Webware for Python

At 03:24 PM 4/30/2001 -0400, Geir Magnusson Jr. wrote:
>It wouldn't surprise me.  People say that you get used to the whitespace
>blocking, but I haven't used python enough to get used to it...

I will totally back that up for programmers but I will keep iterating that 
templates are for non-programmers too. And even in HTML, programmers may 
not necessarily be comfortable with indentation for blocking. You basically 
end up mixing 2 mediums and things get creepy.


>It would be very cool if we could iron out differences and y'all could
>use velocity - that would be the wildest thing - if you could take the
>same set of templates and run it under a python framework or a java
>framework...

I understand your motivation, but I still see a pure Python templating 
solution as the way to go so that I don't have to download and use Java 
(yet another software package).

However, I don't represent the entire Python community. Also nothing would 
stop you from posting examples on how to do this with Webware which does 
not dictate that you use all it's components.

I CCed webware-discuss for their info.

-Chuck

Re: [Webware-discuss] Re: WM difference question From: Terrel Shumway <tshumway@ic...> - 2001-04-30 19:35 Chuck Esterbrook wrote: > >It would be very cool if we could iron out differences and y'all could > >use velocity - that would be the wildest thing - if you could take the > >same set of templates and run it under a python framework or a java > >framework... > > I understand your motivation, but I still see a pure Python templating > solution as the way to go so that I don't have to download and use Java > (yet another software package). > Didn't he mean the other way around?: use Python Velocity to instantiate templates written for Java Velocity. To do that would require a real velocity-as-a-language specification, which would probably improve the robustness of all implementations, because it would remove unstated implementation-language-dependent assumptions. (Don't IEFT standards require at least two reference implementations written in different languages?)

Re: [Webware-discuss] Re: WM difference question From: Geir Magnusson Jr. <geirm@op...> - 2001-04-30 19:49 Terrel Shumway wrote: > > Chuck Esterbrook wrote: > > > >It would be very cool if we could iron out differences and y'all could > > >use velocity - that would be the wildest thing - if you could take the > > >same set of templates and run it under a python framework or a java > > >framework... > > > > I understand your motivation, but I still see a pure Python templating > > solution as the way to go so that I don't have to download and use Java > > (yet another software package). > > > > Didn't he mean the other way around?: use Python Velocity to instantiate templates written for Java Velocity. Either way - a Python user moves to java vel, a java vel user to Python vel. Or templates are shared. > > To do that would require a real velocity-as-a-language specification, which would probably improve the robustness of all implementations, because it would remove unstated implementation-language-dependent assumptions. (Don't IEFT standards require at least two reference implementations written in different languages?) A language spec would be great. We have tossed around the idea a little. However, before that kind of undertaking is attempted, it would also be interesting if you can implement as it is now, and what kind of improvements you come up with, that we in Java-land could fold in. If you every have any questions, feel free to ask - I am very interested in the suggestions and criticisms that would come from someone implementing a Velocity-like template language. Most of our feedback is from template users, or Java servlet/application writers that are using velocity. geir -- Geir Magnusson Jr. geirm@... System and Software Consulting Developing for the web? See http://jakarta.apache.org/velocity/

Re: [Webware-discuss] Re: WM difference question From: Chuck Esterbrook <echuck@mi...> - 2001-04-30 21:32 At 04:46 PM 4/30/2001 -0400, Geir Magnusson Jr. wrote: >A language spec would be great. We have tossed around the idea a >little. A test suite would be a big boon as well. The inputs for each test case, could be the dictionary of data (in some easy to parse text format) and the template. The output of course would be the output. :-) -Chuck

Re: [Webware-discuss] Re: WM difference question From: Geir Magnusson Jr. <geirm@op...> - 2001-04-30 21:37 Chuck Esterbrook wrote: > > At 04:46 PM 4/30/2001 -0400, Geir Magnusson Jr. wrote: > >A language spec would be great. We have tossed around the idea a > >little. > > A test suite would be a big boon as well. The inputs for each test case, > could be the dictionary of data (in some easy to parse text format) and the > template. The output of course would be the output. > > :-) We have a large test suite for the parser, with outputs to compare against. We use JUnit for unit testing the whole pile. Not something you can use in Pythonland of course, but I am sure there is something similar... geir -- Geir Magnusson Jr. geirm@... System and Software Consulting Developing for the web? See http://jakarta.apache.org/velocity/

Re: [Webware-discuss] Re: WM difference question From: Chuck Esterbrook <echuck@mi...> - 2001-04-30 21:47 At 06:33 PM 4/30/2001 -0400, Geir Magnusson Jr. wrote: >We have a large test suite for the parser, with outputs to compare >against. We use JUnit for unit testing the whole pile. Not something >you can use in Pythonland of course, but I am sure there is something >similar... Ahem. Well yes. I was "secretly" planning to rip all your test suites, but in the process, review them as well and see if I could come up with anything new which would then be contributed back. My point earlier was that the inputs could be kept in a language neutral format and therefore shared by any Velocity implementation. Is that the case now, or is the "context data" hard coded in the Java testing code? http://pyunit.sourceforge.net/ is the equivalent of JUnit. It's also included with the latest Python 2.1 release. See, for everything good in Java, there is an equivalent in Python. Also, you cannot say the same thing with s/good/bad/. ;-) (Sorry but after all these exchanges with Java folk, I had to say something.) -Chuck

Re: [Webware-discuss] Re: WM difference question From: Geir Magnusson Jr. <geirm@op...> - 2001-04-30 22:07 Chuck Esterbrook wrote: > > At 06:33 PM 4/30/2001 -0400, Geir Magnusson Jr. wrote: > >We have a large test suite for the parser, with outputs to compare > >against. We use JUnit for unit testing the whole pile. Not something > >you can use in Pythonland of course, but I am sure there is something > >similar... > > Ahem. Well yes. I was "secretly" planning to rip all your test suites, but > in the process, review them as well and see if I could come up with > anything new which would then be contributed back. Well, I guess it's all covered by the ASL, so you don't even have to contribute back :) (but it would be appreciated...) > > My point earlier was that the inputs could be kept in a language neutral > format and therefore shared by any Velocity implementation. Is that the > case now, or is the "context data" hard coded in the Java testing code? The templates are simple ASCII files, as are the comparison output, but the context data is in Java code. Most of the context data could come out, but remember that we can throw living objects into the context for access, so we would have to use some sort of XML <-> object serialization to represent neutrally. I'm game, tho.. > http://pyunit.sourceforge.net/ is the equivalent of JUnit. It's also > included with the latest Python 2.1 release. > > See, for everything good in Java, there is an equivalent in Python. > Also, you cannot say the same thing with s/good/bad/. ;-) > (Sorry but after all these exchanges with Java folk, I had to say something.) You're just pissed cuz we get to use { and } to delimit our blocks.... ;) geir -- Geir Magnusson Jr. geirm@... System and Software Consulting Developing for the web? See http://jakarta.apache.org/velocity/

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