- Zope - Zope lists

REQUEST.form (newbie help)
by Philip Aylesworth 17 Feb '99

17 Feb '99

I am having some trouble getting a list of variables sent to a DTML doc. If i use the URL http://zope/testbr?test=1 I can use  to get a value of test. What I would like to do is get a list of names and values of all variables sent. I tried:  and  These seem to give me an error. Any ideas? Phil. ----------------------------------------------------------- Philip Aylesworth mailto:purple@mnsi.net

1 0

Re: [Zope] DTML Doc property from DTML Method in tag
by Amos Latteier 17 Feb '99

17 Feb '99

At 01:20 PM 2/15/99 -0600, rkw(a)dataplex.net wrote: >this is my structure: > >Folder 'HERE' has Folders within it. >Assume that they have ids ID1, ID2, ID3, etc. >Each of these Folders has a property 'common_name' > >Thus, in HERE/ID1 can reference  and get 'Name 1'. > >In HERE, I execute a form. The checkboxes build a list ids ['ID1', 'ID2']. >Processing the form, > gives me the list. > gives 'ID1' on the >first iteration. But I want to get 'Name 1' Indirection is bit confusing in DTML. Here are examples from a post I made a couple days ago which tries to explain the concept: Suppose that a DTML variable named Category is 'Expensive'. Then  is 'Expensive'. Then  is  which means . Got it? Here's one final example in Python from DocumentTemplate import HTML t=HTML(""", """) print t(foo="bar", bar="spam") which prints "bar, spam". This is analogous to Python. foo.bar is the same things as getattr(foo,'bar')  is the same thing as  In fact, you might think of the DTML underscore namespace mapping as similar to a Python __dict__ namespace mapping. So to get back to your question, > gives me the list. > gives 'ID1' on the >first iteration. But I want to get 'Name 1' You have a list of object ids. You want to iterate through the objects and print each one's common_name, correct? Here's one way to do that:    This assumes that common_name is a string attribute of each object. If common_name were a method, or a DTML method you might want to do something different, for example:      This would allow normal DTML var rendering behavior. So why the two levels of indirection? Let's see. Assume ids is ['ID1','ID2']. The first time through the loop  ==  == 'ID1' What you are looking for is the object named ID1, not the string 'ID1'. So you have to look in the DTML namespace for the object.  ==  It's a little ugly but it works. Make sense? -Amos

3 3

(Fwd) Zope musings
by Tom Jenkins 17 Feb '99

17 Feb '99

Hi folks, We're in the process of evaluating Zope and Cold Fusion. I'm the resident Open Source advocate <g>, so I'd like to make sure that Zope has all its benefits given its due. We're, (well actually Alan is), working on a contract where the client requested Cold Fusion so he has some direct experience. We installed Principia back in December and actually are testing it for another client, but haven't spent much time with it unfortunately. Alan has brought up some usage concerns of his and I hope someone on the list could respond. Please don't think that Alan is against the product; no in fact he is very interested in it but is just unsure of the tools available to help do good work. ------- Forwarded Message Follows ------- Date sent: Wed, 17 Feb 1999 10:08:00 -0500 To: TomJenkins(a)zentuit.com From: Alan Johnston <flames direct to me, please [g]> Subject: Zope musings I don't know why I didn't think of this before. As you can tell from earlier comments, I like Cold Fusion studio, the IDE which allows you to control web page, database, query, and site management all in one window. It is a very slick way to create web pages in general, whether or not they include CF tags and CF queries. But there's no reason you couldn't also use CF Studio to develop Zope/DTML pages. You would have to save your pages on a local drive and use Zope's Upload feature to publish them. But that's a fairly small price in inconvenience. Anyway, the lack of a nice, user-friendly interface for web page creation is the one klunky thing about Zope (Principia) at our site. Typing angle brackets and obscure HTML tags into your browser is a fun way to astound and impress your non-HTML-literate friends, but I wouldn't have wanted to make a Georgetown/CASS questionnaire system that way. And, again, trying to use Netscape Composer, as recommended in the Principia docs is a sub-par solution. So this is really the one (only?) area I've seen so far where CF is way ahead Zope. So if we want to use Zope for real-world tasks, one important need in the near term is to find a better way to do actual web page/HTML design and link that to Zope's cool, object-database approach to Web site management and its apparent (we haven't tried it yet ... I'm going on the testimonials here) speed and technical superiority in database queries and interfaces. (CF uses ODBC ... ugh.) Ironically, using CF Studio would seem to be a good way to make that link. I wonder if there are other (free/cheap) HTML editor/site managers out there? -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- Tom Jenkins DevIS Linux & Samba 2.0 : The best Windows file server http://www.zdnet.com/sr/stories/issue/0,4537,2196106,00.html "Do not be overcome by evil, but overcome evil with good." Romans 12:21

