From roach.list at gmail.com  Wed Jul  2 08:55:09 2008
From: roach.list at gmail.com (Robert Spencer)
Date: Wed, 2 Jul 2008 14:55:09 +0200
Subject: [Moin-user] BadContent = chaos?
Message-ID: <8b8c4c740807020555n32f39c49x453a9961265ee4ca@mail.gmail.com>

Hi,

I'm trying to understand the BadContent page format, but it doesn't
appear to have a consistent one.

What type of regex is it supposed to use?
Who manages it?
Why isn't it sorted in a consistent order?
Who decides on which URLs get added?
Who decides on which URLs get removed?

If I understand things correctly, anyone can edit these pages as long
as it's on the moinmaster wiki. There's errors on a few help pages I'm
itching to correct too.

-- 
Robert Spencer



From tw-public at gmx.de  Wed Jul  2 11:06:17 2008
From: tw-public at gmx.de (Thomas Waldmann)
Date: Wed, 02 Jul 2008 17:06:17 +0200
Subject: [Moin-user] BadContent = chaos?
In-Reply-To: <8b8c4c740807020555n32f39c49x453a9961265ee4ca@mail.gmail.com>
References: <8b8c4c740807020555n32f39c49x453a9961265ee4ca@mail.gmail.com>
Message-ID: <486B9969.40901@gmx.de>


> I'm trying to understand the BadContent page format, but it doesn't
> appear to have a consistent one.

?

> What type of regex is it supposed to use?

Just standard python/perl regex. The regex is matched against all page 
content that someone tries to save (not only against URLs).

> Who manages it?

It is managed in a wiki way on MoinMaster:BadContent.

> Why isn't it sorted in a consistent order?

Because nobody did sort it recently, but everybody just appended at the end.

> Who decides on which URLs get added?

Everyone who cares.

Just be careful, everything matching a regex you added there is 
considered evil - globally.

For example: while mostly it will be spam when you see it, just the word 
Rolex or Casino would make no good regex, because there might be moin 
wikis out there where talk about Rolex or Casino is not spam, but within 
allowed wiki content.

So although the regexes are not limited to URLs, what one usually adds 
there are regexes that match URLs used by spammers.

> Who decides on which URLs get removed?

Same. Usually URLs only get removed if they were wrong and/or gave false 
positives.

> If I understand things correctly, anyone can edit these pages as long
> as it's on the moinmaster wiki.

Yes. You maybe have to read MoinMoin:EditingOnMoinMaster, though.

> There's errors on a few help pages I'm itching to correct too.

Easiest and quickest way is to as on our IRC channel #moin, see 
MoinMoin:MoinMoinChat.


BTW, you maybe want to use a recent moin version and activate the 
textchas. This has proven a better counter measure against spam than 
BadContent is, see HelpOnTextChas for details.



From roach.list at gmail.com  Wed Jul  2 12:06:46 2008
From: roach.list at gmail.com (Robert Spencer)
Date: Wed, 2 Jul 2008 18:06:46 +0200
Subject: [Moin-user] BadContent = chaos?
In-Reply-To: <486B9969.40901@gmx.de>
References: <8b8c4c740807020555n32f39c49x453a9961265ee4ca@mail.gmail.com>
	<486B9969.40901@gmx.de>
Message-ID: <8b8c4c740807020906q9b81e5bsf992e69cc2ed22d7@mail.gmail.com>

On 7/2/08, Thomas Waldmann <tw-public at gmx.de> wrote:
>
>> I'm trying to understand the BadContent page format, but it doesn't
>> appear to have a consistent one.
>
> ?

It appears at first glance to be regex based, but there are high amounts
of wrong entries.

>> What type of regex is it supposed to use?
>
> Just standard python/perl regex. The regex is matched against all page
> content that someone tries to save (not only against URLs).

Cool.

>> Who manages it?
>
> It is managed in a wiki way on MoinMaster:BadContent.

Good.

>> Why isn't it sorted in a consistent order?
>
> Because nobody did sort it recently, but everybody just appended at the end.

I'll re-organise it then.

>> Who decides on which URLs get added?
>
> Everyone who cares.
>
> Just be careful, everything matching a regex you added there is
> considered evil - globally.
<...>

Point taken.

> So although the regexes are not limited to URLs, what one usually adds
> there are regexes that match URLs used by spammers.

I've already seen one URL that should not be there.

>> Who decides on which URLs get removed?
>
> Same. Usually URLs only get removed if they were wrong and/or gave false
> positives.

Okay, I'll remove the URL.

>> If I understand things correctly, anyone can edit these pages as long
>> as it's on the moinmaster wiki.
>
> Yes. You maybe have to read MoinMoin:EditingOnMoinMaster, though.

Just have.

>> There's errors on a few help pages I'm itching to correct too.
>
> Easiest and quickest way is to as on our IRC channel #moin, see
> MoinMoin:MoinMoinChat.

I don't use IRC, I tried once. Didn't work out. Anyone new to IRC should
wear a diaper, have quick access to the toilet and avoid fibre for 7 days
before hand.

If you can suggested a guide for the complete idiot to IRC, I'll try
and read it.
No guarantees though.

> BTW, you maybe want to use a recent moin version and activate the
> textchas. This has proven a better counter measure against spam than
> BadContent is, see HelpOnTextChas for details.

There's a few points that need specifying here:

1. I do not have access to the server that our wiki is on.
2. The wiki has been marked for transfer to a new server, all attempts to
 change the current setup have therefore been rejected.
3. I despise CAPTCHA's.
4. Will always look for an alternative first.
5. Will try textchas on my test setup and then form an opinion.

-- 
Robert Spencer



From rick.vanderveer at gmail.com  Wed Jul  2 15:30:53 2008
From: rick.vanderveer at gmail.com (Rick Vanderveer)
Date: Wed, 2 Jul 2008 14:30:53 -0500
Subject: [Moin-user] linking to a URL in navi-bar broken?
Message-ID: <5c39e1ca0807021230p57a20eahe9ab143a71f4d6b2@mail.gmail.com>

It seems this broke between 1.7.0rc3 and 1.7.0 (final).  It was
working in all prior releases up to and including 1.7.0rc3.

In my tabs (navi-bar configuration in farmconfig.py), I have a link to
an external server. The code is:
        u'[http://mail.example.com Webmail]',

As of 1.7.0, it displays:
//mail.example.com Webmail]

Clicking on this tab takes me to the InterWiki page.  As far as I can
determine, it's trying to find an interwiki named "http:".  Am I
missing something?  Is there something I need to fix?

-Rick



From tw-public at gmx.de  Thu Jul  3 06:19:34 2008
From: tw-public at gmx.de (Thomas Waldmann)
Date: Thu, 03 Jul 2008 12:19:34 +0200
Subject: [Moin-user] BadContent = chaos?
In-Reply-To: <8b8c4c740807020906q9b81e5bsf992e69cc2ed22d7@mail.gmail.com>
References: <8b8c4c740807020555n32f39c49x453a9961265ee4ca@mail.gmail.com>	<486B9969.40901@gmx.de>
	<8b8c4c740807020906q9b81e5bsf992e69cc2ed22d7@mail.gmail.com>
Message-ID: <486CA7B6.30105@gmx.de>

> It appears at first glance to be regex based, but there are high amounts
> of wrong entries.

Define "wrong".

>>> Why isn't it sorted in a consistent order?
>> Because nobody did sort it recently, but everybody just appended at the end.
> 
> I'll re-organise it then.

Be extremely careful with that, because such a change will be hard to 
review.

> I've already seen one URL that should not be there.

Can happen.

>>> Who decides on which URLs get removed?
>> Same. Usually URLs only get removed if they were wrong and/or gave false
>> positives.
> 
> Okay, I'll remove the URL.

If you remove something, please give a edit comment with reason.

>>> There's errors on a few help pages I'm itching to correct too.
>> Easiest and quickest way is to as on our IRC channel #moin, see
>> MoinMoin:MoinMoinChat.
> 
> I don't use IRC, I tried once. Didn't work out. Anyone new to IRC should
> wear a diaper, have quick access to the toilet and avoid fibre for 7 days
> before hand.

X-)

> If you can suggested a guide for the complete idiot to IRC, I'll try
> and read it.
> No guarantees though.

Well, maybe try Wikipedia:IRC (or so) and google for IRC.

If you use linux, X-Chat is a nice irc client for beginners.

The MoinMoin:MoinMoinChat page also links to some web-based clients, so 
you don't have to setup one.

> 3. I despise CAPTCHA's.

TextChas are a bit different from Captchas, that's why they are called 
different. I also don't like Captchas for different reasons.

And there is even a "don't annoy our trusted editors" group. :)




From tw-public at gmx.de  Thu Jul  3 06:26:29 2008
From: tw-public at gmx.de (Thomas Waldmann)
Date: Thu, 03 Jul 2008 12:26:29 +0200
Subject: [Moin-user] linking to a URL in navi-bar broken?
In-Reply-To: <5c39e1ca0807021230p57a20eahe9ab143a71f4d6b2@mail.gmail.com>
References: <5c39e1ca0807021230p57a20eahe9ab143a71f4d6b2@mail.gmail.com>
Message-ID: <486CA955.60705@gmx.de>

> It seems this broke between 1.7.0rc3 and 1.7.0 (final).  It was
> working in all prior releases up to and including 1.7.0rc3.

Hmm, there is no changeset related to that between rc3 and release.

You are maybe rather comparing 1.6 to 1.7?

In 1.6 we changed link markup for wiki pages, but we forgot to change 
the "markup" used for the navibar. So 1.6 was a bit inconsistent as you 
had to use [[target|text]] on a wiki page, but [target text] for the 
navibar setup.

That was fixed in 1.7 (long before release IIRC) to use same markup for 
the navibar.

Note that it still does not use the same code for link parsing, so maybe 
don't expect that everything that works on a wiki page will also work in 
navibar, but the basic stuff will.

> In my tabs (navi-bar configuration in farmconfig.py), I have a link to
> an external server. The code is:
>         u'[http://mail.example.com Webmail]',

[[http://mail.example.com/ | Webmail]]

(BTW, I hope this is also described in docs/CHANGES :)



From lists at theflorys.org  Fri Jul  4 17:38:29 2008
From: lists at theflorys.org (David)
Date: Fri, 04 Jul 2008 17:38:29 -0400
Subject: [Moin-user] Cannot change Settings
Message-ID: <486E9855.2000608@theflorys.org>

I just upgraded from 1.5.x to 1.7.0 and I have a problem with settings.

I login to my home page and click "Settings".  Immediately I get an
error message "You must login to use this action: userprefs." and the
login seems to be canceled.  My user name is a superuser and has full
rights.

I am a relative newbie to Moin configuration.

David Flory




From lists at theflorys.org  Fri Jul  4 18:14:22 2008
From: lists at theflorys.org (David)
Date: Fri, 04 Jul 2008 18:14:22 -0400
Subject: [Moin-user] Login does not stick
Message-ID: <486EA0BE.6000607@theflorys.org>

I just upgraded from 1.5.x to 1.7.0 and I have a second problem with login.

I login to my home page.  My password is accepted and I see my login
name at the top of the page.  However, if I navigate to a new page on
the Wiki my login is gone.  My user name is a superuser and has full
rights.

I am a relative newbie to Moin configuration.

