Webware for Python

Hi!

I am totally amazed of WebWare. Very well done.

On the todo-list I read something about "improving of error
(traceback) handling".  I just like to mention the following link,
which shows Ka-Ping Yee's cgitb.py module:
    http://web.lfw.org/python/

Instead of
    http://web.lfw.org/python/test3.html
one can get
    http://web.lfw.org/python/test4.html
which is quite impressive.

I'm not sure if this can be used, but I just wanted you to know.

Kind regards, Markus

Re: [Webware-discuss] traceback handling From: Chuck Esterbrook <echuck@mi...> - 2001-04-27 13:10 At 03:02 PM 4/27/2001 +0200, Markus Gritsch wrote: >Hi! > >I am totally amazed of WebWare. Very well done. > >On the todo-list I read something about "improving of error >(traceback) handling". I just like to mention the following link, >which shows Ka-Ping Yee's cgitb.py module: > http://web.lfw.org/python/ > >Instead of > http://web.lfw.org/python/test3.html >one can get > http://web.lfw.org/python/test4.html >which is quite impressive. > >I'm not sure if this can be used, but I just wanted you to know. > >Kind regards, Markus I think that TO DO might have been referring more to what happens when Webware internals throw an exception. Especially outside the transaction. In any case, I actually like this better: http://webware.sourceforge.net/Webware/WebKit/Docs/SampleError.html But I could see the point of providing ping's format as an option (or the default if users really love it). Since I'm not inspired by it, someone else will have to provide a patch, although a WEP describing the configuration and implementation approach would be the right first step. -Chuck

Re: [Webware-discuss] traceback handling From: Ian Bicking <ianb@co...> - 2001-04-27 18:01 Markus Gritsch <gritsch@...> wrote: > On the todo-list I read something about "improving of error > (traceback) handling". I just like to mention the following link, > which shows Ka-Ping Yee's cgitb.py module: > http://web.lfw.org/python/ Oooo... that is really pretty. That's much of what I've always wanted in a traceback, but never had (well, I suppose if I used pdb, but that's not an option for CGI anyway). I'm not sure I'd use it when a site was in production -- perhaps a little too verbose (though, heck, what's the harm?), but I'd definately enjoy it while developing. I'm going to look into using it. I also noticed on that page was string-interpolation, somewhat similar to Perl ("there are $x eggs in the basket")... a bit nicer than the sprintf (%s) notation. That might be of interest to someone, as something closer to PSP but with simpler syntax (but only expression-level substitution) htmldoc is nice too, though it looks like you have to have complete doc strings for all methods and functions in order for it to produce good output. But I like the layout. Ian

Re: [Webware-discuss] traceback handling From: Chuck Esterbrook <echuck@mi...> - 2001-04-27 18:51 At 01:01 PM 4/27/2001 -0500, Ian Bicking wrote: >I also noticed on that page was string-interpolation, somewhat similar >to Perl ("there are $x eggs in the basket")... a bit nicer than the >sprintf (%s) notation. That might be of interest to someone, as >something closer to PSP but with simpler syntax (but only >expression-level substitution) I'm working on that as a new version of ReplaceNamedValues that, like the new NamedValueAccess, will provide its functionality via functions, a mix-in and a wrapper, so you can do things how you want. I'll include complete doc strings and a test suite. I thought Tavis had a better version of RNV before he went full steam on templates, but I can't find it on my system. TemplateFiller.py or something? Maybe something else? If someone has something up that alley, please send it my way. The default syntax will be $foo.bar, which is a strict subset of Velocity/webmacro/Tavis. That way if you expand your templates to something more powerful in the future, the existing templates continue to work. Well, assuming you choose Velocity/webmacro/Tavis or something compatible with $foo.bar. -Chuck

1 message has been excluded from this view by a project administrator.