I've developed code on the web since the 90s from cgi to mod_perl to
catalyst, openacs and dozens of frameworks inbetween. Roughly 16
months ago I was faced with a somewhat monumental task rewriting 90658
lines of PHP code based on Smarty with a basic framework.
After four months of working with
find / -name mysql_config
in site.cfg
mysql_config = /usr/local/mysql/bin/mysql_config
Or whereever it exists.
In Debian, it is contained within the package libmysqlclient15-dev and
exists in /usr/bin/mysql_config which may not be on your system.
On Sep 2, 12:33 am, travis+ml-pyl...@subspacef
pymemcompat.h:10:20: error: Python.h: No such file or directory
Python.h, from the python headers is missing. There is no python2.6-
dev package in debian that I see and that file doesn't show up on
packages.debian.org. I suspect apt-get source python2.6 would get the
libraries and you could po
I originally posted this in toscawidgets-discuss, but, in looking a
little further, the error occurs in core.py in Pylons, but, I am not
quite sure where to start poking.
>From http://toscawidgets.org/documentation/tw.forms/tutorials/index.html
and
http://toscawidgets.org/documentation/tw.forms
On Sep 27, 3:48 pm, Mike Orr wrote:
> > �...@validate(movie_form, error_handler=index)
modified to:
@validate(form=movie_form, error_handler=index)
as listed on the docs - I missed the form=, however, even with that,
it doesn't change it.
http://turbogears.org/2.0/docs/main/ToscaWidgets/fo
Using the Helpers: (existing)
${h.form(h.url(action='email'), method='get')}
Email Address: ${h.text('email')}
${h.submit('Submit')}
${h.end_form()}
should be:
${h.form('email', method='get')}
Email Address: ${h.text('email')}
${h.submit('Submit','Submit')}
${h.end_form()}
Using it in your con
On Sep 27, 10:45 pm, cd34 wrote:
> Url at the bottom saying 'Formbuild Manual' points
> tohttp://formbuild.org/pylons.html
> which is a 404. Url should probably behttp://formbuild.org/docs/manual.html
that should read:
Url at the bottom saying 'Formbuil
On Sep 27, 11:12 pm, Graham Higgins wrote:
> http://bitbucket.org/gjhiggins/shabti/src/tip/shabti/templates/micros...
>
> (based on Pylons==0.9.7, ToscaWidgets>=0.9.4,
> tw.forms>=0.9.3dev-20090122)
Downgraded toscawidgets to 0.9.4, tw.forms to 0.9.3, same issue with
your code and my code. Wh
On Sep 28, 2:29 pm, cd34 wrote:
> Going to install a new virtualenv -- though, this one was very
> minimally changed, so, I'm not too hopeful.
$ cat ../lib/python2.5/site-packages/easy-install.pth
import sys; sys.__plen = len(sys.path)
./setuptools-0.6c9-py2.5.egg
./Pylons-0.9.
On Sep 28, 7:49 pm, Graham Higgins wrote:
> >> With this, and the code taken from toscawidget's tutorial
>
> which has explicitly:
>
> """
> In myapp/controllers/movie.py, add the following line:
>
> from tw.mods.pylonshf import validate
which is in my installation -- the entire controller with
On Sep 28, 7:39 pm, Graham Higgins wrote:
> I dunno if this helps but there seem to be three required libraries
> missing from that list. Check Pylons' setup.py for the list of
> install_requires and you'll see it includes:
>
> "Routes>=1.10.3", "WebHelpers>=0.6.4", "Beaker>=1.3dev",
> "Paste
replacing:
@validate(form=movie_form, error_handler=index)
with:
@validate(form=movie_form, error_handler='index')
Allows validation to work as expected. Since error_handler as
specified in the decorator is a function, it is rather obvious why it
fails.
file tw/mods/pylonshf.py (0.9.8-py2.5)
On Sep 29, 2:47 am, Graham Higgins wrote:
> Yes I have the same, needs a string. The issue is not with pylonshf.py
> but with the tw doc you were working from:
>
> >http://toscawidgets.org/documentation/tw.forms/tutorials/validate_pyl...
