[TurboGears] Re: dynamic widget.TableForm
Kevin Dangoor wrote: On 1/4/06, JP Farias [EMAIL PROTECTED] wrote: Is there a way to pass parameters to the callback? I just wanted to pass the **kw the controller method receives to the callback to create the form. No, the callback can't take parameters. However, cherrypy.request should have everything you need. Kevin -- Kevin Dangoor Author of the Zesty News RSS newsreader email: [EMAIL PROTECTED] company: http://www.BlazingThings.com blog: http://www.BlueSkyOnMars.com As a side note, it looks like if you pass in a callable to the decorator, the callable gets passed in your controller object. This is from controllers.py: 104 def validate(form=None, validators=None): 105 def decorator(func): 106 def newfunc(self, *args, **kw): 107 processed = getattr(cherrypy.request, form_processed, False) 108 if not processed and form: 109 if callable(form) and not hasattr(form,'declarative_count'): 110 inputform = form(self) 111 else: 112 inputform = form So, I guess if you needed something that didn't come in on cherrypy.request, you could hang it off of your controller before the form is created. I think it's probably safer to try to get it passed in on cherrypy.request, though. - jmj
[TurboGears] Re: kid problem in turbogear
Enumerate may be your friend here:
?python pages = ['one','two','three','two'] ?
div py:strip=True py:for=i,pagename in enumerate(pages)
a href=${std.url('/%s' % pagename)} py:content=pagenameContent/a
span class=separator py:if=len(pages) i|/span
/div
Now, I can't quite remember, but I think that will work there and
it's that doesn't work. Due to the nature of that loop you could
proably do py:if=not len(pages) == i instead and it would have the
same effect.
This was all written in gmail, so it may or may not work out of the box :)
Lee
On 1/4/06, Ivo Looser [EMAIL PROTECTED] wrote:
Well if you know it's a list or tuple:
?python pages = ['one','two','three'] ?
div py:strip=True py:for=pagename in pages
a href=${std.url('/%s' % pagename)}
py:content=pagenameContent/a·
span class=separator py:if=pagename != pages[-1]|/span
/div
If it's an iterator, I'm not sure.
Thanks Liza it works.
This may not be the best general example. What if pages equaled ['one',
'two', 'three', 'two']?
-- David
its a unique key in the Database. Duplicates arent possible.
May i come to this when i got the solution how i can change unicodes
right to 8859-1 in Database and in the Page. :-)
Best Regards Ivo
[TurboGears] Re: How to avoid importing SQLObjectNotFound
I'm curious as to why you don't want to import SQLObjectNotFound. With SQLObjectNotFound in the except statement it's perfectly clear as to when you want the exception to be handled and when you don't. LookupError may cover other things in your try block and could cause bugs down the road. On 1/5/06, Stephen Thorne [EMAIL PROTECTED] wrote: I've been seeing this in various places around turbogears and example code, and I thought I'd just make a comment about it: except SQLObjectNotFound: Using sqlobject.SQLObjectNotFound isn't actually necessery. SQLObjectNotFound extends LookupError, which is a builtin, so you can just use try: o = foo.byName(bar) except LookupError: raise cherrpy.HTTPRedirect, '/' should work. :) -- Regards, Stephen Thorne Development Engineer Scanned by the NetBox from NetBox Blue (http://netboxblue.com/)
[TurboGears] Re: how to recover from aborted transaction
Note also that TurboGears 0.9 makes the hub.begin(), commit() and end() automatic. Kevin On 1/5/06, Olivier Favre-Simon [EMAIL PROTECTED] wrote: Unless I miss something, this much more simple code should work: n = 'a subject' hub.begin() try: rs = Subject.byName(n) print Found: %s % rs except SQLObjectNotFound: print %s Not Found! Creating... % rs rs = Subject(name=n) hub.commit() hub.end() of course you must have alternateID=True in your model, i.e. something like name = StringCol(alternateID=True, length=200) Olivier. Soni Bergraj wrote: Hello, I try to get a Subject from the database, if this fails i want to create a new one. But how ever i arrange hub.begin(), hub.end() and hub.commit() i always get an AssertionError: This transaction has already gone through ROLLBACK; begin another transaction For the except statement (object does not exist, create a new one) hub.begin() try: subject = Subject.byName(name) hub.commit hub.end() except SQLObjectNotFound: hub.end() hub.begin() subject = Subject(name = name) hub.commit() hub.end() The hub is of type PackageHub, I didn't found documentation for it (deprecated?). Thanks for help, -- Kevin Dangoor Author of the Zesty News RSS newsreader email: [EMAIL PROTECTED] company: http://www.BlazingThings.com blog: http://www.BlueSkyOnMars.com
[TurboGears] Re: Connections with existing SQLObject classes
Hi Randall, You don't *have* to use one of those two hubs unless you want TurboGears' database features (TG config file for the connection URI, automatic connection per thread, automatic transactions in 0.9). Setting _connection on the class to the hub is a fine way to set the hub. You might also try module.__connection__ = hub, which is what TG does in quickstart projects. Kevin On 1/5/06, Randall [EMAIL PROTECTED] wrote: I've got many existing SQLObject classes that I would like to use in TurboGears. They are designed with no connection specified so that they automatically pick up sqlobject.sqlhub. I'm new to TG, so please take into consideration my ignorance. I looked at AutoConnectHub and PackageHub. It looks like I have to use one of these two with my SQLObject classes to get them to work. I've managed to get this to work like so: In Model.py hub = PackageHub(mypackage) MySQLObjectClass._connection = hub Is this a good way to set the connection? Randall -- Kevin Dangoor Author of the Zesty News RSS newsreader email: [EMAIL PROTECTED] company: http://www.BlazingThings.com blog: http://www.BlueSkyOnMars.com
[TurboGears] Re: Announcing Scaffolding - A CMS for Turbogears
Karl Guertin wrote: On 1/4/06, Gábor Farkas [EMAIL PROTECTED] wrote: an interesting discussion about this to trigger or not to trigger issue was spawned based in a blog entry on David Heinemeier Hansson's (ruby on rails...) blog: http://www.loudthinking.com/arc/000516.html You'll notice in that article he qualifies the statement saying that he's talking (only!) about application databases. For most of the stuff I do at work (and apparently for Jorge as well), the database is the lingua franca of the system, i.e. they're integration databases and not app databases. This means that I have to spindle/fold/mutilate SQLObject to fit with the DB relational model but it also means that I want the triggers to fire because that kicks off the rest of the system. What I took away from this article is the sense that while David is a brilliant programmer with an uncanny sense of where to ride the line on many ideas, he's not really a database guy. In fact, I'd say he'd probably happily agree, judging from the tone of the article. I think his view, while interesting in a very abstract we're not talking about real databases with real, important data sense has some merit. However, I think the notion of an application database is a fallacy, unless you are just talking about the database for a blog or some other highly specific, non-enterprise type of application. The minute the data has interest outside the original application, the minute you'll find people wanting to access it by other means and about one minute later his entire fallacy falls apart. Take this with the well-known adage that data outlives applications and David's argument is a recipe for disaster. It's well-known that language to a large extent limits what you can express. I think that extends to other things as well, and I think David's argument reveals more about his MySQL-centric past than anything else. Not only that, but his position is inconsistent with his position on other aspects of the Rails stack. He argues that it's okay to have code in templates as long as it's presented as a domain-specific language, but that it's not okay to have code in databases, that they should rather be dumb storage. He makes a distinction between relational (database) logic and business logic (as do I) but fails to make the same distinction between data logic (relations) and business logic. At the end of the day, even if David's arguments had merit, I'd still be more inclined to trust my relational logic to the database developers and their years of experience in this domain *and* their millions of users who provide an ample test-bed than to one guy who's used one database for a few years, and who, up until a couple of years ago, thought PHP was a good language. Sorry to rant, but he closed off comments on that blog and it's been eating at me ;-) Regards, Cliff
[TurboGears] Re: SQLObject Help
The doc you mention says it's possible. You can use the keyword arguments orderBy to create ORDER BY in the select statements: orderBy takes a string, which should be the database name of the column, or a column in the form Person.q.firstName. You can use -colname to specify descending order, or call MyClass.select().reversed(). Anyway, I don't know how to make ordering by 2 columns, one asc and one desc. Cheers Jarek Zgoda
[TurboGears] TurboStan 0.7
Hot on the tails of 0.5, because 0.6 was boring, TurboStan 0.7 is
released. I probably wouldn't have mentioned it except that there are a
couple important bugfixes and a new feature that everyone (well, me) was
screaming for: template inheritance.