David Flory





From tw-public at gmx.de  Sat Jul  5 13:11:32 2008
From: tw-public at gmx.de (Thomas Waldmann)
Date: Sat, 05 Jul 2008 19:11:32 +0200
Subject: [Moin-user] Login does not stick
In-Reply-To: <486EA0BE.6000607@theflorys.org>
References: <486EA0BE.6000607@theflorys.org>
Message-ID: <1215277892.10277.2.camel@black.firma.waldmann-edv.de>

On Fri, 2008-07-04 at 18:14 -0400, David wrote:
> I just upgraded from 1.5.x to 1.7.0 and I have a second problem with login.
> 
> I login to my home page.  My password is accepted and I see my login
> name at the top of the page.  However, if I navigate to a new page on
> the Wiki my login is gone.  My user name is a superuser and has full
> rights.

There are 2 things to check:

a) if you have some auth = ... wikiconfig setting, please remove it, so
it uses the builtin default (in case you need some special auth stuff,
you can re-add it later).

b) check if your browser has cookies from your wiki site, if yes, delete
them all. then log in and see if you get a MOIN_SESSION cookie (if not,
check if your browser accepts cookies from your wiki site).





From lists at theflorys.org  Sun Jul  6 23:21:59 2008
From: lists at theflorys.org (David)
Date: Sun, 06 Jul 2008 23:21:59 -0400
Subject: [Moin-user] Login does not stick
In-Reply-To: <1215277892.10277.2.camel@black.firma.waldmann-edv.de>
References: <486EA0BE.6000607@theflorys.org>
	<1215277892.10277.2.camel@black.firma.waldmann-edv.de>
Message-ID: <48718BD7.4030403@theflorys.org>


On 7/5/2008 1:11 PM, Thomas Waldmann wrote:
> On Fri, 2008-07-04 at 18:14 -0400, David wrote:
>> I just upgraded from 1.5.x to 1.7.0 and I have a second problem with login.
>>
>> I login to my home page.  My password is accepted and I see my login
>> name at the top of the page.  However, if I navigate to a new page on
>> the Wiki my login is gone.  My user name is a superuser and has full
>> rights.
> 
> There are 2 things to check:
> 
> a) if you have some auth = ... wikiconfig setting, please remove it, so
> it uses the builtin default (in case you need some special auth stuff,
> you can re-add it later).
> 
> b) check if your browser has cookies from your wiki site, if yes, delete
> them all. then log in and see if you get a MOIN_SESSION cookie (if not,
> check if your browser accepts cookies from your wiki site).
> 

I have found the culprit but not the solution.  The problem was my 
firewall.  ZoneAlarm Security Suite v7 was quietly blocking cookies
from localhost:8080.  I had allowed localhost and Firefox but ZoneAlarm 
ignored the settings for the higher port.

Thanks for the suggestion...it gave me the key to troubleshoot.

David
> 
> 
> -------------------------------------------------------------------------
> Sponsored by: SourceForge.net Community Choice Awards: VOTE NOW!
> Studies have shown that voting for your favorite open source project,
> along with a healthy diet, reduces your potential for chronic lameness
> and boredom. Vote Now at http://www.sourceforge.net/community/cca08
> _______________________________________________
> Moin-user mailing list
> Moin-user at lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/moin-user
> 



From Karen.Janssens at ua.ac.be  Mon Jul  7 09:04:47 2008
From: Karen.Janssens at ua.ac.be (Janssens Karen)
Date: Mon, 7 Jul 2008 15:04:47 +0200
Subject: [Moin-user] default search
Message-ID: <4872146F.4060305@ua.ac.be>

Hi

By default hitting enter after entering a keyword in the searchbox leads 
to a title search. for my wiki, I would like to make it do a full-text 
search instead. Is it possible to change this default search behaviour, 
and, if yes, where can I adjust it?

Thanks in advance!

Karen

-- 
Karen Janssens
Universiteit Antwerpen
Anet - Bibliotheekautomatisering
Bibliotheek Stadscampus
Prinsstraat 9
B-2000 Antwerpen
Belgium
Tel    +32 3 220.49.71
Email  karen.janssens at ua.ac.be

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/moin-user/attachments/20080707/d8a20285/attachment.html>

From crosseyedpenguin at yahoo.com  Mon Jul  7 14:56:42 2008
From: crosseyedpenguin at yahoo.com (Roger Haase)
Date: Mon, 7 Jul 2008 11:56:42 -0700 (PDT)
Subject: [Moin-user] default search
In-Reply-To: <4872146F.4060305@ua.ac.be>
Message-ID: <783355.90283.qm@web36203.mail.mud.yahoo.com>


--- On Mon, 7/7/08, Janssens Karen <Karen.Janssens at ua.ac.be> wrote:

> From: Janssens Karen <Karen.Janssens at ua.ac.be>
> Subject: [Moin-user] default search
> To: moin-user at lists.sourceforge.net
> Date: Monday, July 7, 2008, 6:04 AM
> Hi
> 
> By default hitting enter after entering a keyword in the
> searchbox leads 
> to a title search. for my wiki, I would like to make it do
> a full-text 
> search instead. Is it possible to change this default
> search behaviour, 
> and, if yes, where can I adjust it?
> 
> Thanks in advance!
> 
> Karen
> 

There is no configuration option to switch the position of the Titles and Text buttons on the search form.

The easy way is to enter a value and then remember to hit tab twice, then enter.

To do it the hard way, the positions of the buttons are defined in the "searchform" method in /theme/__init__.py.  For release 1.7.0 look for the following near line 760:

<input id="titlesearch" name="titlesearch" type="submit"
    value="%(search_title_label)s" alt="Search Titles">
<input id="fullsearch" name="fullsearch" type="submit"
    value="%(search_full_label)s" alt="Search Full Text">

To change the position of the buttons, cut and paste the fullsearch input tag to be before the titlesearch tag.  If doing that has no effect and you are using a custom theme, then your theme is overriding the searchform method and a similar change must be made within the theme.

Roger Haase 




      



From Greg.Keith at noaa.gov  Tue Jul  8 17:11:52 2008
From: Greg.Keith at noaa.gov (Greg Keith)
Date: Tue, 08 Jul 2008 15:11:52 -0600
Subject: [Moin-user] Which elements containing old user data can I remove?
Message-ID: <4873D818.4060805@noaa.gov>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
 
Hi all-

I recently updated several wikis from MoinMoin 1.5.7 to MoinMoin 1.6.3 
on RHEL 5.1, Python 2.4.3. At the request of our security guy, I also 
changed user authentication so that instead of using their MoinMoin 
CamelCase wiki account usernames, wiki users are authenticated using 
their ActiveDirectory usernames and passwords. So typically, usernames 
like GregButler are now gbutler.

However, I now have some user complaints that they are having error 
messages when they log in, such as:

AttributeError: User instance has no attribute 'aliasname'

...and that when they edit and then save changes to a page, when the 
page refreshes, their edits are lost.

I'd like to make a clean break between the old account names & profiles 
and the new info, if possible. I am wondering if any of these problems 
could be caused by users who are logging in with their old wikiname 
accounts, even after we have set up the new AD accounts for them. Is it 
possible that there are conflicts between cached old usernames and 
profile info in various places? If so, are there old cache items I can 
remove that Moin will re-create automatically? For instance, can I 
remove the:
/data/event-log file
name2id file
entire contents of /data/cache?
any other files?

Thanks for any help in understanding this!

Greg


-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.7 (MingW32)
 
iD8DBQFIc9gY8IR34NeP2BwRAhs9AJ45slwqcSZRek5XXL1f1PtRpewQ/ACggvd9
wKbZpxS3Txx4DzWfcsD6J6Y=
=7Zgg
-----END PGP SIGNATURE-----




From lists at onemanifest.net  Tue Jul  8 17:34:59 2008
From: lists at onemanifest.net (oneman)
Date: Tue, 8 Jul 2008 23:34:59 +0200
Subject: [Moin-user] globally disable icon toolbar on 1.5.3
Message-ID: <3760CA33-8E3D-4306-886A-99635D9A7A0D@onemanifest.net>

Hi All,



How can I globally disable the icon toolbar in Moinmoin 1.5.3? The  
user has a userpreference to turn it on or off. But I'd like to turn  
it of globally for a wiki that is used as a makeshift CMS. I looked  
around but couldn't find a config setting for it.






Groet,


Peter Teunissen

-- Never argue with idiots; they'll drag you down to their own level  
and beat you on experience.







From florian.biree at gmail.com  Tue Jul  8 18:33:56 2008
From: florian.biree at gmail.com (=?UTF-8?B?RmxvcmlhbiBCaXLDqWU=?=)
Date: Wed, 09 Jul 2008 00:33:56 +0200
Subject: [Moin-user] Croquant: a set of MoinMoin plugins for an integration
	with Crunchy
Message-ID: <4873EB54.10702@gmail.com>

Hello,

I proud to announce the 1.0 release of Croquant 
<http://code.google.com/p/croquant/>.

Croquant is a set of MoinMoin plugins (currently one parser, three 
macros and one theme) to allow to use a MoinMoin wiki to write tutorials 
for Crunchy (<http://code.google.com/p/crunchy/>).

Crunchy is a software which transform a static html tutorial to learn 
the python language into a dynamic tutorial with the python interpreter 
embedded inside.
With MoinMoin and Croquant, a tutorial can be written in an easy and 
collaborative way.

This is a part of my Google Summer of Code 2008 for Crunchy. All 
feedback is welcome!

Cheers,
-- 
Thesa ~ Florian Bir?e
e-mail : florian at biree.name
Messagerie Instantan?e Jabber/XMPP/Google Talk : florian.biree at jabber.fr
Site web : http://florian.biree.name/
Carnet web : http://filyb.info/

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 252 bytes
Desc: OpenPGP digital signature
URL: <http://mail.python.org/pipermail/moin-user/attachments/20080709/729d1742/attachment.sig>

From yahoo at jimpop.com  Tue Jul  8 22:43:43 2008
From: yahoo at jimpop.com (Jim Popovitch)
Date: Tue, 8 Jul 2008 22:43:43 -0400
Subject: [Moin-user] globally disable icon toolbar on 1.5.3
In-Reply-To: <3760CA33-8E3D-4306-886A-99635D9A7A0D@onemanifest.net>
References: <3760CA33-8E3D-4306-886A-99635D9A7A0D@onemanifest.net>
Message-ID: <7ff145960807081943r69e5bc63tf3858acf68fb0633@mail.gmail.com>

On Tue, Jul 8, 2008 at 5:34 PM, oneman <lists at onemanifest.net> wrote:
> Hi All,
>
> How can I globally disable the icon toolbar in Moinmoin 1.5.3? The
> user has a userpreference to turn it on or off. But I'd like to turn
> it of globally for a wiki that is used as a makeshift CMS. I looked
> around but couldn't find a config setting for it.


I *think*, but am not 100% sure, that if you set page_iconbar to
nothing you will get what you want.

    page_iconbar = [,]

-Jim P.



