Re: [Zope3-Users] Accessing raw post data

[Jim Washington]

to Zope3-Users list after updating subscription email address...

On Wed, 2010-08-25 at 13:06 +0200, Adam GROSZER wrote:
> Hello Edoardo,
> 
> I think the simplest would be to make the upload a
> multipart/form-data, with some headers added if you can.
> Worst case putting the necessary text manually around the binary?
> Then it's a usual POST request what's easily parsed by zope.
> Otherwise it could be really nasty.
> 

You can use a WSGI middleware filter that grabs the request body and
stores it in the WSGI environment before zope consumes it.

Here's a code example:

from webob import Request

class StashPostBody(object):
"""middleware to stash the post body in the WSGI environment"""
def __init__(self, app):
self.app = app

def __call__(self, environ, start_response):
req = Request(environ)
# filter more here, if desired
need_body = req.method == 'POST'
if need_body:
req.environ['spb.req_body'] = req.body
resp = req.get_response(self.app)
return resp(environ, start_response)

def filter_factory(global_conf):
def filter(application):
return StashPostBody(application)
return filter

Once this is configured in the pastescript .ini file as a filter, you
can get the original request body from zope's request object as
request._environ['spb.req_body'].

- Jim Washington



___
Zope3-users mailing list
Zope3-users@zope.org
https://mail.zope.org/mailman/listinfo/zope3-users

Re: [Zope3-Users] z3c.jsonrpc response content type bug

[Jim Washington]

Roger Ineichen wrote:
> Hi Yuan
> 
>> Betreff: [Zope3-Users] z3c.jsonrpc response content type bug
>>
>> When I use z3c.jsonrpc (svn head) to publish an object, the 
>> response generated gets the content header:
>>
>> Content-Type: application/x-javascript
>>
>> According to the README.txt of the package:
>>
>> This project provides the proposed request type 
>> "application/json". The request type "application/json-rpc" 
>> is supported as long it is not officialy deprecated.
>>
>> And it also expects to receive a request with 'application/json'
>> header. So this should be a bug.
>>
>> To fix it, one just need to modify _prepareResult method in 
>> the file publisher.py to change from
>>
>> # set content type
>> self.setHeader('content-type', 
>> "application/x-javascript;charset=%s" \
>> % charset)
>>
>> to
>>
>> # set content type
>> self.setHeader('content-type', "application/json;charset=%s" \
>> % charset)
>> --
>> Hong Yuan
> 
> Thanks for the hint. I'll take a look at that this week.
> Regards
> Roger Ineichen

A bit of history on this.  The content-type in this bit of code was
determined more than three years ago by testing with several old major
web browsers.  At the time, 'application/x-javascript' was the
content-type that reliably told the web browser to use its javascript
parser on the document so that the rpc data could be used.  Other
settings did not work on all of the web browsers I tested.

So, this content-type was decided empirically (a long time ago), not
from specification. It maybe should continue to be the default for
compatibility with creaky old web browsers.  But maybe it can be
changeable (if you *really* know what you are doing) so that the
outgoing document can meet the expectations of client libraries.

- Jim Washington

___
Zope3-users mailing list
Zope3-users@zope.org
http://mail.zope.org/mailman/listinfo/zope3-users

Re: [Zope3-Users] Dynamic interfaces ?

[Jim Washington]

Thierry Florac wrote:

Hi,

I'm looking for a way to handle what I've called "dynamic interfaces".
My problem is quite simple : I have an application where an
administrator can define "content types" ; for each content type, a set
of common predefined attributes are used, but the administrator can
define for each content type a set of custom attributes, each attribute
being defined by a name, a datatype (text, textline, date, boolean...)
and eventually a list of references values (for choices) ; definition of
these custom attributes is persistent and stored in the ZODB.
When creating data, contributor can select the data type, according to
which the matching set of attributes will be available in forms.

The main point is probably that I'd like to use z3c.form package "as
usual", and so integrate this attributes management into Zope3
interfaces machinery... ; I suppose I should need a king of "interface
factory", but couldn't find any link about that.

Thanks for any help,

  Thierry Florac
  

Hi, Thierry

I was experimenting with dynamic interfaces a while back, and this is 
where I ended up.


Theoretically, Interfaces are just python class objects, not importantly 
different than any others.


You just need to create a python class dynamically:

import new
from zope.interface import Interface

dyn_iface = new.classobj(name,(Interface,),definitions)

where

name is a string, the name of your interface class
Interface is zope.interface.Interface or a descendent
definitions is a dict of class members 
{'__doc__':docstring,'name':Schema,'name2':Schema2}


so,

if your class could be defined in an interfaces file as


class IMyClass(Interface):
   """IMyClass interface"""
   yourName=TextLine(title=u'Your name',required=True)


you can dynamically create it like this:


import new
from zope.interface import Interface
import zope.schema

definitions = {}
# __doc__ needs to be there; empty string is fine
definitions['__doc__'] = 'IMyClass Interface'
definitions['yourName'] = zope.schema.TextLine(title=u'Your 
name',required=True)


iface_name=u'IMyClass'

myiface = new.classobj(iface_name.encode('utf-8'),(Interface,),definitions)


At this point, myiface should be usable as if generated in an interfaces 
file.


According to the python docs, you need to be careful when using "new", 
so appropriate caution is advised. 
http://docs.python.org/lib/module-new.html



HTH,

- Jim Washington

___
Zope3-users mailing list
Zope3-users@zope.org
http://mail.zope.org/mailman/listinfo/zope3-users

Re: [Zope3-Users] Find the zope3 instance home directory (or the /var directory)

[Jim Washington]

Stefan H. Holek wrote:
> I don't think there is any way whatsoever in Zope 3. There is no
> "instance" home to begin with, and the os.pardir hacks don't work
> either because you can't really know where your package is installed.
>
> Stefan
>
>
> On 6. Nov 2007, at 11:02, Eric Br�hault wrote:
>
>> I had a look to different source code, and apparently the solution is
>> to use
>> the path of the current file to get the instance home, so we have
>> code like
>> this:
>> os.path.normpath(
>> os.path.join(os.path.dirname(__file__),
>>  os.pardir, os.pardir, os.pardir, os.pardir))
>>
>> Is there any easier (and cleaner) way to do it ?
>
I have found useful another trick related to __file__ .

If the file system location you are looking for is near code you are
using or can import, python's "inspect" may be your friend.

>>> import inspect

So, if your object is, for example, a view method and you need the
directory where another method is defined for your view, you can

>>> sourcefilepath = inspect.getsourcefile(self.otherMethod)

Here's another example for clarification.  It works, but is not
particularly useful:

Let's say we want to know the file system location where
Decimal.normalize is defined.

>>> from inspect import getsourcefile
>>> from decimal import Decimal

We can get the location directly:

>>> getsourcefile(Decimal.normalize)
'/usr/lib64/python2.5/decimal.py'

Or, if we just want the location of the class:
>>> getsourcefile(Decimal)
'/usr/lib64/python2.5/decimal.py'

or, we can get the location from an instance:

>>> d = Decimal('0.12')
>>> getsourcefile(d.normalize)
'/usr/lib64/python2.5/decimal.py'

>From there, standard os.path functions can get you locations relative to
the location given.

HTH,

-Jim Washington


___
Zope3-users mailing list
Zope3-users@zope.org
http://mail.zope.org/mailman/listinfo/zope3-users

Re: AW: AW: [Zope3-Users] Re: header trouble

[Jim Washington]

Roger Ineichen wrote:
> Hi Jim
>   
> Right now I can see the following usecases:
>
> - File is encoded on file system
> - PageTemplate class defines a encoding
> - HTML template defines encoding
> - is there a encoding option in ZCML?
>
> Are more components involved in this?
>
>   
Hi, Roger

Are you dealing with input or output?

For output, the zope publisher has always (apparently until just
recently) taken care of the meta-content-type tag and header for page
templates.  It checks the client's accept-encoding, and if utf-8 is OK,
the publisher will decide utf-8 is the output charset, otherwise it sets
the charset to whatever the client will accept.  Then, any existing
meta-content-type tag is removed, and a meta-content-type tag is
inserted into the document head (using the document's contentType), and
the appropriate http header is set.  The presumption is that page
template content at publishing time is unicode and gets encoded for
output (with appropriate tags and headers) by the publisher.

So, if this is about outputting page template content, it's a matter of
returning the publisher to its previous behavior, inserting the
meta-content-type tag.

-Jim
___
Zope3-users mailing list
Zope3-users@zope.org
http://mail.zope.org/mailman/listinfo/zope3-users

Re: AW: [Zope3-Users] Re: header trouble

[Jim Washington]

Christian Zagrodnick wrote:

[...]
> But we *are* sending a Content-Type header. And we remove the
> http-equiv thingy from the page template file in favour of the HTTP
> header.
>   
It's a good idea to keep the http-equiv content-type meta tag.  If an
HTML document is stored to disk, and that tag is not present, parsers
and browsers cannot be sure what the character encoding is.

The HTTP header should have precedence when the document is delivered
over the web, but on disk, you do not have that header.

More at:

http://www.trilithium.com/johan/2005/09/meta-http-equiv/

-Jim Washington




___
Zope3-users mailing list
Zope3-users@zope.org
http://mail.zope.org/mailman/listinfo/zope3-users

Re: [Zope3-Users] best way to get started nowadays?