From the Changelog (since 0.5):
- Added std variables and functions (i.e. things like std.tg_js and
std.url are now available)
- Dynamically filter variables passed to Kid for reserved names (this
should be removed when Kid is fixed)
- Added config option (stan.prettyoutput = True) to turn on pretty
html output if utidylib is installed - useful for debugging output
- Changed package name to turbostan from stansupport. You can now
directly import (if you really need to) via turbostan.stansupport
which looks a bit better to my eye than stansupport.stansupport ;-)
- Added inheritance similar to Kid and Django templates
The basic syntax of inheritance is this:
index.stan:
-
html [
body [
div [ slot ('menu') ],
div [ slot ('content') ]
]
]
page.stan:
-
inherits (template = 'index.stan') [
replace (slot = 'content') [
p [ 'this replaces the slot in index.stan' ]
],
replace (slot = 'menu') [
ul (class_ = 'menu') [ li (class_='menuitem')[ m ] for m in
['menu1', 'menu2', 'menu3'] ]
]
]
You may not think it's cool, but I sure did. I also thought it was
damn lucky I got it to work at all, so send me your bug reports.
You can get TurboStan via the Cheese shop (easy_install TurboStan) or
from http://www.develix.com/downloads/TurboStan/TurboStan-0.7.tgz
Regards,
Cliff
[TurboGears] Re: Connections with existing SQLObject classes
Kevin, Is ticket #77 (the SQLObject classes) must be defined in model.py? http://trac.turbogears.org/turbogears/ticket/77 I started my project in a similar situation. I imported my SQLObjects into model.py and it did not work. I ended up, pasting all of my objects into model.py. This gets ugly as your model gets complicated (many SQLObjects). This is not a .9 issue. It would be nice for 1.0. If it is not possible to look for SQLObjects imported into model.py, maybe a model package could be used under the main project dir, and all SQLObjects in that package could be viewed by tg-admin. tg-admin sql features are great, and worth putting all SQLObject's in model.py. Thanks again for your efforts in TurboGears. It is a pleasure to develop in this framework. Happy Coding, Mike
[TurboGears] Re: TurboStan 0.7
Thanks Cliff, That is a great feature. I am stuck in meetings this morning, but look forward to trying it out this afternoon. Mike
[TurboGears] Re: TurboStan 0.7
What python modules we need to install TurboStan? Nevow (zope ???) turbogears
[TurboGears] Re: Announcing Scaffolding - A CMS for Turbogears
On 1/5/06, Cliff Wells [EMAIL PROTECTED] wrote: What I took away from this article is the sense that while David is a brilliant programmer with an uncanny sense of where to ride the line on many ideas, he's not really a database guy. In fact, I'd say he'd probably happily agree, judging from the tone of the article. I think his view, while interesting in a very abstract we're not talking about real databases with real, important data sense has some merit. However, I think the notion of an application database is a fallacy, unless you are just talking about the database for a blog or some other highly specific, non-enterprise type of application. Actually, I do think that both styles (application database, integration database) have merit for certain apps. Here's where application databases make some amount of sense: 1) newfangled, Web 2.0 applications and web services. Things like Backpack (surprise, surprise!), where the application serves a distinct purpose and lives out on the net. Data I/O is all done via web requests that go through the application code, which ensures the integrity of the data (to whatever extent it has been programmed to do so). 2) small office applications. I did a lot of work on billing systems for doctors offices. Those systems are (or, at least, were) quite self-contained and the people in the office tended to use those apps for everything and not be computer literate enough to do much outside of there beyond word processing. (That's a generalization, certainly.) Now, these two points are what you mean by highly specific, non-enterprise type of application, but I wanted to break it out more and emphasize that there's *a lot* of this, and #1 is still a growing category. The larger the organization, the more likely it is to move into the integration database' category. A big organization *could* decide to expose the database entirely via web services and use that layer to maintain the database, or it could opt to move more into stored procedures, triggers and the like to make the database engine do the work. So, even within big organizations, there's still a choice to be made that could enable the application database perspective. Kevin
[TurboGears] Re: SQLObject Help
2006/1/5, zgoda [EMAIL PROTECTED]: The doc you mention says it's possible. You can use the keyword arguments orderBy to create ORDER BY in the select statements: orderBy takes a string, which should be the database name of the column, or a column in the form Person.q.firstName. You can use -colname to specify descending order, or call MyClass.select().reversed(). Anyway, I don't know how to make ordering by 2 columns, one asc and one desc. I guess the docs are not complete. You can pass multiple columnnames in a list: results = Employee.select(orderBy=['-title', 'age']) -- Ksenia
[TurboGears] What is sqlobject-history meant for?
Hello, tg-admin quickstart creates a directory 'sqlobject-history' and I'm really puzzled what it's meant for? Didn't found something useful in the archives. Thanks, -- Soni Bergraj
[TurboGears] Re: Connections with existing SQLObject classes
On 1/5/06, Michael Schneider [EMAIL PROTECTED] wrote: I started my project in a similar situation. I imported my SQLObjects into model.py and it did not work. I ended up, pasting all of my objects into model.py. This gets ugly as your model gets complicated (many SQLObjects). I agree. This will get fixed in some fashion, I'm just not sure what the best way is. (At the moment, you can edit Project.egg-info/sqlobject.txt to put the appropriate list of modules there.) Kevin
[TurboGears] sqlwidgets
I think sqlwidgets is a great idea and I'm determined not to duplicate data and logic in my forms. Referring to formmaker.column_params, couldn't you use column.validator as the validator by default? Also, could there be a use for these attributes? foreignKey: for creating a list or search to populate the element. notNone: validation. Should be done by column.validator, right? tags: Not sure what those are. varchar: If False, set a maxlength attribute in the element. I think that SQLObject should and does to a good degree facilitate detailed descriptions and rules for its data in a way that is reusable and TurboGears should take full advantage. A simple but cool example: An email column is defined in an SQLObject class as notNone. Its FormEncode validator checks to see the address is valid. TurboGears automatically generates the column complete with server side and javascript validation to ensure the value is not null and a valid email address. Randall
[TurboGears] Re: Announcing Scaffolding - A CMS for Turbogears
On 1/5/06, Cliff Wells [EMAIL PROTECTED] wrote: It's well-known that language to a large extent limits what you can express. I'm not entirely sure about 'express', but the definitely affect how you think. This is the Sapir-Whorf Hypothesis. http://lambda-the-ultimate.org/classic/message928.html
[TurboGears] Re: fastdata DataController does not coerce types
True. Where should the limit be defined? I think the select list could be created using convention. In my experience, an id/name combo is the most often used. With SQLObject, even if you don't have a name field, it is easy to add a method for it. I'm implementing this now. Will check back. Randall
[TurboGears] Problems installing pysqlite on Linux
Hello All,
This is a specific question on linux, sqlite, pysqlite.
Sorry to ask such a specific question to this group, but my email to
the pysqlite is awaiting moderator approval.
