Hi,
I think there is a lot a matter of personal taste here.
The main concern, as far as I see it, is maintainability of
the code, which derives its readability. And personal
taste has a lot to do with that.
As it is now, we gave up on if/else tags for readability.
(and later on tag libs in general since we just found
that the servlet engine we use (ServletExec 3.1) still
has problems in their implementation).
-----Original Message-----
From: Borislav Iordanov [mailto:[EMAIL PROTECTED]]
Sent: Thursday, June 07, 2001 4:55 AM
To: [EMAIL PROTECTED]
Subject: RE: RE: If/Else tag?
Hi,
I'm just wondering, what's wrong with scriplets. I'm really failing to find
any value in a tag "replacement" of the usual and natural Java if statement.
It seems to me that the useful conditional tags are the ones with some
predefined and parametrizable (in a simple way) conditions, like the
<req:ifParameter name="..." value="..."> and even then.... what's the point?
Is it just that for you something like:
<utils:if predicate='<%= ... %>'>
</utils:if>
looks better than
<% if (....) %>
<% } %>
Just asking, because I'm really wondering whether there's some inherent
advantage in having tags instead of scriplets, or it's simply a matter of
personal taste and some prefer writing tags.
Borislav
-----Original Message-----
From: Amos Shapira [mailto:[EMAIL PROTECTED]]
Sent: Tuesday, June 05, 2001 2:12 AM
To: '[EMAIL PROTECTED]'
Subject: RE: RE: If/Else tag?
Sorry, but if the purpose of taglibs is, among other things,
to make the code more readable and maintainable then such
tricks are not helping this goal, IMHO.
Also - what about nested "if/else"?
Starts to look like old-fasioned basic to me (the one on
Genie II and XT ROM Basic, not Visual Basic) :-)
I'll stick to scriptlets for now, as much as I'd like to
avoid them.
Cheers,
--Amos
-----Original Message-----
From: Mikko Kurki-Suonio [mailto:[EMAIL PROTECTED]]
Sent: Tuesday, June 05, 2001 8:45 AM
To: '[EMAIL PROTECTED]'
Subject: Re: RE: If/Else tag?
On Tue, 5 Jun 2001, [utf-8] Amos Shapira wrote:
> I considered this but:
>
> 1. It's not always clear (especially if the predicate is complex) that the
> "else" predicate is the opposite of the "if".
> 2. Worse - whenever you change the "if" predicate you have to remember
> to change the "else" too. Very unmaintainable over time.
> 3. It's not clear that the two are actually related to each other.
> 4. You have to consider side-effects (as you said - if the "if" changes
> "x", but also if the evaluation of "x" has side effects).
Simple. Just set a DIFFERENT flag variable whenever you trip the "real IF"
and test the _flag_ for the "ELSE".
In pseudocode:
x_else_flag = true;
if x then x_else_flag = false; blah blah blah;
if x_else_flag then do_the_else_thingy;
//Mikko
