[EMAIL PROTECTED] wrote:
That can be done easily in TurboGears too. Though there is a few things
I don't like about subway, just seeing this blog, that is nothing but
style and preference though.
I don't like tying logic into presentation. The choice of Cheetah and
this form mechanism
thanks.
Yes, I like to build the schema in python code as it looks cleaner to
my eye.
I would rather prefer a way to turn a schema into a presentation. Say
if I have field that is string with only 3 valid choice, make that into
a radio button or dropdown.
This way, when I drop the valid choice
I've tried Mozilla, and I've don't find any options saying that the
code generated will be XHTML compliant. Mozylla generate br, hr,
img, ...
Based on suggestions I've received, I've tried nvu. And nvu has an
options allowing you to generate XHTML files that kid can manage.
Fine
BUT
I
this is great. Testing is really key to getting an understanding
of how a program works. This doc is going to help a lot.
Here is some initial feedback -
I really dig TestGears. I was able to get some tests running last
night in just a few minutes, and it was my first day with the code.
Amazing!
Is there a way to generate an image with some text and numbers for a
registeration page in python/TG?
You are probably looking for something like PIL[1] or PythonMagick[2].
Having used neither myself I can't give you much more info than that
I'm afraid.
Regards,
Lee
[1] http://www.pythonware.com/products/pil/
[2] http://starship.python.net/crew/zack/pymagick/
On 23/10/05, vinjvinj [EMAIL
a long while ago, I would use GD library which allows you to
draw(including putting text) to a GIF. There is a python binding too.
However, the author of that python module moves to PIL so I believe PIL
should give you the same functionality and it is wildly available.
I don't think you need
I just committed the code for the TurboGears identity management
support (revision 89). And because this is such new code, I thought
it might be helpful to include a short How To for getting everything
up and running.
The How To is available at:
On 10/23/05, william [EMAIL PROTECTED] wrote:
I've tried Mozilla, and I've don't find any options saying that the
code generated will be XHTML compliant. Mozylla generate br, hr,
img, ...
Based on suggestions I've received, I've tried nvu. And nvu has an
options allowing you to generate
Rui Carmo's PNGCanvas is quite nice for this, no dependencies,
compact and understandable.
He hasn't implement text/numbers (yet), but you can 'copy/blit' those
from another png containing your 'alphabet'
or do your own low level character drawing.
I just tried this in NVU 1.0 for Windows, and I'm not seeing this
problem. On what platform are you running?
--
I'm on Gentoo.
But, finally, I've found the magic options in Preferences called
Don't encode special characters in attributes. That's the
solution.
To summarize, don't use
But. it depends on RuleDispatch so we need to update setup.py
On 10/23/05, Elvelind Grandin [EMAIL PROTECTED] wrote:
Wonderful. I will start using it right away :)
On 10/23/05, Jeff Watkins [EMAIL PROTECTED] wrote:
I just committed the code for the TurboGears identity management
support
Oops. That's true. Technically nothing actually relies on it yet. But Ian's jsonify code makes use of it, and I have jsonify functions for all the model classes.I suppose I should disable this until Kevin decides whether TurboGears should depend on RuleDispatch.Thanks for finding this.On 23 Oct,
No problems.
On 10/23/05, Jeff Watkins [EMAIL PROTECTED] wrote:
Oops. That's true. Technically nothing actually relies on it yet. But Ian's
jsonify code makes use of it, and I have jsonify functions for all the model
classes.
I suppose I should disable this until Kevin decides whether
I commented out the parts that use jsonify, so this should remove the RuleDispatch requirement (Revision 90).On 23 Oct, 2005, at 12:43 pm, Elvelind Grandin wrote:But. it depends on RuleDispatch so we need to update setup.py -- Jeff Watkins http://metrocat.org/ Computers, they're just a fad.
I'm new to tg and I'm not trying to be rude but ...
Does none of this exist already ?
[ tg philosophy seems to be aimed at reuse ]
Why do you have to add it all ?
I plan to run turbogears behind apache. Most of the laundry list below
seems to be there. The one thing I found missing was a
Is this for a captcha?
If so, a simple google search on captcha python will yield results like
the following:
http://aspn.activestate.com/ASPN/Cookbook/Python/Recipe/440588
http://captchas.net/sample/python/
http://freshmeat.net/projects/pycaptcha/
Regards,
Mamading
On Sun, 2005-23-10 at 13:22 -0400, Jeff Watkins wrote:
snip
Of course, I'm *really* new to Python. So what do I know.
I have a couple of books but I am working with someone else who will
write most of the code (he's buried right now, fixing C++ bugs in
another piece of our project).
Why do
Sounds that what you want is not some infra things of TurboGears(what
Jeff is working on), but more like a generic web apps that allows you
to manipulate some database backend, be it RDBMS/LDAP or even flat
file(like .htaccess) or combination of that.
