Re: Modular Client Code Proposal

2000-08-11 Thread L.C.
The constructs suggested in the original email are similar if not identical to embperl's (or php's for that matter), I was merely pointing out that we shouldn't probably reinvent the wheel with yet another pseudo-scripting language. The embperl page needs not contain the actual perl code to pro

Re: Modular Client Code Proposal

2000-08-11 Thread Joe Rhett
You're missing the point of separating the code and the interface. On Thu, Aug 10, 2000 at 10:00:48AM -0700, L.C. wrote: > > You may want to take a look at Embedded Perl (Embperl), it does EXACTLY what > you are asking and much more, it is a very good tool for building interfaces > in perl. http

RE: Modular Client Code Proposal

2000-08-10 Thread Lynn W. Taylor
It may not be very portable either -- a fair number of us aren't running Apache. -Original Message- From: David Harris Sent: Thursday, August 10, 2000 10:59 AM To: L.C.; [EMAIL PROTECTED] Subject: RE: Modular Client Code Proposal L.C. wrote: > You may want to tak

RE: Modular Client Code Proposal

2000-08-10 Thread David Harris
L.C. wrote: > You may want to take a look at Embedded Perl (Embperl), it does EXACTLY what > you are asking and much more, it is a very good tool for building interfaces > in perl. http://perl.apache.org/embperl/index.html > > You can then include separate header files too, although you realize t

Re: Modular Client Code Proposal

2000-08-10 Thread L.C.
You may want to take a look at Embedded Perl (Embperl), it does EXACTLY what you are asking and much more, it is a very good tool for building interfaces in perl. http://perl.apache.org/embperl/index.html You can then include separate header files too, although you realize that you will eventual

RE: Modular Client Code Proposal

2000-08-10 Thread David Harris
Joe Rhett wrote: > To be honest, one of the huge problems OpenSRS has is the dependancy upon > the template names - template names in templates right beside template names > in the code. > > If you designed it better, you wouldn't need to have any template names in > the code at all -- absolutely

Re: Modular Client Code Proposal

2000-08-09 Thread Joe Rhett
To be honest, one of the huge problems OpenSRS has is the dependancy upon the template names - template names in templates right beside template names in the code. If you designed it better, you wouldn't need to have any template names in the code at all -- absolutely no dependacies. The template

Re: Modular Client Code Proposal

2000-08-09 Thread Joe Rhett
These are all very nice, but I've seen a great many "modular" systems built which end up completely unmaintainable due to the lack of foresight of the programmers. To be honest, I'm not very impressed with OpenSRS code to date: 1. The code must known exactly what's in the templates and where it i

Re: Modular Client Code Proposal

2000-08-09 Thread storage
Charles Daminato <[EMAIL PROTECTED]> said: > I've been working on this for some time - but since I'm not a 'programmer' > per se, don't mind the terms I'm using. I expect full feedback on the > list for this discussion - try to keep it pertinent instead of trying to > get 'new' features added as

Re: Modular Client Code Proposal

2000-08-09 Thread Scott Allan
At 08:44 PM 8/9/00 +, you wrote: >Charles, > >While I think this is a great idea - and I mean that, I'm not just being >obsequious - I also think that you're mixing up your priorities. The most >important next step for OpenSRS should be to document the API completely, >from start to finish. T

Re: Modular Client Code Proposal

2000-08-09 Thread Scott Allan
Points very well taken. A well documented API is first. What we are talking about is beyond that. Regards, sA At 02:32 PM 8/9/00 -0400, you wrote: >My suggestion - and this is a very serious suggestion - is that >Tucows/OpenSRS not spend time developing fancy client code. In fact, >I would p

Re: Modular Client Code Proposal

2000-08-09 Thread Christopher Masto
On Wed, Aug 09, 2000 at 11:34:45AM -0400, Charles Daminato wrote: > With each new feature we add to OpenSRS we find ourselves forced to > release a complete version of the client code. Due to severe > customizations that certain customers have made it becomes increasingly > difficult for you to i

Re: Modular Client Code Proposal

2000-08-09 Thread WebWiz
Comments intertwined below, enclosed in double angle-brackets << like this... >> Regards, Eric Longman Atl-Connect Internet Services +---+ | Atl-Conn

Re: Modular Client Code Proposal

2000-08-09 Thread Charles Daminato
My idea was to have pertinent naming schema for the templates. That way managing several directories wouldn't be a problem. I.E. manage_login.template, base.template, batch_reg_verify.template Stuff like that :) Making a spec for template schema might be just as difficult and involved as creat

Re: Modular Client Code Proposal

2000-08-09 Thread Lance Woodson
Charles Daminato wrote: > > Templates - each new module may need new templates. The proposal is to > have a base template infrastructure (base.html, header.html, footer.html, > etc) and specific APTLY named templates for each new feature and customer > facing screen that encompasses the logic of