Any help would be very apreciated.
Thanks
Mike
--
I develop on windows and will deploy on linux. Python was not
installed on my server,
(and I don't have root access).
suse 9.2 64 bit linux.
sqlite works ok, but I can't seem to get pysqlite to find mysqlite3.so.
All of turbogears runs, but I can't get sqlite going
here is the info
---
I haven't been on UNIX for a couple of years so it is a little rusty.
It looks like I am having problems finding libsqlite3.so when running.
Where I am at:
-compile sqlite
-test sqlite (ok)
-compile pysqlite (ok)
-install pysqlite (no errors)
test, fail, ImportError:
/misc/pdtools/scm/.scmikes/python2.4/dist/lnx64/lib/python2.4/site-
packages/pysqlite2/_sqlite.so: undefined symbol: sqlite3_bind_blob
Is there a place in the python distribution where I can copy sqlite3.so
?
Thanks
Mike
-- install info -.
Python 2,4,2 is installed with:
prefix= /misc/pdtools/scm/.scmikes/python2.4/dist/
exec-prefix= /misc/pdtools/scm/.scmikes/python2.4/dist/lnx64
sqlite3 (latest release) was just installed with the same prefix and
exec prefix
PATH=$PATH:{prefix}/bin;{exec-prefix}/lnx64
sqlite3 works from the command line !!!
Compile/build/install step
I forced the copied sqlite3.h into the python header file area (bad, I
know)
(scmikes)cila6s08:pysqlite-2.0.5python setup.py install
running install
running build
running build_py
creating build
creating build/lib.linux-x86_64-2.4
creating build/lib.linux-x86_64-2.4/pysqlite2
copying lib/__init__.py - build/lib.linux-x86_64-2.4/pysqlite2
copying lib/dbapi2.py - build/lib.linux-x86_64-2.4/pysqlite2
creating build/lib.linux-x86_64-2.4/pysqlite2/test
copying lib/test/__init__.py -
build/lib.linux-x86_64-2.4/pysqlite2/test
copying lib/test/dbapi.py - build/lib.linux-x86_64-2.4/pysqlite2/test
copying lib/test/factory.py -
build/lib.linux-x86_64-2.4/pysqlite2/test
copying lib/test/transactions.py -
build/lib.linux-x86_64-2.4/pysqlite2/test
copying lib/test/types.py - build/lib.linux-x86_64-2.4/pysqlite2/test
copying lib/test/userfunctions.py -
build/lib.linux-x86_64-2.4/pysqlite2/test
running build_ext
building 'pysqlite2._sqlite' extension
creating build/temp.linux-x86_64-2.4
creating build/temp.linux-x86_64-2.4/src
gcc -pthread -fno-strict-aliasing -DNDEBUG -g -O3 -Wall
-Wstrict-prototypes -fPI
C -DPYSQLITE_VERSION=2.0.5 -DPY_MAJOR_VERSION=2 -DPY_MINOR_VERSION=4
-I/misc/p
dtools/scm/.scmikes/python2.4/dist/include/python2.4
-I/misc/pdtools/scm/.scmike
s/python2.4/dist/lnx64/include/python2.4 -c src/module.c -o
build/temp.linux-x86
_64-2.4/src/module.o
In file included from src/connection.h:32,
from src/module.c:24:
/misc/pdtools/scm/.scmikes/python2.4/dist/include/python2.4/sqlite3.h:1254:
warn
ing: function declaration isn't a prototype
gcc -pthread -fno-strict-aliasing -DNDEBUG -g -O3 -Wall
-Wstrict-prototypes -fPI
C -DPYSQLITE_VERSION=2.0.5 -DPY_MAJOR_VERSION=2 -DPY_MINOR_VERSION=4
-I/misc/p
dtools/scm/.scmikes/python2.4/dist/include/python2.4
-I/misc/pdtools/scm/.scmike
s/python2.4/dist/lnx64/include/python2.4 -c src/connection.c -o
build/temp.linux
-x86_64-2.4/src/connection.o
In file included from src/connection.h:32,
from src/connection.c:25:
/misc/pdtools/scm/.scmikes/python2.4/dist/include/python2.4/sqlite3.h:1254:
warn
ing: function declaration isn't a prototype
gcc -pthread -fno-strict-aliasing -DNDEBUG -g -O3 -Wall
-Wstrict-prototypes -fPI
C -DPYSQLITE_VERSION=2.0.5 -DPY_MAJOR_VERSION=2 -DPY_MINOR_VERSION=4
-I/misc/p
dtools/scm/.scmikes/python2.4/dist/include/python2.4
-I/misc/pdtools/scm/.scmike
s/python2.4/dist/lnx64/include/python2.4 -c src/cursor.c -o
build/temp.linux-x86
_64-2.4/src/cursor.o
In file included from src/connection.h:32,
from src/cursor.h:28,
from src/cursor.c:24:
/misc/pdtools/scm/.scmikes/python2.4/dist/include/python2.4/sqlite3.h:1254:
warn
ing: function declaration isn't a prototype
gcc -pthread -fno-strict-aliasing -DNDEBUG -g -O3 -Wall
-Wstrict-prototypes -fPI
C -DPYSQLITE_VERSION=2.0.5 -DPY_MAJOR_VERSION=2 -DPY_MINOR_VERSION=4
-I/misc/p
dtools/scm/.scmikes/python2.4/dist/include/python2.4
-I/misc/pdtools/scm/.scmike
s/python2.4/dist/lnx64/include/python2.4 -c src/cache.c -o
build/temp.linux-x86_
64-2.4/src/cache.o
gcc -pthread -fno-strict-aliasing -DNDEBUG -g -O3 -Wall
-Wstrict-prototypes -fPI
C -DPYSQLITE_VERSION=2.0.5 -DPY_MAJOR_VERSION=2
[TurboGears] Static Site question
Maybe there should be some marker to put on your CherryPy pages, for an option to publish all marked pages, with associated JavaScript, images, CSS, etc, as a static site. Perhaps publish as a tar file or FTP to a site? I haven't seen this feature (am I blind). Seems to me it oughta be added, so one could develop/edit simple sites in TuboGears, and then deploy a static site. Then again, maybe this is Creeping Featurism and not a Missing Feature? Sam Then again, maybe I overlook something.
[TurboGears] Re: Attention: people running TurboGears 0.9 out of SVN
This seemed to work for me (ubuntu/breezy)... cd /path/to/svn/turbogears svn up . python ez_setup.py setuptools python ez_setup.py http://www.turbogears.org/download/eggs/PasteScript-0.4dev_r4192-py2.4.egg python ez_setup.py http://www.turbogears.org/download/eggs/SQLObject-0.7.1dev_r1457-py2.4.egg python setup.py develop The links being from the Download packages section at... http://www.turbogears.org/download/index.html --markc
[TurboGears] Re: fastdata DataController does not coerce types
Check this out. I added this to formmaker.py It creates a select box if your SQLObject class has an attribute named 'option_name'. option_name should be configurable. Really cool I think! # Added to handle foreign keys. def column_widget_fk_col(column): parms = column_parms(column) fk_class_name = column.foreignKey fk_class = sqlobject.classregistry.findClass(fk_class_name) fk_class_data = fk_class.select() if hasattr(fk_class, 'option_name'): options = [[rset.id, rset.option_name] for rset in fk_class_data] return widgets.SelectField(validator=None, options=options, **parms) else: return widgets.TextField(validator=None, **parms) column_widget_fk_col = column_widget.when( isinstance(column, col.SOKeyCol))(column_widget_fk_col)
[TurboGears] Re: sqlwidgets
Forgive me for cross posting, but this really belongs under sqlwidgets. I think sqlwidgets rocks! Check this out. I added this to formmaker.py It creates a select box if your SQLObject class has an attribute named 'option_name'. option_name should be configurable. Really cool I think! # Do this at the top. import sqlobject # Added to handle foreign keys. def column_widget_fk_col(column): parms = column_parms(column) fk_class_name = column.foreignKey fk_class = sqlobject.classregistry.findClass(fk_class_name) fk_class_data = fk_class.select() if hasattr(fk_class, 'option_name'): options = [[rset.id, rset.option_name] for rset in fk_class_data] return widgets.SelectField(validator=None, options=options, **parms) else: return widgets.TextField(validator=None, **parms) column_widget_fk_col = column_widget.when( isinstance(column, col.SOKeyCol))(column_widget_fk_col)
[TurboGears] Re: What is sqlobject-history meant for?