From lists at onemanifest.net  Wed Jul  9 01:06:40 2008
From: lists at onemanifest.net (oneman)
Date: Wed, 9 Jul 2008 07:06:40 +0200
Subject: [Moin-user] globally disable icon toolbar on 1.5.3
In-Reply-To: <7ff145960807081943r69e5bc63tf3858acf68fb0633@mail.gmail.com>
References: <3760CA33-8E3D-4306-886A-99635D9A7A0D@onemanifest.net>
	<7ff145960807081943r69e5bc63tf3858acf68fb0633@mail.gmail.com>
Message-ID: <ACE21E37-BD58-4F43-A368-D260C4818C53@onemanifest.net>


On 9-jul-2008, at 4:43, Jim Popovitch wrote:

> On Tue, Jul 8, 2008 at 5:34 PM, oneman <lists at onemanifest.net> wrote:
>> Hi All,
>>
>> How can I globally disable the icon toolbar in Moinmoin 1.5.3? The
>> user has a userpreference to turn it on or off. But I'd like to turn
>> it of globally for a wiki that is used as a makeshift CMS. I looked
>> around but couldn't find a config setting for it.
>
>
> I *think*, but am not 100% sure, that if you set page_iconbar to
> nothing you will get what you want.
>
>     page_iconbar = [,]
>
No, doesn't work. In the config file it leads to an error or, in the  
case of page_iconbar = [], has no effect. When editing the <theme>.py  
is results in an error too, using page_iconbar = [] results in the  
text 'none' being displayed on the page.

I've got the same issue with the page trail. There don't seem to be  
an easy way to force userprefs globally like you can for the theme.

Maybe a combination of setting the default userprefs in multiconfig  
and hiding them on the prefs page might work, I'll look into that later.


thanks,


Peter



From tw-public at gmx.de  Sun Jul 13 14:26:11 2008
From: tw-public at gmx.de (Thomas Waldmann)
Date: Sun, 13 Jul 2008 20:26:11 +0200
Subject: [Moin-user] Which elements containing old user data can
	I	remove?
In-Reply-To: <4873D818.4060805@noaa.gov>
References: <4873D818.4060805@noaa.gov>
Message-ID: <1215973571.6257.9.camel@black.firma.waldmann-edv.de>


> However, I now have some user complaints that they are having error 
> messages when they log in, such as:
> 
> AttributeError: User instance has no attribute 'aliasname'

This is a bug. Either in the LDAP auth configuration or in the LDAP auth
code. Please file a bug report on the wiki and provide:
 * traceback.html (just save the it in the browser when you see the
AttributeError)
 * your ldap auth related wiki configuration parts (remove secret
information)

> For instance, can I remove the:
> /data/event-log file

Yes, if you don't need statistics, you can truncate this any time.

> name2id file
> entire contents of /data/cache?

Yes, any stuff in the cache/ directory should be rebuilt automatically
(for playing completely safe, stop wiki, delete stuff, start wiki).

BTW, if your user names are different now, your ACLs maybe don't work as
they did before (because ACLs are based on user or group names, groups
have user names). So check if everything is still matching.





From tw-public at gmx.de  Sun Jul 13 14:59:19 2008
From: tw-public at gmx.de (Thomas Waldmann)
Date: Sun, 13 Jul 2008 20:59:19 +0200
Subject: [Moin-user] Croquant: a set of MoinMoin plugins for
	an	integration with Crunchy
In-Reply-To: <4873EB54.10702@gmail.com>
References: <4873EB54.10702@gmail.com>
Message-ID: <1215975559.6257.27.camel@black.firma.waldmann-edv.de>

Hi Florian,

just some quick comments (I don't have any clue about crunchy, so these
comments are just from a moin developer perspective):

macro/ExternalLink
==================
I think everthing you are doing there can be also done by using moin
1.6/1.7 link syntax, e.g.:

[[url|label|title=external_link]]

(not sure about for what that title is useful, I rather would've
expected some class=...)

If you really need an own macro for this, maybe use the argument parser
from wikiutil and don't do your own macro argument parsing).

macro/LoadLocal and LoadRemote
==============================
We have formatter.span().

parser/crunchy
==============
Not sure you need that DummyRequest stuff, see wikiutil.RenderText and
check if that would work for you.

style/crunchy
=============
Should be theme/crunchy. Config stuff should be in different directory.

Both wikiconfigs have indentation errors at line 12.

BTW, do you have some public demo site running your code?

Cheers,

Thomas





From georgeolivergo at yahoo.com  Sun Jul 13 21:32:32 2008
From: georgeolivergo at yahoo.com (George Oliver)
Date: Sun, 13 Jul 2008 18:32:32 -0700 (PDT)
Subject: [Moin-user] add links to header in Balanced theme?
Message-ID: <767702.94010.qm@web52011.mail.re2.yahoo.com>

hi, I'm a new user of MoinMoin trying to add some links to the header using the Balanced theme, so far without real success.

Here is what I've tried:

In the wiki's wikiconfig.py:


   navi_bar = [
        # If you want to show your page_front_page here:
        #u'%(page_front_page)s',
        u'RecentChanges',
        u'FindPage',
        u'[[http://www.matingball/forums | Forum]]'
        u'HelpContents',
              ]

In the wiki's data/plugin/theme/balanced.py I added 

u'<a href="http://www.matingball/forums">Forum</a>', 

in a couple of places, such as the header's html list, and was able to get the Forum link up on the screen, but never really aligned with the layout of the other html.

In balanced.py there is this code for the header:

            # Header
            u'<div id="header">',
            u'<div id="mastwrap"><div id="masthead">',
            #self.logo(),
            self.searchform(d),
            self.username(d),
                        # uncomment to include tabsbar. Also make CSS change to make room for it
            #self.sisternav(),
            u'</div></div>',
            self.msg(d),
            #self.navibar(d),    <------  note this here
            #self.trail(d),
            self.extranav(d),
            self.editbar(d),
            u'</div>',
                        # uncomment to include sidebar. Also make CSS change to make room for it
            #self.sidemenu(d),

            # Post header custom html (not recommended)
            #self.emit_custom_html(self.cfg.page_header2),

            # Start of page
            self.startPage(),
            self.title(d),
        ]
        return u'\n'.join(html)


I tried uncommenting the line with self.navibar, and also changing this to self.navi_bar (as in wikiconfig.py it is navi_bar) but didn't have success down this route. 

Does anyone have some pointers to figuring out how to add links to the header?

Is the balanced theme over-riding or not using the wiki's wikiconfig.py when it comes to the navibar?

Also, a related question -- I've noticed that after making these changes,  I can refresh the wiki multiple times and get a different refreshed page every time, usually cycling through two or three versions of the same page (for example, if I remove an element from the header, refresh, then put it back in, refreshing a few times might make the element appear, disappear, reappear, etc.). It's not at all predictable from my perspective. What is going on here when I do these refreshes? 

The site uses the latest version of Moin and mod_python under Apache.


thanks, George


      



From huangchonggnu at gmail.com  Mon Jul 14 04:26:49 2008
From: huangchonggnu at gmail.com (hch)
Date: Mon, 14 Jul 2008 16:26:49 +0800
Subject: [Moin-user] need help on cgi mode setting up
Message-ID: <cb49ae490807140126x32ea5a97xaf7346e96b749b89@mail.gmail.com>

Hi, all:

My wiki runs as cgi mode, but the apache server complains that "
(13)Permission denied: exec of '/usr/local/share/moin/wiki/cgi-bin/moin.cgi'
failed" when i visit http://ipaddress/wiki <http://218.247.216.21:8000/wiki>.
I've checked the that "-rwxr-x--- 1 apache apache 1737 Jul 11 14:09
moin.cgi" and make sure the files under wiki/ diretory can be executed by
the apache user.  I think the problem must be relate with the selinux rules,
for after i turn off selinux, the moin.cgi runs well.

I check with cmd "getsebool -a |grep httpd"

allow_httpd_anon_write --> off
allow_httpd_bugzilla_script_anon_write --> off
allow_httpd_mod_auth_pam --> off
allow_httpd_nagios_script_anon_write --> off
allow_httpd_squid_script_anon_write --> off
allow_httpd_sys_script_anon_write --> off
httpd_builtin_scripting --> on
httpd_can_network_connect --> off
httpd_can_network_connect_db --> off
httpd_can_network_relay --> off
httpd_disable_trans --> off
httpd_enable_cgi --> on
httpd_enable_ftp_server --> off
httpd_enable_homedirs --> on
httpd_rotatelogs_disable_trans --> off
httpd_ssi_exec --> off
httpd_suexec_disable_trans --> off
httpd_tty_comm --> off
httpd_unified --> on

I can not figure out which item is not right.
My working env is: CentOS release 5 (Final), moinmoin 1.7.0,

Any advices?
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/moin-user/attachments/20080714/023e2742/attachment.html>

From florian.biree at gmail.com  Mon Jul 14 10:27:17 2008
From: florian.biree at gmail.com (=?ISO-8859-1?Q?Florian_Bir=E9e?=)
Date: Mon, 14 Jul 2008 16:27:17 +0200
Subject: [Moin-user] Croquant: a set of MoinMoin plugins for an
 integration with Crunchy
In-Reply-To: <1215975559.6257.27.camel@black.firma.waldmann-edv.de>
References: <4873EB54.10702@gmail.com>
	<1215975559.6257.27.camel@black.firma.waldmann-edv.de>
Message-ID: <487B6245.50000@gmail.com>

Hi Thomas,

Thomas Waldmann a ?crit :
> just some quick comments (I don't have any clue about crunchy, so these
> comments are just from a moin developer perspective):

Many thanks for your comments!

> macro/ExternalLink
> ==================
> I think everthing you are doing there can be also done by using moin
> 1.6/1.7 link syntax, e.g.:
> 
> [[url|label|title=external_link]]

True. This syntax is not documented in the HelpOnMoinWikiSyntax page, it 
is why I made this macro. Now I removed it (and add this moin syntax in 
the doc/syntax.rst file).

> (not sure about for what that title is useful, I rather would've
> expected some class=...)

In fact, it is a design choice of Crunchy. Crunchy parse the title 
attribute of some html elements (yes, it could be class, but it's not) 
to know what it should do.

> macro/LoadLocal and LoadRemote
> ==============================
> We have formatter.span().

Now they use it. Thanks.

> parser/crunchy
> ==============
> Not sure you need that DummyRequest stuff, see wikiutil.RenderText and
> check if that would work for you.

I've just tried to use it, but it seems that wikiutil.renderText call 
the format method of a parser with the inhibit_p=True argument.
But the python parser have not any inhibit_p attribute, so it doesn't work.

> style/crunchy
> =============
> Should be theme/crunchy. Config stuff should be in different directory.
> 
> Both wikiconfigs have indentation errors at line 12.

Changes made.

Here is the archive with the modifications: 
http://croquant.googlecode.com/files/croquant-1.0svn26.tar.bz2

> BTW, do you have some public demo site running your code?

No, not for the moment. I've quickly tried to make one on my server, but 
the MoinMoin packaged (Ubuntu hardy) is an old 1.5. I may install one 
from the source. For the moment, I worked with MoinMoin DesktopEdition.
But in all cases, to really use it, you need to have a local Crunchy 
(Crunchy is not designed - for security reason - to run in a server).

Yours,
-- 
Thesa ~ Florian Bir?e
e-mail : florian at biree.name
Messagerie Instantan?e Jabber/XMPP/Google Talk : florian.biree at jabber.fr
Site web : http://florian.biree.name/
Carnet web : http://filyb.info/

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 252 bytes
Desc: OpenPGP digital signature
URL: <http://mail.python.org/pipermail/moin-user/attachments/20080714/2b1d951c/attachment.sig>

From matthew.brett at gmail.com  Mon Jul 14 11:29:58 2008
From: matthew.brett at gmail.com (Matthew Brett)
Date: Mon, 14 Jul 2008 16:29:58 +0100
Subject: [Moin-user] Cached page permissions?
Message-ID: <1e2af89e0807140829w37156688i40c0167b332706b8@mail.gmail.com>

Hi,

I am using 1.5.6, and have noticed the following:

An unauthorized user attempts to read a page requiring authorization
and fails (you are not allowed to view this page)

I authorize them by adding them to the relevant ACL page, may or may
not restart the server

The user tries again to access the same page, and is again refused.

