phpwiki-talk

Re: [Phpwiki-talk] why preg_match crashes sometimes

[Reini U. <ru...@x-...>]

John Cole wrote:
>   If I recal correctly, I started seeing this bug when I switched from php
> 4.2 to 4.3 on windows.  I agree that your hitting a PHP bug, because when I
> had a php 4.2 server running and an identical setup running 4.3, only the
> 4.3 crashed.
> 
>   If you need me to, I could set up a virtual machine running php 4.2 and
> check again.  Do you think you might be able to get enough info to submit a
> bug report to PHP?  While that wont fix the sf.net version, it would help
> everyone else :-)  Another quick fix is the edit the OldTextFormattingRules
> page so it won't crash a new wiki.  

No, you don't need to setup a vm. I managed to call various php 
versions, httpd servers, and mysql versions on the same OS (Win2000).
In fact every single supported version. Just IIS not yet.

>   I'd prefer to see a good workaround of a PHP bug and have some other small
> issues fixed than spend the next few weeks making the system work with buggy
> php versions.  Expecially if php gets this bug fixed.  You have done so much
> work lately, it really would be nice to do a good stable relese.

The problem is that we cannot wait for this php bug to be fixed, similar 
to the pending mysql bug (> 4.1.x and 5.0.x: search fails BINARY pagename).
We rather must find a workaround for the affected versions.

The new unittest setup runs very fast now, so it's easy to test the 
complete SetupWiki and DumpHtml cycle.

It tells me that the bug comes only with the SQL backends.
The unittest uses the file backend, which runs fine (besides some 
remaining minor wikilens and references problems).
So we have to rewrite the unittests to test more backends.
-- 
Reini Urban
http://xarch.tu-graz.ac.at/home/rurban/

RE: [Phpwiki-talk] why preg_match crashes sometimes

[John C. <joh...@ua...>]

Reini,
  If I recal correctly, I started seeing this bug when I switched from php
4.2 to 4.3 on windows.  I agree that your hitting a PHP bug, because when I
had a php 4.2 server running and an identical setup running 4.3, only the
4.3 crashed.

  If you need me to, I could set up a virtual machine running php 4.2 and
check again.  Do you think you might be able to get enough info to submit a
bug report to PHP?  While that wont fix the sf.net version, it would help
everyone else :-)  Another quick fix is the edit the OldTextFormattingRules
page so it won't crash a new wiki.  

  I'd prefer to see a good workaround of a PHP bug and have some other small
issues fixed than spend the next few weeks making the system work with buggy
php versions.  Expecially if php gets this bug fixed.  You have done so much
work lately, it really would be nice to do a good stable relese.

  My 2 cents ;-)

John

-----Original Message-----
From: php...@li...
[mailto:php...@li...]On Behalf Of Reini
Urban
Sent: Thursday, July 08, 2004 11:43 AM
To: php...@li...
Subject: [Phpwiki-talk] why preg_match crashes sometimes

It crashes more often on linux, esp on sf.net php-4.1.2 than on newer 
versions, esp Windows seems to be much more stable.
Apache2 and IIS crash more often than Apache1.

-------------------------------------
This email and any files transmitted with it are confidential and intended solely for the use of the individual or entity to whom they are addressed. If you have received this email in error please notify the system manager. This message contains confidential information and is intended only for the individual named. If you are not the named addressee you should not disseminate, distribute or copy this e-mail.

[Phpwiki-talk] why preg_match crashes sometimes

[Reini U. <ru...@x-...>]

I've come now this analysis why ConvertOldMarkup() crashes with certain 
pages, and why not always. This is the only remaining showstopper for 
the 1.3.11 release, but it's an ugly and persistent one.

Symptoms:
?action=loadfile&source=pgsrc&overwrite=1
crashes sometimes at the second encounter of the anchored regex loop at 
the page "OldTextFormattingRules", esp. if the CreateToc plugin is used.
Similarily some more large old markup pages, but this is the only one 
from the default pgsrc set.

But, loading the single page will work okay:
?action=loadfile&source=pgsrc/OldTextFormattingRules&overwrite=1

It crashes more often on linux, esp on sf.net php-4.1.2 than on newer 
versions, esp Windows seems to be much more stable.
Apache2 and IIS crash more often than Apache1.

Possibly related symptom:
ADODB requires much more memory since the library upgrade with 1.3.10, 
so that AllPages will fail > ~300 pages, without any given limit 
parameter. Playing with the limit parameter will narrow the max. pageset 
size.
But, PearDB will fail also.

Possible Reason:
Somehow the cached database use too much memory when simply iterating 
over get_all_pages(). Exhausting the available php memory will crash 
anchored PCRE matches, which is a (yet unknown) PHP bug.

