On Fri, May 1, 2009 at 7:57 AM, Massimiliano della Rovere
massimiliano.dellarov...@gmail.com wrote:
One of the the fields describing the wine, the colour of the wine
(field: wine_colour), is implemented like a Choice field.
Red, white and roseé wines (field: wine_type) have different possible
On Thu, Nov 6, 2008 at 8:27 AM, Dragos Petre [EMAIL PROTECTED] wrote:
In this case, an instance of A11 will only accept A3 objects as
children but not A2 ones as one would expect given the definition of
the parent.
You've stated in I11 that only objects providing I3 should be
accepted; I'm not
On Wed, Oct 15, 2008 at 10:46 AM, Hermann Himmelbauer [EMAIL PROTECTED] wrote:
Ist there anything I can do about this?
Require version 3.4.1. We (ZC) have generally found that nailing down
all versions, while tedious for getting updates, is the way to go to
avoid unexpected failures due to too
On Wed, Aug 20, 2008 at 9:46 AM, Marius Gedminas [EMAIL PROTECTED] wrote:
I heard people complaining a lot about find-links in setup.py files.
Indeed; using those find-links (dependency-links?) leads to
potentially using untrusted package sources unintentionally.
Fortunately, buildout offers a
On Tue, Jul 15, 2008 at 6:56 PM, Benji York [EMAIL PROTECTED] wrote:
When one says that a class *provides* an interface they're saying that
instances of that class *implements* the interface.
Sorry, you got that backwards. Instances provide the interfaces their
class implements.
On the other
On Tue, Jul 15, 2008 at 7:36 PM, Tim Cook [EMAIL PROTECTED] wrote:
And in my Maildir example it implements what is defined in IMaildir but
how does it 'provide' what is in IMaildirFactory? Specifically the
__call__ method.
The __call__ of a class is the default constructor (the standard
On Mon, Jul 7, 2008 at 2:25 PM, Tim Cook [EMAIL PROTECTED] wrote:
Okay. But before I do that. Is my approach to initializing an instance
correct or is the problem the way I handled the keywords?
...
for n,v in kw.items():
setattr(self,n,v)
There are people who like this
On Fri, Jun 27, 2008 at 4:10 AM, Jon Hadley [EMAIL PROTECTED] wrote:
./configure MACOSX_DEPLOYMENT_TARGET=10.5 LDFLAGS=-L/opt/local/lib
OPT=-I/opt/local/include
Egads; I'd completely forgotten about that environment variable; I'll
bet I was looking at a build for a different version of Python.
On Wed, Jun 25, 2008 at 4:57 AM, Jon Hadley [EMAIL PROTECTED] wrote:
*As an aside, the worldcookery.com website seem to have been offline
for the past 2 days. I was thinking of installing the finished
This is working for me.
How do I run a python prompt within my application (so that I can
On Thu, Apr 10, 2008 at 2:59 PM, Martin J. Laubach [EMAIL PROTECTED] wrote:
Perhaps I need a radically different approach?
Sounds like you want PageTemplateFile.
-Fred
--
Fred L. Drake, Jr. fdrake at gmail.com
Chaos is the score upon which reality is written. --Henry Miller
On Thu, Mar 6, 2008 at 3:11 AM, Jürgen Kartnaller [EMAIL PROTECTED] wrote:
You can do this on your instance :
del obj.__dict__['attr1']
Of course, if that's *all* you do to it, you'll need to set the
changed flag yourself:
obj._p_changed = True
-Fred
--
Fred L. Drake, Jr.
On Feb 3, 2008 1:12 PM, Philipp von Weitershausen
[EMAIL PROTECTED] wrote:
As this case points out very well, the locked down index isn't necessary
a guarantee for ending up with a well-defined set of packages. Nailing
version numbers in buildout.cfg (see above) yields much more predictable
On Jan 5, 2008 6:43 AM, Christophe Combelles [EMAIL PROTECTED] wrote:
Since I did not find a Firefox search plugin for PyPI, I quickly made one:
Very cool!
Thanks for sharing.
-Fred
--
Fred L. Drake, Jr.fdrake at gmail.com
Chaos is the score upon which reality is written. --Henry
On Nov 14, 2007 1:40 PM, Christophe Combelles [EMAIL PROTECTED] wrote:
That's my understanding, however that does not really explain why looping over
non-persistent objects in a btree should absolutely raise everything in the
memory.