The fix is to edit the page in question with some trivial edit, at
which point the permissions become active.

This seems to be independent of the user's browser, because I get the
same behavior when I log in as the user and try and access the page.

I have tried logging on as the admin user and using the 'Delete cache'
action - this has no effect.

I conclude that there is some kind of user-specific cache that is only
cleared by a page edit.   At a first pass I can't see where this is,
and I was wondering what the correct way to clear this cache might be.

Thanks a lot,

Matthew



From georgeolivergo at yahoo.com  Tue Jul 15 21:47:34 2008
From: georgeolivergo at yahoo.com (George Oliver)
Date: Tue, 15 Jul 2008 18:47:34 -0700 (PDT)
Subject: [Moin-user] add links to header in Balanced theme?
Message-ID: <531780.59804.qm@web52010.mail.re2.yahoo.com>

Just wanted to post up my answer to this earlier question, as I hadn't found a short and clear answer in the archives when I checked. Of course the answer is a fairly easy one :). 

In balanced.py there are two functions, header() and footer(), each returning a list called html (among other items). The list html is composed of a number of functions which display things like the search form, editing links, and so forth, and to add a new link you just need to choose which function you want to add to. 

I chose the function username(); that function returns a list called userlinks, and to add a link all you do is

userlinks.append(u'<a href="<your link goes here>">Your link name</a>') 

Do this just before the list comprehension at the end of the username() defintion (or wherever is appropriate if you choose a different function). 

Your link probably will be relative, like "/<relative link". 

As to the problem of refreshing, I'm not totally clear on the issues but what I needed to do was delete the compiled balanced.pyc and restart the Apache server after a change. This seems to clear up the problem, by I'm guessing generating a new .pyc from the edited balanced.py.  


      



From jason at petsod.org  Thu Jul 17 10:25:34 2008
From: jason at petsod.org (Jason Petsod)
Date: Thu, 17 Jul 2008 10:25:34 -0400
Subject: [Moin-user] Tracebacks for exceptions in macros
Message-ID: <20080717142534.GA12581@petsod.org>

Hello,

I am trying to write a couple macros and actions on MM 1.7 and would like to
enable tracebacks to see how badly I screw up. So I uncommented the
following line in my moin.fcgi but I don't seem to get tracebacks whenever
my macro breaks and raises an exception.
----
os.environ['MOIN_DEBUG'] = '1'
----

I've also explicitly enabled "show_traceback" in my wikiconfig.py.

As an example, I've written a macro that will always raise an exception:
----
elysium /var/www/wikitest.quadco.net/data/plugin/macro $ cat BrokenMacro.py
def macro_BrokenMacro(macro):
    raise Exception("foo")
----

See http://wikitest.quadco.net/BrokenMacro for an example. I would expect to
see some sort of traceback of "Exception: foo".

If it matters, I'm running lighttpd 1.5.18.

Thanks,
-Jason



From jason at petsod.org  Thu Jul 17 13:05:33 2008
From: jason at petsod.org (Jason Petsod)
Date: Thu, 17 Jul 2008 13:05:33 -0400
Subject: [Moin-user] Tracebacks for exceptions in macros
In-Reply-To: <20080717142534.GA12581@petsod.org>
References: <20080717142534.GA12581@petsod.org>
Message-ID: <20080717170533.GB12581@petsod.org>

On 17/07/2008 at 10:25:34 -0400, Jason Petsod wrote:
> os.environ['MOIN_DEBUG'] = '1'

So apparently the trick is to leave this line commented. I commented it out
and tracebacks showed up.

-Jason



From jpenguin at gmail.com  Sun Jul 20 06:10:12 2008
From: jpenguin at gmail.com (Zhenlei Cai (Jerry))
Date: Sun, 20 Jul 2008 18:10:12 +0800
Subject: [Moin-user] MoinMoin 1.7.0 AttachFIle broken under mod_python 3.3.1?
Message-ID: <7035af30807200310w24c5c8cag53a8c35d38dadf06@mail.gmail.com>

Once I upgrade to 1.7.0, attaching file always fail with a "I/O operation on
closed file"  error (stack attached at end). After tracing the code, I found
I am using mod_python 3.3.1. Its util.Field class will close the uploaded
temp file in its destructor.  Apparently MoinMoin save this file in its form
but since the original Field object has been garbage collected when Moin
later trys to copy the saved file the file has been closed and hence the
error.
Please confirm if this is a bug.



/usr/lib/python2.5/site-packages/MoinMoin/request/__init__.py in run
(self=<MoinMoin.request.request_modpython.Request object at 0x2aaabe10c950>)

 1318                          self.page.send_page()
 1319                      else:
 1320                          handler(self.page.page_name, self)
 1321
 1322              # every action that didn't use to raise MoinMoinFinish
must call this now:
handler = <function execute at 0x2aaabea905f0> self =
<MoinMoin.request.request_modpython.Request object at 0x2aaabe10c950>
self.page = <MoinMoin.Page.Page object at 0x2aaabea8f610>
self.page.page_name = u'MileStones'
/usr/lib/python2.5/site-packages/MoinMoin/action/AttachFile.py in execute
(pagename=u'MileStones', request=<MoinMoin.request.request_modpython.Request
object at 0x2aaabe10c950>)

  511      handler = globals().get('_do_%s' % do[0])
  512      if handler:
  513          msg = handler(pagename, request)
  514      else:
  515          msg = _('Unsupported AttachFile sub-action: %s') %
(wikiutil.escape(do[0]), )
msg undefined handler = <function _do_upload at 0x2aaabea907d0> pagename =
u'MileStones' request = <MoinMoin.request.request_modpython.Request object
at 0x2aaabe10c950>
/usr/lib/python2.5/site-packages/MoinMoin/action/AttachFile.py in _do_upload
(pagename=u'MileStones', request=<MoinMoin.request.request_modpython.Request
object at 0x2aaabe10c950>)

  590      # add the attachment
  591      try:
  592          target, bytes = add_attachment(request, pagename, target,
filecontent, overwrite=overwrite)
  593          msg = _("Attachment '%(target)s' (remote name
'%(filename)s')"
  594                  " with %(bytes)d bytes saved.") % {
target = u'netbank.pdf' bytes undefined global add_attachment = <function
add_attachment at 0x2aaabea90140> request =
<MoinMoin.request.request_modpython.Request object at 0x2aaabe10c950>
pagename = u'MileStones' filecontent = <closed file '<fdopen>', mode 'w+b'
at 0x2aaabe1135d0> overwrite = 0
/usr/lib/python2.5/site-packages/MoinMoin/action/AttachFile.py in
add_attachment (request=<MoinMoin.request.request_modpython.Request object
at 0x2aaabe10c950>, pagename=u'MileStones', target=u'netbank.pdf',
filecontent=<closed file '<fdopen>', mode 'w+b' at 0x2aaabe1135d0>,
overwrite=0)

  208          stream = open(fpath, 'wb')
  209          try:
  210              _write_stream(filecontent, stream)
  211          finally:
  212              stream.close()
global _write_stream = <function _write_stream at 0x2aaabea900c8>
filecontent = <closed file '<fdopen>', mode 'w+b' at 0x2aaabe1135d0> stream
= <closed file '/var/opt/wiki/gaocandev/data/pages...hments/netbank.pdf',
mode 'wb' at 0x2aaabeaab7b0>
/usr/lib/python2.5/site-packages/MoinMoin/action/AttachFile.py in
_write_stream (content=<closed file '<fdopen>', mode 'w+b' at
0x2aaabe1135d0>, stream=<closed file
'/var/opt/wiki/gaocandev/data/pages...hments/netbank.pdf', mode 'wb' at
0x2aaabeaab7b0>, bufsize=8192)

  175      if hasattr(content, 'read'): # looks file-like
  176          import shutil
  177          shutil.copyfileobj(content, stream, bufsize)
  178      elif isinstance(content, str):
  179          stream.write(content)
shutil = <module 'shutil' from '/usr/lib64/python2.5/shutil.pyc'>
shutil.copyfileobj = <function copyfileobj at 0x2aaabdb30ed8> content =
<closed file '<fdopen>', mode 'w+b' at 0x2aaabe1135d0> stream = <closed file
'/var/opt/wiki/gaocandev/data/pages...hments/netbank.pdf', mode 'wb' at
0x2aaabeaab7b0> bufsize = 8192
/usr/lib64/python2.5/shutil.py in copyfileobj (fsrc=<closed file '<fdopen>',
mode 'w+b' at 0x2aaabe1135d0>, fdst=<closed file
'/var/opt/wiki/gaocandev/data/pages...hments/netbank.pdf', mode 'wb' at
0x2aaabeaab7b0>, length=8192)

   19      """copy data from file-like object fsrc to file-like object
fdst"""
   20      while 1:
   21          buf = fsrc.read(length)
   22          if not buf:
   23              break
buf undefined fsrc = <closed file '<fdopen>', mode 'w+b' at 0x2aaabe1135d0>
fsrc.read = <built-in method read of file object at 0x2aaabe1135d0> length =
8192
ValueError
I/O operation on closed file

args = ('I/O operation on closed file',)
message = 'I/O operation on closed file'
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/moin-user/attachments/20080720/99567910/attachment.html>

From crosseyedpenguin at yahoo.com  Sun Jul 20 11:25:59 2008
From: crosseyedpenguin at yahoo.com (Roger Haase)
Date: Sun, 20 Jul 2008 08:25:59 -0700 (PDT)
Subject: [Moin-user] MoinMoin 1.7.0 AttachFIle broken under mod_python
	3.3.1?
In-Reply-To: <7035af30807200310w24c5c8cag53a8c35d38dadf06@mail.gmail.com>
Message-ID: <960833.85554.qm@web36203.mail.mud.yahoo.com>




--- On Sun, 7/20/08, Zhenlei Cai (Jerry) <jpenguin at gmail.com> wrote:

> From: Zhenlei Cai (Jerry) <jpenguin at gmail.com>
> Subject: [Moin-user] MoinMoin 1.7.0 AttachFIle broken under mod_python 3.3.1?
> To: moin-user at lists.sourceforge.net
> Date: Sunday, July 20, 2008, 3:10 AM
> Once I upgrade to 1.7.0, attaching file always fail with a
> "I/O operation on
> closed file"  error (stack attached at end). After
> tracing the code, I found
> I am using mod_python 3.3.1. Its util.Field class will
> close the uploaded
> temp file in its destructor.  Apparently MoinMoin save this
> file in its form
> but since the original Field object has been garbage
> collected when Moin
> later trys to copy the saved file the file has been closed
> and hence the
> error.
> Please confirm if this is a bug.
> 

This bug with a possible fix is discussed here:
  http://moinmo.in/MoinMoinBugs/1.7ModPythonAttachmentUploadFails

Some/many of the former mod-python users (including me) have switched to mod_wsgi.  See:
  http://master17.moinmo.in/HelpOnInstalling/ApacheWithModWSGI

Roger Haase


      



From jbardin at bu.edu  Sun Jul 20 13:06:09 2008
From: jbardin at bu.edu (james bardin)
Date: Sun, 20 Jul 2008 13:06:09 -0400
Subject: [Moin-user] MoinMoin 1.7.0 AttachFIle broken under mod_python
	3.3.1?
In-Reply-To: <a3b675320807200948r5431f310jfba8b89ae2b1b4bb@mail.gmail.com>
References: <7035af30807200310w24c5c8cag53a8c35d38dadf06@mail.gmail.com>
	<960833.85554.qm@web36203.mail.mud.yahoo.com>
	<a3b675320807200948r5431f310jfba8b89ae2b1b4bb@mail.gmail.com>
Message-ID: <a3b675320807201006g4520bff4x27384d1f887dbee1@mail.gmail.com>

On Sun, Jul 20, 2008 at 12:48 PM, james bardin <jbardin at bu.edu> wrote:

>
>> I also just upgraded to 1.7, and I fully tested unset cgi, but failed to
> test attachments with mod_python.
> This is a BIG problem for me now, and the above patch doesn't help. Does
> any one have any other ideas? I would be fine moving to wsgi, but I haven't
> gotten that to work with one of out auth systems yet because it needs to be
> in an apache <Location> or <Directory>.
>
> Thanks
> -jim
>
>
>
Sorry, I did get the patch working.
If anyone has ideas on how to run a wikifarm, with separate <location>
directives for each wiki, under wsgi, I'd love to convert.