[Jim Washington]

Chris Withers wrote:
> Benji York wrote:
>> Chris Withers wrote:
>>> So, what *is* the "standard" way now and where can I read about how
>>> to do it?
>>
>> That's in flux at the moment.  I think Philipp's zopeproject is a
>> step in the right direction, but haven't had a chance to look at it
>> closely yet.
>
> Where's zopeproject's home?
>
> Chris
>
Start at pypi (cheeseshop). You easy_install it.

-Jim Washington
___
Zope3-users mailing list
Zope3-users@zope.org
http://mail.zope.org/mailman/listinfo/zope3-users

Re: [Zope3-Users] Re: Zope3 still not compatible with Python2.5

[Jim Washington]

Philipp von Weitershausen wrote:
> This is a rather confusing post. I'm not sure what you're trying to
> say with it.
>
> Is Python 2.5 officially supported? No. We never said it was.
>
> Does Zope 3.4 actually work on Python 2.5.1? Yes, it works just fine
> for me and others. For example, Grok runs on Python 2.5.1 no problem [1].
>
This is very good to hear.  I'm not on the grok list.

> Is this a platform issue? Maybe, but your post isn't giving *any*
> useful information for tracking this down. All you're basically saying
> is that you were "experimenting with several things all at once" and
> that that was a bad idea. You certainly got that right.
>

Yup.  That's pretty much all I was saying - plus that in my case, the
quickest solution was to go back to python2.4.  I was also hoping I
might hear some time frame when I might expect to try again successfully. 

> I hope the rest isn't taken for FUD. I certainly encourage people to
> try out Zope 3.4 on Python 2.5 and report any problems *properly* so
> that we can chase down the bugs.
>
No FUD intended. 

I have attached a couple of tracebacks.  Traceback1 occurs when asking
for /index.html.  Traceback2 occurs when the instance from traceback1 is
stopped and restarted.

Is this more helpful?

Regards,

-Jim Washington


2007-09-11T22:42:56 ERROR SiteError http://localhost:8080/@@index.html
Traceback (most recent call last):
  File 
"/home/jwashin/buildout-eggs/tmprA8xJD/zope.publisher-3.5.0a1.dev_r78838-py2.5.egg/zope/publisher/publish.py",
 line 133, in publish
  File 
"/home/jwashin/buildout-eggs/tmpNVNVrW/zope.app.publication-3.4.0a1_2-py2.5.egg/zope/app/publication/zopepublication.py",
 line 167, in callObject
  File 
"/home/jwashin/buildout-eggs/tmprA8xJD/zope.publisher-3.5.0a1.dev_r78838-py2.5.egg/zope/publisher/publish.py",
 line 108, in mapply
   - __traceback_info__: >
  File 
"/home/jwashin/buildout-eggs/tmprA8xJD/zope.publisher-3.5.0a1.dev_r78838-py2.5.egg/zope/publisher/publish.py",
 line 114, in debug_call
  File 
"/home/jwashin/buildout-eggs/tmpBqALB1/zope.app.container-3.5.0a1-py2.5-linux-x86_64.egg/zope/app/container/browser/contents.py",
 line 439, in index
  File 
"/home/jwashin/buildout-eggs/tmpW6ie0m/zope.app.pagetemplate-3.4.0b1dev_r75616-py2.5.egg/zope/app/pagetemplate/viewpagetemplatefile.py",
 line 83, in __call__
  File 
"/home/jwashin/buildout-eggs/tmpW6ie0m/zope.app.pagetemplate-3.4.0b1dev_r75616-py2.5.egg/zope/app/pagetemplate/viewpagetemplatefile.py",
 line 51, in __call__
  File 
"/home/jwashin/buildout-eggs/tmpYyeA98/zope.pagetemplate-3.4.0a1-py2.5.egg/zope/pagetemplate/pagetemplate.py",
 line 115, in pt_render
  File 
"/home/jwashin/buildout-eggs/tmpG5P6SX/zope.tal-3.4.0b1-py2.5.egg/zope/tal/talinterpreter.py",
 line 271, in __call__
  File 
"/home/jwashin/buildout-eggs/tmpG5P6SX/zope.tal-3.4.0b1-py2.5.egg/zope/tal/talinterpreter.py",
 line 346, in interpret
  File 
"/home/jwashin/buildout-eggs/tmpG5P6SX/zope.tal-3.4.0b1-py2.5.egg/zope/tal/talinterpreter.py",
 line 891, in do_useMacro
  File 
"/home/jwashin/buildout-eggs/tmpG5P6SX/zope.tal-3.4.0b1-py2.5.egg/zope/tal/talinterpreter.py",
 line 346, in interpret
  File 
"/home/jwashin/buildout-eggs/tmpG5P6SX/zope.tal-3.4.0b1-py2.5.egg/zope/tal/talinterpreter.py",
 line 536, in do_optTag_tal
  File 
"/home/jwashin/buildout-eggs/tmpG5P6SX/zope.tal-3.4.0b1-py2.5.egg/zope/tal/talinterpreter.py",
 line 521, in do_optTag
  File 
"/home/jwashin/buildout-eggs/tmpG5P6SX/zope.tal-3.4.0b1-py2.5.egg/zope/tal/talinterpreter.py",
 line 516, in no_tag
  File 
"/home/jwashin/buildout-eggs/tmpG5P6SX/zope.tal-3.4.0b1-py2.5.egg/zope/tal/talinterpreter.py",
 line 346, in interpret
  File 
"/home/jwashin/buildout-eggs/tmpG5P6SX/zope.tal-3.4.0b1-py2.5.egg/zope/tal/talinterpreter.py",
 line 861, in do_defineMacro
  File 
"/home/jwashin/buildout-eggs/tmpG5P6SX/zope.tal-3.4.0b1-py2.5.egg/zope/tal/talinterpreter.py",
 line 346, in interpret
  File 
"/home/jwashin/buildout-eggs/tmpG5P6SX/zope.tal-3.4.0b1-py2.5.egg/zope/tal/talinterpreter.py",
 line 957, in do_defineSlot
  File 
"/home/jwashin/buildout-eggs/tmpG5P6SX/zope.tal-3.4.0b1-py2.5.egg/zope/tal/talinterpreter.py",
 line 346, in interpret
  File 
"/home/jwashin/buildout-eggs/tmpG5P6SX/zope.tal-3.4.0b1-py2.5.egg/zope/tal/talinterpreter.py",
 line 949, in do_defineSlot
  File 
"/home/jwashin/buildout-eggs/tmpG5P6SX/zope.tal-3.4.0b1-py2.5.egg/zope/tal/talinterpreter.py",
 line 346, in interpret
  File 
"/home/jwashin/buildout-eggs/tmpG5P6SX/zope.tal-3.4.0b1-py2.5.egg/zope/tal/talinterpreter.py",
 line 822, in do_loop_tal
  File 
"/home/jwashin/buildout-eggs/tmpD0OTbe/zope.tales-3.4.0a1-py2.5.egg/zope/tales/tales.p

[Zope3-Users] Zope3 still not compatible with Python2.5

[Jim Washington]

For now, anyway.  There is work going on, but it is not ready yet.

Well, it was an interesting experiment, and I thought I would write it
up for the group.

Gentoo recently made 2.5 the system default python.  I have been a
couple of days trying to track down why things had broken.  Yes, I know
-- system default python is bad, but I compile it all (it's Gentoo!), so
the UCS2 vs UCS4 thing is not a problem, etc., etc.

Anyway, I turned out to be experimenting with several things all at once:

1.  gcc 4.2.0

2.  UCS4 default vs. UCS2 (has is always been this way in Gentoo?
Apparently, you can set this to UCS2 if you know what you are doing.  I
wonder how you would know that you know what you are doing?)

3.  using zopeproject instead of my own attempt at a buildout recipe
(highly recommended!  Thanks, Philipp!)

4.  Plus, what turned out to be the culprit, python2.5.1

5.  And, of course, I am using a 64-bit OS, which you cannot discount as
a possible source for things going wrong.

I was getting tracebacks that ended with

MemoryError

-or-

an error where UserDict does not have attribute "data".

So, if you see these errors, fret not. Zap your broken new Data.fs files
and go back to python-2.4 for your zope3 instances.

With hopes that this saves someone else some frustration,

-Jim Washington


___
Zope3-users mailing list
Zope3-users@zope.org
http://mail.zope.org/mailman/listinfo/zope3-users

Re: [Zope3-Users] Calling a view in a doc test

[Jim Washington]

Florian Lindner wrote:
> Hello,
> in a doctest I have an object which has a view registered.
> I want to call this view and test for the XML it returns.
> How can I call the view so that it is being rendered, just like called by a 
> browser?
>
> Thanks,
>
> Florian
> ___
> Zope3-users mailing list
> Zope3-users@zope.org
> http://mail.zope.org/mailman/listinfo/zope3-users
>   
Hi, Florian

Search for zope.testbrowser.

It's sometimes used in functional doctests.

from zope.testbrowser.testing import Browser
browser = Browser('http://localhost/')
#browser.handleErrors = False
browser.open('/somepath/someview.html')

-Jim Washington
___
Zope3-users mailing list
Zope3-users@zope.org
http://mail.zope.org/mailman/listinfo/zope3-users

Re: [Zope3-Users] Blog naming proposals

[Jim Washington]

Pablo Ambrosio wrote:
> "Bitacora", spanish for log. Since all names with "blog" in it seem
> already taken.
>
hehe.  I wondered where the word "bitakora" came from.

as in http://www.codesyntax.com/bitakora

But that seems to be taken, too.  Sort-of...