When using non-persistent objects, the records holding the
On 10/12/07, Raphael Ritz [EMAIL PROTECTED] wrote:
Last time this came up here (which is quite a while ago) this was
not supported. Don't know if that changed.
This isn't currently part of zope.schema. Exactly what the right
thing is isn't clear (at least to me).
Sometimes, when filling in
On 8/24/07, Christophe Combelles [EMAIL PROTECTED] wrote:
My advice would be to use a simple persistent object and store some stuffs in
it
(sql query?), so you can easily traverse to it and to its view.
As long as the object doesn't store any data in the ZODB, there's no
reason to store it
On 7/24/07, Otto van der Meer [EMAIL PROTECTED] wrote:
The idea is that many residents of a house have a single user account
(nature of app).
and that the names of the residents are associated with the account so
they can be used in the application.
I guess it depends on what you mean by user
On 8/2/07, Hermann Himmelbauer [EMAIL PROTECTED] wrote:
mylist = []
@button.buttonAndHandler(_(u'Suchen'), name='suchen')
def handle_suchen(self, action):
self.mylist.append(123)
When I output mylist in my page template, the number 123 is appended to
mylist if I
On 8/2/07, Jeff Shell [EMAIL PROTECTED] wrote:
How common is this problem?
Too common, though improvements have been made recently.
If you're using buildout, it's easy to use a faster index that doesn't
go down so easily. Set the index to Jim Fulton's ppix like this:
[buildout]
index =
On 7/19/07, Alek Kowalczyk [EMAIL PROTECTED] wrote:
Thanks! List should be list, whether it is persistent or not, but it's the
simplest solution...
You probably want to check the results of the edit widgets; they might
produce list values instead of PersistentList values. Widgets are not
On 7/19/07, Leonardo Rochael [EMAIL PROTECTED] wrote:
They seem to do the right thing (I fixed one of the cases when it didn't).
Check zope/app/form/browser/itemswidgets.py:MultiDataHelper._toFieldValue()
...
if hasattr(self.context, '_type'):
_type = self.context._type
On 5/3/07, Maciej Wisniowski [EMAIL PROTECTED] wrote:
AFAIK for textarea html fields you can't set something like
maxlength. You have to use javascript for that.
That's correct; maxlength is for input with type=text or type=password.
-Fred
--
Fred L. Drake, Jr.fdrake at gmail.com
Chaos
On 5/3/07, Darryl Cousins [EMAIL PROTECTED] wrote:
My approach is different to Fred's because I rarely call setUpWidgets in
my form code. I have a bunch of custom widgets in a known place and use
custom_widget. Some widgets are for special fields or needs and some are
as simple as this one. I
On 5/3/07, Hermann Himmelbauer [EMAIL PROTECTED] wrote:
Well, when looking at the template, I roughly have to write this for every
widget:
Another approach is to use a template that uses a view on each widget;
that view can be whatever it needs to be, and let's you have per-field
bits
On 5/2/07, Hermann Himmelbauer [EMAIL PROTECTED] wrote:
Moreover I wonder if it's possible to customize the attributes size
and maxlength of HTML input fields via the zope schema. Schemas offer the
attribute max_length, however this does not affect the rendered HTML - size
seems to be always set
On 4/19/07, Christian Theune [EMAIL PROTECTED] wrote:
Do you consider it a bug? Is it tracked somewhere already?
Not a bug, but missing a useful feature. I may have filed an issue
about it at some point, but really don't remember.
-Fred
--
Fred L. Drake, Jr.fdrake at gmail.com
Chaos
On 4/18/07, Leonardo Rochael [EMAIL PROTECTED] wrote:
I was looking at formlib and I found that If i want to customize widgets per
skin, I just have to register their view to a different type, e.g.:
As you can for all views.
However I found no such way of customizing the render_submit_button
On 4/6/07, Philipp von Weitershausen [EMAIL PROTECTED] wrote:
How is there a ZCML pain? Simply register the adapter for all
IAnnotatable objects. Typically your content objects are annotatable
anyway because you want DublinCore stuff etc.