Thanks
-jim
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/moin-user/attachments/20080720/dc056568/attachment.html>

From tw-public at gmx.de  Sun Jul 20 13:41:42 2008
From: tw-public at gmx.de (Thomas Waldmann)
Date: Sun, 20 Jul 2008 19:41:42 +0200
Subject: [Moin-user] moin 1.7.1 released!
Message-ID: <1216575702.13696.4.camel@black.firma.waldmann-edv.de>


See http://moinmo.in/MoinMoinDownload for the release archive and the
change log.

Some system text translations and help pages still need improvement - if
you know a non-english language, you can also help translating and/or
fixing stuff on master17, see: http://moinmo.in/MoinDev/Translation -
your work will get into 1.7.2 release then.





From bernd.bartmann at gmail.com  Mon Jul 21 08:23:54 2008
From: bernd.bartmann at gmail.com (Bernd Bartmann)
Date: Mon, 21 Jul 2008 14:23:54 +0200
Subject: [Moin-user] Problem with upgrade to moin-1.7.1 and AD auth
Message-ID: <6c18a4f0807210523y3f500ae1rbaba19722a609102@mail.gmail.com>

Hi,

I'm trying to upgrade from moin-1.6.2 to moin-1.7.1. After following
the HelpOnUpdating guide to the point where I should run  the
"migation data" command I get an error message when trying to access
the wiki:


ConfigurationError
ImportError: cannot import name http
Check that the file is in the same directory as the server script. If
it is not, you must add the path of the directory where the file is
located to the python path in the server script. See the comments at
the top of the server script.
Check that the configuration file name is either "wikiconfig.py" or
the module name specified in the wikis list in farmconfig.py. Note
that the module name does not include the ".py" suffix.
Error in your configuration file "D:/moin/rd_wiki\wikiconfig.py"
around line 189.

The problematic line seems to be related to the http auth code that I
use for AD authentication:

from MoinMoin.auth.http import http
from MoinMoin.auth import moin_session
auth = [http, moin_session]

Any ideas what I can do to solve the issue?

Best regard,
Bernd.



From bernd.bartmann at gmail.com  Mon Jul 21 08:53:13 2008
From: bernd.bartmann at gmail.com (Bernd Bartmann)
Date: Mon, 21 Jul 2008 14:53:13 +0200
Subject: [Moin-user] Problem with upgrade to moin-1.7.1 and AD auth
In-Reply-To: <6c18a4f0807210523y3f500ae1rbaba19722a609102@mail.gmail.com>
References: <6c18a4f0807210523y3f500ae1rbaba19722a609102@mail.gmail.com>
Message-ID: <6c18a4f0807210553q41ac900fxca0639c29ceb30e6@mail.gmail.com>

On Mon, Jul 21, 2008 at 2:23 PM, Bernd Bartmann wrote:
> Hi,
>
> I'm trying to upgrade from moin-1.6.2 to moin-1.7.1. After following
> the HelpOnUpdating guide to the point where I should run  the
> "migation data" command I get an error message when trying to access
> the wiki:
>
>
> ConfigurationError
> ImportError: cannot import name http
> Check that the file is in the same directory as the server script. If
> it is not, you must add the path of the directory where the file is
> located to the python path in the server script. See the comments at
> the top of the server script.
> Check that the configuration file name is either "wikiconfig.py" or
> the module name specified in the wikis list in farmconfig.py. Note
> that the module name does not include the ".py" suffix.
> Error in your configuration file "D:/moin/rd_wiki\wikiconfig.py"
> around line 189.
>
> The problematic line seems to be related to the http auth code that I
> use for AD authentication:
>
> from MoinMoin.auth.http import http
> from MoinMoin.auth import moin_session
> auth = [http, moin_session]
>
> Any ideas what I can do to solve the issue?

Looks like it now has to be:

from MoinMoin.auth.http import HTTPAuth
auth = [HTTPAuth()]

Why does this have to change with every new version?

Best regards,
Bernd.



From bernd.bartmann at gmail.com  Mon Jul 21 09:12:31 2008
From: bernd.bartmann at gmail.com (Bernd Bartmann)
Date: Mon, 21 Jul 2008 15:12:31 +0200
Subject: [Moin-user] Problem with upgrade to moin-1.7.1 and AD auth
In-Reply-To: <6c18a4f0807210553q41ac900fxca0639c29ceb30e6@mail.gmail.com>
References: <6c18a4f0807210523y3f500ae1rbaba19722a609102@mail.gmail.com>
	<6c18a4f0807210553q41ac900fxca0639c29ceb30e6@mail.gmail.com>
Message-ID: <6c18a4f0807210612s6753cfa3qffda349ff9e9d17c@mail.gmail.com>

Hi again,

after digging around some more I found that the HelpOnAuthentication
page has the correct instructions, but
ApacheOnWin32withDomainAuthentication still contains the old and now
broken instructions.

Best regards,
Bernd



From tw-public at gmx.de  Mon Jul 21 10:41:44 2008
From: tw-public at gmx.de (Thomas Waldmann)
Date: Mon, 21 Jul 2008 16:41:44 +0200
Subject: [Moin-user] Problem with upgrade to moin-1.7.1 and AD auth
In-Reply-To: <6c18a4f0807210553q41ac900fxca0639c29ceb30e6@mail.gmail.com>
References: <6c18a4f0807210523y3f500ae1rbaba19722a609102@mail.gmail.com>
	<6c18a4f0807210553q41ac900fxca0639c29ceb30e6@mail.gmail.com>
Message-ID: <4884A028.4060904@gmx.de>

>> The problematic line seems to be related to the http auth code that I
>> use for AD authentication:

Note that you can also access AD directly by using ldap auth plugin 
(gives some more features than just getting the auth user name, like 
updating the email addr from AD, for example).

>> from MoinMoin.auth.http import http
>> from MoinMoin.auth import moin_session
>> auth = [http, moin_session]
>>
>> Any ideas what I can do to solve the issue?
> 
> Looks like it now has to be:
> 
> from MoinMoin.auth.http import HTTPAuth
> auth = [HTTPAuth()]

Exactly.

> Why does this have to change with every new version?

Sorry about that, but sometimes you don't find the right (and most 
flexible) way from the beginning, but you have to gradually improve it.

The latest change from auth functions to auth objects was necessary to 
support stuff like OpenID. And, as a side effect, it made other auth 
methods also much better configurable and much more flexible.

For example, if you have a domain controller and a backup domain 
controller (both AD/LDAP servers), you can now configure them both and 
have failover in case your DC stops working.

I think it can stay now like it is for a while, it is hopefully flexible 
enough now.




From crosseyedpenguin at yahoo.com  Mon Jul 21 11:05:58 2008
From: crosseyedpenguin at yahoo.com (Roger Haase)
Date: Mon, 21 Jul 2008 08:05:58 -0700 (PDT)
Subject: [Moin-user] Missing 1.7.1 Underlay Pages on moinmo.in
Message-ID: <608961.35822.qm@web36208.mail.mud.yahoo.com>

moin.in SystemInfo says release 1.7.1, but some underlay pages that are in the release distribution, such as HelpOnInstalling/ApacheWithModWSGI, are not available.  

Roger Haase


      



From tw-public at gmx.de  Mon Jul 21 11:00:00 2008
From: tw-public at gmx.de (Thomas Waldmann)
Date: Mon, 21 Jul 2008 17:00:00 +0200
Subject: [Moin-user] Missing 1.7.1 Underlay Pages on moinmo.in
In-Reply-To: <608961.35822.qm@web36208.mail.mud.yahoo.com>
References: <608961.35822.qm@web36208.mail.mud.yahoo.com>
Message-ID: <4884A470.5030400@gmx.de>

Roger Haase schrieb:
> moin.in SystemInfo says release 1.7.1, but some underlay pages that are in the release distribution, such as HelpOnInstalling/ApacheWithModWSGI, are not available.  

Ah, thanks for the hint.

But if you just want to read them, it is all on 
http://master17.moinmo.in/ :)



From bernd.bartmann at gmail.com  Mon Jul 21 11:15:20 2008
From: bernd.bartmann at gmail.com (Bernd Bartmann)
Date: Mon, 21 Jul 2008 17:15:20 +0200
Subject: [Moin-user] Problem with upgrade to moin-1.7.1 and AD auth
In-Reply-To: <20080721131553.GF61826@fw.wtp3.org>
References: <6c18a4f0807210523y3f500ae1rbaba19722a609102@mail.gmail.com>
	<20080721131553.GF61826@fw.wtp3.org>
Message-ID: <6c18a4f0807210815y1d8dd95dx970e657ecb9c0d7@mail.gmail.com>

On Mon, Jul 21, 2008 at 3:15 PM, Krzysztof Stryjek wrote:
> If you want to setup AD auth in moinmoin 1.7.x you need another
> configuration. I spent some time to make it work correctly. But
> unfortunatelly Moinmoin group membership with AD is broken, so I
> came back to 1.6.3 version. I posted question to the list about 3 weeks
> ago, but have no answer.

This is interesting as I'm seeing a similar issue now. From your other
email I see that you're using LDAP for AD auth. In my case it's
mod_auth_sspi.
The issue that I'm seeing is that I (admin account) have full access
to the Wiki, but other users cannot access any Wiki page. The weird
thing is that another admin user also cannot access the Wiki. The ACLs
reference a Wiki page listing all the allowed user accounts and are
not build around AD group membership.

Best regards,
Bernd.



From tw-public at gmx.de  Mon Jul 21 12:05:54 2008
From: tw-public at gmx.de (Thomas Waldmann)
Date: Mon, 21 Jul 2008 18:05:54 +0200
Subject: [Moin-user] Moinmoin 1.7.0,
 LDAP (ActiveDirectory) and Group based acls
In-Reply-To: <20080630213517.GE55232@fw.wtp3.org>
References: <20080630213517.GE55232@fw.wtp3.org>
Message-ID: <4884B3E2.20200@gmx.de>

Hi Krzysztof,

> I had some problems, because LDAP authorization has changed. But this is
> not problem: I've read module file and made configuration with failover
> (I've configured two Acitve Directory Controlers). That's working
> correcltly.