I think that's used for database backup and restore (upgrade?).On 1/5/06, Soni Bergraj [EMAIL PROTECTED] wrote: Hello,tg-admin quickstart creates a directory 'sqlobject-history' and I'm really puzzled what it's meant for?Didn't found something useful in the archives.Thanks,--Soni Bergraj-- I never gave anybody hell. I just told the truth and the Republicans thought it was hell. - Harry S. Truman.Best,Jeff
[TurboGears] Re: What is sqlobject-history meant for?
On 1/5/06, Jeff Grimmett [EMAIL PROTECTED] wrote: I think that's used for database backup and restore (upgrade?). My guess is that eventually it'll be a db migration feature like rails has.
[TurboGears] Re: tg-admin sql create failing with MySQL 3.23
I just double checked this, and I think you must be using an AutoConnectHub rather than a PackageHub. This behavior is actually implemented in PackageHub. Kevin On 1/4/06, Mike Kent [EMAIL PROTECTED] wrote: I decided to try TurboGears 0.9 (svn) on another linux box at work, which has MySQL 3.23. After installing 0.9 by following the instructions on the 'Contributing to TurboGears' page, and defining my model (all of which works fine on another machine with MySQL 4), I run 'tg-admin sql create' and get: Exception exceptions.TypeError: 'exceptions must be classes, instances, or strings (deprecated), not NoneType' in bound method Transaction.__del__ of sqlobject.dbconnection.Transaction object at 0x40a4ddcc ignored OK, I know this version of MySQL doesn't have transaction. Checking 'dev.cfg', I see a new comment: # if you are using a database or table type without transactions # (MySQL default, for example), you should turn off transactions # by prepending notrans_ on the uri # sqlobject.dburi=notrans_mysql://username:[EMAIL PROTECTED]:port/databasename So I do so. Then when I run 'tg-admin sql create' I get a long exception dump that ends in: AssertionError: No SQLObject driver exists for notrans_mysql (only sqlite, sapdb, postgresql, firebird, maxdb, sybase, interbase, psycopg, mysql, postgres) Examining the code leads me to believe that while TurboGears understand 'notrans_', SQLObject does not. Is this a known issue? -- Kevin Dangoor Author of the Zesty News RSS newsreader email: [EMAIL PROTECTED] company: http://www.BlazingThings.com blog: http://www.BlueSkyOnMars.com
[TurboGears] Re: What is sqlobject-history meant for?
usage: tg-admin sql record [options] Record historical information about the database status Record state of table definitions. The state of each table is written out to a separate file in a directory, and that directory forms a version. A table is also added to you datebase (sqlobject_db_version) that reflects the version the database is currently at. Use the upgrade command to sync databases with code. On 1/5/06, Soni Bergraj [EMAIL PROTECTED] wrote: Hello, tg-admin quickstart creates a directory 'sqlobject-history' and I'm really puzzled what it's meant for? Didn't found something useful in the archives. Thanks, -- Soni Bergraj -- Kevin Dangoor Author of the Zesty News RSS newsreader email: [EMAIL PROTECTED] company: http://www.BlazingThings.com blog: http://www.BlueSkyOnMars.com
[TurboGears] Re: Proposal: Visitor Tracking
Kevin Dangoor wrote: On 1/3/06, Michele Cella [EMAIL PROTECTED] wrote: Can this be a starting point? http://www.cherrypy.org/changeset/908 Yeah, that's good stuff. Unfortunately, that is likely CherryPy 3.0. It would be nice if we can implement a stopgap before then. Good news (if I'm not mistake), it seems as this (or sort of) has found its way into CP trunk (2.2): http://www.cherrypy.org/ticket/145 Ciao Michele
[TurboGears] Re: Problems installing pysqlite on Linux
Michael, I had a similar problem (the exact same message from the import, actually), and I eventually solved it by fixing my LD_LIBRARY_PATH to include 'libsqlite3.so.0' I think you're having the same problem. When I do ldd on _sqlite.so, I get this: [phedex] /usr/local/lib/python2.4/site-packages/pysqlite2 ldd _sqlite.so libsqlite3.so.0 = /opt/cms/lcg/external/sqlite/3.0.8/slc3_ia32_gcc323/lib/libsqlite3.so.0 (0x00ede000) libpthread.so.0 = /lib/tls/libpthread.so.0 (0x002e2000) libc.so.6 = /lib/tls/libc.so.6 (0x001a8000) libgcc_s.so.1 = /lib/libgcc_s.so.1 (0x0030) /lib/ld-linux.so.2 = /lib/ld-linux.so.2 (0x00878000) Notice how that is linking to libsqlite3.so.0, but yours is not. That's why I think your LD_LIBRARY_PATH is incorrect.
[TurboGears] Re: tg-admin sql create failing with MySQL 3.23
On 1/5/06, Mike Kent [EMAIL PROTECTED] wrote: Now, it's true that this model.py, and my entire project, was originally created for TG 0.8, and moved over to a new project using TG 0.9. Could this be a factor? I just quickstarted a project from scratch (r457) and I'm still seeing the issue. I haven't looked much at the db code, so I didn't trace it fully. It looks like it should be pulled out but somehow it's not. Steps: 1. tg-admin quickstart 2. configure dburi from another project using 'mysql://' 3. copy a sqlobject from that class into quickstarted project 4. tg-admin shell 5. Object.get(1) -- is ok 6. prepend 'notrans_' onto dburi 7. tg-admin shell 8. Object.get(1) -- stack trace as listed above
[TurboGears] 1000+ Users Subscribed
Some time between Tuesday and today we passed the 1000 users mark on the mailing list! Congratulations to all involved, especially Kevin The Danger Dangoor! -- [EMAIL PROTECTED]
[TurboGears] escaping text for kid templates...
