Webware for Python

I would recommend that you first test with Mozilla or some browser that 
lets you monitor cookies.  (From Mz click Tools -> cookie manager -> 
manage cookies).
First of all make sure that you indeed blocking the cookies. 

Next dump the session contents for each page and try to find where the 
session is being lost.

-Aaron


Jeffrey Clement wrote:

>Good Morning,
>
>I've been working with webware 0.8 and have an app that is using a
>variant on   the Securepage used in the Examples.  On one of these
>secure pages I have a button which through some horrible javascript,
>pops up another Window for some dataentry.  Both of these forms are
>subclassed from my SecurePage.  So as a user I can login to the mainpage
>and it all works fine.  Then I click the button, enter stuff and it all
>works fine.  But then it seems to be clearing my cookies (including
>session ID) which is logging me out.
>
>It's to the point that it very consistent.  I have page 1:
>http://.../UserMonitor
>click the button which just opens next page in a window
>http://.../UserSetProxy
>enter some stuff in the setProxy window.  Hit submit which sends to
>UserSetProxy for storage and then has a JS window.close().
>Now if I refresh UserMonitor it nolonger has any cookies and so it
>starts a new session.
>
>I am certain I have no code that should be invalidating the session and
>the timeout is defaults (15min).
>
>Does anyone have any idea what could cause this.  The hostname in the
>URL is staying the same,both servlets are in the same path.  I can't see
>why the browser is forgetting these.  It works fine in Mozilla and IE<6
>but IE6 and Konqi both have this problem.
>
>Thank you,
>Jeff
>
>  
>

I am out of the office from the 21rst of March through 
the 26th of March.

If you need to reach me emergently please call our corporate 
cell  phone: 228-342-8395.

I will be checking my email periodically while on travel.

Thank You.

I am out of the office from the 21rst of March through 
the 26th of March.

If you need to reach me emergently please call our corporate 
cell  phone: 228-342-8395.

I will be checking my email periodically while on travel.

Thank You.

Am I the only one looking forward to the 27th? :)

On Sun, 2003-03-23 at 15:09, jlawhead@... wrote:
> I am out of the office from the 21rst of March through 
> the 26th of March.
> 
> If you need to reach me emergently please call our corporate 
> cell  phone: 228-342-8395.
> 
> I will be checking my email periodically while on travel.
> 
> Thank You.

Scott Russell wrote:
> Am I the only one looking forward to the 27th? :)

Yeah. Please unsubscribe this guy. Or temporarily disable mail delivery 
in the Mailman admin dialog. But the first option will have a better 
educational effect :->

-- Gerhard

Offending content follows:

> On Sun, 2003-03-23 at 15:09, jlawhead@... wrote:
> 
>>I am out of the office from the 21rst of March through 
>>the 26th of March.
>>
>>If you need to reach me emergently please call our corporate 
>>cell  phone: 228-342-8395.
>>
>>I will be checking my email periodically while on travel.
>>
>>Thank You.
> 
> 
> 
> 
> 
> -------------------------------------------------------
> This sf.net email is sponsored by:ThinkGeek
> Welcome to geek heaven.
> http://thinkgeek.com/sf
> _______________________________________________
> Webware-discuss mailing list
> Webware-discuss@...
> https://lists.sourceforge.net/lists/listinfo/webware-discuss

I am out of the office from the 21rst of March through 
the 26th of March.

If you need to reach me emergently please call our corporate 
cell  phone: 228-342-8395.

I will be checking my email periodically while on travel.

Thank You.

On Saturday 22 March 2003 17:06, Tripp Lilley wrote:
> On Wed, 19 Mar 2003, Tavis Rudd wrote:
> > Why don't you create a temporary development branch of the cvs?
>
> I gotta throw a +1 onto this fire :) I synced the other night to the CVS
> head having not seen this message and was bummed when it barfed on me.
> Since I was actively working on something else, I didn't bother to chase
> down the barf, but reverted to 0.8. But I agree with Tavis... radical work
> should (IMHO) go on in a dev branch until it's "close" to stable. Some
> oddball corner cases are okay, but general stability is desirable in the
> mainline.

Considering the history of several refactorings of WebKit's core over the last 
two years - the developers eventually lost interest in them or became too 
busy to finish them off and push for a release - I agree with Ian.  If his 
refactoring of Application.py is going to be in the next release, it should 
be in the main branch. Otherwise it'll never get the exposure and serious 
testing required.

> > On Wednesday 19 March 2003 04:04, Ian Bicking wrote:
> > > CVS has been very stable previously.  You should not expect that, at
> > > least for the next couple weeks.  0.8 is has all the features that CVS
> > > does (prior to these commits), so use that if you want stability.  But
> > > if you don't mind bugs, please help me test!

I'm new to this group, but not to distributed development - in my
experience, normally the cvs HEAD is unstable, and stable branches get
formed off of it, providing release platforms.  Good features and
bugfixes get backported to stable branches if someone cares enough to do
so, ususally by a release maintenance team, not the dev team.