1 0

Zope/IIS and IE5
by Phil Harris 17 Feb '99

17 Feb '99

Hi all, After playing about with Zope on Linux, (works great btw), I started looking at it on NT with IIS. After struggling for a while with the permissions using IE5b I decided to try NS4.5 and lo and behold it works. In a nutshell, the authentication system doesn't work when used from IE5b, but it does work with NS4.5, obviously IE5b is at fault here. Anybody got any ideas? TIA Phil wmlph(a)d032.ml.uwcm.ac.uk

2 1

RE: [Zope] DigiCool Attribution
by Paul Everitt 17 Feb '99

17 Feb '99

Martijn wrote: > Version 1.10.0 prerelease 1 has a builtin method > 'ZopeAttributionButton' > which is used in the standard_html_footer. By the way, I'm remiss in letting people know the schedule for 1.10. We wrapped it up yesterday, checking in Amos' changes for officially blessing import/export and putting it in the interface. We are going to use 1.10 in a consulting gig this week for which the customers are onsite here. If it passes muster we'll make binaries early next week and get them on the website. Since 1.10 corrects a lot of bugs, we hope to annoint it as the official stable release for quite a while. This will allow us to roll out, as experimental work, the advanced but risky stuff we're finishing up in Zope Classes, WebDAV, and ZServer. --Paul Paul Everitt Digital Creations paul(a)digicool.com 540.371.6909

1 0

DigiCool Attribution
by Oleg Broytmann 17 Feb '99

17 Feb '99

Hello! Is there any simple way I can satisfy Attribution, as it stated in Zope License? It's say: "web site must provide attribution by placing the accompanying "button" and a link to the accompanying "credits page" on the website's main entry point." Would it be enough if I create a button pointing to a copy of copyright (as on manage_copyright screen)? I think there should be standard tag for this, like standard_html_header:  or such... Oleg. ---- Oleg Broytmann National Research Surgery Centre http://sun.med.ru/~phd/ Programmers don't die, they just GOSUB without RETURN.

4 4

more help with externs, please . . .
by David Wagle 17 Feb '99

17 Feb '99

Ok, I thought I had a handle on this after Amos' message yesterday . . .but I clearly don't have a clue as to what is going on. I have the following external method: def dwtest(self,REQUEST=None): "find out what the hell is going on with REQUEST headers" if REQUEST is None and hasattr(self,'REQUEST'): REQUEST=self.REQEUEST return "REQUEST is None and hasattr(self,'REQUEST')" elif REQUEST is None and not hasattr(self,'REQUEST'): return "REQUEST is None. Self does not have attr REQUEST" elif REQUEST is not None and hasattr(self,'REQUEST'): return "REQUEST is not None. self has attr REQUEST" elif REQUEST is not None and not hasattr(self,'REQUEST'): return "REQUEST is not none. self does not have attr" If I call it with the URL, I get the expected: REQUEST is not None. self has attr REQUEST If I call it with the DTML  I get: REQUEST is None. Self does not have attr REQUEST" ??? I am missing somethign really fundamental here, and I don't know what it is. Help please. ... daw --------------------------------------------- A computer without Microsoft products is like a dog without bricks chained to its head ---------------------------------------------

3 2

RE: [Zope] DigiCool Attribution
by Paul Everitt 17 Feb '99

17 Feb '99

Oleg wrote: > Is there any simple way I can satisfy Attribution, as it > stated in Zope > License? It's say: "web site must provide attribution by placing the > accompanying "button" and a link to the accompanying "credits > page" on the > website's main entry point." The license has changed -- check out: http://www.zope.org/License/ With that in mind, we encourage everyone to help us spread the good word for Zope. To facilitate this we made a change for Zope 1.10 to add a standard method you can insert into your documents, similar to your proposal. Needless to say, I'd love to see you display Zope as prominently as possible without it becoming intrusive. --Paul Paul Everitt Digital Creations paul(a)digicool.com 540.371.6909