If you like the 'z' thing, something like expressionz or noize might be
fun.  Certainly better than anything with a z prepended.

If you really want your product to have a stand-out name, check whatever
you are evaluating on google.  If there are no matches, all the better.

For example: junglefig .  (It's a planet earth reference.)

-Jim Washington

___
Zope3-users mailing list
Zope3-users@zope.org
http://mail.zope.org/mailman/listinfo/zope3-users

[Zope3-Users] Initial File Releases - Zif Collective

[Jim Washington]

On behalf of the Zif Collective we are pleased to announce the release
of the following packages:

zif.gzipper-0.2
WSGI middleware providing gzip compression for HTTP server output. Does
what mod_gzip/mod_deflate does in apache, but if your Zope 3 is not
behind apache, this may be useful.  Highly configurable for choosing
what gets compressed, but does not compress unless the client allows.

zif.headincludes-0.2
Zope 3-enabled WSGI middleware to manipulate css and javascript elements
in the HTML header.  It's essentially the same utility as
zc.resourcelibrary, except implemented as WSGI middleware.

zif.jsmin-0.2
WSGI middleware for javascript compression. Configurable levels of
compression.

zif.jsonserver-0.1
JSON/JSON-RPC support for Zope 3. Now with zif namespace.  Employs
python-cjson (fast and compliant) for coding/encoding JSON if
python-cjson is installed.

zif.xtemplate-0.1
Document-oriented XHTML templating for Zope 3. Use lxml's elementtree
API and XPath functionality to generate pages.  This is fairly alpha --
interfaces and scope are still in flux. 

The Zif Collective is a growing collection of useful packages for Zope
3, although it is intended that some of the packages will find utility
in other python efforts. Our software packages are available as eggs or
tar.gz sources at the Python Cheese Shop and Sourceforge. We welcome
your feedback and participation to improve our offering. You will find
our landing page at:

zif.sourceforge.net

Regards,
Jim Washington
David Pratt

The Zif Collective contains no harmful chemicals.
___
Zope3-users mailing list
[EMAIL PROTECTED]
http://mail.zope.org/mailman/listinfo/zope3-users

Re: [Zope3-Users] Is DTML Deprecated In Zope3?

[Jim Washington]

Mark, Jonathan (Integic) wrote:
> Phillip says in his book not to use DTML any more. However, I am finding that 
> what is easy in DTML, producing an Atom 1.0 feed, for instance, is harder in 
> ZPT. I get peculiar errors telling me that my XML boilerplate is not valid. 
> DTML doesn't care. 
>
> I notice that in the Zope 3.3 ZMI you can create a DTML page. So is DTML 
> going to be supported for the next few years?
>   
If you are publishing xml-ish content in zope 3, you might consider
using something like the techniques in zif.xtemplate.  zif.xtemplate
makes XHTML (with appendix c) documents, but you could easily do
something similar for Atom (or RSS) feeds.

zif.xtemplate uses lxml to create XHTML documents using the elementtree
API.  This means using python functions, classes and methods to get
around in your data, do looping, etc.  You can load and parse a template
from the file system as a starter piece.  Additional processing is
available with lxml's xpath and xslt support. 

I find this document-centric method a very nice thing to develop with. 
On the bad side, view code using this technique needs a z3 restart (or
z3reload) for changes, but on the good side, all error tracebacks are
standard python tracebacks.

zif.xtemplate is sort-of alpha for the moment.  I'm still evaluating
methods that need to be included, and those that are YAGNI.  We have not
done any benchmarks vs. ZPT, but it seems fast enough. It includes an
HTMLSanitizer for dealing with untrusted HTML.  In the default
pretty-print mode, zif.xtemplate puts out web pages with very pretty source.

zif.xtemplate is at http://zif.sourceforge.net .  We have not made any
releases yet, but to see/use the code, it's available in sourceforge's
svn repository.  I'm interested in feedback.

svn co
http://zif.svn.sourceforge.net/svnroot/zif/zif.xtemplate/trunk/src/zif/xtemplate

-Jim Washington
___
Zope3-users mailing list
Zope3-users@zope.org
http://mail.zope.org/mailman/listinfo/zope3-users

Re: [Zope3-Users] Crash

[Jim Washington]

Maciej Wisniowski wrote:
>> The last entry in the access.log is the following:
>> 218.160.132.172 - - [10/Mar/2007:22:11:45 -0500] "CONNECT
>> sextw.com.tw:25 HTTP/1.0" 404 0 "-" "-"
>>
>> There is no other information to indicate why as far as we can tell.
>> No exceptions, nothing.
>> 
> Anything in /var/log/messages? Segfault or something like that maybe?
>
> We had similiar issues with Zope2 and there were segfault errors in
> messages file
> (due to 64 bit architecture and DCOracle2)
>
>   
I have no idea whether this is relevant, but I see port 25, which is
SMTP.  The implementation of SMTP is synchronous in Zope 3, AFAICT.  I
found last week that sending out emails through a very slow SMTP server
tied up Zope 3's server threads waiting for that server's response,
which led to in turn to Zope 3 being nonresponsive.

-Jim Washington

___
Zope3-users mailing list
Zope3-users@zope.org
http://mail.zope.org/mailman/listinfo/zope3-users

Re: [Zope3-Users] Menus

[Jim Washington]

David Johnson wrote:
I am curious what methods people use to implements simple web site 
menus?  It seems to be this must be a common task.


So far I've done this as follows:
1. Create a skin
2. Create a content type Menu as Ordered Container
3. Create a content type MenuItem as contained by Menu
4. Add a Menu and fill it with MenuItems
4. Register Menu
5. Create a macro which returns the Menu using zapi.getUtility()
6. Use the macro in the skin, and display all the MenuItems.


Hi, David

You can do it that way.

I've been successful with the browser:menu... directives in zcml.  The 
following, as part of my browser/configure.zcml file, actually creates 
an IBrowserMenu utility with a name of "myMenu".  I have another set of 
menuItems for the same menu for different interfaces, so the same named 
menu has different items at different parts of the site.


I find it handy to have this configuration in the same file where I 
configure the names of the pages.



  
   



   

   


- Jim Washington
___
Zope3-users mailing list
Zope3-users@zope.org
http://mail.zope.org/mailman/listinfo/zope3-users

Re: [Zope3-Users] getId in zope3?

[Jim Washington]

Peter Bengtsson wrote:
> In Zope2 every object has an 'id' which meant that you can use
> someobject.getId() to find out what name the object is stored under in
> its container.
> Is there an equivalent in zope3?
> ...or am I looking for a solution to a problem that doesn't exist?
Hi, Peter

Have a look at

zope.traversing.api

If you have an older source tree, look for

zope.app.traversing.api

which is deprecated, but OK to use until 3.5.

-Jim Washington
___
Zope3-users mailing list
Zope3-users@zope.org
http://mail.zope.org/mailman/listinfo/zope3-users

Re: [Zope3-Users] App server tracebacks for functional tests

[Jim Washington]

John Maddison wrote:
> Hi all,
>
> Firstly, I have written some tests for an application, and with both
> "./bin/test -u" and "./bin/test -f", they pass successfully.  However,
> "./bin/test" fails in the functional tests (on a form submit) with a
> rather uninformative "HTTP Error 500: Internal Server Error".
> Performing as-near-as-I-can-make-it identical actions in a browser
> works fine.  I'm sure I'm missing something obvious - has anyone seen
> something like this before?  [I'm not attaching source code since I've
> not yet observed this problem in anything less than the entire
> (unfortunately) proprietary application I'm working on.]
>
> Secondly, and leading on from this, is it possible to get the
> traceback the application server side of things produces in functional
> (zope.app.testing.functional.FunctionalDocFileSuite) tests?  At the
> moment the only information I'm seeing is the aforementioned 500
> error.
>
If you are using zope.testbrowser,  you can set

browser.handleErrors=False

to get a more informative traceback.

-Jim Washington
___
Zope3-users mailing list
Zope3-users@zope.org
http://mail.zope.org/mailman/listinfo/zope3-users

[Zope3-Users] ANN: JSONView

[Jim Washington]

JSONView is now available as part of the jsonserver package.

A JSONView responds to HTTP GET and returns a snippet of JSON.  This is 
useful for AJAX/XHR, things like repopulating a select list without a 
full page refresh.


Usage is fairly simple.  Make a view class subclassing JSONView from 
jsonserver.  In your view class, make a method, doResponse, that returns 
a python dict or list that is the data you wish to return to the 
client.  One other method is always called before publication, 
doCacheControl, which may be overridden/extended for setting caching 
headers.  Content-type, which is by default 'application/json', may be 
overridden in your class by setting an instance variable named 
contentType.  Register your view with the  ZCML directive, 
giving it a name, for (interface), class, and permission.


JSONView takes care of matching named parameters in the request with 
named parameters in the doResponse method, so a view named "sum" with a 
method signature like doResponse(a,b=10) can be requested like http://x 
ample.com/sum?a=4&b=9


jsonserver with JSONViews is available (ATM) only on z3labs svn:

http://codespeak.net/svn/z3/jsonserver/trunk

For more information the README (ftest) is at 
http://codespeak.net/svn/z3/jsonserver/trunk/JSONViews.txt .