I am seeing a bug in a TurboGears application that I have in production where people are entering text in textile format. I am saving this text in the database, and on the way out textile encoding it, and passing it into a kid template, which then includes the string in the template using the XML() function. The bug I am seeing is when people use certain characters in their text (such as and , for example) that cause invalid XML. Typically, I would just run my string through cgi.escape() before I pass it into textile, but this will use named entity definitions, which I don't believe XML supports. Is there an equivalent to cgi.escape out there that will generate XML- friendly entity definitions, or a better solution to my problem? I am considering just doing a string replacement on the two offending characters for now, but this seems a bit of a hack, and doesn't handle other characters that could potentially cause problems... Thanks. -- Jonathan LaCour http://cleverdevil.org
[TurboGears] Re: 1000+ Users Subscribed
On 1/5/06, Jared Kuolt [EMAIL PROTECTED] wrote: Some time between Tuesday and today we passed the 1000 users mark on the mailing list! Congratulations to all involved, especially Kevin The Danger Dangoor! Boy, I sure hope The Danger that doesn't stick. I haven't been planning to enter pro wrestling, but with a name like that I might just have to. For a bit more on the 1,000 member milestone: http://www.blueskyonmars.com/2006/01/05/1000-members/ Thank you to all of the members of this community who have helped it grow and flourish! Kevin
[TurboGears] Re: tg-admin sql create failing with MySQL 3.23
On 1/5/06, Mike Kent [EMAIL PROTECTED] wrote: From the top of my model.py: from sqlobject import * from turbogears.database import PackageHub hub = PackageHub(books) __connection__ = hub Now, it's true that this model.py, and my entire project, was originally created for TG 0.8, and moved over to a new project using TG 0.9. Could this be a factor? Thankfully, no. This is standard Python... so if you've got a PackageHub, you've got a PackageHub :) I'll have to look a bit further then. For me, the uri that went into the AutoConnectHub that's running behind the scenes had the notrans_ removed when I was playing at the Python prompt. Kevin -- Kevin Dangoor Author of the Zesty News RSS newsreader email: [EMAIL PROTECTED] company: http://www.BlazingThings.com blog: http://www.BlueSkyOnMars.com
[TurboGears] Re: Proposal: Visitor Tracking
On 1/5/06, Michele Cella [EMAIL PROTECTED] wrote: Good news (if I'm not mistake), it seems as this (or sort of) has found its way into CP trunk (2.2): http://www.cherrypy.org/ticket/145 Gadzooks. All hail the mighty Fumanchu! This puts me in a bit of a quandry as far as CP 2.2 is concerned. IMHO, this is a hugely important feature for a variety of reasons. I think I'll work on clearing the TurboGears 0.9 queue in Trac and then decide whether CP 2.2 arrives in TurboGears 0.9 or 1.0. Kevin
[TurboGears] Re: escaping text for kid templates...
Hi Jonathan, You might want to try cgi.escape. There was a change made to Kid that gives it (some|all) of the common HTML entities. Kevin On 1/5/06, Jonathan LaCour [EMAIL PROTECTED] wrote: I am seeing a bug in a TurboGears application that I have in production where people are entering text in textile format. I am saving this text in the database, and on the way out textile encoding it, and passing it into a kid template, which then includes the string in the template using the XML() function. The bug I am seeing is when people use certain characters in their text (such as and , for example) that cause invalid XML. Typically, I would just run my string through cgi.escape() before I pass it into textile, but this will use named entity definitions, which I don't believe XML supports. Is there an equivalent to cgi.escape out there that will generate XML- friendly entity definitions, or a better solution to my problem? I am considering just doing a string replacement on the two offending characters for now, but this seems a bit of a hack, and doesn't handle other characters that could potentially cause problems... Thanks. -- Jonathan LaCour http://cleverdevil.org -- Kevin Dangoor Author of the Zesty News RSS newsreader email: [EMAIL PROTECTED] company: http://www.BlazingThings.com blog: http://www.BlueSkyOnMars.com
[TurboGears] Re: Proposal: Visitor Tracking
Kevin Dangoor wrote: Gadzooks. All hail the mighty Fumanchu! :-) This puts me in a bit of a quandry as far as CP 2.2 is concerned. IMHO, this is a hugely important feature for a variety of reasons. I think I'll work on clearing the TurboGears 0.9 queue in Trac and then decide whether CP 2.2 arrives in TurboGears 0.9 or 1.0. I'm not an expert but I'm under the impression (after reading all these blog posts regarding python and RoR) that Paste Deploy+WSGI can really become a winning factor for python (and TG) future. But when will CP 2.2 arrive? Their release schedule is time or feature based? Ciao Michele
[TurboGears] Re: Catwalk hates me
hi, i got the same empty page comming from catwalk request solved with: tg.allow_json = True in dev.cfg and prod.cfg andre
[TurboGears] Re: Proposal: Visitor Tracking
I would also consider the flexibility and re-usability of middleware to be a big assist too since it will allow a variety of talented developers using multiple frameworks ways to consolidate efforts. Some examples of current middleware in Paste: - Eval Exception (that awesome AJAX-based interactive Python traceback explorer) - HTTPExceptions (a standard 404/500/etc HTTP catcher) - Auth functions (Secure Cookies, OpenID, CAS, Digest, Basic) Obviously the middleware just takes part of one side of most the problems, but it does push re-usable portions into a place where we can all actually re-use them. Being able to attach small WSGI apps, plug-and-play style, also is rather powerful. If you take a look at the end of the middleware file I use: http://pylons.groovie.org/svn/trunk/pylons/templates/paster_template/+package+/config/middleware.py_tmpl With the knowledge that the Cascade app tries each app, and moves on if it gets a 404, its pretty obvious how you could extend your app to support multiple static doc roots (I hope). A basic amount of working knowledge immediately becomes quite powerful. Several parts of TurboGears would work great as smaller WSGI apps used by default on a new TG project. It's also be easy to toggle them on/off. Ie, turn on the Toolbox during dev mode, but off during production. Or even package Catwalk as a small Paste compatible app, and anyone working on a SQLObject-based project can toss up a nice Catwalk explorer just by pointing a conf file to their models file... The possibilities are powerful, extensible, and quickly approaching viability. The middleware file I use could be used almost directly with TurboGears by replacing the pylons.make_app thing, with the command needed to create a CherryPy WSGI app. This isn't pie-in-the-sky stuff, its very close at hand, especially with the CherryPy patch. Given the inevitability that different frameworks will appeal to different people, being able to unite on hard problems wherever possible is a big win in my book.
[TurboGears] Re: Dynamic setting of config options (e.g., database URI)
This is the very first time I have ever messed with python code and only then with help from #python. I know there are probably a zillion righteous ways to do it but this actually works, for me. Aim; to provide a dynamic default database value to TurboGears so that no config file(s) have to be edited to get a simple example working. Using this method, or something similar, means that a *simple* tutorial can be distributed and unzipped and immediately used without being forced to edit dev.cfg. . in dev.cfg comment out all sqlobject.dburi values . mv your_sqlite.db to default.sqlite . add to *-start.py - from os.path import * . add to *-start.py - tg_path = dirname(realpath(__file__)) . add to /your_tg_install_path/turbogears/database.py line line ~10 - import __main__ as main . change /your_tg_install_path/turbogears/database.py line line ~142 from - dburi = cherrypy.config.get(sqlobject.dburi, None) to - dburi = cherrypy.config.get(sqlobject.dburi, sqlite://%s/default.sqlite % main.tg_path) . start your *-start.py server Bingo! Hopefully somebody who knows how to will add something like this properly. --mark(what's a global between friends)c
[TurboGears] Re: Made a filter similar to variablesdecode()
Sorry to double post, but a cleaned up and tested version is here, in the cherrypy wiki: http://www.cherrypy.org/wiki/NestedListFilter There is also an example of a FormEncode schema.. very easy to set them up now that this filter does a lot of grunt work :).
[TurboGears] Re: Stripping 'submit' argument from validated form input inconsistent
Kevin Dangoor wrote: On 1/4/06, Keir Mierle [EMAIL PROTECTED] wrote: The devcast says that the submit argument is removed from the form parameters when passed via the inputform; hence there is no 'submit' parameter to signup_confirm(). This works great, but only when there is no validation errors. When there is errors, somehow the 'submit' parameter is sent the second time (i.e. after showing the validation errors, the user puts in valid data, clicks submit again). This causes a 'unexpected keyword argument submit' error. I suspect this is not the intended behaviour. I looked through the validation code briefly but did not see an obvious answer. Sounds like a bug to me. Can you submit a ticket? http://trac.turbogears.org Regarding sumbit arguments being stripped during the form validation, if we change the SubmitButton widget to not use the name attribute (who need this?) there is no need to special case it for removing. Ciao Michele
[TurboGears] Re: escaping text for kid templates...
