w3svc!STR::Copy(char const *,unsigned long) + 0x9A2
w3svc!CLIENT_CONN::OnSessionStartup(int *,void *,unsigned long,int) +
0x642
w3svc!HTTP_HEADERS::Reset(void) + 0x1CA
w3svc!IIS_SERVER_INSTANCE::LockThisForWrite(void) + 0x63C
w3svc!IIS_SERVER_INSTANCE::LockThisForWrite(void) + 0x5AF
ISATQ!_AtqWriteSocket@16 + 0x218
ISATQ!CDirMonitor::RemoveEntry(class CDirMonitorEntry *) + 0x13A
+ 0x7B0EEF7A
Peter Blaise responds:
I just did some MORE reading at Google:
http://www.google.com/search?hl=en&q=HTTP+server+encountered+an+unhandle
d+exception+while+processing+the+ISAPI+Application&btnG=Google+Search
http://www.google.com/search?hl=en&q=wam%21DllCanUnloadNow&btnG=Google+S
earch
http://www.google.com/search?hl=en&q=w3svc%21STR%3A%3A&btnG=Google+Searc
h
... and so on. You could read, read, read everyone else's reports on
seeing similar error message lines, but I do not think you'll find a
quick and simple answer like, "Put the balls on the other side."!
What system log? Who's error report? This feels the same as above [RE:
Debugging MediaWiki crash in PHP] above. You may be in for a serious
reinstall ... AFTER carefully and thoroughly confirming that every piece
and part in the system is perfect, not intermittent, and not part of a
cascade failure (like an overheating video chip overloading the power
supply and causing the hard drive to miswrite data - I'm on my 4th
computer and now have a 9 inch fan flowing on it!).
The first think I do is remove the hard driven, copy it to a spare, and
then work from there. Really, though, installing fresh on another
machine and migrating may be worth consideration. 25 wikis, eh? What's
the plan for bringing them back up after the server fries? Now's the
time to develop one and implement it!
Let us know how it goes.
Hi,
Our set up is: MediaWiki: 1.9.2 (r2132) PHP: 5.2.2 (cgi-fcgi)
MySQL: 5.0.15-nt running on IIS
Our users sometimes author in MS Word and then copy and paste their
article text into the MediaWiki edit box. Word often automatically
changes some characters as you type such as turning a dash into an
mdash, or a single quote into an apostrophe. These things generally
display fine in MediaWiki, however, we have to export the contents of
the wiki to be imported into a final environment for client approval.
Unfortunately I don't have the authority to change the rather odd
process. Since I think the users will continue to paste things in from
Word, does anyone know of an extension or tool which would "clean" the
text input into the wiki? I was unable to find anything like this in
the Extensions pages.
- Courtney Christensen