What I have done so far:
I've tried to improve the sql calls, and to purge the cache from 
unneeded data on such page iterations.
* _cached_html should be kept in the cache for the requested page. This 
new policy is not yet enforced globally, just for some API calls.
* pagedata[%content'] should only be kept for pages which really need 
the content. e.g. for the size column on pagelists the content can be 
safely purged after having done the strlen(). Sorting on size makes it 
problematic.
Nevertheless, despite these efforts the memory footprint increased 
dramatically lately, though it really should have been decreased.
   https://sourceforge.net/forum/message.php?msg_id=2654786

I'm investigating this further. Maybe someone heard about similar 
problems elsewhere.
-- 
Reini Urban
http://phpwiki.sf.net/

Re: [Phpwiki-talk] PearDB consistency: pending cvs rename will break CVS snapshot for one day

[Reini U. <ru...@x-...>]

Reini Urban wrote:
> I just submitted an sf.net CVS rename request, which will most likely 
> break the nightly CVS snapshot and the demo and test site for this one 
> day. (If I cannot fix it before midnight)
> https://sourceforge.net/tracker/index.php?func=detail&aid=987179&group_id=1&atid=200001 

> I think this will happen in about two days.

Uuh, that was fast!

Already done and tested. So no conflicts will arise.
Just do a fresh cvs up for the WikiDB/backends.

-- 
Reini Urban
http://xarch.tu-graz.ac.at/home/rurban/

[Phpwiki-talk] rename pages...

[John C. <joh...@ua...>]

Reini,
  That worked great.  I'm still a little foggy on the regexp option, as the
page does say that no regexp are used :-)  Perhaps a RenameHowTo page is in
order.

  I did find one bug, however.  I had a Chapter04 page and needed to rename
it to Chapter04/Part01.  Well, it deleted both pages.  So I would say that
renameing to a subpage doesn't work (at least a subpage of itself).

Thanks,

John

-----Original Message-----
From: php...@li...
[mailto:php...@li...]On Behalf Of Reini
Urban
Sent: Thursday, July 08, 2004 4:42 AM
To: php...@li...
Subject: Re: [Phpwiki-talk] <div> tag for plugins?


First rename
"chapter1/*" from "chapter1/" to "chapter01/".

Then rename
"chapter01/*" from "part" to "part0".

PS: The new regex option will help in the second step.
"(\d)" => "0\1"
if the parts are named like "part<number>"
-- 

-------------------------------------
This email and any files transmitted with it are confidential and intended solely for the use of the individual or entity to whom they are addressed. If you have received this email in error please notify the system manager. This message contains confidential information and is intended only for the individual named. If you are not the named addressee you should not disseminate, distribute or copy this e-mail.

[Phpwiki-talk] PearDB consistency: pending cvs rename will break CVS snapshot for one day

[Reini U. <ru...@x-...>]

I just submitted an sf.net CVS rename request, which will most likely 
break the nightly CVS snapshot and the demo and test site for this one 
day. (If I cannot fix it before midnight)
https://sourceforge.net/tracker/index.php?func=detail&aid=987179&group_id=1&atid=200001

* cd lib/WikiDB/backend
* rename the four PearDB $backend_type.php files to PearDB_$backend_type.php
* rename the classnames to have the "WikiDB_backend_PearDB_" prefix

Rationale:
lib/WikiDB/backend has for historical reasons the PearDB (=SQL), ADODB 
and the other supported backends (file, dba, cvs) as single files.
PearDB just includes its specializers with the name ($backend_type, e.g. 
mysql.php).
ADODB came after that and does it right, using ADODB_$backend_type.php.
The specializer should have its base class in the filename (and the 
class name also). ADODB does it right, PearDB not.

Affected Files:
mysql.php, sqlite.php, oci8.php and pgsql.php

Required source changes:
rename the classnames
fix the loader in lib/WikiDB/SQL.php.

Required developer changes:
#wait for the cvs rename
#wait for the source changes
cd lib/WikiDB/backend
cvs up

I think this will happen in about two days.
burley has about 25 pending tracker requests, and fixes about 10 per day.
https://sourceforge.net/tracker/index.php
But with the recent file release problems it can last longer also
("FRS issue"), though regs is supposed to handle this.
-- 
Reini Urban
http://xarch.tu-graz.ac.at/home/rurban/

Re: [Phpwiki-talk] <div> tag for plugins?