Thanks Kevin, this worked like a champ once I upgraded to TurboGears 0.8.7 (it didn't seem to work before). Thanks for letting me know about this, I seemed to miss the change during my holiday vacation! -- Jonathan On Jan 5, 2006, at 1:39 PM, Kevin Dangoor wrote: Hi Jonathan, You might want to try cgi.escape. There was a change made to Kid that gives it (some|all) of the common HTML entities. Kevin On 1/5/06, Jonathan LaCour [EMAIL PROTECTED] wrote: I am seeing a bug in a TurboGears application that I have in production where people are entering text in textile format. I am saving this text in the database, and on the way out textile encoding it, and passing it into a kid template, which then includes the string in the template using the XML() function. The bug I am seeing is when people use certain characters in their text (such as and , for example) that cause invalid XML. Typically, I would just run my string through cgi.escape() before I pass it into textile, but this will use named entity definitions, which I don't believe XML supports. Is there an equivalent to cgi.escape out there that will generate XML- friendly entity definitions, or a better solution to my problem? I am considering just doing a string replacement on the two offending characters for now, but this seems a bit of a hack, and doesn't handle other characters that could potentially cause problems... Thanks. -- Jonathan LaCour http://cleverdevil.org -- Kevin Dangoor Author of the Zesty News RSS newsreader email: [EMAIL PROTECTED] company: http://www.BlazingThings.com blog: http://www.BlueSkyOnMars.com -- Jonathan LaCour http://cleverdevil.org
[TurboGears] nl2br for Kid?
Is there a simple solution for newline-to-br (nl2br) in Kid? Simple explantation, I need something that will escape all HTML entities then add a br/ at every newline. Thanks in advance, Jared Kuolt -- [EMAIL PROTECTED]
[TurboGears] Re: nl2br for Kid?
Something like this, I guess:
span py:for=line in line.splitlines()${line}br//span
On Jan 5, 2006, at 12:58 PM, Jared Kuolt wrote:
Is there a simple solution for newline-to-br (nl2br) in Kid?
Simple explantation, I need something that will escape all HTML
entities then add a br/ at every newline.
Thanks in advance,
Jared Kuolt
--
[EMAIL PROTECTED]
[TurboGears] Re: nl2br for Kid?
Works for me, thanks!
On 1/5/06, Bob Ippolito [EMAIL PROTECTED] wrote:
Something like this, I guess:
span py:for=line in line.splitlines()${line}br//span
On Jan 5, 2006, at 12:58 PM, Jared Kuolt wrote:
Is there a simple solution for newline-to-br (nl2br) in Kid?
Simple explantation, I need something that will escape all HTML
entities then add a br/ at every newline.
Thanks in advance,
Jared Kuolt
--
[EMAIL PROTECTED]
--
[EMAIL PROTECTED]
[TurboGears] Re: TurboGears new site look - let's get to it!
Just going to throw this on the table quickly for some feedback http://koorb.co.uk/static/images/tglogos/layouts/blue-light-2.png
[TurboGears] Re: TurboGears new site look - let's get to it!
Hi, Short comments: - The download box should left align with the 'Learn fast column - Green for the download is ok (kind of a convention this days), but orange will be better for the other links so they could play up to to the golden gear. - What about a blue bar at the bottom repeating the top menu (and place for copyright notices and other legaleze) - Search box - Join mail list box -(?) a rss icon for subscribing to planet TG (?) Great work! Cheers Ronald On Jan 5, 2006, at 11:34 PM, Richard (koorb) wrote: Just going to throw this on the table quickly for some feedback http://koorb.co.uk/static/images/tglogos/layouts/blue-light-2.png Ronald Jaramillo mail: ronald AT checkandshare DOT com blog: http://www.checkandshare.com/blog
[TurboGears] Re: TurboGears new site look - let's get to it!
I like most of it, except that green candy-looking Download Now button makes the whole amalgamation remind me of KDE. On 1/5/06, Richard (koorb) [EMAIL PROTECTED] wrote: Just going to throw this on the table quickly for some feedback http://koorb.co.uk/static/images/tglogos/layouts/blue-light-2.png -- [EMAIL PROTECTED]
[TurboGears] Re: TurboGears new site look - let's get to it!
Ronald Jaramillo wrote: Hi, Short comments: - The download box should left align with the 'Learn fast column - Green for the download is ok (kind of a convention this days), but orange will be better for the other links so they could play up to to the golden gear. - What about a blue bar at the bottom repeating the top menu (and place for copyright notices and other legaleze) - Search box - Join mail list box -(?) a rss icon for subscribing to planet TG (?) Agreed, I really like the 1) 2) 3) slogan but I think we can find better words for Run and Visit (Quickstart is ok)... by the way I don't know what we can use! :P Really great work! Ciao Michele
[TurboGears] Re: TurboGears new site look - let's get to it!
On 1/5/06, Richard (koorb) [EMAIL PROTECTED] wrote: Just going to throw this on the table quickly for some feedback http://koorb.co.uk/static/images/tglogos/layouts/blue-light-2.png Excellent grid and hiearchy. The following are my nits, I don't expect them to be mocked up, but please incorporate them when you're doing the actual design: Please center the layout, it feels unbalanced to me when flush left. I venture that anybody visiting this site will have at least 1024 screens. I know that 800 is the current standard width because quite a few people still run 800x600, but I would be very surprised if anybody interested in python web development were running a screen that small. I'd say you can safely go up to 990px if you need the space (which your layout really doesn't). The line height on the body copy is too high and is breaking up the text, please drop it to betweeen 1.2em and 1.3 em. The current spacing is fine on the sidebars. I'd like to see the download box changed: +---+ | DOWNLOAD || | |\/-| | Turbogears 0.8a6 | - 1.3 em or so | Released Dec 26, 2005 | - .8 em or so, it should fit under the 'Turbogears' | * Validation Bugfixes| | * OtherFeatures | | * OtherFeatures2 | +---+ As you can see from my amazing ascii art drawing, I think you should drop the 'now' so that it reads 'Download Turbogears 0.8a6'. The bullets display the key features of the release. I feel the redundant download and upgrade links at the bottom don't add anything and that people know to click the download header or the product name to get download or upgrade info. Now to design a release page and a decent start page. ;]
[TurboGears] Re: TurboGears new site look - let's get to it!
On 1/5/06, Karl Guertin [EMAIL PROTECTED] wrote: Excellent grid and hiearchy. The following are my nits, I don't expect them to be mocked up, but please incorporate them when you're doing the actual design: One last one -- and I know this is really out there and I think I'm the only one that will be bothered by this: I don't like that the gear is facing away from the page, makes me feel like the web page is tense.
[TurboGears] devcasts - DataController
Hi all, Is there a way to control the access to FastData/Crud thru turbogears.identity??? Something like: class Root(controllers.RootController): @turbogears.expose(template=wgstock.templates.welcome) def index(self): import time return dict(now=time.ctime()) class Admin(controllers.Controller): @identity.require( identity.in_group( admin ) ) User = DataController(sql_class=TG_User) @identity.require( identity.in_group( admin ) ) Group = DataController(sql_class=TG_Group) Thanks, Helio Pereira
[TurboGears] Re: TurboGears new site look - let's get to it!