This sounds like this will result in huge numbers of
On 4/5/07, David Pratt [EMAIL PROTECTED] wrote:
The UUID generator that I have been looking at is with the Chandler
project (currently under Apache license) that could be easily wrapped:
Python 2.5 comes with a uuid module that works just fine with Python
2.4; that comes in handy as well, and
On 2/23/07, Svenn Helge Grindhaug [EMAIL PROTECTED] wrote:
This is what I thought as well, but I got stuck leading me to think I should
implement the information as meta data. I get a forbiddenAttribute error (see
below) after adding a course and trying to enter the newly created folder.
Could
On 2/23/07, Svenn Helge Grindhaug [EMAIL PROTECTED] wrote:
I tried different ways of adding the permissions for IWriteContainer and
IReadContainer without any luck. Can you tell me how to combine those and the
permissions for ICourse schema.
This is where I find it easier to keep the
On 2/23/07, Svenn Helge Grindhaug [EMAIL PROTECTED] wrote:
Are you talking about the same separation as Christophe did, create a
ICourseDetail in addition to ICourse. That do not change the way the
configuration is done, or does it?
It can be used to make describing access easier, and can also
On 1/30/07, Thierry Florac [EMAIL PROTECTED] wrote:
My problem is that the StateLabels are translated correctly in several
templates, but not when used inside this table (I can only see the
status-... translation key in generated output).
You'll need to use a SortableColumn subclass that
On 11/5/06, Stephan Richter [EMAIL PROTECTED] wrote:
This is officially an implementation detail; zope.traversing.api.getName() is
the API call.
Interesting; I'd never known it was anything other than __name__.
These APIs just keep changing...
(Er, this indicates a serious documentation
On 10/29/06, Perry Smith [EMAIL PROTECTED] wrote:
But since emacs seems to be happy and not really do any checking, can
you suggest a way to test whatever I come up with? If not, I'll
browse the net some more to see what I can find.
If you're only concerned about what Emacs does, then you're
On 10/29/06, Perry Smith [EMAIL PROTECTED] wrote:
I bumped into two problems that stumped me. First, it is general practice
to do:
metal:arbitrary_tag define-macro=page
Constructs for this can be written in RELAX NG, but I've not tried
doing that myself though. The Atom specification
On 10/27/06, Michael Haubenwallner [EMAIL PROTECTED] wrote:
what should one suggest for development ?
- using zope3 checkout with its own python interpreter
- or remove site-packages from zope3 sys.path
I use a pristine Python installation that doesn't have anything else
installed into it.
On 10/27/06, Jim Fulton [EMAIL PROTECTED] wrote:
Lets not forget: or what might have been left out by the system packager.
Good point. We've often encountered systems where distutils has been
left out, mistaken for something only needed by developers.
-Fred
--
Fred L. Drake, Jr.fdrake
On 10/22/06, Perry Smith [EMAIL PROTECTED] wrote:
I don't know if they apply to Zope 3. Also, last night, I was reading
about Content Providers and Viewlets and Viewlet Managers which
are TAL statements not listed in Appendix C or the TAL 1.4 Spec.
TAL statements, or expression types? The
On 10/21/06, Perry Smith [EMAIL PROTECTED] wrote:
It is easy to find Zope 2 documentation for these items but I have not seen
how much Zope 3 kept or changed.
For Zope 3 are these the right places to look:
Yes, those are the right place. I've added links to these documents
in the
On 10/18/06, Thierry Florac [EMAIL PROTECTED] wrote:
What I'd like in fact is to be able to display and enter dates in a
format following user's locale.
That doesn't seem to work or is not supported by current implementation
of zc.datetimewidget...
This is at least partly an issue of being
On 10/16/06, Oliver Petznick [EMAIL PROTECTED] wrote:
i'm trying to overwrite the DateWidget so the user can select day/
month/year out of 3 dropdowns. my problem is that zope searches for
'form.field' and not for 'form.field.day' and so on.. so i get the
no input error.. how i can solve this?
On 10/2/06, Christophe Combelles [EMAIL PROTECTED] wrote:
How can I obtain the absolute_url of the context object in a content provider or
in a viewlet ?
Is there some reason the usual context/@@absolute_url doesn't work?