>
> It's wrong.
>
> I was lucky, I picked up tw a while a
The very minimal controller you need to do the ToscaWidgets movie
tutorial is:
from pylons import request, response, session, tmpl_context
from tw.mods.pylonshf import validate
import tw.forms as twf
from cp.lib.base import BaseController, render
movie_form = twf.TableForm('movie_form', actio
Here's a controller that does a chained validator to ensure that the
password and password verification string match along with a user-
supplied FancyValidator on the old password field to make sure it
matches the string 'test'. In production, you would change the
condition check to send the subm
On Oct 1, 5:17 am, Krishnakant wrote:
> > from cp.lib.base import BaseController, render
> I did not see any code specifying that the form will be rendered using
> mako.
My default render function is defined in project/lib/base.py and is
set to mako. In base.py:
from pylons.templating import r
On Oct 2, 7:32 pm, Dan06 wrote:
> rendering the template it was trying to create the 'data' directory.
mako keeps a compiled version of the template in data/templates. If
you use filebased sessions, they are written in data/sessions.
--~--~-~--~~~---~--~~
You rec
is lockd/statd running or an alternative dlm?
NFS v4?
On Oct 2, 7:21 am, Alvaro wrote:
> when doing the setup for an aplication with paster (i.e: paster setup-
> app test.ini) on a directory mounted over NFS the process is stuck
> forever, hanging.
> I believe there is a syscall for locking fil
On Oct 15, 4:43 am, Arun Tomar wrote:
> in my experience nginx and mod_wsgi works best and the performance is also
> better than apache+mod_wsgi. by the way which OS are you using.
You'll really want to benchmark this. There is a reason that apache-
mpm-worker/mod_wsgi is faster than nginx/mod_
On Oct 14, 6:54 am, Pete Heybrock
wrote:
> My team at work are thinking of using Pylons to provide a GUI for an
> existing suite of command line utilites. Although I'm impressed with
> what it offers, I've found the learning curve pretty steep (my
> previous experience being with relatively simpl
On Oct 22, 12:42 pm, "Thomas G. Willis" wrote:
> I coudl be wrong here, but I was under the impression that fastcgi was on
> it's way out of favor and that mod_wsgi was the new hottness. this may be
> why you can't find any information on it.
Here's why you probably want to at least consider n
I have an application where I have a url of:
/widgets
>From that list, you can view the detail of an individual widget with:
/widget/1
This page displays widget_header and widget_detail lines. The form at
the bottom allows the user to add a new detail line.
@validate(widget_form, error_handl
On Nov 11, 5:09 pm, gsandorx wrote:
> PS: I'm trying to manage a Postfix mail server. The commands I want to
> get the output from is "postfix reload", "postmap hash:/etc/postfix/
> virtual"
I don't have an answer for your original question, but,
http://www.postfix.org/MYSQL_README.html and
ht
On Nov 12, 8:32 am, gsandorx wrote:
> However, even I made my question related to Postfix, I have the same
> problem with other app I've developed that need to call external
> programs/scripts and it would be nice to get the output - just like
import os
os.popen("ls -al /var/www").readlines()
If the error is indeed an Internal Server Error, and you are running
IE6 with Smart Errors turned on, since Internal Server Error is not
>540 bytes, IE6 will display its generic error page.
In Tools, Options, Advanced Settings, there is a checkbox to disable
smart error messages. IE8 handles it a
On Mar 13, 10:02 am, gazza wrote:
> 1) How do you remove the pylons page eg
>
> http://AA.BB.CC.DD:5000
public/index.html You can remove it and it'll display your root
controller
may want to look at
config/routing.py
perhaps add:
map.connect('/{action}', controller='root')
you might
you can override:
def setup_model(model, metadata, **p)
and I believe the only other modification I made was to use my own
model with the following arguments:
__table_args__ = {'useexisting':True}
populate_existing=reload
I still maintained the Group and Role tables, but, added one fiel
in development.ini
authkit.users.sqlalchemy_04_driver:UsersFromDatabase
authkit.form.authenticate.user.type =
cp.model.ccmauthkit:UsersFromDatabase
authkit.form.authenticate.user.data = cp.model
in cp.model/__init__.py
def setup_model(model, metadata, **p):
pass
and then I overloaded authki
PC clock on the IE6 machine considerably wrong and expiring the CRSF
cookie before the post request?