[Reini U. <ru...@x-...>]

John Cole wrote:

> Reini,
>   The rename multiple seems to act odd ;-)
> 
>   Here is my situation, I have a bunch of pages:
> 
> chapter1/part1
> chapter1/part2
> chapter1/part3
> 
> etc...
> 
>   I realised that I need to have things named chapter01/part01, so I went to
> the admin menu and tried the admin select on chapter1/part* and it selected
> all of the pages I wanted to rename.  Great!
> 
>   Now, I made sure all were selected and pressed the rename button.  It came
> back with a page with a rename to->from box at the top.  This page will not
> let me go on to the next one without putting something into the top rename
> from->to boxes, so I fill one out from: chapter1/part1 to:chapter01/part01
> and press rename selected pages (all are selected)
> 
>   Next I'm back at a similar page, but with a 'Rename to' box next to each
> selected page.  Looking good.
> 
>   I make all of my changes and press yes.
> 
>   And, it changes only the one I filled in the first from->to box, and the
> rest is returns Couldn't rename page 'chapter1/part2' to 'chapter01/part02'
> 
>   Hopefully it's something simple :-)

First rename
"chapter1/*" from "chapter1/" to "chapter01/".

Then rename
"chapter01/*" from "part" to "part0".

PS: The new regex option will help in the second step.
"(\d)" => "0\1"
if the parts are named like "part<number>"
-- 
Reini Urban
http://xarch.tu-graz.ac.at/home/rurban/

Re: [Phpwiki-talk] ADOdb backend -- the return

[Reini U. <ru...@x-...>]

Philippe Vanhaesendonck wrote:
> The main driver to use FETCH_NUM is performance, but we revert to 
> FETCH_ASSOC for the iterators, since they are using hashes.
> 
> I was thinking the following: if we build the column list when we build 
> the query and pass it to the iterator (via the backend class or via the 
> new() constructor), then we wouldn't need to FETCH_ASSOC at all!
> 
> I can code it and submit a patch if you think it is a good idea...

Good idea. So I have time to catch our remaining preg_match crashes.

And it's only needed for special border cases (plugins).
All our internal core functions should use FETCH_NUM.
-- 
Reini Urban
http://phpwiki.sf.net/

[Phpwiki-talk] UnfoldSubpages word=10 broken...

[John C. <joh...@ua...>]

Reini,
  UnfoldSubpages is no longer taking the words parameter.  If I set it to 10
(or any other number) it returns the entire sub page.

Thanks,

John Cole

-------------------------------------
This email and any files transmitted with it are confidential and intended solely for the use of the individual or entity to whom they are addressed. If you have received this email in error please notify the system manager. This message contains confidential information and is intended only for the individual named. If you are not the named addressee you should not disseminate, distribute or copy this e-mail.

RE: [Phpwiki-talk] <div> tag for plugins?

[John C. <joh...@ua...>]

Reini,
  The rename multiple seems to act odd ;-)

  Here is my situation, I have a bunch of pages:

chapter1/part1
chapter1/part2
chapter1/part3

etc...

  I realised that I need to have things named chapter01/part01, so I went to
the admin menu and tried the admin select on chapter1/part* and it selected
all of the pages I wanted to rename.  Great!

  Now, I made sure all were selected and pressed the rename button.  It came
back with a page with a rename to->from box at the top.  This page will not
let me go on to the next one without putting something into the top rename
from->to boxes, so I fill one out from: chapter1/part1 to:chapter01/part01
and press rename selected pages (all are selected)

  Next I'm back at a similar page, but with a 'Rename to' box next to each
selected page.  Looking good.

  I make all of my changes and press yes.

  And, it changes only the one I filled in the first from->to box, and the
rest is returns Couldn't rename page 'chapter1/part2' to 'chapter01/part02'

  Hopefully it's something simple :-)

Thanks,

John Cole

-------------------------------------
This email and any files transmitted with it are confidential and intended solely for the use of the individual or entity to whom they are addressed. If you have received this email in error please notify the system manager. This message contains confidential information and is intended only for the individual named. If you are not the named addressee you should not disseminate, distribute or copy this e-mail.

[Phpwiki-talk] ADOdb backend -- the return

[Philippe V. <Phi...@to...>]

Me again :-\

Reini,

The main driver to use FETCH_NUM is performance, but we revert to 
FETCH_ASSOC for the iterators, since they are using hashes.

I was thinking the following: if we build the column list when we build 
the query and pass it to the iterator (via the backend class or via the 
new() constructor), then we wouldn't need to FETCH_ASSOC at all!

I can code it and submit a patch if you think it is a good idea...

--
Philippe

[Phpwiki-talk] ADOdb backend again...

[Philippe V. <Phi...@to...>]

Reini,

In my quest to get ADOdb/Oracle running, i struggled quite some time 
over a php warning, and I enventually found the issue...