This is new code, so comments/suggestions are welcome.

-Jim Washington

___
Zope3-users mailing list
Zope3-users@zope.org
http://mail.zope.org/mailman/listinfo/zope3-users

Re: [Zope3-Users] Re: Reportlab threadsafe solution?

[Jim Washington]

Chris Withers wrote:

Chris McDonough wrote:

Why not just use a mutex (a recursive lock makes things easier too)?


Yeah, Big Fat Lock has been my solution with Reportlab, worked well 
under high load for several years.


I don't use RLocks myself, paranoia says they're not big or fat enough 
;-)


Chris

Thanks, all, for the variety of interesting methodologies for addressing 
this.


I am going with the threading.Lock idea.

RLock would be overkill, since the module does not do anything 
foreseeably reentrant.  And, I would not want to think myself 
insufficiently paranoid.  :)


-Jim Washington

___
Zope3-users mailing list
Zope3-users@zope.org
http://mail.zope.org/mailman/listinfo/zope3-users

[Zope3-Users] Reportlab threadsafe solution?

[Jim Washington]

Hi

I need to produce PDFs with reportlab.

reportlab is not threadsafe. 
(http://two.pairlist.net/pipermail/reportlab-users/2006-June/005037.html)


I think I need a way to queue or generate a "lockfile" on requests for 
PDFs so that only one-at-a-time is generated.


I'm hoping that there is a utility for this already, but search in the 
source for queue or serialize or lock gets me a lot of stuff that does 
not seem to relate to what I want.


This looks like it would be a relatively simple utility.  Has this been 
done yet?  Any hints?


-Jim Washington
___
Zope3-users mailing list
Zope3-users@zope.org
http://mail.zope.org/mailman/listinfo/zope3-users

Re: [Zope3-Users] How to make a new namespace for pagetemplates?

[Jim Washington]

FB wrote:

Hi,

On Fri, Sep 08, 2006 at 12:54:14PM +0200, Martijn Pieters wrote:
  

On 9/8/06, FB <[EMAIL PROTECTED]> wrote:


my employer want to have all external links marked with a small icon telling
anonymous users from the internet that everything behind given links is
beyond our responsibility.
  

Why not use a piece of javascript to do this? See the linkpopper
product on plone.org for a way to process all links in a page and
process them. That product makes external links open in a new window,
but the code should be easy to alter.



Thank you for the hint.

But there are several reasons for not using JS:
   * One of the constraints of that site is javascript being optional.
 Problem ist: marked links are mandatory - they have to be marked
 even with javascript turned off.
   * I'd like to have a tag-postprocessing namespace for some other
 reasons, too - e.g. for a printing-view that automatically creates
 a list of links at the end of the page.
   * I'd like to know, how to make a new pagetemplate namespace :-).

  

As a follow-on to Philipp's comment about WSGI middleware:

It could be fairly easy (to be sure, not *really* easy) to do a 
search-and-replace for link tags and do appropriate alterations in WSGI 
middleware.


If I were doing this, I might re-parse the document in middleware using 
lxml, then do an xpath query to find the locations in the document that 
need changing.  Then, it would be a matter of using the elementtree API 
to make the changes.  lxml.etree.tounicode() would put the page back 
together for output.  Since outgoing pages generated with ZPT are 
generally parseable, this might work OK.  N.B., You may have to use the 
DTD at some point with this method.  Lxml will attempt to xml-minimize 
(e.g., ) all tags without text content, so you will have to assure 
that this does not happen where it shouldn't.


If you are not too married to generating pages with ZPT, and if you like 
the elementtree API, there is also a possibility of a lxml/elementtree 
(python-only) method for generating pages, which can have a method for 
post-processing the page, in a manner like the above, except that the 
document stays in a parsed state until it is output by Zope.  I have a 
working base class for this.  I call it XTemplate.  It's 
not-quite-ready-for-prime-time (no docs or tests, yet, and you have to 
do skinning within this framework), but if there is interest, I should 
be able to release a preview in the near future.


-Jim Washington
___
Zope3-users mailing list
Zope3-users@zope.org
http://mail.zope.org/mailman/listinfo/zope3-users

Re: [Zope3-Users] Re: zc.tables integration with Zope3

[Jim Washington]

Luis De la Parra wrote:

Hello

this is how indexes are implemented in RDBMs.. don't know if sqllite has
such a feature, but in oracle the table content is just stored in files
(structured, but unordered) and a set of sorted indexes is kept for the
primary key and all other secundary indexes. 


each index has the search fields like index0 = "lastname/firstname" and
index1 = "telephone" and a  file/segment/row pointer to the information.
every update to the table triggers an update to all indices in the same
transaction, but you get searching, sorting and batching in return.

  
Yup.  But if your data is in a zodb,  you only have one (maybe two, with 
an OrderedFolder)  canonical ordering of things in a container.  Any 
other order requires sorting the entire set on the fly.


... At the moment. ;)

For anyone following this thread, I have updated my factoradic 
implementation code this morning.  It now memoizes to an sqlite db, and 
is much much faster overall.  http://blog.hill-street.net


-Jim Washington
___
Zope3-users mailing list
Zope3-users@zope.org
http://mail.zope.org/mailman/listinfo/zope3-users

Re: [Zope3-Users] zc.tables integration with Zope3

[Jim Washington]

David Pratt wrote:
Hi Jim. This approach is completely new to me. It is interesting and 
I'll keep an eye on your blog. How expensive is updating indexes with 
data modification. You are talking about sqlite, does this mean you 
figure the indexes would be a significant drag on the ZODB due their 
size or is this idea a performance consideration. Many thanks.


The expense is in recalculating the new factoradic for each stored 
index, which involves a lot of division and factorial operations.  But 
this might be done in a smart way.  Maybe instead of on-change, we could 
recalculate an index on first access to that index after change.


Sqlite would be mainly for performance, ATM.  A lookup is less expensive 
than calculating a factoradic of a very large number.


Size may also be an issue if the index is stored in the zodb.  An 
integer (long as it might be) is the most compact representation of the 
factoradic. I'm pretty sure that longs end up as base ten strings in a 
pickle, but that is still smaller than the pickle of the factoradic list 
representation.  I still need to experiment more, but it's not out of 
the question that the indexes could end up as sqlite-backed annotations 
to the container/data source, particularly when sqlite may already be in 
the mix for memoizing factorial and/or factoradic calculations.


It may still turn out that this is all not worth the effort in practice.

-Jim Washington
___
Zope3-users mailing list
Zope3-users@zope.org
http://mail.zope.org/mailman/listinfo/zope3-users

Re: [Zope3-Users] zc.tables integration with Zope3

[Jim Washington]

Martijn Faassen wrote:

Jim Washington wrote:
This is probably a bit premature, but I have been doing some thinking 
about reordering/batching of large sets.


My current (not-quite-ready-for-prime-time) solution involves 
factoradics, and I have done a bit of a write-up on my blog, 
http://blog.hill-street.net/?p=5 .


I don't have time to read your article right now, but I will do so 
later. I'm quite interested in making the presentation of tabular data 
scale better. Right now, zc.table excepts to have access to the whole 
list in order to sort and batch it, but with huge amounts of data this 
may not scale so well. Relational databases have limit queries that 
seem to help here, though I'm not sure how much optimization is 
happening on the backend level.


Anyway, just wanted to register my interest in this topic, if indeed I 
understand you correctly.




Hi, Martijn

Good to hear that there may be some interest.  As usual, soon after 
publishing, I came up with some ideas for improving the implementation 
algorithms.  It's on my to-do list. 

BTW, I have recently been having some fun generating Zope 3 pages using 
lxml.  I plan to do another article about that, soon.


-Jim Washington
___
Zope3-users mailing list
Zope3-users@zope.org
http://mail.zope.org/mailman/listinfo/zope3-users

Re: [Zope3-Users] zc.tables integration with Zope3

[Jim Washington]

This is probably a bit premature, but I have been doing some thinking 
about reordering/batching of large sets.


My current (not-quite-ready-for-prime-time) solution involves 
factoradics, and I have done a bit of a write-up on my blog, 
http://blog.hill-street.net/?p=5 .


The idea is that you can have indexes to several permutations 
(first-name-sort, last-name sort, last-modified-date-sort, etc) of a 
set, and refer to those permutations by name, which refers to an integer 
index.  The factoradic representation of that integer can be used in a 
quick algorithm to make a list of the ids in the set in that specific 
order, and simply slicing that list would give easy batching capability 
without needing to access the objects at all.  There would, however, be 
a need to update all of the permutation indexes when the data change.


One thing I have not dealt with yet to make this really useful would be 
a utility that memoizes the factorials and factoradics, which are 
expensive to calculate, probably using an sqlite database.


The big trade-off is doing a sort each time the data is accessed vs. 
doing all of the named sorts each time the data is changed.


I am not sure at the moment whether this solves a real need yet, but I 
thought I would bring it up since we are discussing batching.  I think 
it is just at the "interesting idea" stage for right now.