-Fred
--
Fred L. Drake, Jr.fdrake at gmail.com
Every sin is the
On 9/27/06, Darryl Cousins [EMAIL PROTECTED] wrote:
browser:menuItem
menu=zmi_actions title=Add sample gallery
for=zope.app.folder.interfaces.IFolder
action=addSampleGallery.html
permission=zope.ManageSite
/
The install method obviously does a lot more including
On 9/26/06, John Maddison [EMAIL PROTECTED] wrote:
Right, that's basically what I want to run, but I can't seem to figure
out how to have that run at ZODB creation time. Should I be listening
for the event that Fred added (thanks!) and run that in the
subscriber?
There are a couple of ways to
On 9/26/06, Baiju M [EMAIL PROTECTED] wrote:
If this is a _bug fix_ and not a _feature_, can we port it to 3.3 also.
Of cource in 3.3.1 release only.
I think so. I've got to get some sleep now, but feel free to go ahead
and backport it if like.
-Fred
--
Fred L. Drake, Jr.fdrake at
On 9/6/06, Stephan Richter [EMAIL PROTECTED] wrote:
I still use vocabularies. I still don't get sources. Vocabularies have not
been deprecated yet. I think sources have to become much more
developer-friendly before finding wide-spread support.
Odd that you should mention this. ;-) We have a
On 9/1/06, Sreeram Raghav [EMAIL PROTECTED] wrote:
Since is Zope3 is still new and not much documentation exists on
certain topic, i have decided to compile a small, HOWTO on the top of
indexing and searching of PDF files in Zope3. I have been working on
Thanks for sharing!
More of the work
On 8/24/06, Benji York [EMAIL PROTECTED] wrote:
If zc were to become zope community, I'd be *totally* for it (but I
don't speak for ZC here, so that may not be cool with others, like say
Jim). We'd either keep using zc under the new meaning, or switch to
something else.
z3c clearly stands for
On 7/18/06, Darryl Cousins [EMAIL PROTECTED] wrote:
This gets me back to the form, but all fields have been reset. All else
that I have tried has come to nought.
Add 'self.form_reset = False' before the return. This suppresses
resetting the form data.
-Fred
--
Fred L. Drake, Jr.fdrake
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
--
Fred L. Drake, Jr.fdrake at gmail.com
Education is hanging around until you've caught on. -- Robert
[Re: zf as a namespace package]
On 5/12/06, Benji York [EMAIL PROTECTED] wrote:
As I said earlier, that sounds good for things that have been given to
the foundation, but non-foundation code needs another name space (IMO).
Aside from legacy, why would any code being committed to the ZF
On 5/12/06, Cliff Ford [EMAIL PROTECTED] wrote:
This is an example mextURL function that I used to add a folder
containing a page, and then go to the folder. Going to the page is just
an addition to the last line.
This is very confusing. Why are you creating new content in
nextURL()?
On 5/8/06, Benji York [EMAIL PROTECTED] wrote:
I must have been unclear. zf sounds great to me as a name space for
code that's covered by the contributer agreement. z (or something
else) as a default name space for non-foundation code that doesn't
otherwise have a name space.
I'm not sure
On 5/5/06, Florian Lindner [EMAIL PROTECTED] wrote:
Now I want to place some additional text above the form stuff. But I want to
keep all this usefull JavaScript stuff that form lib inserts.
Can you use the extra_info slot?
-Fred
--
Fred L. Drake, Jr.fdrake at gmail.com
Don't let
On 5/4/06, luis [EMAIL PROTECTED] wrote:
I'm having problems writing !-- comments -- in my zcml configuration
files...
for example, if during development I want to comment out some of the
registrations, or just write some note on what an adapter (for example) is
used for, I get a
On 5/4/06, luis [EMAIL PROTECTED] wrote:
yep. that was it.. thanks!
No problem. Just be glad we're not using SGML, where the comment
rules are *much* more subtle. :-)
-Fred
--
Fred L. Drake, Jr.fdrake at gmail.com
Don't let schooling interfere with your education. -- Mark Twain
On 5/4/06, Benji York [EMAIL PROTECTED] wrote:
You /can/ use -- in HTML comments, but the rules for doing so are so
complex that most people don't bother.
That's because HTML is SGML. XHTML is not, and has the same
restrictions as any other XML-based language.