In the WikiDB_backend_ADODB_iter class, we have:

        if (substr($backend->_dbh->databaseType,0,5) == 'mysql')
            $result->fetchMode = 1;
        else
            $result->fetchMode = 2;

Maybe I am missing something, but beside the fact that 'else' part 
(non-mysql) does not fly, I think this statement is not necessary:

The fetchMode in the ADORecordSet is set when that class is 
instanciated, and does not change afterwards. Since for the iterators we 
do a
    $dbh->SetFetchMode(ADODB_FETCH_ASSOC);
the ADORecordSet fetchMode will be set to MYSQL_ASSOC (== 1) when we 
'Execute' the query, and will not be changed when we reset the 
ADOConnection fetchMode to ADODB_FETCH_NUM.

So basically, for the MySQL case, we set $result->fetchMode to what it 
is already (no impact), and for any other DBMS we do something random, 
since the ADORecordSet fetchMode property is DBMS specific...

Does this makes sense or am I working too late?

--
Philippe

Re: [Phpwiki-talk] Problem with pagelist column sorting

[Dan F. <dfr...@cs...>]

Scott Yilek wrote:

> I've been working on sorting pagelist columns and have been running 
> into a slight difficulty.  I am defining a new _getSortableValue 
> function for each new custom pagelist column, however sorting does not 
> work.  There is a line in the _pageCompare function of pagelist that, 
> for my custom columns, always returns before sorting is done: 
> if (!isset($pagelist->_columns_seen[$colNum])) return 0;
>
> Is this line correct?  I changed it to isset instead of !isset and 
> sorting worked!  However, I am worried this could have affected 
> something else.
> Scott


Reini,

My question: can we check in this fix? We need it (or some fix like it) 
to make pagelist sorting work.

Dan

[Phpwiki-talk] Problem with pagelist column sorting

[Scott Y. <sy...@cs...>]

I've been working on sorting pagelist columns and have been running into 
a slight difficulty.  I am defining a new _getSortableValue function for 
each new custom pagelist column, however sorting does not work.  There 
is a line in the _pageCompare function of pagelist that, for my custom 
columns, always returns before sorting is done:  

if (!isset($pagelist->_columns_seen[$colNum])) return 0;

Is this line correct?  I changed it to isset instead of !isset and 
sorting worked!  However, I am worried this could have affected 
something else. 

Scott

[Phpwiki-talk] Re: [phpwiki-checkins] CVS: phpwiki/tests/unit test.php,1.5,1.6

[Dan F. <dfr...@cs...>]

Reini,

Thanks for checking in more unit tests! I hope this will be a convenient 
way to check over time that things haven't broken.

Question: how do you get this latest test to pass? When I run the unit 
tests, I now get the following:

 % php test.php
Run tests .. 
/scratch/gibson/apache2/doc/dfrankow/phpwiki-next/lib/WikiDB/backend/file.php: 
97: error# 2: 
fopen(/scratch/gibson/apache2/doc/dfrankow/phpwiki-next/tests/unit/testbox/page_data/AddCommentPlugin): 
failed to open stream: No such file or directory
Traceback:
  
/scratch/gibson/apache2/doc/dfrankow/phpwiki-next/lib/WikiDB/backend/file.php:97
  
/scratch/gibson/apache2/doc/dfrankow/phpwiki-next/lib/WikiDB/backend/file.php:187
  
/scratch/gibson/apache2/doc/dfrankow/phpwiki-next/lib/WikiDB/backend/file.php:275
  /scratch/gibson/apache2/doc/dfrankow/phpwiki-next/lib/WikiDB.php:1773
  /scratch/gibson/apache2/doc/dfrankow/phpwiki-next/lib/WikiDB.php:1112
  /scratch/gibson/apache2/doc/dfrankow/phpwiki-next/lib/loadsave.php:626
  /scratch/gibson/apache2/doc/dfrankow/phpwiki-next/lib/loadsave.php:873
  /scratch/gibson/apache2/doc/dfrankow/phpwiki-next/lib/loadsave.php:934
  /scratch/gibson/apache2/doc/dfrankow/phpwiki-next/lib/loadsave.php:1020
  
/scratch/gibson/apache2/doc/dfrankow/phpwiki-next/tests/unit/lib/SetupWiki.php:21
  /export/scratch/apache/2.0.46/php/lib/php/PHPUnit/TestCase.php:155
  /export/scratch/apache/2.0.46/php/lib/php/PHPUnit/TestCase.php:139
  /export/scratch/apache/2.0.46/php/lib/php/PHPUnit/TestResult.php:207
  /export/scratch/apache/2.0.46/php/lib/php/PHPUnit/TestCase.php:127
  /export/scratch/apache/2.0.46/php/lib/php/PHPUnit/TestSuite.php:141
  /export/scratch/apache/2.0.46/php/lib/php/PHPUnit/TestSuite.php:141
  /export/scratch/apache/2.0.46/php/lib/php/PHPUnit.php:71
  /scratch/gibson/apache2/doc/dfrankow/phpwiki-next/tests/unit/test.php:212


I poked into this a little bit. It looks like before saving page, it 
tries to check the page metadata, doesn't file the page (file), and 
throws an error.

