- Zope-Dev - Zope lists

FW: [Zope] Zope stability
by Alexander Limi 13 Dec '99

13 Dec '99

Sorry, I had a spelling error in the mail-address to the dev. list, so this never showed up. I think some of the points detailed could be interesting for the next version of Zope. Alexander Limi http://mp3.no > > Also, on a side note, is there a way to do offline development of Zope, and > > commit those changes to the main site at a later date? I know this raises > > important concurrency concerns, but has it ever been considered as a part > > of Zope's feature set? A CVS of sorts, where you check out and in files? > > Well, you can run a developoment server with an identical environmen. To > update the live site, you export the relevant parts, and import them > into the production server. This simply won't work, as we have users from everywhere that want to do changes 24/7. So the main site can't be updated from a production site, because users change the main site all the time. So unless one can update selected parts in a satisfactory way (by exporting only the relevant changes) this is no option. And I can imagine it would get messy, as one had to manually keep track of what changes are made to the test server, export those bits, and then import them on the production server. Would it be an idea to implement this in an easy-to-use way in Zope 2.2? To be able to create a "diff-file" of sorts, that can be exported into the main site, and exclude certain hierarchies, like the ones the users access? While we're at it, my own wishlist for the next version: - Better handling of inheritance in ZClasses, see my earlier post (ZClasses, inheritance & stuff, on the zope(a)zope.org list) (Basically, the ability to use superclasses effectively) - Zope should produce html/dtml in compliance with w3.org's recommendations: in all lowercase - Access to the built-in debug screens from the contol panel (in production mode or not) - An option to turn on "open in new window" as default when using the preview mode (I know, mozilla and all, but as a stop-gap solution until mozilla is stable enough to use :) (small changes that would make me happy, except for the inheritance bit, which may be more time-consuming to implement) Alexander Limi http://mp3.no

2 1

__cmp__ methods being ignored?
by Benno Rice 13 Dec '99

13 Dec '99

Hi, I posted about this to -dev a little while ago but received no response. I've got a Python class that I'm using in Zope, which defines it's own __cmp__ method. However this method doesn't appear to be being called when Zope tests for equality. Is this a bug or a feature? Is there some way I can get around this? The situation is that I have some of these objects in an array attribute of another object. If I try and compare an object that I Acquired to one that's in the list, it fails as I'm comparing an object of my type to one of ImplicitAcquirerWrapper. However, if my __cmp__ method were being used in the equality test, they'd match (I checked this). I also tried setting __repr__. This worked as I'd expect it to but didn't affect the equality test. Any help would be most appreciated, I really need to get this working. Many thanks. -- Benno Rice "No, no. We're *sweet* and XNFP Aries Dark Subculture- *innocent* evil bastards." friendly Internet Geek benno(a)netizen.com.au "Defend your joy"

1 0

Re: [Zope-dev] local/set
by Anthony Baxter 12 Dec '99

12 Dec '99

>>> "Phillip J. Eby" wrote > Let: allows read-only variables to be computed whose values are usable > within a DTML nesting scope. > > Local: allows writable variables to be computed, whose values are usable > and changeable within a DTML nesting scope. > > Set: singleton tag that changes the value of a "local" variable, as long as > it was defined in the innermost surrounding "local" block. Given the ongoing discussions over the difficulty of learning DTML, the general ugliness of DTML, and the discussion over what DTML is actually good for, isn't adding local/set just taking us further and further down the path of horror? It's already _way_ too easy to start with a simple DTML page, and then extend it and extend it beyond any sane limits until you end up with a perl-like morass of code. Utterly unmaintainable. Surely using PythonMethods or ExternalMethods is a much much better approach here, rather than bolting yet another bag onto the side of DTML. dtml-let solves a problem: it's a much nicer way to spell _.namespace() I'm not convinced that _.namespace is such a great idea in the first place. I've seen too much DTML that makes me cry when I have to go and maintain it - I'd actually like to see it made a bit more obvious when you should stop doing things in DTML. -- Anthony Baxter <anthony(a)interlink.com.au> It's never too late to have a happy childhood.