Great that someone tested that in real life. :))

> 
> But now members of SomeGroup are not "recognized" correctly. On pages,
> where acl are like:
> 
> #acl SomeGroup: read,write,admin All:

This is not correct (and won't work), there must not be a space after 
SomeGroup:.

> auth_methods_trustes = [ ???, ??? ]

Quoting from MoinMoin/config/multiconfig.py:

     # default to http and xmlrpc_applytoken to get old semantics
     # xmlrpc_applytoken shall be removed once that code is changed
     # to have proper session handling and use request.handle_auth()
     auth_methods_trusted = ['http', 'xmlrpc_applytoken']

This is the correct setting name, you only need that one. :)

> I've tried many methods: http, values set to LDAPAuth.name attribute,
> but still administrative logins are not recognized.

I think this is unrelated. The trusted stuff is only for the builtin 
"Trusted:" group (similar to "Known:" and "All:" groups).

If you use your own group anyway (and you don't use "Trusted:" in ACLs), 
you don't need to care for that.

> # we have ta change name, because name = ldap as default. And with 2
> # instances of LDAPAuth class Moinmoin comlains about duplicate name
> # values.
> # Also LDAPAuth.name is set as auth_method...
> #
> ldap_auth_dc03.name = 'dc03'

Good hint, we try to fix that.

Cheers, Thomas





From tw-public at gmx.de  Mon Jul 21 11:58:27 2008
From: tw-public at gmx.de (Thomas Waldmann)
Date: Mon, 21 Jul 2008 17:58:27 +0200
Subject: [Moin-user] Problem with upgrade to moin-1.7.1 and AD auth
In-Reply-To: <6c18a4f0807210815y1d8dd95dx970e657ecb9c0d7@mail.gmail.com>
References: <6c18a4f0807210523y3f500ae1rbaba19722a609102@mail.gmail.com>	<20080721131553.GF61826@fw.wtp3.org>
	<6c18a4f0807210815y1d8dd95dx970e657ecb9c0d7@mail.gmail.com>
Message-ID: <4884B223.8030300@gmx.de>

Bernd Bartmann schrieb:
> On Mon, Jul 21, 2008 at 3:15 PM, Krzysztof Stryjek wrote:

Interesting. I did not get Krzysztof's post.

Krzysztof: can you post it again?

>> If you want to setup AD auth in moinmoin 1.7.x you need another
>> configuration.

If you mean when comparing it to 1.6, yes - you mostly have to move the 
configuration params to the LDAPAuth() object constructor.

>> I spent some time to make it work correctly. But
>> unfortunatelly Moinmoin group membership with AD is broken,

Moin does not support "LDAP groups" yet, but you can put the group 
member names (exactly like they are shown at the top border of the 
screen) onto a wiki group page.

>> came back to 1.6.3 version. I posted question to the list about 3 weeks
>> ago, but have no answer.

/me looks. OK, I got that one, maybe just missed it...





From wtp at wtp3.org  Mon Jul 21 15:27:58 2008
From: wtp at wtp3.org (Krzysztof Stryjek)
Date: Mon, 21 Jul 2008 21:27:58 +0200
Subject: [Moin-user] Moinmoin 1.7.0,
	LDAP (ActiveDirectory) and Group based acls
In-Reply-To: <4884B3E2.20200@gmx.de>
References: <20080630213517.GE55232@fw.wtp3.org> <4884B3E2.20200@gmx.de>
Message-ID: <20080721192757.GJ61826@fw.wtp3.org>

Hello,

Dnia Mon, Jul 21, 2008 at 06:05:54PM +0200, Thomas Waldmann napisa?(a):
> Hi Krzysztof,
> 
> >where acl are like:
> >#acl SomeGroup: read,write,admin All:
> 
> This is not correct (and won't work), there must not be a space after SomeGroup:.
> 
Yes, it's my mistake on this post. At moin pages there is:
#acl MitGroup:read,write,admin All:

and as I wrote (I think so) I have "MitGroup" page with only accounts:
 * acc1
 * acc2
 * acc3
(non of this accounts are my wiki "superuser")

> >auth_methods_trustes = [ ???, ??? ]
> 
> Quoting from MoinMoin/config/multiconfig.py:
> 
>     # default to http and xmlrpc_applytoken to get old semantics
>     # xmlrpc_applytoken shall be removed once that code is changed
>     # to have proper session handling and use request.handle_auth()
>     auth_methods_trusted = ['http', 'xmlrpc_applytoken']
> 
> This is the correct setting name, you only need that one. :)
> 
Sorry, but I don't understand. So now wich auth_methods I shold have?
As I've written when I logged in as acc1 membership of MitGroup is not
honoured :-(((

> >I've tried many methods: http, values set to LDAPAuth.name attribute,
> >but still administrative logins are not recognized.
> 
> I think this is unrelated. The trusted stuff is only for the builtin "Trusted:" group (similar to "Known:" and "All:" groups).
> 
> If you use your own group anyway (and you don't use "Trusted:" in ACLs), you don't need to care for that.
> 
I think now it is much clear. 'All:' is not only predefined Groups :-)))

I hope that now my problem is explained more clear.

I would like to say, that your work is great and moinmoin wiki is very
useful at my work to build documentation very quicly and easy. I hope
my some posts will be helpful to make moinmoin better.

Greetings
-- 
/~\ The ASCII                  Krzysztof Stryjek
\ / Ribbon Campaign            wtp (at) wtp3.org
 X  Against HTML           http://fw.wtp3.org/~wtp/
/ \ Email!               GG: 3608113 JID:wtp at chrome.pl

War doesn't prove who's right, just who's left.



From bernd.bartmann at gmail.com  Tue Jul 22 03:28:48 2008
From: bernd.bartmann at gmail.com (Bernd Bartmann)
Date: Tue, 22 Jul 2008 09:28:48 +0200
Subject: [Moin-user] Moinmoin 1.7.0,
	LDAP (ActiveDirectory) and Group based acls
In-Reply-To: <20080721192757.GJ61826@fw.wtp3.org>
References: <20080630213517.GE55232@fw.wtp3.org> <4884B3E2.20200@gmx.de>
	<20080721192757.GJ61826@fw.wtp3.org>
Message-ID: <6c18a4f0807220028q53ebdfb7l8aea47f9896abf2@mail.gmail.com>

In my case with moin-1.7.1 the problem looks like this:

superuser = [u"Bba114Sb", u"Bpe135Sb"]
acl_rights_before = (u"Bba114Sb:read,write,delete,revert,admin"
u"Bpe135Sb:read,write,delete,revert,admin")
acl_rights_default = (u"RDUsersGroup:read,write,delete,revert" u"All:")


The page RDUsersGroup contains:

#acl -All:read

 * Bba114Sb
 * Tgo156Sb
 * Bpe135Sb

The User Bba114Sb can access everything, but Bpe135Sb and every other
user gets "Sie d?rfen diese Seite nicht ansehen".
ACLs were working correctly with moin-1.6.2

Best regards,
Bernd.



From GregNoel at tigris.org  Tue Jul 22 04:06:19 2008
From: GregNoel at tigris.org (Greg Noel)
Date: Tue, 22 Jul 2008 01:06:19 -0700
Subject: [Moin-user] Moinmoin 1.7.0,
	LDAP (ActiveDirectory) and Group based acls
In-Reply-To: <6c18a4f0807220028q53ebdfb7l8aea47f9896abf2@mail.gmail.com>
References: <20080630213517.GE55232@fw.wtp3.org> <4884B3E2.20200@gmx.de>
	<20080721192757.GJ61826@fw.wtp3.org>
	<6c18a4f0807220028q53ebdfb7l8aea47f9896abf2@mail.gmail.com>
Message-ID: <D7EE5E49-8969-4B5C-8DBC-3BCDBC0502F3@tigris.org>

On Jul 22, 2008, at 12:28 AM, Bernd Bartmann wrote:

> acl_rights_default = (u"RDUsersGroup:read,write,delete,revert"  
> u"All:")

These two strings are run together with no intervening space.

Hope this helps,
-- Greg Noel, retired UNIX guru




From tw-public at gmx.de  Tue Jul 22 04:40:27 2008
From: tw-public at gmx.de (Thomas Waldmann)
Date: Tue, 22 Jul 2008 10:40:27 +0200
Subject: [Moin-user] Moinmoin 1.7.0,
	LDAP (ActiveDirectory) and Group	based acls
In-Reply-To: <20080721192757.GJ61826@fw.wtp3.org>
References: <20080630213517.GE55232@fw.wtp3.org> <4884B3E2.20200@gmx.de>
	<20080721192757.GJ61826@fw.wtp3.org>
Message-ID: <1216716027.18083.10.camel@black.firma.waldmann-edv.de>


> Yes, it's my mistake on this post. At moin pages there is:
> #acl MitGroup:read,write,admin All:

OK, that looks better.

So the problem is likely at not up-to-date definition of
page_group_regex in your wiki config (this had to be changed in 1.7.0,
see docs/CHANGES).

> >     auth_methods_trusted = ['http', 'xmlrpc_applytoken'] 
> > This is the correct setting name, you only need that one. :)
> > 
> Sorry, but I don't understand. So now wich auth_methods I shold have?

I wasn't referring to your auth methods. I just pointed out that the
correct NAME for this setting is auth_methods_trusted (as you have used
2 different names).






From tw-public at gmx.de  Tue Jul 22 04:49:58 2008
From: tw-public at gmx.de (Thomas Waldmann)
Date: Tue, 22 Jul 2008 10:49:58 +0200
Subject: [Moin-user] Moinmoin 1.7.0,
	LDAP (ActiveDirectory) and Group	based acls
In-Reply-To: <6c18a4f0807220028q53ebdfb7l8aea47f9896abf2@mail.gmail.com>
References: <20080630213517.GE55232@fw.wtp3.org> <4884B3E2.20200@gmx.de>
	<20080721192757.GJ61826@fw.wtp3.org>
	<6c18a4f0807220028q53ebdfb7l8aea47f9896abf2@mail.gmail.com>
Message-ID: <1216716598.18083.19.camel@black.firma.waldmann-edv.de>


> superuser = [u"Bba114Sb", u"Bpe135Sb"]

OK.

> acl_rights_before = (u"Bba114Sb:read,write,delete,revert,admin"
> u"Bpe135Sb:read,write,delete,revert,admin")
> acl_rights_default = (u"RDUsersGroup:read,write,delete,revert"
> u"All:")

That's incorrect. You can concatenate strings in Python that way, but
you forgot the blank between the 2 ACL entries. Look:

>>> (u"foo" u"bar")
u'foobar'

So if you want to keep that way of doing things (and not just use 1 long
unicode string), you have to add one blank at the end of each string
(except the last string, but it does no harm if you do it there also).