Michele Cella wrote: Agreed, I really like the 1) 2) 3) slogan but I think we can find better words for Run and Visit (Quickstart is ok)... by the way I don't know what we can use! :P 1) Quickstart 2) Run 3) Gear Up !? Ok, time to sleep. :-) Ciao Michele
[TurboGears] Re: Problems installing pysqlite on Linux
Thank you very much. I tried your trick, and I could not get it to go. I bit the bullet, and did a static link against sqlite3, and that worked. I wish that I could get the dynamic lib going, but I am going to start with this. Thank you very much, Mike
[TurboGears] Re: TurboStan 0.7
[EMAIL PROTECTED] wrote: What python modules we need to install TurboStan? TurboGears = 0.9 Nevow (I think Nevow may have a dependency on zope.interfaces) Regardless, you should be able to use easy_install TurboStan and not worry about this too much. Cliff Nevow (zope ???) turbogears
[TurboGears] Extending TG objects
I'm having a bit of problem with inheritance. Because of @turbogears.expose() I can't seem to be able to do this: class MyClass(ParentClass): @turbogears.expose(self, args): def foo(self): d = super(MyClass, self).foo() d.update(dict(bar='baz')) return d because of the double-decoration going on. Is this kind of behaviour even possible? How do I access the original, pre-decorated method in the parent? -- Regards, Stephen Thorne Development Engineer Scanned by the NetBox from NetBox Blue (http://netboxblue.com/)
[TurboGears] Re: Stripping 'submit' argument from validated form input inconsistent
I agree. There is no reason for the submit button to have a name unless it's given an explicit value that the back-end wants to track. Only in that case should it be given a name (but then it should be sent to the method as well).
[TurboGears] Re: how to recover from aborted transaction
Just out of curiosity why you want a transaction that affects only one table?
[TurboGears] Re: Made a filter similar to variablesdecode()
That's a good question. You could if you want, but I suppose that would make javascript harder. document.form_name['var.name'][0] is slightly better then document.form_name['var-1.name']. Plus, I haven't tried NestedVariables in complement with @turbogears.expose. If you do it like that, then there's no need to unpack your parameters, now is there?
[TurboGears] Re: Extending TG objects
On Jan 5, 2006, at 4:17 PM, Stephen Thorne wrote: I'm having a bit of problem with inheritance. Because of @turbogears.expose() I can't seem to be able to do this: class MyClass(ParentClass): @turbogears.expose(self, args): def foo(self): d = super(MyClass, self).foo() d.update(dict(bar='baz')) return d because of the double-decoration going on. Is this kind of behaviour even possible? How do I access the original, pre-decorated method in the parent? That should work perfectly fine, given a correct turbogears.expose call of course. How about you show a full example that doesn't work and also post the traceback? -bob
[TurboGears] Re: Made a filter similar to variablesdecode()
On 1/5/06, Igor Murashkin [EMAIL PROTECTED] wrote:
That's a good question. You could if you want, but I suppose that would
make javascript harder. document.form_name['var.name'][0] is slightly
better then document.form_name['var-1.name']. Plus, I haven't tried
NestedVariables in complement with @turbogears.expose. If you do it
like that, then there's no need to unpack your parameters, now is there?
I ask because I'm working on nested forms and I currently handle the
output using the decorator. I'd ideally like to make it so that you
didn't have to set NestedVariables as your validator so the container
validators would act like formencode Schema's chained_values.
It shows up on the form as:
foo.bar-0 : '1'
foo.bar-1 : '2'
foo.bar-2 : '3'
baz :'4'
which would turn into:
params = {
'foo':{bar:['1','2','3']},
'baz':'4'}
using the code snippet I supplied.
[TurboGears] Re: sqlwidgets
you have to be carefull on how automated the data is going in and out of the DB, because that could lead to SQLinjection attacks.
[TurboGears] Re: Extending TG objects
Bob Ippolito wrote: That should work perfectly fine, given a correct turbogears.expose call of course. How about you show a full example that doesn't work and also post the traceback? This was a PEBKAC, sorry. -- Regards, Stephen Thorne Development Engineer Scanned by the NetBox from NetBox Blue (http://netboxblue.com/)
[TurboGears] Re: Static Site question
I'm sorry I don't think I undestand well. Do you mean generating a dinamic page and saving the output to a html file so you could later on use that standalone? if so it's should be really easy (i'm especulating but) I bet all you have to do is intercept the file-like object that kid returns and dump that to a file, which in python is a day in the park I believe what you need is this http://www.turbogears.org/docs/api/kid.serialization.Serializer-class.html#serialize then dump that stream to a html file.
[TurboGears] customization of datacontroller messages and translation
Hi, I wonder what would be the preferred way to customize datacontroller messages (things like Record deleted!)? Maybe all messages should go to a class variable: class DataController(BaseDataController): messages = dict( delete='Record deleted!', add='Record added!', . ) Than we can use: class News(DataController): messages = dict( delete='Newsitem is deleted', add='Newsitem is added, don't forget to a href=/pages/link ik to a page/a' ) And the translation: it looks like turbogears internal modules don't use i18n yet. Since datacontroller and complex widgets are the first components exposed to the end user, it is probably a good idea not to hardcode english strings in it. Oh, and maybe translations for those components should be contributed as plugins / language packs :) ? --Ksenia
[TurboGears] Re: how to recover from aborted transaction
Just out of curiosity why you want a transaction that affects only one table? Because of the concurrency of several web clients. -- Soni Bergraj
[TurboGears] Re: customization of datacontroller messages and translation
Hello, is somebody aware of a way to get Cheetah templates working with 0.8.7? I tried http://trac.turbogears.org/turbogears/ticket/214 but the patch is rejected. Any ideas? -- Soni Bergraj
[TurboGears] Getting Cheetah templates to work with 0.8.7?
Sorry for the wrong subject of the previous message:) Hello, is somebody aware of a way to get Cheetah templates working with 0.8.7? I tried http://trac.turbogears.org/turbogears/ticket/214 but the patch is rejected. Any ideas? -- Soni Bergraj
[TurboGears] Re: Getting Cheetah templates to work with 0.8.7?
On 1/5/06, Soni Bergraj [EMAIL PROTECTED] wrote: but the patch is rejected. Any ideas? Move to SVN or wait a week or two for 0.9. Kevin has said that he wants 1.0 by pycon, which is at the end of Feb, so that doesn't leave too much time for two major point releases.
[TurboGears] [Test] Please ignore this
not getting mails from tg list since december 27.