Is there some setup option I am missing?

Thanks for your attention.

Dan


Reini Urban wrote:

>Update of /cvsroot/phpwiki/phpwiki/tests/unit
>In directory sc8-pr-cvs1.sourceforge.net:/tmp/cvs-serv23978
>
>Modified Files:
>	test.php 
>Log Message:
>cache groups, added load and save tests 
>  (all pgsrc files must be able to be transformed)
>
>
>Index: test.php
>===================================================================
>RCS file: /cvsroot/phpwiki/phpwiki/tests/unit/test.php,v
>retrieving revision 1.5
>retrieving revision 1.6
>diff -u -2 -b -p -d -r1.5 -r1.6
>--- test.php	1 Jul 2004 15:38:39 -0000	1.5
>+++ test.php	6 Jul 2004 09:15:23 -0000	1.6
>@@ -1,3 +1,4 @@
>-<?php  // #! /usr/local/bin/php -Cq
>+#! /usr/local/bin/php -Cq
>+<?php  
> /* Copyright (C) 2004, Dan Frankowski <dfr...@cs...>
>  * Copyright (C) 2004, Reini Urban <ru...@x-...>
>@@ -43,4 +44,5 @@ ini_set('include_path', ini_get('include
> # This quiets a warning in config.php
> $HTTP_SERVER_VARS['REMOTE_ADDR'] = '127.0.0.1';
>+define('GROUP_METHOD', 'NONE');
> 
> # Other needed files
>@@ -122,4 +124,5 @@ class MockRequest {
>         $this->_dbi = WikiDB::open($dbparams);
>         $this->_user = new MockUser("a_user", true);
>+        $this->_group = WikiGroup::getGroup();
>         $this->_args = array('pagename' => 'HomePage', 'action' => 'browse');
>         //$this->Request();
>@@ -159,5 +162,5 @@ class MockRequest {
>     }
>     function getGroup() {
>-        return WikiGroup::getGroup();
>+        return $this->_group;
>     }
> }
>@@ -185,5 +188,7 @@ require_once 'PHPUnit.php';
> # lib/config.php might do a cwd()
> require_once dirname(__FILE__).'/lib/InlineParserTest.php';
>+require_once dirname(__FILE__).'/lib/SetupWiki.php';
> require_once dirname(__FILE__).'/lib/PageListTest.php';
>+require_once dirname(__FILE__).'/lib/DumpHtml.php';
> require_once dirname(__FILE__).'/lib/plugin/ListPagesTest.php';
> require_once dirname(__FILE__).'/lib/plugin/AllPagesTest.php';
>@@ -200,4 +205,6 @@ $suite->addTest( new PHPUnit_TestSuite("
> $suite->addTest( new PHPUnit_TestSuite("PageListTest") );
> $suite->addTest( new PHPUnit_TestSuite("ListPagesTest") );
>+$suite->addTest( new PHPUnit_TestSuite("SetupWiki") );
>+$suite->addTest( new PHPUnit_TestSuite("DumpHtml") );
> $suite->addTest( new PHPUnit_TestSuite("AllPagesTest") );
> $suite->addTest( new PHPUnit_TestSuite("AllUsersTest") );
>
>
>
>-------------------------------------------------------
>This SF.Net email sponsored by Black Hat Briefings & Training.
>Attend Black Hat Briefings & Training, Las Vegas July 24-29 - 
>digital self defense, top technical experts, no vendor pitches, 
>unmatched networking opportunities. Visit www.blackhat.com
>_______________________________________________
>phpwiki-checkins mailing list
>php...@li...
>https://lists.sourceforge.net/lists/listinfo/phpwiki-checkins
>  
>

RE: [Phpwiki-talk] <div> tag for plugins?

[John C. <joh...@ua...>]

Reini,
  Thanks for the change!  However, I tried it this morning (and I do have
your change) but it doesn't seem to work.  :-(  I put a few echo's around
the code and they didn't appear either.  Maybe it was just a long weekend
;-) but I can't seem to make it work.

Thanks,

John Cole

-----Original Message-----
From: php...@li...
[mailto:php...@li...]On Behalf Of Reini
Urban
Sent: Saturday, July 03, 2004 2:42 AM
To: php...@li...
Subject: Re: [Phpwiki-talk] <div> tag for plugins?


