Re: [Zope] Stopping a <dtml-in>
"Jonothan Farr" <jfarr@real.com> wrote:
http://www.zope.org/Members/jfarr/Patches/dtml-break
I requested that this page be cataloged but it never was.
I don't know why about that one.
I also submitted it as a patch to the distribution but it was refused. Something about a new dtml-in implementation that I haven't seen materialize in the last 5 versions or so.
I seem to recall that this one was me; I likely closed out your Collector issue (May last year, was it), expecting that Evan's (?) "loop variable" patch for the '<dtml-in>' tag would get folded in, which never gelled. The sentiment against "DTML-as-scripting-language" also makes introducing more control structures difficult to sell; I don't buy that, myself, but have to listen to folks who do. Anyway, I'm sorry that your proposal fell through the cracks; the "hot money" these days is all on HiperDOM/XHTML stuff, rather than DTML, so it may stay there (especially if the patch is hard to apply to a recent Zope). Tres. -- =============================================================== Tres Seaver tseaver@digicool.com Digital Creations "Zope Dealers" http://www.zope.org
At 1/17/01 10:24 PM, Tres Seaver wrote:
"Jonothan Farr" <jfarr@real.com> wrote: I also submitted it as a patch to the distribution but it was refused. Something about a new dtml-in implementation that I haven't seen materialize in the last 5 versions or so.
The sentiment against "DTML-as-scripting-language" also makes introducing more control structures difficult to sell; I don't buy that, myself, but have to listen to folks who do.
Some 8 months ago, I thought "break" was missing from <dtml-in> and wrote to this list. Within hours, clever solutions were suggested, the most impressive of which was Jonothan's patch mentioned above. I'm happy to see Tres saying that he doesn't buy the anti-DTML-scripting sentiment. To me, it makes little sense to create a language and then make it hard to use. Even less so when refusing an offered change that is more of a correction than it is an extension. Now just don't get me started on "_['index-whatever']" :-) -- Dennis Nichols nichols@tradingconnections.com
Jonothan Farr wrote:
Now just don't get me started on "_['index-whatever']" :-)
Funny. I patched that too, and I know I'm not the only one. It also "fell through the cracks". *sigh*
I've never understood the reaction that always comes from DC w.r.t. patches that fix that. There always seems to be a huge 'no!' for no particular reason, even though that would save _so_ many people _so_ much time and effort :-( Chris
Now just don't get me started on "_['index-whatever']" :-)
Funny. I patched that too, and I know I'm not the only one. It also "fell through the cracks". *sigh*
I've never understood the reaction that always comes from DC w.r.t. patches that fix that.
There always seems to be a huge 'no!' for no particular reason, even though that would save _so_ many people _so_ much time and effort :-(
Yeah, so hey, DC, what gives? Why don't we have sequence_item and family or some reasonable alternative? This has been dragging out for too long! --jfarr
On Thursday 18 January 2001 17:37, Dennis Nichols wrote:
At 1/17/01 10:24 PM, Tres Seaver wrote: Some 8 months ago, I thought "break" was missing from <dtml-in> and wrote to this list. Within hours, clever solutions were suggested, the most impressive of which was Jonothan's patch mentioned above. I'm happy to see Tres saying that he doesn't buy the anti-DTML-scripting sentiment. To me, it makes little sense to create a language and then make it hard to use. Even less so when refusing an offered change that is more of a correction than it is an extension. Now just don't get me started on "_['index-whatever']" :-)
Count me as another vote of support. As much as people diss DTML, my first few sites were entirely DTML because, before PythonScripts or PythonMethods, it was the simplest solution when you needed TTW editing. It would take just a little nudge (the sequence-* to sequence_* patch, and perhaps this dtml-break) to make DTML much easier to use. I consider DTML as a relatively unthreatening step for people between newbie status and moving into full Python methods. Now, if you'll excuse me, I've got to go back and start writing my first ZClass...
-- Dennis Nichols nichols@tradingconnections.com
Have a better one, Curtis Maloney
Tres Seaver wrote:
The sentiment against "DTML-as-scripting-language" also makes introducing more control structures difficult to sell; I don't buy that, myself, but have to listen to folks who do.
Well, DTML doesn't strike me as being designed to script in, so using it as such is going to cause problems. In my experience, any DTML with "" in it should have been done as a Python Script, even if it's one line long ;-) However, that doesn't mean something like <dtml-break> shouldn't exist, although it's really treading that fine line, and it doesn't mean that 'DTML is bad' as I used to think. DTML is great, if you use it like it was designed to be used; as a templating language. That requries a bit of iteration, and maybe conditionals are okay, maybe dtml-break falls in there? However, if you start writing things like ZClass constructors in DTML, you very quickly get unreadable, obtuse code that's hard to maintain. I dunno, I wish I could get my head around HiperDOM/XHTML but it just doesn't strike me as very approachable right now... My $0.02... Chris
Anyway, I'm sorry that your proposal fell through the cracks; the "hot money" these days is all on HiperDOM/XHTML stuff, rather than DTML, so it may stay there (especially if the patch is hard to apply to a recent Zope).
It would be easy to apply the patch. The work would be in testing it. It seems like getting it into the new release while it's still alpha would get a lot more eyeballs looking at it. HiperDOM/XHTML may be the way of the future, but DTML is what we've got. Why not make it a little bit easier to use if it doesn't take much effort? --jfarr
participants (5)
-
Chris Withers -
Curtis Maloney -
Dennis Nichols -
Jonothan Farr -
Tres Seaver