1 0

Zope lobbyists, anyone?
by Chris Larson 17 Feb '99

17 Feb '99

I'm trying to get Zope running at my ISP, and want to be able to lobby them as best I can. To this end, I'm trying to get as best a handle on the, um, environmental impact, let's say, of running Zope as I can. Can someone help me out here, mebbe point me to an overview of what exactly is required, vis-a-vis system resources? My site at this point is fairly low traffic, and fairly simple. I've posted a couple of short responses I've had in discussion with the admin below. I'd like to be able to respond to him with as much technical impact info as possible. How much memory is expected to be used, is the program gonna be launching multiple processes, etc. As I understand it, Zope launches one process, and that's it. Part of the problem right now is that I'm playing with it in a user account that launches processes as user 'www', and that Zope is very unstable in the current environment. Apparently, nfs causes rapid corruption of the database, causing 'serious application errors' following which I have a hard time tracking down and/or killing the process(es) left behind. This situation should change soon, as far as the 'user' running the processes at least. In a nutshell, I'm a relatively uninformed newbie, with an intermittent relationship with unix (linux), and I'm trying to figure out why nfs kills Zope, while I wait for my ISP to migrate my Apache virtual account to a machine with Python 1.5.1, after which at least processes will be launched as my user. I desperately want to implement Zope, but am faced with the issues of system impact, of which I have no idea, with the probability that I'll be faced with a no-threads Python (RedHat default, see below), with an ISP who may be reluctant to allow me to implement Zope, depending on the above-mentioned impact, and with the fact that I'm probably be stuck dealing with nfs. If any of the above are killers, I'd appreciate a heads-up, and if anyone can provide me with a concise overview of what Zope'll do to the systemm and/or other issues that might frighten an ISP away, I would be eternally (well, indefinitely, at least) indebted. Actually, it would be nice if the resource stuff were covered in the ISP FAQ, no? Much, much thanks, Chris Larson ----- On Fri, 12 Feb 1999, Chris Larson wrote: > Does the system kill long-running processes initiated by users who are > not logged in? If I have a cgi-wrapper that launches a process, will that > process remain running, as it should, or will it be killed? It only kills them when the parent process dies. A long-term process may never die if the parent's don't die. BTW, I have had to kill copies of your python program on www-01 and www-03 earlier today and last Saturday. They seem to never die (because our webserver does not die and they don't exit on their own). Have you ever looked at other versions of this program? It seems that the one you are using is trying to use an awful lot of resident memory on the systems. If it gets to be repetitive that I am killing your processes, then I may have to disable your program alltogether (removing execute bits). --- On Fri, 12 Feb 1999, Chris Larson wrote: > I need access to a Python 1.5.1 or higher, compiled with threads. We're just going to be using the default redhat-supported version of python (their compile). I don't know if it is suitable for what you need. >I need to be running everything on a machine where nfs will not come into >play ... I've been able to get my application running on www-01, through my > user account, but within a few minutes, everytime, the db becomes > corrupted and I need to reinstall from scratch. You might consider a database such as MySQL which we do have setup on our system and you can use it along with php or perl to handle a database. Those databases are located on local directories and each virtual server has a mysql server running on it.

3 2

MailHost patch, final version
by Martijn Pieters 17 Feb '99

17 Feb '99

Right, On the now well know locations: http://www.xs4all.nl/~zmj/Zope/MailHost.1.9.0.patch and http://www.xs4all.nl/~zmj/Zope/MailHost.1.10.0pr1.patch you'll find a patch that will actually give you a fully RFC821 compliant MailHost object. - All datalines sent to the SMTP server use \r\n line-endings. - The body (and only the body) of the message is quoted-printable encoded. - The appropriate MIME headers are added. This way, all SMTP servers should be able to process the email, and any clients receiving it will decode the quoted-printable encoding. Yes I know, the SendMail.send method has an argument 'subj' that is actually never used, but this was allready the case with the original DC code. I'll leave it up to them to remove it. -- M.J. Pieters, Web Developer | Antraciet http://www.antraciet.nl | Tel: +31-35-6254545 Fax: +31-35-6254555 | mailto:mj@antraciet.nl http://www.antraciet.nl/~mj | PGP: http://wwwkeys.nl.pgp.net:11371/pks/lookup?op=get&search=0xA8A32149 ------------------------------------------

1 0