-Jim Washington
___
Zope3-users mailing list
Zope3-users@zope.org
http://mail.zope.org/mailman/listinfo/zope3-users

Re: [Zope3-Users] Products/Ordered Container

[Jim Washington]

David Johnson wrote:
> I've tried to follow the SVN, but I'm not quite sure where products are
> for Zope 3.  Where should I look? The link on the zope.org site for
> zope-3 products points to the root of the Zope SVN.  
>
> In particular, I'm interested in an ordered container in which the order
> can be manipulated (rather than add order or alphabetical).
>
> In general I'm curious to see what's out there.  As well as possible
> contribute some small products I've written.
>   
For an ordered container implementation, look in zope.app.container.ordered

I would not use it for a huge container (it maintains a separate list of
item ids for ordering), but it suffices for reasonably-sized containers.

-Jim Washington
___
Zope3-users mailing list
Zope3-users@zope.org
http://mail.zope.org/mailman/listinfo/zope3-users

Re: [Zope3-Users] Re: Dojo Support

[Jim Washington]

Rocky Burt wrote:
> On Tue, 2006-30-05 at 14:49 -0400, Jim Washington wrote:
>   
>> I am just starting to try-out Dojo (http://dojotoolkit.org). 
>>
>> Interestingly, the RPC it does is JSON-RPC. Since I am well-acquainted 
>> with JSON-RPC support in Zope 3, I have put together a package for 
>> making Dojo and its JSON-RPC client work with Zope 3.
>>
>> http://zif.hill-street.net/dojosupport
>>
>> At the moment, Dojo is just a new toy for me, but I see a lot of 
>> potential for rich user interfaces using Zope 3 and Dojo.
>> 
>
> Perhaps the z3c.javascript project in svn.zope.org would benefit from
> inclusion with dojo (rather than you making your own prj).
>   
Thanks, Rocky.  I sometimes need to be reminded about things like this.

I see some dojo support is already there in z3c.javascript, and it is
for most purposes identical to what I worked out.

-Jim Washington

___
Zope3-users mailing list
Zope3-users@zope.org
http://mail.zope.org/mailman/listinfo/zope3-users

[Zope3-Users] Dojo Support

[Jim Washington]

I am just starting to try-out Dojo (http://dojotoolkit.org). 

Interestingly, the RPC it does is JSON-RPC. Since I am well-acquainted 
with JSON-RPC support in Zope 3, I have put together a package for 
making Dojo and its JSON-RPC client work with Zope 3.


http://zif.hill-street.net/dojosupport

At the moment, Dojo is just a new toy for me, but I see a lot of 
potential for rich user interfaces using Zope 3 and Dojo.


-Jim Washington
___
Zope3-users mailing list
Zope3-users@zope.org
http://mail.zope.org/mailman/listinfo/zope3-users

Re: [Zope3-Users] New Request Variable?

[Jim Washington]

Fred Drake wrote:

On 5/19/06, Jim Washington <[EMAIL PROTECTED]> wrote:

Request is also the wrong place.  It's slotted; I cannot add variables.


Isn't this what the request annotations are for?


 -Fred

Thanks, Fred.  That is, of course, the correct (and quite elegant) 
solution. 


-Jim Washington

___
Zope3-users mailing list
Zope3-users@zope.org
http://mail.zope.org/mailman/listinfo/zope3-users

[Zope3-Users] New Request Variable?

[Jim Washington]

I have an expensive function (the result is a long list of IntIds) that 
I only want to call once in a request.  It only gets calculated the 
first time it is called, and the results may be used anytime after 
that.  But only for the current request.  A subsequent request has to 
recalculate from the beginning.


Session is the wrong place to stash this.  I do not want want it to 
persist beyond the current request.


Request is also the wrong place.  It's slotted; I cannot add variables.

So where is the proper place to hold on to something like this just for 
the duration of one request?


Suggestions (or solutions), anyone?  I fear I may have missed something 
obvious.


-Jim Washington

___
Zope3-users mailing list
Zope3-users@zope.org
http://mail.zope.org/mailman/listinfo/zope3-users

Re: [Zope3-Users] How to make catalog working?

[Jim Washington]

> The problem is not that objects have been created before the IntIDs utility.
>
> The IntID utility says that 3 objects are registered, but the catalog indexes 
> are still zero count.
>
>   
This is a bit baffling.  Did you name your IntIDs utility?  The IntIDs
utility works best when unnamed, I think.

-Jim Washington


___
Zope3-users mailing list
Zope3-users@zope.org
http://mail.zope.org/mailman/listinfo/zope3-users

Re: [Zope3-Users] Re: [Zope3-dev] Re: Zope 3 lacks Ajax capability?

[Jim Washington]

Benji York wrote:
> Jim Washington wrote:
>> I have a public example (not open-sourced, sorry - line-of-business app)
>> of jsonserver in action at http://www1.vtdata.org . It's fast and does
>> the async communication with little fuss.
>
> Very cool.
Thanks!  (and apologies to non-US-ians about the US-centric questions :(
) The quiz app still has a few soon-to-be-resolved edge-case buglets,
but we have been using it for high school contests around the state
(yes, through cranky school district firewalls/proxies, etc.), and for
occasional exams around the university.

An upcoming jsonserver release will have its own JSON-RPC javascript
included, based on Roger's JSON-RPC javascript in Tiks.  jsolait, while
still OK to use, will no longer be an external "recommended" dependency,
which I think will make a lot of people happy.

The new semantics of JSON-RPC1.1 will be available.  Keyword parameters
are cool!

Best of all, the new release will be a unified release that will operate
in both Zope 2/Five and Zope 3.  Balazs Ree is the genius behind that
amazing feat.

Stay tuned!

-Jim Washington
___
Zope3-users mailing list
Zope3-users@zope.org
http://mail.zope.org/mailman/listinfo/zope3-users

Re: [Zope3-Users] Re: [Zope3-dev] Re: Zope 3 lacks Ajax capability?

[Jim Washington]

Marco Mariani wrote:
> Jim Washington wrote:From the docs:
>
>> jsonserver is available at http://zif.hill-street.net/jsonserver
>>
> From the docs:
>
>> jsolait from http://jsolait.net is the recommended client-side
>> javascript library.
>>  
>>
>
> I guess I won't have problems with Mochikit instead, but why is it
> "recommended" ?
It is what we provide directions for.  MochiKit was not around when we
started jsonserver.

JSON-RPC is a flavor of JSON with specific semantics, like XML-RPC for XML.

The last time I looked, MochiKit did JSON, but not JSON-RPC.

If you wish to use plain JSON, jsonserver also provides an IJSONWriter
utility.

-Jim Washington

-Jim Washington
___
Zope3-users mailing list
Zope3-users@zope.org
http://mail.zope.org/mailman/listinfo/zope3-users

[Zope3-Users] Re: [Zope3-dev] Re: Zope 3 lacks Ajax capability?

[Jim Washington]

:: moved to zope3-users

Jeff Rush wrote:
> Tarek Ziadé wrote:
>> Jeff Rush wrote:
>>
>> what does your code actually do ?
>
> Provide a chat window at the bottom of a page, in which a student
> interacts with a teaching app and members of his team.  In the upper
> portion of the page, the teaching app alternately presents proficency
> questionaires and lessons.
>
> My idea was to construct a Zope widget component that can be dropped
> over, using widget= in ZCML, a TextArea/Lines Zope widget and
> transform it into an async typing window into a chat space.

You might consider using jsonserver (JSON-RPC) for the ajax-y
communication.  Soon, JSON-RPC (1.1-specification) will have native
support for keyword (named) parameters, which provides a python-ish feel
for the javascript calls.

I have a public example (not open-sourced, sorry - line-of-business app)
of jsonserver in action at http://www1.vtdata.org . It's fast and does
the async communication with little fuss.

jsonserver is available at http://zif.hill-street.net/jsonserver

-Jim Washington
___
Zope3-users mailing list
Zope3-users@zope.org
http://mail.zope.org/mailman/listinfo/zope3-users

Re: [Zope3-Users] How to make catalog working?

[Jim Washington]

Frank Burkhardt wrote:
> Hi,
>
> On Mon, May 15, 2006 at 11:16:09PM +0200, Florian Lindner wrote:
>   
>> Hello,
>> I've added some content objects of interface IFoo to my site.
>> Then I added a catalog to my site (and also a IntID utility). I registered 
>> both. To the catalog I've added a FieldIndex and a TextIndex, set the 
>> interface to IFoo and the fieldname to a field that IFoo has. I registered 
>> both indexes.
>> 
>
> The IntID utility has to be registered *before* all the objects you
> want to find. No object created before you had a registered IntID
> will ever be found.
>
> Have a look at
>
>  http://zope3.mpg.de/suchen ("Das Prinzip")
>
>   
You can get the IntIDs utility to register objects after they are created.

Something like the below will register and catalog a bunch of items:

def catalog_items(self):
intids = zapi.getUtility(IIntIds,[name])
catalog = zapi.getUtility(ICatalog,[name])
for item in some_function_that_returns_the_items():
catalog.index_doc(intids.register(item), item)

The most important part is the intids.register(item) statement.