Of course, every project forms it's own culture, and it looks like
running critical webware servers directly out of CVS is the norm for
some people.  I'm forced to wonder how you maintain release control,
perform defect tracking, prove code coverage and regression testing and
ensure code stability with this policy, though.  Aren't these issues
what releases and release branches are for?

On Sat, 2003-03-22 at 20:16, Tavis Rudd wrote:
> On Saturday 22 March 2003 17:06, Tripp Lilley wrote:
> > On Wed, 19 Mar 2003, Tavis Rudd wrote:
> > > Why don't you create a temporary development branch of the cvs?
> >
> > I gotta throw a +1 onto this fire :) I synced the other night to the CVS
> > head having not seen this message and was bummed when it barfed on me.
> > Since I was actively working on something else, I didn't bother to chase
> > down the barf, but reverted to 0.8. But I agree with Tavis... radical work
> > should (IMHO) go on in a dev branch until it's "close" to stable. Some
> > oddball corner cases are okay, but general stability is desirable in the
> > mainline.
> 
> Considering the history of several refactorings of WebKit's core over the last 
> two years - the developers eventually lost interest in them or became too 
> busy to finish them off and push for a release - I agree with Ian.  If his 
> refactoring of Application.py is going to be in the next release, it should 
> be in the main branch. Otherwise it'll never get the exposure and serious 
> testing required.
> 
> > > On Wednesday 19 March 2003 04:04, Ian Bicking wrote:
> > > > CVS has been very stable previously.  You should not expect that, at
> > > > least for the next couple weeks.  0.8 is has all the features that CVS
> > > > does (prior to these commits), so use that if you want stability.  But
> > > > if you don't mind bugs, please help me test!
> 
> 
> 
> -------------------------------------------------------
> This SF.net email is sponsored by:Crypto Challenge is now open! 
> Get cracking and register here for some mind boggling fun and 
> the chance of winning an Apple iPod:
> http://ads.sourceforge.net/cgi-bin/redirect.pl?thaw0031en
> _______________________________________________
> Webware-discuss mailing list
> Webware-discuss@...
> https://lists.sourceforge.net/lists/listinfo/webware-discuss

I don't think this made it to the list. Since it's support of an
argument I made, I feel the need to ensure rdg gets proper
representation. :)

On Sun, 2003-03-23 at 21:00, rdg wrote:
> >From my open source experience, I must aggree w/ Scott Russell
> on the CVS HEAD being unstable.
> 
> -rdg
> 
> On Sun, 2003-03-23 at 03:23, Scott Russell wrote:
> > I'm new to this group, but not to distributed development - in my
> > experience, normally the cvs HEAD is unstable, and stable branches get
> > formed off of it, providing release platforms.  Good features and
> > bugfixes get backported to stable branches if someone cares enough to do
> > so, ususally by a release maintenance team, not the dev team.
> > 
> > Of course, every project forms it's own culture, and it looks like
> > running critical webware servers directly out of CVS is the norm for
> > some people.  I'm forced to wonder how you maintain release control,
> > perform defect tracking, prove code coverage and regression testing and
> > ensure code stability with this policy, though.  Aren't these issues
> > what releases and release branches are for?
> > 
> > On Sat, 2003-03-22 at 20:16, Tavis Rudd wrote:
> > > On Saturday 22 March 2003 17:06, Tripp Lilley wrote:
> > > > On Wed, 19 Mar 2003, Tavis Rudd wrote:
> > > > > Why don't you create a temporary development branch of the cvs?
> > > >
> > > > I gotta throw a +1 onto this fire :) I synced the other night to the CVS
> > > > head having not seen this message and was bummed when it barfed on me.
> > > > Since I was actively working on something else, I didn't bother to chase
> > > > down the barf, but reverted to 0.8. But I agree with Tavis... radical work
> > > > should (IMHO) go on in a dev branch until it's "close" to stable. Some
> > > > oddball corner cases are okay, but general stability is desirable in the
> > > > mainline.
> > > 
> > > Considering the history of several refactorings of WebKit's core over the last 
> > > two years - the developers eventually lost interest in them or became too 
> > > busy to finish them off and push for a release - I agree with Ian.  If his 
> > > refactoring of Application.py is going to be in the next release, it should 
> > > be in the main branch. Otherwise it'll never get the exposure and serious 
> > > testing required.
> > > 
> > > > > On Wednesday 19 March 2003 04:04, Ian Bicking wrote:
> > > > > > CVS has been very stable previously.  You should not expect that, at
> > > > > > least for the next couple weeks.  0.8 is has all the features that CVS
> > > > > > does (prior to these commits), so use that if you want stability.  But
> > > > > > if you don't mind bugs, please help me test!