John Cole wrote:
>   I'm working on my docbook conversion and things are going pretty good so
> far.  I'm rendering out htmlhelp and pdf's now.
> 
>   I'd like to suggest a small change that I hope won't impact anything,
but
> would make converting xhmtl with xsl stylesheets easier (as far as I know
> how with xsl stylesheets).
>   Would it be possible to put a <div> tag around each plugin with an id or
> class name of the plugin?  

Good idea.

Not the classname and not another div tag, but the ID of the plugin DIV 
is now the plugin->getName().
Like so:
<div class="plugin tightenable" id="AddComment">

Note, that we don't enumerate the id as we do in anchors for CreateToc, 
so duplicate plugins get the same id.
-- 
Reini Urban
http://xarch.tu-graz.ac.at/home/rurban/



-------------------------------------------------------
This SF.Net email sponsored by Black Hat Briefings & Training.
Attend Black Hat Briefings & Training, Las Vegas July 24-29 - 
digital self defense, top technical experts, no vendor pitches, 
unmatched networking opportunities. Visit www.blackhat.com
_______________________________________________
Phpwiki-talk mailing list
Php...@li...
https://lists.sourceforge.net/lists/listinfo/phpwiki-talk

-------------------------------------
This email and any files transmitted with it are confidential and intended solely for the use of the individual or entity to whom they are addressed. If you have received this email in error please notify the system manager. This message contains confidential information and is intended only for the individual named. If you are not the named addressee you should not disseminate, distribute or copy this e-mail.

Re: [Phpwiki-talk] plugin help

[Dan F. <dfr...@cs...>]

Dmitry M. wrote:

>On Wed, Jun 09, 2004 at 02:29:04PM -0700, Dmitry M. wrote:
>  
>
>>I'm trying to see if a external "category" association will be useful at
>>all. Basically every wiki page can belong any number of "categories".
>>    
>>
>
>
>Hi all. I actually sort of did what I was planning to. If you don't mind
>taking a look: http://www.lazymonkey.org/wiki/CategorizerPlugin
>I've finished it yesterday and I'd like some opinions. Some of the stuff
>is bound to be broken, so don't worry about it too much :)
>  
>

Dmitry,

I wish some sort of similar functionality. I was planning to implement 
it through various back-links tricks. Something like:

- Some pages (e.g., Book, Restaurant) are special 'category pages' by 
linking to a special 'Category' page.

- Other pages link to the category pages (e.g., The Firm links to Book, 
La Bodega links to Restaurant). These category links are link your 
category label. However, they aren't entered in a separate editor, or 
displayed or searched separately.

I'm not sure how my proposal related to your implementation. That is, I 
don't know what I like about mine versus yours. Mine is probably more 
traditional in the sense that people have been using back-links in 
Phpwiki to represent categories since before either of us. Thus, mine 
perhaps requires fewer code mods as well.

Yours has a different UI. Possibly better? Don't know.

Comments?

Dan

[Phpwiki-talk] ih

[hughchaloner <hug...@ma...>]

hi

[Phpwiki-talk] safe_html snippet in PhpWiki

[Reini U. <ru...@x-...>]

Hi
phpwiki uses now your fine safe_html library for our RawHtml plugin.
   http://chxo.com/scripts/safe_html-test.php

We just enabled table, tr and td and maybe later more presentational 
markup, because the only purpose for the raw html mode is to allow 
limited enhancements of presentational markup. (and to disable malicious 
code).

http://phpwiki.sf.net/demo/en/RawHtmlPlugin and 
http://phpwiki.sf.net/demo/en/RawHtmlPluginLocked
-- 
Reini Urban
http://phpwiki.sf.net/

[Phpwiki-talk] Re: [Linux-7110-psion] phpwiki

[Brian D. <du...@ap...>]

Reini,
   I've installed our own phpwiki and moved our pages over to it.  I'm not
entirely sure if I did it all correctly, so if you wouldn't mind, please
give us a week or so before deleting the pages.  I'll add a "TO BE DELETED"
to the top of the PsilinuxNetbook page to warn anyone who goes there away.
   Sorry for the mix up!

B.D.