-Jim Washington

___
Zope3-users mailing list
Zope3-users@zope.org
http://mail.zope.org/mailman/listinfo/zope3-users

Re: [Zope3-Users] More fun with WSGI/zope.paste

[Jim Washington]

Gary Poster wrote:
>
>
> Checking in the code is an assertion of provenance/license: for
> instance, I wouldn't have known about the Plone code, which is
> potentially a problem because of GPL vs. ZPL (see below).
>
Hold-off on checking-in jsmin.  The original author of the packer has
not decided whether allow us a BSD-ish license or ZPL.
> Wanna get commit privileges? :-)  It's the easiest way for you to
> assert the code's status.
>
Can't at the moment.  My note to Benji explains.  Maybe after I see the
new contributor agreement...
>>> - putting them in a namespace?
>> Probably a good idea.  If it was only one... well, but I do seem to have
>> gotten prolific. :)
>
> :-)
>
>> Go for it (on whichever namespace gets decided).  These three projects,
>> I feel a need to reiterate, need zope.paste and Paste.Deploy (or a
>> similar stack), to use with zope3, so deprecating zc.resourcelibrary may
>> not be a good idea until more folks are on-board with the wsgi filters
>> idea.
>
> I think the project is on board with wsgi.  paste is maybe not as
> mainstream in the Zope world yet, so yes, maybe we need to let that
> settle out.  If there are no issues with the paste-based version,
> though, I'd like zc to use it.
>
No further issues.  There is some code from Python Cookbook (Python
License, presumably, and presumably acceptable) but the rest is
substantially mine.
>> gzipper and jsmin really have no particular ties to zope at all,
>> except that I used Zope3 for developing them, and they probably work OK
>> in Zope3 as a consequence. (PS. er, actually, the packer in jsmin came
>> from Plone.)
>
> eek!  GPL can't go in zope.org.  Do you know what the license is to
> that particular component?
>
Of course, I went "eek!" first when Balazs wanted to include the packer
in jsonserver / concatresource when it was GPL.  This will be handled in
a most appropriate and satisfactory manner.
> With the caveats above, sounds great. :-)
>
Good.  You will find that my code has the very liberal Academic Free
License referenced.  If you need different, I will be happy to relicense.

-Jim Washington

___
Zope3-users mailing list
Zope3-users@zope.org
http://mail.zope.org/mailman/listinfo/zope3-users

Re: [Zope3-Users] More fun with WSGI/zope.paste

[Jim Washington]

Oops.  Forgot to cc the list.

Gary Poster wrote:
>
> On May 7, 2006, at 4:47 PM, Jim Washington wrote:
>
>> Following along from Gary's idea that zc.resourcelibrary could be done
>> as WSGI middeware, I am now previewing headincludes, a wsgi middleware
>> filter with an alternative implementation of zc.resourcelibrary.  It
>> usurps a lot of zc.resourcelibrary configuration for compatibility, so
>> they cannot be used at the same time.  But reconfiguring is just a
>> matter of a few files.
>>
>> For more information go to http://zif.hill-street.net/headincludes .
>>
>> The readme is at http://zif.hill-street.net/headincludes/README.txt .
>>
>> development status: "works for me"
>
> :-) awesome!
>
Hi, Gary

Thanks!

> What do you think about some or all of the following:
>
> - moving development of all three of these to zope.org svn?
OK.  I do not have commit privileges on zope.org svn.  But you do, and I
do not mind if you put these packages up there.
> - putting them in a namespace?
Probably a good idea.  If it was only one... well, but I do seem to have
gotten prolific. :)
> - merging zc.resourcelibrary and headincludes?
That would require a commitment to wsgi filters, or maybe some fun with
imports, but also OK.
> - the possibility that we might want to include things not only in the
> head later (some old JS code wanted to be at the end, for instance)
> and so "resourcelibrary", or at least something less specific than
> "headincludes", might be a better name?
>
That could be fun as well.  Naming is not all that important to me.  For
this iteration, I just needed to distinguish headincludes from
zc.resourcelibrary to avoid name clashes.
> More concretely, I suggest three new projects in zope.org:
> z3c.resourcelibrary (deprecating zc.resourcelibrary)
> z3c.gzipper
> z3c.jsmin
>
Go for it (on whichever namespace gets decided).  These three projects,
I feel a need to reiterate, need zope.paste and Paste.Deploy (or a
similar stack), to use with zope3, so deprecating zc.resourcelibrary may
not be a good idea until more folks are on-board with the wsgi filters
idea.  gzipper and jsmin really have no particular ties to zope at all,
except that I used Zope3 for developing them, and they probably work OK
in Zope3 as a consequence. (PS. er, actually, the packer in jsmin came
from Plone.)
> You could also choose "zc"--that could stand for "zope community" as
> much as "zope corporation"--but it might cause confusion.  "z" has
> also been proposed. :-)  The "zope" namespace means "from the Zope
> project", not specifically for zope (see zope.interface, for instance)
> so there's ample precedent for general things going in a "z*"
> namespace...
>
> What do you think?  Getting the code in a publicly-accessible repo,
> ideally svn.zope.org, is my primary desire--everything else is
> peripheral.
>
I agree.
> (I don't have as much use personally for your JSON-server right now,
> btw, but maybe that would be good to put in the repo too?)
>
jsonserver is sitting happily in z3labs, where Balazs Ree and I are
working (slowly) on a unified package for zope2 /five and zope3.  It's a
moving target, and we keep going off on interesting tangents. :)

So, to recap, I have largely positive thoughts about your suggestions.
And please feel free to put whichever of these packages you feel are
appropriate wherever you wish on svn.zope.org.  How's that for flexibility?

Regards,

-Jim Washington

___
Zope3-users mailing list
Zope3-users@zope.org
http://mail.zope.org/mailman/listinfo/zope3-users

[Zope3-Users] More fun with WSGI/zope.paste

[Jim Washington]

Following along from Gary's idea that zc.resourcelibrary could be done
as WSGI middeware, I am now previewing headincludes, a wsgi middleware
filter with an alternative implementation of zc.resourcelibrary.  It
usurps a lot of zc.resourcelibrary configuration for compatibility, so
they cannot be used at the same time.  But reconfiguring is just a
matter of a few files.

For more information go to http://zif.hill-street.net/headincludes .

The readme is at http://zif.hill-street.net/headincludes/README.txt .

development status: "works for me"

-Jim Washington
___
Zope3-users mailing list
Zope3-users@zope.org
http://mail.zope.org/mailman/listinfo/zope3-users

Re: Re : Re : [Zope3-Users] formlib problem

[Jim Washington]

Stéphane Brault wrote:
Hi Jim, 
 the Items object is my object, linked to my items, table which implements the IItems interface I use for my form:
 
 from neteven.interfaces.items import IItems
 from zope.formlib import form
 
 class ItemsForm(form.EditForm):

 form_fields = form.Fields(IItems)
 form_fields = form_fields.omit('dateLastUpdate')
 
 Items is defined this way:
 
 from zope.interface import implements

 from sqlobject import *
 from sqlos import SQLOS
 from neteven.interfaces.items import IItems
 
 class Items(SQLOS):

 implements(IItems)
 
 class sqlmeta:

 table = 'items'
 
 
  When I add this zcml declaration :

 
 everything works fine from the ZMI (I created simple containers to test that).
 
 My JSON server side code is:
 
 def getItemEdit(self, itemId):

 item = Items.get(itemId) #a sqlos function to get an Item instance 
from the table given its Id
 return ItemsForm(item, self.request)()
 
The fact is I don't use any page template right now ( though I might do this later to have a nice GUI).
 
  

Actually, you are using a page template.

Calling ItemsForm() invokes the render() method of 
zope.formlib.form.FormBase, which uses pageform.pt in the zope.formlib 
folder by default.  This is where title_or_id is requested.


You may wish to set a different template for your ItemsForm class.

Have you tried subclassing form.SubPageEditForm instead of form.EditForm?

-Jim Washington


___
Zope3-users mailing list
Zope3-users@zope.org
http://mail.zope.org/mailman/listinfo/zope3-users

Re: Re : [Zope3-Users] formlib problem

[Jim Washington]

ales.py line 696 in evaluate
 => 'return expression(self)'
 ** exceptions.TypeError: ('Could not adapt', , 
)
 C:\Python24\Lib\site-packages\zope\tales\expressions.py line 249 in __call__
 => 'v = var(econtext)'
 ** exceptions.TypeError: ('Could not adapt', , 
)
 C:\Python24\Lib\site-packages\zope\tales\expressions.py line 205 in __call__
 => 'return self._eval(econtext)'
 ** exceptions.TypeError: ('Could not adapt', , 
)
 C:\Python24\Lib\site-packages\zope\tales\expressions.py line 199 in _eval
 => 'return ob()'
 ** exceptions.TypeError: ('Could not adapt', , 
)
 C:\Python24\Lib\site-packages\zope\app\pagetemplate\talesapi.py line 73 in 
title_or_name
 => "return getattr(self, 'title', '') or zapi.name(self.context)"
 ** exceptions.TypeError: ('Could not adapt', , 
)
 C:\Python24\Lib\site-packages\zope\app\traversing\api.py line 149 in getName
 => 'return IPhysicallyLocatable(obj).getName()'
 ** exceptions.TypeError: ('Could not adapt', , 
)
 C:\Python24\Lib\site-packages\zope\interface\interface.py line 682 in __call__
 => 'raise TypeError("Could not adapt", obj, self)'
 ** exceptions.TypeError: ('Could not adapt', , 
)
 127.0.0.1 - - [19/Apr/2006:16:13:32 +0200] "POST /netevenTest/test/neteven.Accounts.1/ HTTP/1.1" 
200 163 "-" "Mozilla/5.0 (Windows; U; Windows NT 5.1; fr-FR; rv:1.7.12) Gecko/20050919 
Firefox/1.0.7"
 
 It seems that retrieving my object from code doesn't give me an address for it, which would cause the problem with IPhysicallyLocatable (my uninformed guess).

