Re: Debugging failed tests with pylons globals
On Wed, Jul 9, 2008 at 10:42 PM, Graham Dumpleton [EMAIL PROTECTED] wrote: The only reference I can find to SOP and mod_wsgi together is a prior comment by you: I tried to find the original message earlier too but I couldn't. So maybe I misunderstood it, or maybe the original person was wrong about it. -- Mike Orr [EMAIL PROTECTED] --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups pylons-discuss group. To post to this group, send email to pylons-discuss@googlegroups.com To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/pylons-discuss?hl=en -~--~~~~--~~--~--~---
Re: Debugging failed tests with pylons globals
Ben, do you remember? Wasn't there an environment that couldn't use SOPs and you added self.request et al to the controllers because of it? Or did I dream the whole thing? -- Mike Orr [EMAIL PROTECTED] --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups pylons-discuss group. To post to this group, send email to pylons-discuss@googlegroups.com To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/pylons-discuss?hl=en -~--~~~~--~~--~--~---
Re: Debugging failed tests with pylons globals
On Jul 10, 2008, at 12:27 AM, Mike Orr wrote: Ben, do you remember? Wasn't there an environment that couldn't use SOPs and you added self.request et al to the controllers because of it? Or did I dream the whole thing? Only async environments can't use SOP's, though since eventlet will monkey patch thread-locals, even under those conditions the SOP's work. Unless twisted monkey patches thread locals, they wouldn't work under that async environment. Cheers, Ben smime.p7s Description: S/MIME cryptographic signature
Mercurial version cannot be installed
Hi,
regularly I've been updating my pylons installation from the mercurial
version.
Now I 'm having trouble when issuing sudo python setup.py develop
from the mercurial repository. It tells me about a Paste conflict.
...after some advance in the installaton process occurs this
Running FormEncode-1.0.1/setup.py -q bdist_egg --dist-dir /tmp/
easy_install-Q7IYWW/FormEncode-1.0.1/egg-dist-tmp-5ZA1pw
warning: no files found matching '*.py' under directory '.'
warning: no files found matching '*.html' under directory 'docs'
no previously-included directories found matching '**/.svn'
zip_safe flag not set; analyzing archive contents...
formencode.api: module references __file__
formencode.fields: module references __file__
formencode.util.doctest24: module references __file__
formencode.util.doctest24: module MAY be using inspect.getsourcefile
Adding FormEncode 1.0.1 to easy-install.pth file
Installed /Library/Frameworks/Python.framework/Versions/2.5/lib/
python2.5/site-packages/FormEncode-1.0.1-py2.5.egg
Searching for Paste=1.7.2
Reading http://pypi.python.org/simple/Paste/
Reading http://pythonpaste.org
No local packages or download links found for Paste=1.7.2
error: Could not find suitable distribution for
Requirement.parse('Paste=1.7.2')
Before this I indeed update with easy_install Paste==dev , then
easy_install PasteDeploy==dev and easy_install PasteScript==dev
succesfully. That is I have the most up to date versions of the Paste
pieces.
Any hint?
Domingo Aguilera
--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups
pylons-discuss group.
To post to this group, send email to pylons-discuss@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at
http://groups.google.com/group/pylons-discuss?hl=en
-~--~~~~--~~--~--~---
Re: Mercurial version cannot be installed
I experienced the same problem yesterday, but gave up.
On Thu, Jul 10, 2008 at 8:23 AM, batok [EMAIL PROTECTED] wrote:
Hi,
regularly I've been updating my pylons installation from the mercurial
ve
Now I 'm having trouble when issuing sudo python setup.py develop
from the mercurial repository. It tells me about a Paste conflict.