On Sun, 4 Jul 2004, Reini Urban wrote:

> Hi,
> You link from your sf.net site
> http://linux-7110.sourceforge.net/howtos/netbook_new/index.htm
> to our phpwiki, where you added a lot of pages linked to the start page 
> "http://phpwiki.sourceforge.net/phpwiki/PsilinuxNetbook" for your 
> documentation purposes.
> "[Wiki for Psilinux netBook] Please add information, questions, FAQ 
> (with answers?), using this Wiki" ...
> 
> But this phpwiki instance is only for documentation of phpwiki itself, 
> and not a general wiki which can be used for other projects.
> 
> Would you please so kind and setup your own version of phpwiki at your 
> own sf.net account and move your wiki pages over there. I can help with 
> that if you want.
> Someome sooner or later will remove your PsilinuxNetbook pages from
> http://phpwiki.sourceforge.net/phpwiki/ because they have nothing to do 
> with phpwiki.
> 
> gforge (as opensourced software version of sf.net) has lately integrated 
> phpwiki as wiki module, so that all gforge (and maybe sf.net some day) 
> projects will have their own instance of phpwiki for documentation 
> purposes. But sf.net has no wiki module yet.
> -- 
> Reini Urban
> http://phpwiki.sf.net/
> 
> 
> 
> 
> -------------------------------------------------------
> This SF.Net email sponsored by Black Hat Briefings & Training.
> Attend Black Hat Briefings & Training, Las Vegas July 24-29 - 
> digital self defense, top technical experts, no vendor pitches, 
> unmatched networking opportunities. Visit www.blackhat.com
> _______________________________________________
> Linux-7110-psion mailing list
> Lin...@li...
> https://lists.sourceforge.net/lists/listinfo/linux-7110-psion
> 
> 

[Phpwiki-talk] Re: [Linux-7110-psion] phpwiki

[Brian D. <du...@ap...>]

Oops!  My apologies...merely my ignorance.  At the moment I would not
know how to set up and move our wiki pages to psilinux....I'll look
into it, but certainly a brief sketch of how to do that would be
appreciated.

Since the page was at sourceforge, I assumed it was a Wiki for all...
Well, we'll soon set it right.  Thanks for letting me know about the
problem, rather than just deleting the pages!

Apologies again,

B.D.

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

Brian Dushaw
Applied Physics Laboratory
University of Washington
1013 N.E. 40th Street
Seattle, WA  98105-6698
(206) 685-4198   (206) 543-1300
(206) 543-6785 (fax)
du...@ap...

Web Page:  http://faculty.washington.edu/dushaw/index.html

On Sun, 4 Jul 2004, Reini Urban wrote:

> Hi,
> You link from your sf.net site
> http://linux-7110.sourceforge.net/howtos/netbook_new/index.htm
> to our phpwiki, where you added a lot of pages linked to the start page 
> "http://phpwiki.sourceforge.net/phpwiki/PsilinuxNetbook" for your 
> documentation purposes.
> "[Wiki for Psilinux netBook] Please add information, questions, FAQ 
> (with answers?), using this Wiki" ...
> 
> But this phpwiki instance is only for documentation of phpwiki itself, 
> and not a general wiki which can be used for other projects.
> 
> Would you please so kind and setup your own version of phpwiki at your 
> own sf.net account and move your wiki pages over there. I can help with 
> that if you want.
> Someome sooner or later will remove your PsilinuxNetbook pages from
> http://phpwiki.sourceforge.net/phpwiki/ because they have nothing to do 
> with phpwiki.
> 
> gforge (as opensourced software version of sf.net) has lately integrated 
> phpwiki as wiki module, so that all gforge (and maybe sf.net some day) 
> projects will have their own instance of phpwiki for documentation 
> purposes. But sf.net has no wiki module yet.
> -- 
> Reini Urban
> http://phpwiki.sf.net/
> 
> 
> 
> 
> -------------------------------------------------------
> This SF.Net email sponsored by Black Hat Briefings & Training.
> Attend Black Hat Briefings & Training, Las Vegas July 24-29 - 
> digital self defense, top technical experts, no vendor pitches, 
> unmatched networking opportunities. Visit www.blackhat.com
> _______________________________________________
> Linux-7110-psion mailing list
> Lin...@li...
> https://lists.sourceforge.net/lists/listinfo/linux-7110-psion
> 
> 

[Phpwiki-talk] phpwiki

[Reini U. <ru...@x-...>]

Hi,
You link from your sf.net site
http://linux-7110.sourceforge.net/howtos/netbook_new/index.htm
to our phpwiki, where you added a lot of pages linked to the start page 
"http://phpwiki.sourceforge.net/phpwiki/PsilinuxNetbook" for your 
documentation purposes.
"[Wiki for Psilinux netBook] Please add information, questions, FAQ 
(with answers?), using this Wiki" ...

But this phpwiki instance is only for documentation of phpwiki itself, 
and not a general wiki which can be used for other projects.