 Is there a way to solve the problem, knowing that my object is retrieved 
through sqlos, and thus has no true address ? There might be something else but 
being
 quite new to zope 3 (and enjoying it, thanks to you all guys), I can't figure 
it out.
  

Maybe these questions will help.

What is an Items object? 
Why are you calling a page template that needs title_or_name with an 
Items object as the context?


-Jim Washington
___
Zope3-users mailing list
Zope3-users@zope.org
http://mail.zope.org/mailman/listinfo/zope3-users

Re: [Zope3-Users] url traversal and jsonserver

[Jim Washington]

Pete Taylor wrote:

hi all,
i've been looking at jsonserver recently, and am very impressed. 
however, my question doesn't really involve jsonserver functionality,

so much as it does a bit of url traversal indirection.

i have a simple "dashboard" like approach to a particular user
interface design, where most of the content displayed to the user is
determined by content associated with her/his principal id.  like
principal annotations sorta, only I needed/wanted a little more direct
control.  also, all of the content for a user is held in a
registration utility, in ++etc++site.  i've been using viewlets and
formlib to give the users access to their data without actually having
it in the standard traversal path.  that way no smart user can think
"hmm, my url is some/path/my_user_id, let me ry some/path/johnsid"
etc.  to them my.site/Consumer" is all they need to see, which I like.

what i would like to do is set up jsonrpc:view's for a number of the
different content objects that the user has access to in her/his
content directory.  the trouble, for me, comes in when i need to give
the javascript a ServiceProxy address.

while i can probably figure out how to pass a particular url path to
the javascript through some tales construction, it occurs to me that
the object that jsonrpc view is actually providing a json-rpc view ON
is under ++etc++site.  i can't really pass that to the javascript,
since I don't think ++etc++site is going to let the javascript (with
the current principl's authority) connect.

so what I'm wondering about is how to expose content held in a
traversal path like
"http://my.site/++etc++site/MyRegistrationUtility/ConsumerStuff/some_user_id/stuff";
to a json-rpc ServiceProxy.

is it possible?  is there a way i could get at this more indirectly? 
would principal annotations provide a better means of url exposure for

picking up a serviceproxy?
  
I would leave the jsonrpc view as a view of the content-space object 
(/Consumer).


jsonrpc requests are in many ways just like other http requests.  In 
particular, request.principal is available to the view.


Since the data you want is keyed to the user, you can do a utility 
lookup or an adaptation inside your view code using the principal object 
as a key to get (or set) the data.  Then you do not need to deal with 
troublesome urls at all.


-Jim Washington
___
Zope3-users mailing list
Zope3-users@zope.org
http://mail.zope.org/mailman/listinfo/zope3-users

[Zope3-Users] Announce: httpgz - gzip compression

[Jim Washington]

Announcing httpgz, a Zope 3 package that performs gzip compression of 
most responses.


More information and download is available at 
http://zif.hill-street.net/httpgz . There's not much to it, but 
bandwidth savings of up to 80% are possible.


httpgz requires Zope 3 version 3.2 or greater.

Enjoy!

-Jim Washington


___
Zope3-users mailing list
Zope3-users@zope.org
http://mail.zope.org/mailman/listinfo/zope3-users

Re: [Zope3-Users] Re: Please Guido, pick me, pick me!

[Jim Washington]

Martin Aspeli wrote:

Lennart Regebro <[EMAIL PROTECTED]> writes:

  

But having a name that signifies that there is something a little more than a
new major-version release going on here would go a long way in giving people
something to fix their minds on.
  

Not if that name changes with every release, becuase then people will
see it as a release code name, because that's what it will be.



Absolutely - we seem to be on the same page. :)

The branding needs to be consistent and carried forward, until such time we
refactor the whole thing again and call it Zope 4. :)
  
My university, Virginia Tech, is doing something similar right now, and 
I think it is in part due to recent negative publicity involving 
athletics (I could be wrong). 

The vision includes a new trademarked tagline (Invent the Future) and a 
new logo incorporating the tagline. 
http://www.vtnews.vt.edu/story.php?relyear=2006&itemno=53


We could do something similar:

Zope 3: Excellence in Web Component Design

OK, I'm not a marketer, but something in that vein could help to gain 
community standing.


-Jim Washington
___
Zope3-users mailing list
Zope3-users@zope.org
http://mail.zope.org/mailman/listinfo/zope3-users

Re: [Zope3-Users] Problem: "Could not locate Zope software installation!" - solution

[Jim Washington]

My (local) fix was to make a symlink, lib64->lib, in the instance 
directory.  This also works.

Oops, I meant swhome directory.

-Jim Washington
___
Zope3-users mailing list
Zope3-users@zope.org
http://mail.zope.org/mailman/listinfo/zope3-users

Re: [Zope3-Users] Problem: "Could not locate Zope software installation!" - solution

[Jim Washington]

Martin Kudlvasr wrote:

Ronald L Chichester wrote:


This, of course, begs the question of some 64-bit compatibility issue.
  
That is a possibility, but we do have a 64-bit Athalon running here used 
  in daily development with no problems (running Ubuntu).



the problem is really 64 compatibility issue.
When I look into mkzopeinstance script:


***
for parts in [("src",), ("lib", "python"), ("Lib", "site-packages")]:
d = os.path.join(swhome, *(parts + ("zope", "app", "appsetup")))
if os.path.isdir(d):
d = os.path.join(swhome, *parts)
sys.path.insert(0, d)
break
else:
try:
import zope.app.server
except ImportError:
print >>sys.stderr, "Could not locate Zope software
installation!"
sys.exit(1)
***
this means that first of directories $PREFIX/src, $PREFIX/lib/python,
$PREFIX/Lib/site-packages (suffixed by zope/app/appsetup) is added to
sys.path

BUT !!! when compiled on 64 architecture, zope compiles its files into
$PREFIX/lib64 directory, so the mkzopeinstance cannot import
zope.app.server and fails.

I succesfully used:

***
for parts in [("src",), ("lib", "python"), ("lib64", "python"), ("Lib",
"site-packages")]:
***

3.2.0-final has the same problem

I please anybody competent to fix this.

  

This has been reported to the collector:

http://www.zope.org/Collectors/Zope3-dev/528

My (local) fix was to make a symlink, lib64->lib, in the instance 
directory.  This also works.


-Jim Washington
___
Zope3-users mailing list
Zope3-users@zope.org
http://mail.zope.org/mailman/listinfo/zope3-users

Re: [Zope3-Users] More MySQL

[Jim Washington]

David Johnson wrote:


Thanks to everyone for the MySQL assistance.  We use Debian and I just 
installed the python-mysqlda (seems obvious in retrospect).


 

In any case, after I install the python adapater, I get the following 
when starting zope:


 

ConfigurationError: ('Invalid value for', 'menu', "ImportError: 
Couldn't import add_connection, No module named add_connection")


 

I think there used to be an "add_connection" menu provided by 
zope.app.rdb, and I imagine it had a purpose for ttw development.  It's 
gone as of Zope 3.2 (maybe earlier), so what has worked for me is 
commenting-out or removing the configuration in mysqldbda/configure.zcml:






BTW, psycopgda has a similar configuration entry, and might need the 
same fix.


-Jim Washington
___
Zope3-users mailing list
Zope3-users@zope.org
http://mail.zope.org/mailman/listinfo/zope3-users

[Zope3-Users] Eric3 unit tests and debugger?

[Jim Washington]

I like the Eric3 python IDE. http://www.die-offenbachs.de/detlev/eric3.html

I've been using it for a while, mainly for file organization and its SVN 
and Bicycle Repair Man integration.  It claims support for unit testing 
and debugging, but the setup for those seems to be more oriented toward 
full PyQt applications than packages within another framework.  Has 
anyone tried to get those working for z3 project development? 


-Jim Washington

___
Zope3-users mailing list
Zope3-users@zope.org
http://mail.zope.org/mailman/listinfo/zope3-users

Re: [Zope3-Users] Problem: "Could not locate Zope software installation!"

[Jim Washington]

Ronald L Chichester wrote:

I downloaded ZopeX3-3.0.1 and Zope3.1 (final).  The ./configure | make 
| make install (as root) went without a hitch.  However, when I went 
to make an instance (with either 3.0.1 or 3.1) I got a "Could not 
locate Zope software installation!"


In both cases, I used the default settings (i.e., Zope was made in 
/usr/local/Zopex/


I used /usr/local/ZopeX3x/bin/mkzopeinstance -u manager:secret

... and got the error message.

I've seen this before in the bug track list, but has anyone got a 
workaround?  I'm using Zope on Gentoo Linux (2.6.13-r2 kernel) with 
python 2.3.5 and and GCC 3.4.4.



Just a guess.  Do you have net-zope/zopeinterface installed?  It
provides a package named "zope" in /usr/lib/[system
python]/site-packages.  This might make weird namespace issues.

-Jim Washington

___
Zope3-users mailing list
Zope3-users@zope.org
http://mail.zope.org/mailman/listinfo/zope3-users

Re: [Zope3-Users] pau and zope.manager

[Jim Washington]

I wrote:

I want to use pau, with session (cookie) based authentication.  No 
basic authentication.