...after some advance in the installaton process occurs this
Running FormEncode-1.0.1/setup.py -q bdist_egg --dist-dir /tmp/
easy_install-Q7IYWW/FormEncode-1.0.1/egg-dist-tmp-5ZA1pw
warning: no files found matching '*.py' under directory '.'
warning: no files found matching '*.html' under directory 'docs'
no previously-included directories found matching '**/.svn'
zip_safe flag not set; analyzing archive contents...
formencode.api: module references __file__
formencode.fields: module references __file__
formencode.util.doctest24: module references __file__
formencode.util.doctest24: module MAY be using inspect.getsourcefile
Adding FormEncode 1.0.1 to easy-install.pth file
Installed /Library/Frameworks/Python.framework/Versions/2.5/lib/
python2.5/site-packages/FormEncode-1.0.1-py2.5.egg
Searching for Paste=1.7.2
Reading http://pypi.python.org/simple/Paste/
Reading http://pythonpaste.org
No local packages or download links found for Paste=1.7.2
error: Could not find suitable distribution for
Requirement.parse('Paste=1.7.2')
Before this I indeed update with easy_install Paste==dev , then
easy_install PasteDeploy==dev and easy_install PasteScript==dev
succesfully. That is I have the most up to date versions of the Paste
pieces.
Any hint?
Domingo Aguilera
--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups
pylons-discuss group.
To post to this group, send email to pylons-discuss@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at
http://groups.google.com/group/pylons-discuss?hl=en
-~--~~~~--~~--~--~---
Bumping Mako requirement version to 2.0
Hi folks... Would that be a problem for pylons users if we bump the Mako requirement version to = 2.0.0 for Toscawidgets and tw.forms projects ? I intent to start converting all Genshi templates of dbsprockets and probably other tw.x projects to Mako templates... And been able to use `id` instead of context.get(id) is something I can only do with Mako 2.0.x. [1] and [2]. For tw.forms I've used `context.get(id)`, but I can run over all templates and simplify them as well... Is there any objection ? :) Cheers, Roger [1] - http://www.makotemplates.org/trac/changeset/375 [2] - http://www.makotemplates.org/trac/browser/mako/trunk/CHANGES#L82 --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups pylons-discuss group. To post to this group, send email to pylons-discuss@googlegroups.com To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/pylons-discuss?hl=en -~--~~~~--~~--~--~---
Re: Debugging failed tests with pylons globals
So SOPs are here to stay for now but nobody thinks they're ideal. The trouble is, we don't know what would be better. I understand that pylons 1.0 will be breaking backwards compatibility. That might be a good time to look at what might be a better option. If 98% of cases are served by threadlocals, that might be the way to go. It's rarely wise to introduce this kind of complexity for 2% of the use cases. Of course, there's also the possibility that SOPs aren't that complicated and the problem is really that nobody knows how to deal with them. The documentation for SOPs is somewhat lackluster; the API is described in the paste docs, but not the motivation or concepts. In the end, there is no technical reason why a program should not be able to access an SOP from anywhere. What I would like to see is a way of registering a thread to access the SOP that another thread is refering to. This obviously isn't safe, and shouldn't be the default, but for testing it's quite valuable to be able to access SOPs that were registered to a thread that is not your own. Justin --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups pylons-discuss group. To post to this group, send email to pylons-discuss@googlegroups.com To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/pylons-discuss?hl=en -~--~~~~--~~--~--~---
Expired Beaker Values and Memory
Hello, I'm using the pylons global cache object to cache a variety of data. I've noticed that beaker never clears expired values. It will refresh expired values with the passed create function when the are accessed, but expired values are never removed. This makes it hard to use beaker as a general cache as things that aren't touched for some time just build up in memory. I've run into this a number of times, but I'll give one solid example. My application uses facebook to get information on a user. We are allowed to cache this information for up to 24 hours, and for performance and stability reasons, its definitely in our interest to do so. However, using beaker just leads to unbounded memory usage. If a user hasn't logged in in a few days, there's no reason for us to still have his expired data lying around. Of course, this is fairly easy to fix. We can just run a cleanup thread every few minutes to clear out expired values. However, this seems like a bit of an oversight in a piece of software designed for long running processes. Is there anything in beaker that I've missed that addresses this? Should there be? Also, does this apply to sessions? Are old sessions kept around indefinitely? I haven't looked too much at them, but I have to imagine the situation is similar. Thanks, Justin --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups pylons-discuss group. To post to this group, send email to pylons-discuss@googlegroups.com To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/pylons-discuss?hl=en -~--~~~~--~~--~--~---