--
You received this message because you are subscribed to the Google Groups
"pylons-discuss" group.
To post to this group, send email to pylons-disc...@googlegroups.com.
To unsubscribe from this
I did notice after upgrading to the beta that the sun now smiles at
you in the favicon.ico
--
You received this message because you are subscribed to the Google Groups
"pylons-discuss" group.
To post to this group, send email to pylons-disc...@googlegroups.com.
To unsubscribe from this group, se
I've noticed that since upgrading from 0.9.7 that sometimes rather
than the interactive trackback, I get a white screen with just:
Internal Server Error
This can occur with something as simple as:
${some_undefined_string}
When I upgraded, I used the -t switch, merged the configurations, so,
hav
0.9.7 appeared to catch everything. I've never seen the or text/
plain error message prior to 1.0rc1
And I did -t pylons based on your initial posting, merged in the .bak
files using diff to identify the sections.
paster console log:
Exception happened during processing of request from ('xx.xx
At this point, it appears to be something I've done wrong in the
upgrade as a freshly installed 1.0rc1 displays the trackback properly.
I'll debug this a bit further.
--
You received this message because you are subscribed to the Google Groups
"pylons-discuss" group.
To post to this group, send
Something else is considerably broken with my upgraded installation.
webhelpers pager() on pages that were previously working and have been
untouched are also throwing the 'Internal Server Error'
I think something related to the change in the config handling
probably broke in my installation. I'm
Last night I set up a brand new 1.0rc1 installation to test my
trackback problems. While installing ToscaWidgets, I got an odd
error:
File "/var/www/pylons/lib/python2.5/site-packages/ToscaWidgets-0.9.9-
py2.5.egg/tw/mods/pylonshf.py", line 11, in
from pylons.templating import render
Impor
Right now, the issue appears related to ToscaWidgets. As soon as I
put ToscaWidgets into my test environment, the undefined variable
changes from the mako traceback to the 'Internal Server Error'
Here's the error prior to ToscaWidgets being installed:
Error !
NameError: Undefined
1 asdf
2 ${asd
My error. That explains the difference between the two and I missed
the note regarding ToscaWidgets.
http://toscawidgets.org/trac/tw/ticket/48 is the ticket with Graham
Higgins's proposed patch.
--
You received this message because you are subscribed to the Google Groups
"pylons-discuss" group
My error. I missed the note regarding ToscaWidgets and 0.10rc1.
http://toscawidgets.org/trac/tw/ticket/48 is the ticket with Graham
Higgins's proposed patch.
--
You received this message because you are subscribed to the Google Groups
"pylons-discuss" group.
To post to this group, send email
The scanning machines are looking for an open Socks proxy server that
they can use to send Spam. As long as your apache config is
configured properly, you should be ok. Normally you'll see those when
Apache is answering on 3128, 8000, 8080 and you'll see them less
frequently if you only answer on
Can you paste your .wsgi script?
and the full system path of where you installed the virtualenv?
--
You received this message because you are subscribed to the Google Groups
"pylons-discuss" group.
To post to this group, send email to pylons-disc...@googlegroups.com.
To unsubscribe from this gr
I think that the documentation should be updated. This same question
has been asked a few times on serverfault/stackoverflow recently.
Perhaps a deployment recipes section should be added that includes
multiple installation recipes.