But maybe using just this is easier:

acl_rights_default = u"RDUsersGroup:read,write,delete,revert All:"





From bernd.bartmann at gmail.com  Tue Jul 22 06:30:13 2008
From: bernd.bartmann at gmail.com (Bernd Bartmann)
Date: Tue, 22 Jul 2008 12:30:13 +0200
Subject: [Moin-user] Moinmoin 1.7.0,
	LDAP (ActiveDirectory) and Group based acls
In-Reply-To: <1216716598.18083.19.camel@black.firma.waldmann-edv.de>
References: <20080630213517.GE55232@fw.wtp3.org> <4884B3E2.20200@gmx.de>
	<20080721192757.GJ61826@fw.wtp3.org>
	<6c18a4f0807220028q53ebdfb7l8aea47f9896abf2@mail.gmail.com>
	<1216716598.18083.19.camel@black.firma.waldmann-edv.de>
Message-ID: <6c18a4f0807220330u1f062cc2vd665156176bc460b@mail.gmail.com>

On Tue, Jul 22, 2008 at 10:49 AM, Thomas Waldmann wrote:
>
>> superuser = [u"Bba114Sb", u"Bpe135Sb"]
>
> OK.
>
>> acl_rights_before = (u"Bba114Sb:read,write,delete,revert,admin"
>> u"Bpe135Sb:read,write,delete,revert,admin")
>> acl_rights_default = (u"RDUsersGroup:read,write,delete,revert"
>> u"All:")
>
> That's incorrect. You can concatenate strings in Python that way, but
> you forgot the blank between the 2 ACL entries. Look:
>
>>>> (u"foo" u"bar")
> u'foobar'
>
> So if you want to keep that way of doing things (and not just use 1 long
> unicode string), you have to add one blank at the end of each string
> (except the last string, but it does no harm if you do it there also).
>
> But maybe using just this is easier:
>
> acl_rights_default = u"RDUsersGroup:read,write,delete,revert All:"

Ok. So now I've changed the acl_rights to:

acl_rights_before = (u"Bba114Sb:read,write,delete,revert,admin
Bpe135Sb:read,write,delete,revert,admin")
acl_rights_default = (u"RDUsersGroup:read,write,delete,revert All:")

Access for the admin user Bpe135Sb now works fine, but still no go for
users in the RDUsersGroup.

Best regards,
Bernd.



From tw-public at gmx.de  Tue Jul 22 07:28:21 2008
From: tw-public at gmx.de (Thomas Waldmann)
Date: Tue, 22 Jul 2008 13:28:21 +0200
Subject: [Moin-user] Moinmoin 1.7.0,
	LDAP (ActiveDirectory) and Group	based acls
In-Reply-To: <6c18a4f0807220330u1f062cc2vd665156176bc460b@mail.gmail.com>
References: <20080630213517.GE55232@fw.wtp3.org> <4884B3E2.20200@gmx.de>
	<20080721192757.GJ61826@fw.wtp3.org>
	<6c18a4f0807220028q53ebdfb7l8aea47f9896abf2@mail.gmail.com>
	<1216716598.18083.19.camel@black.firma.waldmann-edv.de>
	<6c18a4f0807220330u1f062cc2vd665156176bc460b@mail.gmail.com>
Message-ID: <1216726101.6010.1.camel@black.firma.waldmann-edv.de>

> Ok. So now I've changed the acl_rights to:
> 
> acl_rights_before = (u"Bba114Sb:read,write,delete,revert,admin
> Bpe135Sb:read,write,delete,revert,admin")
> acl_rights_default = (u"RDUsersGroup:read,write,delete,revert All:")
> 
> Access for the admin user Bpe135Sb now works fine, but still no go for
> users in the RDUsersGroup.

Just read my other posts of today, you'll likely find the solution
there.




From bernd.bartmann at gmail.com  Tue Jul 22 07:53:27 2008
From: bernd.bartmann at gmail.com (Bernd Bartmann)
Date: Tue, 22 Jul 2008 13:53:27 +0200
Subject: [Moin-user] Moinmoin 1.7.0,
	LDAP (ActiveDirectory) and Group based acls
In-Reply-To: <1216726101.6010.1.camel@black.firma.waldmann-edv.de>
References: <20080630213517.GE55232@fw.wtp3.org> <4884B3E2.20200@gmx.de>
	<20080721192757.GJ61826@fw.wtp3.org>
	<6c18a4f0807220028q53ebdfb7l8aea47f9896abf2@mail.gmail.com>
	<1216716598.18083.19.camel@black.firma.waldmann-edv.de>
	<6c18a4f0807220330u1f062cc2vd665156176bc460b@mail.gmail.com>
	<1216726101.6010.1.camel@black.firma.waldmann-edv.de>
Message-ID: <6c18a4f0807220453o1ade496jc4c2388aca27b2d2@mail.gmail.com>

On Tue, Jul 22, 2008 at 1:28 PM, Thomas Waldmann wrote:
> Just read my other posts of today, you'll likely find the solution
> there.

In case you refer to the page_group_regex change: I already copied the
new defaults from the wikiconfig.py delivered with moin-1.7.1 to my
own wikiconfig.py:

    page_category_regex = ur'(?P<all>Category(?P<key>\S+))'
    page_dict_regex = ur'(?P<all>(?P<key>\S+)Dict)'
    page_group_regex = ur'(?P<all>(?P<key>\S+)Group)'
    page_template_regex = ur'(?P<all>(?P<key>\S+)Template)'

Or do you mean something else?

Best regards,
Bernd.



From tw-public at gmx.de  Tue Jul 22 10:03:50 2008
From: tw-public at gmx.de (Thomas Waldmann)
Date: Tue, 22 Jul 2008 16:03:50 +0200
Subject: [Moin-user] Moinmoin 1.7.0,
 LDAP (ActiveDirectory) and Group based acls
In-Reply-To: <6c18a4f0807220453o1ade496jc4c2388aca27b2d2@mail.gmail.com>
References: <20080630213517.GE55232@fw.wtp3.org>
	<4884B3E2.20200@gmx.de>	<20080721192757.GJ61826@fw.wtp3.org>	<6c18a4f0807220028q53ebdfb7l8aea47f9896abf2@mail.gmail.com>	<1216716598.18083.19.camel@black.firma.waldmann-edv.de>	<6c18a4f0807220330u1f062cc2vd665156176bc460b@mail.gmail.com>	<1216726101.6010.1.camel@black.firma.waldmann-edv.de>
	<6c18a4f0807220453o1ade496jc4c2388aca27b2d2@mail.gmail.com>
Message-ID: <4885E8C6.7020100@gmx.de>

> In case you refer to the page_group_regex change:

Yes, I meant that.

> I already copied the
> new defaults from the wikiconfig.py delivered with moin-1.7.1 to my
> own wikiconfig.py:
> 
>     page_category_regex = ur'(?P<all>Category(?P<key>\S+))'
>     page_dict_regex = ur'(?P<all>(?P<key>\S+)Dict)'
>     page_group_regex = ur'(?P<all>(?P<key>\S+)Group)'
>     page_template_regex = ur'(?P<all>(?P<key>\S+)Template)'

If you just want the default, you don't need to put them into wiki 
config, btw.

If you still have problems with groups, please file a bug report on the 
wiki, giving all details.




From bernd.bartmann at gmail.com  Wed Jul 23 02:17:12 2008
From: bernd.bartmann at gmail.com (Bernd Bartmann)
Date: Wed, 23 Jul 2008 08:17:12 +0200
Subject: [Moin-user] Moinmoin 1.7.0,
	LDAP (ActiveDirectory) and Group based acls
In-Reply-To: <4885E8C6.7020100@gmx.de>
References: <20080630213517.GE55232@fw.wtp3.org> <4884B3E2.20200@gmx.de>
	<20080721192757.GJ61826@fw.wtp3.org>
	<6c18a4f0807220028q53ebdfb7l8aea47f9896abf2@mail.gmail.com>
	<1216716598.18083.19.camel@black.firma.waldmann-edv.de>
	<6c18a4f0807220330u1f062cc2vd665156176bc460b@mail.gmail.com>
	<1216726101.6010.1.camel@black.firma.waldmann-edv.de>
	<6c18a4f0807220453o1ade496jc4c2388aca27b2d2@mail.gmail.com>
	<4885E8C6.7020100@gmx.de>
Message-ID: <6c18a4f0807222317v39754f79k62648869eda75bb0@mail.gmail.com>

On Tue, Jul 22, 2008 at 4:03 PM, Thomas Waldmann wrote:
>> In case you refer to the page_group_regex change:
>
> Yes, I meant that.
>
>> I already copied the
>> new defaults from the wikiconfig.py delivered with moin-1.7.1 to my
>> own wikiconfig.py:
>>
>>     page_category_regex = ur'(?P<all>Category(?P<key>\S+))'
>>     page_dict_regex = ur'(?P<all>(?P<key>\S+)Dict)'
>>     page_group_regex = ur'(?P<all>(?P<key>\S+)Group)'
>>     page_template_regex = ur'(?P<all>(?P<key>\S+)Template)'
>
> If you just want the default, you don't need to put them into wiki
> config, btw.
>
> If you still have problems with groups, please file a bug report on the
> wiki, giving all details.

Ok, problem solved. I forgot to "moin ... maint cleancache". Now
everything is working again.
Thanks for your help and all the efforts that you've put into moin!

Best regards,
Bernd.



From rick.vanderveer at gmail.com  Wed Jul 23 10:10:45 2008
From: rick.vanderveer at gmail.com (Rick Vanderveer)
Date: Wed, 23 Jul 2008 09:10:45 -0500
Subject: [Moin-user] maint cleanpage
Message-ID: <5c39e1ca0807230710v69008c92ld7a4a66f23d0bf4e@mail.gmail.com>

I'm trying to use the "moin --config-dir=... --wiki-url=... maint
cleanpage" command (on Windows), as I have several pages effected by
the bug mentioned in the 1.7.1 release notes. Is this supposed to do
perform the action, or simply output the pages question?  I can take
the output and modify it so that it runs on Windows, but just want to
make sure I'm not missing anything.

-Rick



From tw-public at gmx.de  Wed Jul 23 10:53:50 2008
From: tw-public at gmx.de (Thomas Waldmann)
Date: Wed, 23 Jul 2008 16:53:50 +0200
Subject: [Moin-user] maint cleanpage
In-Reply-To: <5c39e1ca0807230710v69008c92ld7a4a66f23d0bf4e@mail.gmail.com>
References: <5c39e1ca0807230710v69008c92ld7a4a66f23d0bf4e@mail.gmail.com>
Message-ID: <488745FE.9030002@gmx.de>

> I'm trying to use the "moin --config-dir=... --wiki-url=... maint
> cleanpage" command (on Windows), as I have several pages effected by
> the bug mentioned in the 1.7.1 release notes. Is this supposed to do
> perform the action, or simply output the pages question?

It currently outputs a shell script that does the action when you 
execute the script.

For Windows you likely need to replace mv by rename, mkdir by md, ...

> I can take
> the output and modify it so that it runs on Windows, but just want to
> make sure I'm not missing anything.

Maybe modifying the cleanpage script to have some --dryrun --deleted 
--trash params would be good for platform independence.





From cpghost at cordula.ws  Wed Jul 23 15:52:10 2008
From: cpghost at cordula.ws (cpghost)
Date: Wed, 23 Jul 2008 21:52:10 +0200
Subject: [Moin-user] Displaying HTML source code as attachment
Message-ID: <20080723215210.33114a84@epia-2.farid-hajji.net>

Hello,

after upgrading from MoinMoin 1.5.8 to 1.7.0, files ending
with .html (and .xml), attached with {{attachment:file.html}}
don't appear as source code anymore, but are rendered by the
browser.

Is there a way to turn off this behavior and revert back to
the old way of displaying the content of those files as
plain boring text/plain inline attachment (short of renaming
them by appending a .txt as in file.html.txt which I'd prefer
to avoid)?