-Fred
--
Fred L. Drake, Jr.
On 4/12/06, Lorenzo Gil Sanchez [EMAIL PROTECTED] wrote:
I'm trying to use formlib and while looking at the source code of this
package for a better understanding I realized that the form.Action class
has two methods with the same name: 'submitted'. You can see it at:
...
Is that ok or is it a
On 4/5/06, David Pratt [EMAIL PROTECTED] wrote:
Hi Fred. Thanks for this. The packaging code for Z3 for Windows is a
good example. I will be adding other components as well as Zope but this
could provide a good reference. Is the packaging code in svn somewhere
where I can look at this. Many
On 3/14/06, David Johnson [EMAIL PROTECTED] wrote:
I am experimenting with formlib, and I'm stuck on one point, how do you set
the values of fields before they are rendered?
...
How do I set the value of field before it is rendered? I've been playing
with things like:
On 2/17/06, Chris Withers [EMAIL PROTECTED] wrote:
Oh well, not too attached to DocBook then...
Though I'm LaTeX-friendly, I'd certainly rather DocBook over
OpenDocument. Not only are there better toold for working with it,
but the markup offered is more interesting for the subject domain.
On 2/17/06, Brett Haydon [EMAIL PROTECTED] wrote:
I am just starting with zope 3.2 and am using the zope.formlib to generate a
simple form. Where and how do I override the presentation aspect of
the input field length in html.
e.g. size=20
div class=widgetinput class=textType
On 2/15/06, Stephan Richter [EMAIL PROTECTED] wrote:
So with that in mind, if you do not know how to do professional desktop
publishing, then LaTeX is a much better options, since it results look really
professional.
We also have lots of cool support for Python documentation in LaTeX,
which
On 2/15/06, Philipp von Weitershausen [EMAIL PROTECTED] wrote:
I've never tried using OpenOffice and I'm not sure whether
its changes would actually leave existing whitespace intact (this would
be crucial for decent diffs).
I've worked a bit with generating OpenDocument documents for use with
On 11/14/05, Roger Ineichen [EMAIL PROTECTED] wrote:
This is a bug, please add a issue to the bugtracker!
The parser engine tries to interprete the div tag.
Though this is most certainly a contentious feature, this is not a bug
in the parser, but a bug in the template itself. The parser is a
On 10/24/05, Chris Withers [EMAIL PROTECTED] wrote:
Well, if we're all in agreement, what's required for it to actually
change? ;-)
Someone with time, motivation, a detailed plan, and the ability to
convince Guido.
Most of the difficulty is in actually removing the NotImplemented name
from
On 9/28/05, Gary Poster [EMAIL PROTECTED] wrote:
A Zope object versioning system is more appropriate for this case.
One take is in the zope trunk now--maybe zope.app.versioning?
zope.app.versioncontrol
-Fred
--
Fred L. Drake, Jr.fdrake at gmail.com
Zope Corporation
On 8/4/05, Alec Munro [EMAIL PROTECTED] wrote:
We want to allow a choice field with an other option, that allows
them to type in an arbitrary value. So far, we have this working up
until the point of validation. At the point of validation, it notices
that the other value is not in the
On 7/26/05, Florian Lindner [EMAIL PROTECTED] wrote:
Mmmhh, I'm wondering why it used to work about 2 weeks ago...
Good question; I'd have been surprised if they did.
Why are these
class declarations necessary? Everything is file-based code, so AFAIK there
should not be a problem with
On 7/25/05, Florian Lindner [EMAIL PROTECTED] wrote:
without change of my codebase my objects which use the partial annotations
adapter are not working anymore. I use it like that:
contentitem.py:
from zope.app.dublincore import annotatableadapter
factory =
On 7/9/05, Santi Camps [EMAIL PROTECTED] wrote:
I'm not able to find any way to internationalize this alert message.
Anybody knows some way to do it ?
I haven't tried this myself, but you should be able to put the message
in an HTML element that is hidden by the CSS, and internationalize
that
On 5/10/05, Florian Lindner [EMAIL PROTECTED] wrote:
how can I automatically create a addform that allows the user to enter the
metadate zope:description and the __name__. Using addform-fields has not
worked.
There are a couple of ways.
1. Use a custom class and template. This isn't so
73 matches
Mail list logo