The documentation on Graham's site (while not PEP-8 standard), d
If your model relates the comments to the movie, SQLAlchemy's lazy-
loading would handle this automatically. In your comments template:
% if tmpl_movie.comments:
% for comment in tmpl_context.movie.comments:
{$comment.text}
% endfor
You could wrap your include so that it wouldn't
updated this in
http://groups.google.com/group/toscawidgets-discuss/t/abbf98633fd3c89c
I've taken Graham Higgins's patch, modified it slightly to do the
version math and to handle the 0.10rc1 and 1.0rc1 version identifiers
with a hack. Toscawidgets loads in both 0.10rc1 and 1.0rc1 and does
handl
When writing a pylons app, how much code do you put into your
controller?
One thought I had was to put enough code in to handle basic
operations, but, if the controller required more than 20 or so lines
of code, put that into a class or function so that the controller
itself was reasonably lightwe
On Apr 3, 12:19 pm, Ben Bangert wrote:
> The nice thing about keeping a controller small, is when it has a lot of
> actions. I have a few controllers in one of my projects that are very large.
> Though I consider this less than ideal and when I have some time I'll likely
> go in and refactor to
My feeling is that refactoring code for the purpose of shrinking
something to an arbitrary 10-23 lines of code may also negatively
impact maintainability. Breaking a procedure into parts just for the
sake of making small pieces where you aren't violating DRY principals
just seemed counterproductiv
My feeling is that refactoring code for the purpose of shrinking
something to an arbitrary 10-23 lines of code may also negatively
impact maintainability. Breaking a procedure into parts just for the
sake of making small pieces where you aren't violating DRY principals
just seemed counterproductiv
On Apr 4, 1:48 pm, Jonathan Vanasco wrote:
> I could be wrong, but this looks just like the stuff that i'd put into
> a helper library to keep my controllers a bit cleaner...
>
> (task_id,status) = h.newtask('add')
>
> helpers/tasks.py
> def newtask( method ):
> t= task().bigtask().
While th
On Apr 5, 9:36 am, Tim Black wrote:
> I'm beginning to think that when I write controllers named things like
> getTaskHTML() or setTaskHTML() when I have a model.Task object but it
> doesn't have an attribute named HTML, I should refactor those methods to
> this form: model.Task.HTML where HTML i
controller/whatever.py
contains the functions that do the actual work, populating
tmpl_context.*, deals with the database, etc.
model/whatever.py
contains the schema for forms
model/database.py
contains the schema for the 1170 tables in this project
lib/whatever.py
contains the classes t
On Apr 6, 10:54 am, Jonathan Vanasco wrote:
> Usually I go for application maintainability. Looking at the original
> code , this bigtask() class looked to be a logging functionality that
> is used across multiple controllers and actions. i'm inclined to
> refactor code like that into helper f
Have you looked at the logs to determine whether you are under a D/DOS
attack?
for mod_evasive: http://www.apachelounge.com/viewtopic.php?t=917
contains a pretty decent guide.
However, you'll need to set up some decent firewall rules if you are
under attack once you determine the type of attack.
On Apr 7, 10:55 pm, Matt Feifarek wrote:
> If indeed it is small numbers of IPs, don't forget about tcpwrappers
> (hosts.allow and hosts.deny); probably easier to get going than iptables.
tcpwrappers only works for services started from inetd or that have
explicit support built in (i.e. some ssh
With ToscaWidgets.org, you have the ability to override the default
TableForm template with your own template and I believe you can
override individual elements. In tw.forms, there are templates
lib/python2.5/site-packages/tw.forms-0.9.9-py2.5.egg/tw/forms/templates
$ ls
calendar.html inp
When you start a Pylons project, what packages do you usually start
with in your software stack?
I think I'm one of the few using ToscaWidgets - if I didn't have so
much developed in TG2 before I switched to Pylons, I think I would
have used FormAlchemy. I have debated moving to FA. I gave Sprox
model:
class cp_ticket(DeclarativeBase):
__tablename__ = 'cp_ticket'
ticket_id = Column(mysql.MSBigInteger(20, unsigned = True),
primary_key=True, autoincrement = True)
subject = Column(Unicode(100), default = 'Trouble Report')
ticket_detail = relation('cp_ticket_detail', backref
On Apr 15, 5:02 pm, Ben Bangert wrote:
> Using a name that's not defined in a Mako template should *always* throw an
> error. What is happening when you use an undefined variable, and what are you
> expecting?