I am not sure what platform you are on but
I just started using TurboGears yesterday, and saw your document. It's
well written, even for such an early draft.
Unfortunately I couldn't get your method of testing to work with
'python setup.py testgears'. When I ran it with the verbose option it
looked as though the only place it was
Oh, and here's the template that it calls. Sorry:
!DOCTYPE html PUBLIC -//W3C//DTD XHTML 1.0 Transitional//EN
http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd;
?python import sitetemplate ?
html xmlns=http://www.w3.org/1999/xhtml;
xmlns:py=http://purl.org/kid/ns#; py:extends=sitetemplate
On 10/23/05, Jeff Watkins [EMAIL PROTECTED] wrote:
Oops. That's true. Technically nothing actually relies on it yet. But Ian's
jsonify code makes use of it, and I have jsonify functions for all the model
classes.
I suppose I should disable this until Kevin decides whether TurboGears
should
On 10/23/05, loki77 [EMAIL PROTECTED] wrote:
Jeez- and as I post that last bit of code I realize where I screwed up.
My template was wrong. Sorry for the noise, and thanks if you took
the time to read through this.
There's nothing like talking it out to solve a problem sometimes :)
By the
Hi!
I was following the discussion about form validation and I think
validation should go into the model.
I'm not sure, but I think FormEncode integrates with SQLObject. Maybe
Ian can provide more info on this.
If that is true, we could think a way of integrating that with cherrypy,
and make
On 10/23/05, Guy Hulbert [EMAIL PROTECTED] wrote:
I'm new to tg and I'm not trying to be rude but ...
Does none of this exist already ?
[ tg philosophy seems to be aimed at reuse ]
Why do you have to add it all ?
I plan to run turbogears behind apache. Most of the laundry list below
On 10/23/05, matthew clark [EMAIL PROTECTED] wrote:
this is great. Testing is really key to getting an understanding of how a
program works. This doc is going to help a lot. Here is some initial
feedback -
I really dig TestGears. I was able to get some tests running last night in
just
TestGears looks for modules named test-* anywhere inside your project's package.
There's a little bit more info about TestGears on its page:
http://www.turbogears.org/testgears/
(Note again that TestGears will be deprecated in favor of Nose sometime soon.)
Kevin
On 10/23/05, loki77 [EMAIL
On 10/22/05, [EMAIL PROTECTED] [EMAIL PROTECTED] wrote:
Sounds interesting but does it belong to controllers.py ? That seems
like presentation to me and say if one day, I want to put a layer of
dojo on top of the web page, will it mean lots of change to the python
code ?
Otherwise, it is a
On Sun, 2005-23-10 at 15:21 -0400, Kevin Dangoor wrote:
On 10/23/05, Guy Hulbert [EMAIL PROTECTED] wrote:
snip
I plan to run turbogears behind apache. Most of the laundry list below
snip
I'm sure you won't be the only person running behind Apache. I want
What if your authentication is
On 10/23/05, Phillip J. Eby [EMAIL PROTECTED] wrote:
Not at the moment; just 'rm -rf' anything you're not using. In the 0.7
or 0.8 timeframe, there will be a 'nest' tool for managing eggs, which
will support listing and deleting eggs, along with other
to-be-determined package management
Awesome. Thanks for the pointer, Kevin.
I'm going to sound like a broken record here soon, but thanks Kevin :)
I thought I remembered something about the variable provider, but
couldn't remember if it was a cherrypy or turbogears thing.
On Sun, 2005-23-10 at 15:50 -0400, Kevin Dangoor wrote:
On 10/22/05, [EMAIL PROTECTED] [EMAIL PROTECTED] wrote:
found the reason.
I am using svk to mirror svn repo. That don't have .svn in the
checkout.
But svk is much nicer to use than svn :-)
I'm glad you found that. I never
Jeff Watkins wrote:
I just committed the code for the TurboGears identity management
support...
Looks good!
Thanks for all your hard work on this.
Some of the things I still have to add:
Can I suggest that you change the model to use a PackageHub instead of an
AutoConnectHub to
On 23 Oct, 2005, at 4:47 pm, Robert Leftwich wrote:
Can I suggest that you change the model to use a PackageHub instead
of an AutoConnectHub to allow the identity data to be sourced from
a different db if required?
Excellent idea. I need to look at PackageHub a little closer, but it
Can be found at http://www.awaretek.com/python/index.html
I've just switched to working from an svn checkout as described in 'Developing
with eggs' from http://www.turbogears.org/community/contributing.html. Although
the latest TG code is used when running apps it doesn't appear to use the code
from the thirdparty directory of the checkout, e.g. the
Is it just me, or are the podcasts listed there only a small part of
the whole podcast?