The problem is, when the pau is activated, the zope.manager defined in 
zcml seems to be no longer accessible, effectively locking me out of 
the zmi.


What I think is happening is the pau appends a prefix to the principal 
name, so that the principal, instead of being "zope.manager", becomes 
"prefixzope.manager", which has no permissions anywhere.


I think my choices are the following.

1.  make pau always look (last) in principalRegistry and return a 
non-prefixed principal if found and validated
2.  have my authentication plugin look in principalRegistry and assign 
the same roles for the principals found in principalRegistry, but with 
the pau prefix.  This would happen when the plugin is created or on 
demand.
3.  provide methods for my authentication plugin to generate an 
emergency user for one of its valid principals


Or did I miss something in the documentation that gets around this?


Apparently not?  So, I am going to choose door #3.  It should be pretty 
simple.  The main hazard is getting it wrong, which will require some 
amusing spelunking with the debugger to deactivate the utility if there 
is anything important in the ZODB.  On the good side, it will prep me 
for the next project, which I think will require ldap.


-Jim Washington

___
Zope3-users mailing list
Zope3-users@zope.org
http://mail.zope.org/mailman/listinfo/zope3-users

[Zope3-Users] pau and zope.manager

[Jim Washington]

I want to use pau, with session (cookie) based authentication.  No basic 
authentication.


The problem is, when the pau is activated, the zope.manager defined in 
zcml seems to be no longer accessible, effectively locking me out of the 
zmi.


What I think is happening is the pau appends a prefix to the principal 
name, so that the principal, instead of being "zope.manager", becomes 
"prefixzope.manager", which has no permissions anywhere.


I think my choices are the following.

1.  make pau always look (last) in principalRegistry and return a 
non-prefixed principal if found and validated
2.  have my authentication plugin look in principalRegistry and assign 
the same roles for the principals found in principalRegistry, but with 
the pau prefix.  This would happen when the plugin is created or on demand.
3.  provide methods for my authentication plugin to generate an 
emergency user for one of its valid principals


Or did I miss something in the documentation that gets around this?

-Jim Washington
___
Zope3-users mailing list
Zope3-users@zope.org
http://mail.zope.org/mailman/listinfo/zope3-users

Re: [Zope3-Users] Announce: jsonserver-1.0beta3 and jsonserver-1.1alpha

[Jim Washington]

PS.  username "anonymous" and no password are needed for read access to
the svn repository.

-Jim

___
Zope3-users mailing list
Zope3-users@zope.org
http://mail.zope.org/mailman/listinfo/zope3-users

[Zope3-Users] Announce: jsonserver-1.0beta3 and jsonserver-1.1alpha

[Jim Washington]

Announcing jsonserver1.0 beta3 and 1.1alpha

jsonserver provides some JSON-RPC (http://json-rpc.org) support for zope3. 


jsonserver-1.0beta3 is available at http://zif.hill-street.net/jsonserver

or at svn repository

http://brigadoon.hill-street.net/svn/repos/jsonserver/tags/jsonserver-1.0beta3/

The major change for beta3 is improved safety for the minjson.py JSON 
parser.  It now pre-parses and inspects the incoming JSON to make rogue 
client exploits less likely.  They were not likely before, but now it's 
more belt-and-suspenders.  Unlike beta2, beta3 works with python-2.3.


jsonserver trunk (1.1alpha) is keeping up with zope3 trunk.  It now does 
the wsgi thing, too, which I presume to be zope3's behavior for 3.2+.  
IMPORTANT: jsonserver trunk is now likely incompatible with zope3 3.1 or 
lower. 


jsonserver trunk is available at svn repository

http://brigadoon.hill-street.net/svn/repos/jsonserver/trunk

Why use JSON-RPC when there is a perfectly good XML-RPC?  Client 
support.  jsonserver with jsolait works with most gecko browsers 
(Firefox, Mozilla, Netscape 6.1+, etc.), khtml browsers (Safari 1.2+ and 
konqueror 3.4+), recent IEs, and Opera 8.1. And I understand it is very 
easy to write your own client in javascript, should you choose not to 
use jsolait.  JSON-RPC also has a "notify" concept, where the client may 
send a message to the server and not expect a return response.  This is 
handy for methods that do things but do not return things.


Why does jsonserver replace the zope3 HTTP server?  Client support.  We 
need to sniff at the incoming data stream in case the client is Opera 
and the content-type is not set the way we want.  Actually, jsonserver 
uses much the same code as zope3 trunk, and at the moment I don't think 
there is any other way to add another RPC listener. There should not be 
any noticeable difference other than adding support for JSON-RPC.


So this is an AJAX technology? Well, it's more AJAJ,  JSON instead of 
XML for the last letter of the acronym.  Web browser clients use 
XMLHTTPRequest to send and receive data without a complete page refresh. 
You may transport HTML or XML snippets as javascript strings, so there 
is little difference technically.


-Jim Washington
___
Zope3-users mailing list
Zope3-users@zope.org
http://mail.zope.org/mailman/listinfo/zope3-users

[Zope3-Users] Announcement - jsonserver 1.0 beta1, JSON-RPC for Zope3

[Jim Washington]

Announcing:  jsonserver 1.0 beta1.

jsonserver is an implementation of JSON-RPC for Zope 3.  It enhances
Zope 3's http server to listen for content-type of 'text/x-json' and
allows json-rpc methods.  json-rpc methods are used like xml-rpc methods.

jsonserver is licensed ZPL2.1.

This is a beta release.  It does everything I need it to do, but there
may still be some bugs or features I missed that people want.  It works
with current Zope 3 svn trunk.  It may work with earlier versions of
Zope 3. Tests are included, just to be sure.

Since the alpha announcement, there have been many bugfixes, and the
JSON implementation is now a replaceable component.  jsonserver also now
provides feature="jsonrpc", so you can write conditional zcml that takes
effect only if jsonrpc is available (thanks, Roger Ineichen!).

The README.txt file has installation instructions and information about
dependencies.

jsonserver is available at http://zif.hill-street.net/jsonserver

For more about JSON and links about JSON-RPC, see
http://en.wikipedia.org/wiki/JSON .

-Jim Washington

___
Zope3-users mailing list
Zope3-users@zope.org
http://mail.zope.org/mailman/listinfo/zope3-users

Re: [Zope3-Users] Announcement - jsonserver 1.0 alpha, JSON-RPC for Zope3

[Jim Washington]

Stephan Richter wrote:

>Congratulations, though I know its late! :-) There might be some interest in 
>the Z3-ECM community of a JSON server.
>
>  
>
I would hope there would be interest in Z3-ECM.  JSON's almost-python
notation is very fast to read and write, and particularly so with
minjson.py, which is part of the project.  Of course, JSON(-RPC) only
transports text-ish things; images and such need a different transport,
I think.

And I probably should have been a bit more explanatory on a couple of
things in my announcement:

1.  Since it is in my svn repository, and still alpha, it will change
(presumably improve) as I see fit.  If anyone sees anything that needs
improving, let me know; I'm still enthusiastic about the project, so
updates can happen quickly.  In fact, I just did a minor update.  Debug
info now uses the logging facility instead of simply printing on the
console :) .  I'll wait until I hear positive things from anyone needing
unicode (still a nagging worry) with it before I make a beta announcement.

2.  My subversion repository is http only:

svn co http://brigadoon.hill-street.net/svn/repos/jsonserver/trunk 
jsonserver

will work OK.  I did not bother poking another hole in the firewall for
svn://

3.  minjson.py (also updated an hour or so ago) seems to work OK in
pythons 2.3 and 2.4

4.  If there's interest, I could probably put together and include a
simple demonstration application. I am open for suggestions on what that
app might be.

>BTW, I have also deicded that I will tackle the HTTPRequestFactory issue for 
>3.2 by developing some sort of registry based on subscribers that decides 
>based on registered subscribers which request to create. This way all you 
>need to do is to register a new subscriber to hook up a new HTTP extension. 
>Maybe I make it a utility, I don't know. :-)
>  
>
Thanks, Stephan. That sounds appropriate.  It seemed unpythonic to
duplicate the code for all content-type listeners to add one
content-type listener.

Regards,

-Jim Washington
___
Zope3-users mailing list
Zope3-users@zope.org
http://mail.zope.org/mailman/listinfo/zope3-users

[Zope3-Users] Announcement - jsonserver 1.0 alpha, JSON-RPC for Zope3

[Jim Washington]

It had to happen eventually...

Announcing:  jsonserver 1.0 alpha.

The code is at subversion repository:

http://brigadoon.hill-street.net/svn/repos/jsonserver .

Username 'anonymous' with blank password has read-only access.

jsonserver is an implementation of JSON-RPC for Zope 3.  It enhances
Zope 3's http server to listen for content-type of 'text/x-json' and
allows json-rpc methods.  json-rpc methods are used like xml-rpc methods.

jsonserver is licensed ZPL2.1 .

This is an alpha release.  It does everything I need it to do, but there
may be some bugs and features I missed.  It works with current Zope 3
svn trunk.  It may work with earlier versions of Zope 3. Tests are
included, just to be sure.

The README.txt file has installation instructions and information about
dependencies.

For more about JSON and links about JSON-RPC, see
http://en.wikipedia.org/wiki/JSON .

-Jim Washington

___
Zope3-users mailing list
Zope3-users@zope.org
http://mail.zope.org/mailman/listinfo/zope3-users