>   Myself (tvon) and Brian Skahan (polin8) are always in there, and while
> neither of us is an expert on Webware (relative newbies really), we are
> always happy to try to help out or to discuss related topics.

Screw qualifications (or not, ask if you care :), but I'm always in #python on
freenode so I'll join in on the possible fun as (LoppEar).

- Luke

On Friday 21 March 2003 10:15, Aaron Held wrote:

> To render a form I use something like:
>
> <tr>
> #for $field in $fields::
>     <td>$field.label</td>
> #end
> #for $row in $ResultSet:
>     <tr>
>           #for $field in $row:
>              <td>$field.value</td>
>           #end
>     </tr>
> #end

though, Cheetah doesn't need the colons.

Now you made me look at old code.

fieldList = [0,1,16,2,21,3,4,5,18,6,12,13,14,15,17]

template->

    #for $record in $rs
    #set $rowNum += 1
    <tr 
onmouseover="this.style.background='#ffffff';this.style.cursor='hand'"
        onmouseout="this.style.background='#D9DCC2'">
        <td>$rowNum</td>
        #for $i in $fieldList
        <td class="data" align="center">$record[$i]</td>
        #end for
    #end for


Tavis Rudd wrote:

>On Friday 21 March 2003 10:15, Aaron Held wrote:
>
>  
>
>>To render a form I use something like:
>>
>><tr>
>>#for $field in $fields::
>>    <td>$field.label</td>
>>#end
>>#for $row in $ResultSet:
>>    <tr>
>>          #for $field in $row:
>>             <td>$field.value</td>
>>          #end
>>    </tr>
>>#end
>>    
>>
>
>though, Cheetah doesn't need the colons.
>
>
>
>-------------------------------------------------------
>This SF.net email is sponsored by:Crypto Challenge is now open! 
>Get cracking and register here for some mind boggling fun and 
>the chance of winning an Apple iPod:
>http://ads.sourceforge.net/cgi-bin/redirect.pl?thaw0031en
>_______________________________________________
>Webware-discuss mailing list
>Webware-discuss@...
>https://lists.sourceforge.net/lists/listinfo/webware-discuss
>  
>

You can use a combination of a function and the class's __init__ to turn a
class into a singleton. It doesn't matter where you define it. Try something
like this:

MySingleton.py:
---------
import threading

def mySingleton():
    if MySingleton._instance is None:
        try:
            MySingleton()
        except MySingleton, inst:
            return inst
    return MySingleton._instance

class MySingleton:
    _instance = None
    _lock = threading.RLock()

    def __init__(self):
        MySingleton._lock.acquire()
        try:
            if not MySingleton._instance is None:
                raise MySingleton._instance
            MySingleton._instance = self
            # Any other init code here
        finally:
            MySingleton._lock.release()

---------

To access the singleton, use:
from MyModule.MySingleton import mySingleton
singleton = mySingleton()


You don't necessarily need to use an RLock to handle the locking. I like to
use them because a singleton's initialization code will often call other
methods on the singleton which require the lock, and it can be a real pain
to have multiple locks for the same object. (RLock allows the same thread to
acquire without blocking.)

- Ben


-----Original Message-----
From: webware-discuss-admin@...
[mailto:webware-discuss-admin@... Behalf Of Martin
Stoufer
Sent: Thursday, March 20, 2003 3:32 PM
To: webware-discuss@...
Subject: [Webware-discuss] Writing your own singletons


Can someone in the know offer up a few pointers or links to code that
implements a singleton inside a WebKit context?

I've attempted to incorporate the instantiation of a singleton class in
the __init__ of the context, but that doesn't seem to be doing the
trick. It gets instantiated when the context is created, but
subsequently so each time a script in that dir is called as well.



--
* Martin C. Stoufer              *
* DIDC/DSD/ITG                   *
* Lawrence Berkeley National Lab *
* MS 50B-2215 510-486-8662       *





-------------------------------------------------------
This SF.net email is sponsored by: Tablet PC.
Does your code think in ink? You could win a Tablet PC.
Get a free Tablet PC hat just for playing. What are you waiting for?
http://ads.sourceforge.net/cgi-bin/redirect.pl?micr5043en
_______________________________________________
Webware-discuss mailing list
Webware-discuss@...
https://lists.sourceforge.net/lists/listinfo/webware-discuss

Hallo,
Martin Stoufer hat gesagt: // Martin Stoufer wrote:

> Can someone in the know offer up a few pointers or links to code that
> implements a singleton inside a WebKit context?
> 
> I've attempted to incorporate the instantiation of a singleton class in
> the __init__ of the context, but that doesn't seem to be doing the
> trick. It gets instantiated when the context is created, but
> subsequently so each time a script in that dir is called as well.

I don't know if it's a singleton the the technical sense, but Ian
Bicking's ParamFactory.py saved my *ss in several threading-related
issues by providing unique objects.

ciao
-- 
 Frank Barknecht                               _ ______footils.org__