3 2

RE: I have a dream
by Sin Hang Kin 12 Dec '99

12 Dec '99

Thank you for pointing out PIA. I love its syntax (a little bit too lengthly). I have another idea, right now, zope is target html-oriented output. However, it shoudl be much easier if it just publish the oodb into xml and let the xsl and javascript to do the actual job for html presentation. for example, the tree tag could be much easier if it just export the tree in a xml segment. We could develope a set of convention and tools to render the tree into html. The respond would be much faster. What can be best is zope can be more structural oriented instead of layout oriented. Rgs, kentsin

1 0

RE: I have a dream
by Sin Hang Kin 12 Dec '99

12 Dec '99

Thank you for pointing out PIA. I love its syntax (a little bit too lengthly). I have another idea, right now, zope is target html-oriented output. However, it shoudl be much easier if it just publish the oodb into xml and let the xsl and javascript to do the actual job for html presentation. for example, the tree tag could be much easier if it just export the tree in a xml segment. We could develope a set of convention and tools to render the tree into html. The respond would be much faster. What can be best is zope can be more structural oriented instead of layout oriented. Rgs, kentsin

1 0

RE: I have a dream
by Sin Hang Kin 12 Dec '99

12 Dec '99

Thank you for pointing out PIA. I love its syntax (a little bit too lengthly). I have another idea, right now, zope is target html-oriented output. However, it shoudl be much easier if it just publish the oodb into xml and let the xsl and javascript to do the actual job for html presentation. for example, the tree tag could be much easier if it just export the tree in a xml segment. We could develope a set of convention and tools to render the tree into html. The respond would be much faster. What can be best is zope can be more structural oriented instead of layout oriented. Rgs, kentsin

1 0

RE: I have a dream
by Sin Hang Kin 12 Dec '99

12 Dec '99

Thank you for pointing out PIA. I love its syntax (a little bit too lengthly). I have another idea, right now, zope is target html-oriented output. However, it shoudl be much easier if it just publish the oodb into xml and let the xsl and javascript to do the actual job for html presentation. for example, the tree tag could be much easier if it just export the tree in a xml segment. We could develope a set of convention and tools to render the tree into html. The respond would be much faster. What can be best is zope can be more structural oriented instead of layout oriented. Rgs, kentsin

1 0

OFSP/Help problems
by Robin Becker 11 Dec '99

11 Dec '99

After some futzing around with CVS I managed to get the OFSP/Help directory to download. So I now see the help product in the Products Folder, but when I press on any of the help buttons I get an attribute error Help is unknown. -- Robin Becker

2 1

A simply question for help?
by Victor.Zhai＠ogilvy.com 11 Dec '99

11 Dec '99

Hi,all I want to display a new sub windows when click the following hyper link: <a href="#" onClick="window.open('<dtml-var absolute_url>/viewInfo>','','scrollbars=yes,width=640,height=430')"><dtml-var jobtitle></a> But you know that <dtml-var absolute_url> is like http://www.mydomain.com/object... ,And it is not a physical web page. How can I realize this? Give me some advice. Thanks in advance. "WorldSecure Server <ogilvy.com>" made the following annotations on 12/11/99 03:00:07 ------------------------------------------------------------------------------ Privileged/Confidential Information may be contained in this message. If you are not the addressee indicated in this message (or responsible for delivery of the message to such person), you may not copy or deliver this message to anyone. In such case, you should destroy this message and kindly notify the sender by reply email. Please advise immediately if you or your employer does not consent to email for messages of this kind. Opinions, conclusions and other information in this message that do not relate to the official business of the Ogilvy Group shall be understood as neither given nor endorsed by it. ==============================================================================

2 1

Re: [Zope-dev] local/set
by Anthony Baxter 11 Dec '99

11 Dec '99

>>> Lalo Martins wrote > > dtml-let solves a problem: it's a much nicer way to spell _.namespace() > > You answer yourself. Hm, I should have made it clear that the post I made wasn't necessarily advocating one thing over another, but was more pointing out some things that I see as issues. > dtml-local is IMHO wrong. But dtml-set should be added, it fits > naturally. Or, in your words, is a much nicer way to spell > REQUEST.set. Is it? Not according to Phillip's original post - set is for changing values in the current local namespace. REQUEST.set is a different thing altogether. It doesn't worry me that REQUEST.set has to be spelled out explicitly, as this highlights quite clearly what's going on... (Aside: It seems to me that one of the more common uses of REQUEST.set that I've seen is as a sort-of global namespace that's useful for stuffing temporary values into. It's the reason I did the session code - (ab)using REQUEST in this way seemed bad. Ok, so the sessions also suffer from this a bit, but they're a nice fit for so many problems.) local/set seem like they need to both be in, or neither. Allowing set to modify any namespace strikes me as dangerous. Anthony -- Anthony Baxter <anthony(a)interlink.com.au> It's never too late to have a happy childhood.

1 0