Lee
On 23/10/05, [EMAIL PROTECTED] [EMAIL PROTECTED] wrote:
Can be found at http://www.awaretek.com/python/index.html
Hmmm, should be the whole podcasts...
Maybe try http://www.awaretek.com/index.xml
or worst case go to http://www.podcastalley.com/podcast_details.php?
pod_id=5786 and click on the podcast title you want...
But I am perturbed. I get the whole podcasts when I click on the
It's odd. I think it may have something to do with a combination of
Firefox, Quicktime and possibly the host of the MP3. I just
subscribed in iTunes (searching in the apple podcast directory for
python411 brought it straight up) and the podcasts are downloading in
full.
On 23/10/05, Ron
OK. Sorry for the trouble. As I say, every once in a while, something
strange happens with the podcast feeds causing incomplete downloads. I
do not know what it is yet.
Ron Stephens
No problem. I've managed to get them now and I'm uploading them to my
phone to listen to at work tomorrow. :)
On 23/10/05, [EMAIL PROTECTED] [EMAIL PROTECTED] wrote:
OK. Sorry for the trouble. As I say, every once in a while, something
strange happens with the podcast feeds causing incomplete
On 10/23/05, william [EMAIL PROTECTED] wrote:
I just tried this in NVU 1.0 for Windows, and I'm not seeing this
problem. On what platform are you running?
--
I'm on Gentoo.
But, finally, I've found the magic options in Preferences called
Don't encode special characters in
prmiarily mirroring. I don't need to remember all the urls of modules
used in TurboGears. And I do have some local changes to some of them.
Also becaue I have some linux kernel patches collection too, I need
the star merge of svk. It means I can seperate my version with upstream
and merge
On 10/23/05, Ian Bicking [EMAIL PROTECTED] wrote:
And then extract these into a schema. But while you might want this
validation on a form, I'm of the opinion that it's better to be more lax
when doing programmatic updates, especially because there might not be a
feedback cycle to correct
At 07:07 PM 10/23/2005 -0500, Ian Bicking wrote:
Kevin Dangoor wrote:
I thought that was why Ian made paster?
Yes, there would be overlap, and it would be good to consider how best
to deal with that. If the paster create code goes into nest, I have no
problem with deprecating its presence in
I have this data model for an amateur hockey league. they want to track
the stats by year as well. I thought about making the entries by player
ID and year, but I am unsure how to make that happen.
Any help will be appreciated.
Phillip J. Eby wrote:
At 07:07 PM 10/23/2005 -0500, Ian Bicking wrote:
Kevin Dangoor wrote:
I thought that was why Ian made paster?
Yes, there would be overlap, and it would be good to consider how best
to deal with that. If the paster create code goes into nest, I have no
problem
On 10/23/05, Jeff Watkins [EMAIL PROTECTED] wrote:
Excellent idea. I need to look at PackageHub a little closer, but it
seems like the right thing to do.
Luckily, you don't have to look at it a little closer... just replace
AutoConnectHub with PackageHub and you're golden :)
On that note,
I've just updated the contributing page to have a section about
dealing with dependencies. The short of it is that CherryPy does not
use setuptools and therefore doesn't generate an egg. (The other
projects all use setuptools.)
I *believe* that, from the thirdparty directory, you can run
On 10/23/05, Jeff Watkins [EMAIL PROTECTED] wrote:
I just committed the code for the TurboGears identity management
support (revision 89). And because this is such new code, I thought
it might be helpful to include a short How To for getting everything
up and running.
The How To is
On 23 Oct, 2005, at 11:05 pm, Tim Lesher wrote:
This looks really nice, but I'm having a spot of difficulty with the
how-to, running against an otherwise-happy svn installation (rev 91).
Adding the @identity.require( group=admin ) decorator to any
exposed method (secured, index, what have
Kevin Dangoor wrote:
On 10/23/05, Ian Bicking [EMAIL PROTECTED] wrote:
And then extract these into a schema. But while you might want this
validation on a form, I'm of the opinion that it's better to be more lax
when doing programmatic updates, especially because there might not be a
feedback
On 10/23/05, Jeff Watkins [EMAIL PROTECTED] wrote:
What it's choking on is, strangely, the py:extends attribute of the
html element:
-
html xmlns=http://www.w3.org/1999/xhtml;
xmlns:py=http://purl.org/kid/ns#;
py:extends='master.kid'
-
I'll ask the obvious
Robin Munn wrote:
As for primary keys made of multiple columns, SQLObject doesn't
support that, and I believe it never will, due to a deliberate design
decision. See the note at the bottom of
http://sqlobject.org/SQLObject.html#many-to-many-relationships for a
brief discussion as to why.
55 matches
Mail list logo