0.9.7 returned a mako template error in the browser which allowed you
to see the cont
On Apr 15, 7:36 pm, Ben Bangert wrote:
> Ah, yes, there was a ToscaWidgets bug which masked the proper error handling.
> I mean in my existing projects I've moved to 1.0rc1, I haven't seen this
> issue. Were you saying that this was happening with a non-toscawidgets
> project using 0.10/1.0?
I
Might want to take a look at this thread:
http://groups.google.com/group/pylons-discuss/browse_thread/thread/2e88433546d97410
--
You received this message because you are subscribed to the Google Groups
"pylons-discuss" group.
To post to this group, send email to pylons-disc...@googlegroups.com
On Apr 19, 8:32 am, Martin Stein wrote:
> Thanks for the hint. I've already read about Marcus' way of using
> Django forms (which is really nice), but his method means that you
> have the complete Django project as a dependency. I think some people
> might like a more light-weight approach. That's
On Apr 19, 6:28 pm, Alex wrote:
> Does anyone have any advice on hosting multiple Pylons apps in a low
> memory environment like a VPS. I've written a few apps using Pylons
> and it's great, so I'd like to use it for more. However the stuff I've
> used it for so far has justified running the apps
On Apr 19, 5:59 pm, Ben Bangert wrote:
> I'm using the formlib approach that Mike Bayer wrote about
> here:http://techspot.zzzeek.org/?p=28
>
> I'm really rather liking it. It's substantially faster than toscawidgets or
> the formencode htmlfill method (it still uses Formencode for validation /
On Apr 20, 2:34 am, Ben Bangert wrote:
> On Apr 19, 2010, at 7:27 PM, cd34 wrote:
>
> > I looked at that, but, when you're developing a webapp that has 315
> > forms, that lack of automation seems to fly in the face of why I chose
> > a framework.
>
> Framewo
On Apr 21, 2:51 pm, Jonathan Vanasco wrote:
> also, as i remember from my mod_perl days, apache doesn't like to free
> memory on a "restart". you need to explicitly "stop" then "start".
There is still an issue with the OOM killer that kills an apache child
and leaves the shared memory segments a
http://code.google.com/p/isapi-wsgi/
I just ran across this the other day. The project seems to be active
and lists that it should be production quality. SQLAlchemy supports
obdc on Windows machines fairly seamlessly.
Might take a look at that project. It appears that it should work.
Perhaps s
It has been fixed:
http://groups.google.com/group/mako-discuss/msg/f06184b0ee507c63
I just pulled the latest tip and it works as expected.
--
You received this message because you are subscribed to the Google Groups
"pylons-discuss" group.
To post to this group, send email to pylons-disc...@go
Offhand, since upgrading to the latest tip of Mako, .pager() appears
to have started working for me.
tmpl_context.paginator = paginate.Page(
databases,
page=int(request.params.get('page', 1)),
items_per_page = 40,
)
${tmpl_context.paginator.page
ok, that makes sense. After I couldn't get ToscaWidgets working in
0.10rc1, I did try a fresh 1.0rc1, but, never changed pagination after
switching back to the 0.10rc1 virtualenvironment as it was 'less
broken' than 1.0rc1 with ToscaWidgets.
--
You received this message because you are subscribe
SQLAlchemy (which can be installed with Pylons in the paster create -t
pylons packagename) handles interaction with Oracle.
http://www.sqlalchemy.org/docs/reference/dialects/oracle.html
--
You received this message because you are subscribed to the Google Groups
"pylons-discuss" group.
To post
I recall a similar issue in TurboGears when a print statement was left
in a package that was used for debugging. I would suspect something
similar here. Something is sending output to stdio/stderr prior to
the start of the request.
Try running it in paster and look for any output that is reporte
LogLevel error
or, at worst:
LogLevel debug
in your apache config might produce more results in your error logs.
Your config looks right, but, the error logs are probably set to crit
which is why you're not seeing anything of value in the logs.