Would you please so kind and setup your own version of phpwiki at your 
own sf.net account and move your wiki pages over there. I can help with 
that if you want.
Someome sooner or later will remove your PsilinuxNetbook pages from
http://phpwiki.sourceforge.net/phpwiki/ because they have nothing to do 
with phpwiki.

gforge (as opensourced software version of sf.net) has lately integrated 
phpwiki as wiki module, so that all gforge (and maybe sf.net some day) 
projects will have their own instance of phpwiki for documentation 
purposes. But sf.net has no wiki module yet.
-- 
Reini Urban
http://phpwiki.sf.net/

Re: [Phpwiki-talk] Bug report: can't have just a 0 in a cell in a table

[Reini U. <ru...@x-...>]

Davide Andrea wrote:
> Edit a page
> And to it a table that has a cell (in a column other than the 1st one) whose
> contents is just the number zero.
> Hit preview
> The page refreshes but is totally blank
> 
> For example:
> A|
>  0

This was fixed recently. (with 1.3.8 if I remember correctly)

See http://phpwiki.sourceforge.net/phpwiki/TableBugZero (1.3.10)
  or http://phpwiki.sourceforge.net/demo/en/TableBugZero (latest CVS)

-- 
Reini Urban
http://xarch.tu-graz.ac.at/home/rurban/

[Phpwiki-talk] Bug report: can't have just a 0 in a cell in a table

[Davide A. <1d...@at...>]

Edit a page
And to it a table that has a cell (in a column other than the 1st one) whose
contents is just the number zero.
Hit preview
The page refreshes but is totally blank

For example:
A|
 0

Thanks
Davide Andrea
DAVIDE
PO Box 4825, Boulder CO 80306
1-303-413-1500
1d...@at...
http://DavideAndrea.com/
http://partnumber.com/

Re: [Phpwiki-talk] ADOdb backends...

[Reini U. <ru...@x-...>]

Philippe Vanhaesendonck wrote:
> Reini,
> The ADOdb/Oracle backend is working more or less (still have to fix a 
> couple of things), but I am not too happy with the current state...

Yes, me too... :)

> So here are a couple of questions about WikiDB and backends:
> 
> 1. Table names / database objects: where do the 'abstraction' happens?
> I was initially thinking that the backends were responsible for the 
> 'physical' implementation -- including table naming (e.g. _table_names 
> hash), but it looks it is not the case, since we find back table names 
> in WikiDB.php.
> Given the fact that WikiDB is not using the same names ('links' io 
> 'link') and does not uses prefixes, I am missing some 'translation 
> table' here...

The backend constructor holds the hash $this->_table_names, which is 
extracted for each method. Certain extensions (session, rating, dbprefs, 
...) add to this hash.

The "links" in WikiDB which you spotted is a typo. But it was for the 
intermediate upgrade step from locking to transactions.

I added the list of tables which need to be locked to the backend method 
of _lock_tables, so that the application lock can be generated (simply 
concat the tablenames).
I know, this is a bad hack, but better than locking all tables, and 
needed till we will have converted it to transactions where we really 
need them. At the beginning of each db transaction, and not within each 
and every submethod.

I wanted to have this finished for the 1.4.0 release.

> 2. locking mechanism
> This is looking a bit weird to me: we are locking all over the place: 
> WikiDB, backends, inside and outside transactions, ...
> Is there a rationale behind this?
> It does not look very compatible with a transactional DBMS which does 
> not hold locks longer than a transaction (and cannot unlock without 
> ending the transaction). The notion of transaction and locking are 
> tighly bound.

I know. Sigh.
The first version was for mysql only, which needed locking ALL tables 
(for MyISAM). A complete mess.
I tried to loose the locking a bit by slowly converting to use 
transactions and application locks instead, but this conversion is not 
yet finished. So better DBMS still have to suffer from the mysql 
architectural decisions.

For ADODB it's a bit better, since I rewrote it for 1.3.10 completely.
But we still have the mess from old WikiDB.php and it's _lock_tables() 
calls throughout all nested methods. All these have to be converted to 
transactions, and mysql as stupid case has to translate 
BeginTransaction() to lock_tables.

> 3. ADOdb generic backend
> We do have a lot of mysql specifics testing in there...
> eg:
>    if ((substr($dbh->databaseType,0,5) == 'mysql')....
> 
> Shouldn't we try to have a backend which is really generic, and have all 
> the database specifics in the ADODB_<db>.php subclass?
> 
> Current situation is relatively messy (no offense -- just my impressions)

That's absolutely right. For simplicity I left it as is, but the
dynamic $backend->_dbh->databaseType checks should be moved into the db 
specific backends. mysqli and mysqlt should be tested, ...

The real problem is mysql, which is a mess.
ADODB is also a mess, as you can see with the fetchMode hack.
-- 
Reini Urban
http://xarch.tu-graz.ac.at/home/rurban/