[TurboGears] Re: Announcing Scaffolding - A CMS for Turbogears
Kevin Dangoor wrote: Actually, I do think that both styles (application database, integration database) have merit for certain apps. Here's where application databases make some amount of sense: 1) newfangled, Web 2.0 applications and web services. Things like Backpack (surprise, surprise!), where the application serves a distinct purpose and lives out on the net. Data I/O is all done via web requests that go through the application code, which ensures the integrity of the data (to whatever extent it has been programmed to do so). As far as things like Basecamp go, at least part of the reason I'm not using it myself is because of this attitude about data integrity. Again, I think David is a brilliant guy, but that doesn't make up for the years of domain-specific experience and wide user-base available to the database developers. Any problem at the application layer (which is where most problems seem to occur) can cause data integrity issues unless the database itself is enforcing those rules. To be certain, I can understand his reluctance to use stored procedures and would even grant him that part of the argument. But foreign keys? Really. Those belong in the database: they are *not* business logic nor are they as easily (or efficiently, I might add) enforced from outside the database. If the app fails I can restart it. If it takes my data with it, I've got problems. If my company depends on project management software, it won't be written by a MySQL guy, no matter how smart he is. 2) small office applications. I did a lot of work on billing systems for doctors offices. Those systems are (or, at least, were) quite self-contained and the people in the office tended to use those apps for everything and not be computer literate enough to do much outside of there beyond word processing. (That's a generalization, certainly.) I have almost the exact opposite experience. The small office is where I've almost always found a mishmash of Access databases, Excel spreadsheets, etc, some of them sharing data, some not (and most poorly designed, because the people who designed them really should have stuck to word-processing). When migrating these systems to a centralized system, there is always a large overlap where the centralized system starts taking over functionality but parts remain elsewhere until they are eventually brought into the fold. I think David's argument falls apart there as well. If it were always starting from scratch or there were a specific cut-over date, then I'd be more likely to agree, but I've rarely encountered that. They always want the old system left intact just in case the new system has flaws, or they need some feature of the new system right away, etc, but the problem of data syncronization requires some integration of the existing infrastructure (or creation of infrastructure which brings stray applications into the fold during the migration period, which can often be months). Now, these two points are what you mean by highly specific, non-enterprise type of application, but I wanted to break it out more and emphasize that there's *a lot* of this, and #1 is still a growing category. Absolutely. Still, I'd wager that even these databases will one day outgrow this category. Data outlives the app that created it, almost every single time. These relationships need to be defined along with the data *in the database* if you are at all concerned about the poor sap who will have to one day retire your app or integrate it with other software. The larger the organization, the more likely it is to move into the integration database' category. A big organization *could* decide to expose the database entirely via web services and use that layer to maintain the database, or it could opt to move more into stored procedures, triggers and the like to make the database engine do the work. So, even within big organizations, there's still a choice to be made that could enable the application database perspective. Yes, but then so much for DRY ;-) David's argument puts you in the position to choose between DRY and data integrity. I don't think this choice needs to be made or should be made. At least one friend of mine does just this. He defines the constraints in both PostgreSQL and also in the Rails model. Not a big deal in terms of a workaround, but not very DRY either. If this were an insurmountable issue, I'd call it fair and move on. Since it's not even remotely insurmountable, but rather just one guy's rather poorly thought-out opinion about where the line between business logic and relational logic lay, I'll pass ;-) Regards, Cliff
[TurboGears] Re: devcasts - DataController
Helio MC Pereira, el jueves 5 de enero a las 23:18 me escribiste: Hi all, Is there a way to control the access to FastData/Crud thru turbogears.identity??? Something like: class Root(controllers.RootController): @turbogears.expose(template=wgstock.templates.welcome) def index(self): import time return dict(now=time.ctime()) class Admin(controllers.Controller): @identity.require( identity.in_group( admin ) ) User = DataController(sql_class=TG_User) @identity.require( identity.in_group( admin ) ) Group = DataController(sql_class=TG_Group) If this doesn't work this should do it: class UserDataController(DataController, identity.SecureResource): sql_class = TG_User require = identity.in_group(admin) class Root(controllers.RootController): # ... class Admin(controllers.Controller): user = UserDataController() (untested!) -- Leandro Lucarella (luca) | Blog colectivo: http://www.mazziblog.com.ar/blog/ ., \ GPG: 5F5A8D05 // F8CD F9A7 BF00 5431 4145 104C 949E BFB6 5F5A 8D05 / '' Me encanta el éxito; por eso prefiero el estado de progreso constante, con la meta al frente y no atrás. -- Ricardo Vaporeso. Punta del Este, Enero de 1918.
[TurboGears] Re: customization of datacontroller messages and translation
Ksenia Marasanova, el viernes 6 de enero a las 02:20 me escribiste: Hi, I wonder what would be the preferred way to customize datacontroller messages (things like Record deleted!)? Maybe all messages should go to a class variable: I gess messages in DataController just have to user _() to be internationalized... On the other hand, maybe some people just want to use a different message, in which case the dictionary stuff can be a good idea. -- Leandro Lucarella (luca) | Blog colectivo: http://www.mazziblog.com.ar/blog/ ., \ GPG: 5F5A8D05 // F8CD F9A7 BF00 5431 4145 104C 949E BFB6 5F5A 8D05 / '' Damian_Des Me anDa MaL eL CaPSLoCK
[TurboGears] Re: sqlwidgets
Understood, but at what level is input checked? Traditionally, I always used the parameter feature of database drivers, which is a good way to prevent injection. I believe the best place to check for sql injection is at the db api level. Also, the field validators, FormEncode, act as another barrier. The reason I'm focusing on building forms from the data objects is the Don't Repeat Yourself principle. It's better to define logic or data only once in your application. I've been writing web apps long enough to become sick of defining a database field as NOT NULL, wiriting server side logic to enforce NOT NULL, then writing javascript to check a field is NOT NULL. After seeing this pattern too frequently, it's time for something better.
[TurboGears] Re: Made a filter similar to variablesdecode()
Hey Karl, you can do that too, I made this filter earlier, and with
your example you would pass almost the same values as you have there:
foo.bar: '1'
foo.bar: '2'
foo.bar: '3'
baz : '4'
result = { 'foo' : {'bar' : ['1', '2', '3']}, 'baz' : '4'} }
Although I'm not as sure as to how you would make a validation schema
for that as easily, but once again it will be easy to access the values
in javascript with a documents.my_form['foo.bar'][0], [1], [2], etc.
### Code below
##
class BuildTreeFilter(BaseFilter):
Will build a tree for request params with periods in the name,
for an example look at _buildTree().
def beforeMain(self):
cherrypy.request.paramMap =
self._buildTree(cherrypy.request.paramMap)
def _buildTree(self, oldDict):
Given a dictionary with keys like 'foo.bar.joo', this will
return a new tree
with a key of foo = { 'bar': {'joo' : oldValue}}, for every
key.
Example:
print _buildTree({'company': 'Unexistential Inc',
'names.firstName': ['Joe', 'Boe', 'Moe'],
'names.lastName': ['Smith', 'Myth',
'Keith'],
'phone': '555-8980'})
{'company': 'Unexistential Inc',
'names': {'firstName': ['Joe', 'Boe', 'Moe'],
'lastName': ['Smith', 'Myth', 'Keith']},
'phone': '555-8980'}
stack = []
y = {}
for key, value in oldDict.iteritems():
splitList = key.split(., 1)
if len(splitList) 1: #there was a period in the key
firstLevel = splitList[0]
secondLevel = splitList[1]
firstValue = y.get(firstLevel)
if firstValue:
if type(firstValue) != dict:
y[firstLevel] = {}
y[firstLevel].update({secondLevel : value})
else:
y[firstLevel] = {secondLevel : value}
stack.append( firstLevel ) #add to stack only once
else:
y[key] = value
for i in iter(stack):
y[i] = self._buildTree( y[i] )
return y
[TurboGears] Rewriting static URLs to serve them dynamically
Newbie-Q. I have a gaggle of URLs of form /archive/number.html I want to update serve their content dynamically; it's obvious how to get almost-there with URLs like /archive/?number=number but I want the old URLs to work. It's a MovableType site, so there are probably other people with my problem. Is there a plausible answer in TG, or do I need to (say) run behind Apache and use mod_rewrite on those URLs?
[TurboGears] Re: Rewriting static URLs to serve them dynamically
On 1/6/06, Clank [EMAIL PROTECTED] wrote:
Newbie-Q. I have a gaggle of URLs of form
/archive/number.html
I want to update serve their content dynamically; it's obvious how to
get almost-there with URLs like
/archive/?number=number
but I want the old URLs to work.
It's a MovableType site, so there are probably other people with my
problem. Is there a plausible answer in TG, or do I need to (say) run
behind Apache and use mod_rewrite on those URLs?
What you want is to expose a default() method. Take a look at page 4
of the 20-minute Wiki tutorial, on
http://www.turbogears.org/docs/wiki20/page4.html, for an example. Have
/archive point at a class that looks something like the following:
class ArchiveByNumber(object):
@turbogears.expose(html=...)
def default(self, number):
if number.endswith('.html'):
number = number[:-5]
if number.endswith('.htm'):
number = number[:-4]
number = int(number) # In actual code, you'd use a validator here
# Fetch the appropriate page from the database
The default() method will take any remaining URL components and
receive them as arguments. If you had, for example:
class Archive(object):
@turbogears.expose(html=...)
def default(self, *args):
logDebug(Received args %s..., args)
And your user fetched the URL /archive/2004/11/30/my-blog-post.htm,
your function would log Received args ('2004', '11', '30',
'my-blog-post.htm')...
I hope this helps.
--
Robin Munn
[EMAIL PROTECTED]
GPG key 0xD6497014