Remember you can also specify an ErrorLog directiv
After dealing with a DDOS today on a site (non-pylons, but the issue
is still the same), it became very evident that there are very few
hooks in apache that allow you to drop a connection where it needs to
be dropped. mod_security/mod_evasive don't hook apache early enough
in the request phase to
You might find this thread might solve the issue:
http://groups.google.com/group/pylons-discuss/browse_thread/thread/744a1bea7928a223/
The URL in the error is also incorrect and should probably point to:
http://wiki.pylonshq.com/display/pylonsfaq/Warnings
--
You received this message because you
On Apr 26, 5:03 am, Haron Media wrote:
> Also, it was not necessarily distributed since the nature of attack
> could allow single machine used to attack. What do your logs say, how
> many IPs were involved?
Directly involved from slowloris, 2241 based on firewall counters.
Total DDOS had 77861 or
On Apr 26, 4:19 pm, Haron Media wrote:
> Anyways, so I was wondering how soon before the botnets start deploying
> slowloris. Seems like they already have.
So far, every machine we've checked that was hitting the machine with
slowloris is running linux. Since Windows can't handle more than 130
s
I've been using nginx with uwsgi: http://projects.unbit.it/uwsgi/ as
documented here:
http://tonylandis.com/python/deployment-howt-pylons-nginx-and-uwsgi/
I had problems using using 0.9.5.rc1, but, 0.9.4.4 worked fine.
I would not recommend nginx's mod_wsgi as it has problems with
blocking appli
On Apr 27, 10:00 pm, Terry Schmitt wrote:
> Like Weixi Yen, my preference is the simplicity of using reverse proxy
> with Nginx. I've used this in the past for Java based apps. It's fast,
> easy to configure and I will probably use that as my first choice as I
> venture into the Python world.
If
On Apr 28, 8:55 pm, Haron Media wrote:
> > While paster is great for development, at least in simple benchmarking
> > I was never able to get it to go reliably beyond a few hundred
> > concurrent connections.
>
> Per how many processes / threads / paste instances (and on what hardware)?
X3220, 8g
Are you connecting to a mysql server and processing data from
sqlalchemy on your page?
add
?use_unicode=1&charset=utf8
to your sqlalchemy url:
sqlalchemy.url =
mysql://username:passw...@localhost/databasename?use_unicode=1&charset=utf8
I think that is the condition that generated a similar er
On May 7, 12:53 pm, Mike Orr wrote:
> The CherryPy server also works with Pylons and is supposed to have
> better performance than PasteHTTPServer. There should be an INI
> configuration in the Pylons wiki or list archive.
I'm reasonably happy with apache2-mpm-worker/mod_wsgi or nginx/uwsgi.
I am
You can use daemon mode which won't run into quite that much memory
growth if you're not running multiple instances. Embedded mode has
some different memory requirements.
You might post over on the mod_wsgi list as Graham Dumpleton is very
responsive, and knows the ins and outs and probably will
You can use daemon mode which won't run into quite that much memory
growth if you're not running multiple instances. Embedded mode has
some different memory requirements.
You might post over on the mod_wsgi list as Graham Dumpleton is very
responsive, and knows the ins and outs and probably will
Do you mean protect the actual html that generates the site from being
saved? or protecting static content from being able to be hotlinked?
or preventing scraping your site's content for resyndication
somewhere?
Basically, if it is published on the web, it can be saved. Sites
disable the right m
maybe take a look at .expunge() or .expunge_all() in sqlalchemy. I
believe that frees up the internal resources of the session, but, I'm
not 100% sure.
http://www.sqlalchemy.org/docs/reference/orm/sessions.html#sqlalchemy.orm.session.Session.expunge
--
You received this message because you ar
I believe invalidate() will do what you need:
http://pylonshq.com/docs/en/0.9.7/thirdparty/beaker/#beaker.cache.CacheManager.invalidate
--
You received this message because you are subscribed to the Google Groups
"pylons-discuss" group.
To post to this group, send email to pylons-disc...@google
85 matches
Mail list logo