I've tried to add the following to the file
$PYTHON_PATH/site-packages/MoinMoin/wikiutil.py
but while it works for the additional extensions
like '.dat', '.cxx' etc. it doesn't work for '.html'
nor '.xml':

MIMETYPES_MORE = {
  # ...
  '.dat': 'text/plain',
  '.cxx': 'text/x-c++src',
  '.i': 'text/x-c++src',
  '.schema': 'text/plain',
  '.rpy': 'text/x-python',
  '.wsgi': 'text/x-python',
  '.php': 'text/plain',   # see MIMETYPES_sanitize_mapping
  '.html': 'text/plain',  # don't render as HTML! (won't work)
  '.xml': 'text/plain',   # don't render as HTML! (won't work)
}

MIMETYPES_sanitize_mapping = {
  # ...
  ('application', 'x-httpd-php'): ('text', 'plain'),
  ('text', 'html'): ('text', 'plain'), # (won't work)
}

Thanks,
-cpghost.

-- 
Cordula's Web. http://www.cordula.ws/



From tw-public at gmx.de  Thu Jul 24 09:33:39 2008
From: tw-public at gmx.de (Thomas Waldmann)
Date: Thu, 24 Jul 2008 15:33:39 +0200
Subject: [Moin-user] Displaying HTML source code as attachment
In-Reply-To: <20080723215210.33114a84@epia-2.farid-hajji.net>
References: <20080723215210.33114a84@epia-2.farid-hajji.net>
Message-ID: <488884B3.30308@gmx.de>

> after upgrading from MoinMoin 1.5.8 to 1.7.0,

BTW (unrelated to your post) 1.7.1 is out.

> files ending
> with .html (and .xml), attached with {{attachment:file.html}}
> don't appear as source code anymore, but are rendered by the
> browser.

Moin 1.5 didn't have that syntax.

{{...}} means transclude ("render here")
[[...]] means link.

Maybe you are confusing it with {{{ ... }}} (3 vs. 2 curly braces)?

> Is there a way to turn off this behavior and revert back to
> the old way of displaying the content of those files as
> plain boring text/plain inline attachment (short of renaming
> them by appending a .txt as in file.html.txt which I'd prefer
> to avoid)?

Well, appending .txt might be the only way currently, because it detects 
the mimetype from the suffix (and then, also sends that mimetype to the 
browser).

Of course you can use a parser section / pre section and paste the stuff 
there, but maybe you don't want to handle it that way.

> I've tried to add the following to the file
> $PYTHON_PATH/site-packages/MoinMoin/wikiutil.py
> but while it works for the additional extensions
> like '.dat', '.cxx' etc. it doesn't work for '.html'
> nor '.xml':
> 
> MIMETYPES_MORE = {
>   # ...
>   '.dat': 'text/plain',
>   '.cxx': 'text/x-c++src',
>   '.i': 'text/x-c++src',
>   '.schema': 'text/plain',
>   '.rpy': 'text/x-python',
>   '.wsgi': 'text/x-python',
>   '.php': 'text/plain',   # see MIMETYPES_sanitize_mapping
>   '.html': 'text/plain',  # don't render as HTML! (won't work)
>   '.xml': 'text/plain',   # don't render as HTML! (won't work)
> }
> 
> MIMETYPES_sanitize_mapping = {
>   # ...
>   ('application', 'x-httpd-php'): ('text', 'plain'),
>   ('text', 'html'): ('text', 'plain'), # (won't work)
> }

That SHOULD work. Did you restart moin after changing configuration?

Your python version?





From rick.vanderveer at gmail.com  Thu Jul 24 17:18:57 2008
From: rick.vanderveer at gmail.com (Rick Vanderveer)
Date: Thu, 24 Jul 2008 16:18:57 -0500
Subject: [Moin-user] maint cleanpage
In-Reply-To: <488745FE.9030002@gmx.de>
References: <5c39e1ca0807230710v69008c92ld7a4a66f23d0bf4e@mail.gmail.com>
	<488745FE.9030002@gmx.de>
Message-ID: <5c39e1ca0807241418u370cae7bs1b7c6d2a8594e633@mail.gmail.com>

Thanks Thomas, that's what I suspected, and just wanted to confirm.

For those Windows users out there, here's what I did to get it to work:

http://moinmo.in/RickVanderveer/MigratingFromMoin16ToMoin17#cleanpage

-Rick


On Wed, Jul 23, 2008 at 9:53 AM, Thomas Waldmann <tw-public at gmx.de> wrote:
>> I'm trying to use the "moin --config-dir=... --wiki-url=... maint
>> cleanpage" command (on Windows), as I have several pages effected by
>> the bug mentioned in the 1.7.1 release notes. Is this supposed to do
>> perform the action, or simply output the pages question?
>
> It currently outputs a shell script that does the action when you
> execute the script.
>
> For Windows you likely need to replace mv by rename, mkdir by md, ...
>
>> I can take
>> the output and modify it so that it runs on Windows, but just want to
>> make sure I'm not missing anything.
>
> Maybe modifying the cleanpage script to have some --dryrun --deleted
> --trash params would be good for platform independence.
>
>
>
> -------------------------------------------------------------------------
> This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
> Build the coolest Linux based applications with Moblin SDK & win great prizes
> Grand prize is a trip for two to an Open Source event anywhere in the world
> http://moblin-contest.org/redirect.php?banner_id=100&url=/
> _______________________________________________
> Moin-user mailing list
> Moin-user at lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/moin-user
>



From wtp at wtp3.org  Sun Jul 27 09:56:37 2008
From: wtp at wtp3.org (Krzysztof Stryjek)
Date: Sun, 27 Jul 2008 15:56:37 +0200
Subject: [Moin-user] Moinmoin 1.7.0,
	LDAP (ActiveDirectory) and Group based acls
In-Reply-To: <6c18a4f0807222317v39754f79k62648869eda75bb0@mail.gmail.com>
References: <20080630213517.GE55232@fw.wtp3.org> <4884B3E2.20200@gmx.de>
	<20080721192757.GJ61826@fw.wtp3.org>
	<6c18a4f0807220028q53ebdfb7l8aea47f9896abf2@mail.gmail.com>
	<1216716598.18083.19.camel@black.firma.waldmann-edv.de>
	<6c18a4f0807220330u1f062cc2vd665156176bc460b@mail.gmail.com>
	<1216726101.6010.1.camel@black.firma.waldmann-edv.de>
	<6c18a4f0807220453o1ade496jc4c2388aca27b2d2@mail.gmail.com>
	<4885E8C6.7020100@gmx.de>
	<6c18a4f0807222317v39754f79k62648869eda75bb0@mail.gmail.com>
Message-ID: <20080727135637.GC67421@fw.wtp3.org>

Hello,

Dnia Wed, Jul 23, 2008 at 08:17:12AM +0200, Bernd Bartmann napisa?(a):
> On Tue, Jul 22, 2008 at 4:03 PM, Thomas Waldmann wrote:
> >> In case you refer to the page_group_regex change:
> >
> > Yes, I meant that.
> >
> >> I already copied the
> >> new defaults from the wikiconfig.py delivered with moin-1.7.1 to my
> >> own wikiconfig.py:
> >>
> >>     page_category_regex = ur'(?P<all>Category(?P<key>\S+))'
> >>     page_dict_regex = ur'(?P<all>(?P<key>\S+)Dict)'
> >>     page_group_regex = ur'(?P<all>(?P<key>\S+)Group)'
> >>     page_template_regex = ur'(?P<all>(?P<key>\S+)Template)'
> >
> > If you just want the default, you don't need to put them into wiki
> > config, btw.
>
Well, it was a problem in my case. But this regex were copied during
making a wiki instance (FreeBSD ports Makefile), so this "old"
definitions were in wikiconfig.py. When I've made upgrade to 1.7.0
version and modified wikiconfig.py everything works correctly.

So once again thanks for your help to me. And your efforts to moin.

Greetings from Poland
-- 
/~\ The ASCII                  Krzysztof Stryjek
\ / Ribbon Campaign            wtp (at) wtp3.org
 X  Against HTML           http://fw.wtp3.org/~wtp/
/ \ Email!               GG: 3608113 JID:wtp at chrome.pl

Labor, n.:
	One of the processes by which A acquires property for B.
		-- Ambrose Bierce, "The Devil's Dictionary"
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 195 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/moin-user/attachments/20080727/f7ef7791/attachment.sig>

From clement at hudson-trading.com  Mon Jul 28 18:17:03 2008
From: clement at hudson-trading.com (clement at hudson-trading.com)
Date: Mon, 28 Jul 2008 18:17:03 -0400
Subject: [Moin-user] free linking in moinmoin 1.7.0_1 doesnt work
Message-ID: <b6173c130807281517q65ac9ccch4d9447cd6db93e1c@mail.gmail.com>

Hi,

Im currently using moinmoin 1.5.8 and when doing free linking on the front
page it works great, as expected.
But Ive upgraded to moinmoin 1.7.0_1 and even after referencing to the
following link:

http://moinmoin.wikiwikiweb.de/HelpOnLinking

the free link, doesnt work with a single word or multiple words example
pattern - ["String"].
Is there any/some way of getting this fixed. Please explain the solution.

Thanks Clement
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/moin-user/attachments/20080728/b4665708/attachment.html>

From roach.list at gmail.com  Mon Jul 28 18:32:01 2008
From: roach.list at gmail.com (Robert Spencer)
Date: Tue, 29 Jul 2008 00:32:01 +0200
Subject: [Moin-user] free linking in moinmoin 1.7.0_1 doesnt work
In-Reply-To: <b6173c130807281517q65ac9ccch4d9447cd6db93e1c@mail.gmail.com>
References: <b6173c130807281517q65ac9ccch4d9447cd6db93e1c@mail.gmail.com>
Message-ID: <8b8c4c740807281532w6b2887fp4aea5343776d0978@mail.gmail.com>

On 7/29/08, clement at hudson-trading.com <clement at hudson-trading.com> wrote:
> Hi,
>
> Im currently using moinmoin 1.5.8 and when doing free linking on the front
> page it works great, as expected.
> But Ive upgraded to moinmoin 1.7.0_1 and even after referencing to the
> following link:
>
> http://moinmoin.wikiwikiweb.de/HelpOnLinking

That's old, go to http://moinmo.in/HelpOnLinking or reference your new
moinmoin help pages.

> the free link, doesnt work with a single word or multiple words example
> pattern - ["String"].
> Is there any/some way of getting this fixed. Please explain the solution.

The syntax changed quite some time back. It's now:

[[free link]]

Please read the page I've linked to, as there has been a lot of
changes including changes to the syntax for macros.

-- 
Robert Spencer



From tw-public at gmx.de  Tue Jul 29 05:10:28 2008
From: tw-public at gmx.de (Thomas Waldmann)
Date: Tue, 29 Jul 2008 11:10:28 +0200
Subject: [Moin-user] free linking in moinmoin 1.7.0_1 doesnt work
In-Reply-To: <b6173c130807281517q65ac9ccch4d9447cd6db93e1c@mail.gmail.com>
References: <b6173c130807281517q65ac9ccch4d9447cd6db93e1c@mail.gmail.com>
Message-ID: <488EDE84.7040401@gmx.de>

> Im currently using moinmoin 1.5.8 and when doing free linking on the 
> front page it works great, as expected.
> But Ive upgraded to moinmoin 1.7.0_1 and even after referencing to the 
> following link:

Additionally to what Robert already said, please note that there are 
migration scripts converting all your data - you need to run them.

See docs/CHANGES for more details, please read it completely until you 
hit 1.5.8 - there are also some pointers to more documentation at the 
top of CHANGES file.