- Zope - Zope lists

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

Re: [Zope] Feature Request Withdrawal(RE Non-English User Names)
by Simon Michael 17 Feb '99

17 Feb '99

"LEE, Kwan Soo" <kslee(a)plaza1.snu.ac.kr> writes: > *I have posted almost same message using www.egroups.com/list/Zope from an internet cafe just before the New Year's Day Holiday Season on Lunar Calendar began. But interestingly, that message does not seem get to the list and I I had the same problem (posting to the zope list via egroups didn't work). I joined the list at zope.org instead. Regards, -Simon

1 0

RE: [Zope] something like , but outside  tag
by Jeffrey Shell 17 Feb '99

17 Feb '99

> From: Service Informatique CHSR [mailto:chsrinfo@guetali.fr] > Hello, > > I wonder if something behaving like  but outside > the  tag exist? There's the REQUEST variable QUERY_STRING  Part of the CGI Environment actually. It's everything that occurs after the ? in the URL of a GET. .jPS

1 0

Location of sequence-previous, sequence-next
by Shaw, Howard 17 Feb '99

17 Feb '99

I have not looked at the internal code, so I don't know what is involved, but I would very much prefer it if the 'sequence-previous' data and the 'sequence-next' data were each available at every phase of the  tag. I would like to do:  <tr> <td>Blah </td> <td>Blah </td> <td>Blah </td> <td>Blah </td> </tr>  Previous href Next href   where basically both the previous pointer and the next pointer show up at the end of the table. At the moment I am using this kludge:   <tr> <td></td> <td></td> <td><a href="mailto:"></td> <td><a href="mailto:"></td> </tr>    <tr> <td></td> <td></td> <td><a href="mailto:"></td> <td><a href="mailto:"></td> </tr>   <tr><td colspan=4><table width=100%><tr>    <td><center><a href="query_start=">Previous</a></center></td>  <td><center>Previous</center></td>     <td><center><a href="query_start=">Next</a></center></td>  <td><center>Next</center></td>    I don't know what the overhead is for performing that  twice, but I don't like it. It makes the code look kludgey and is a pain to work with. This code would be much cleaner and more understandable if the 'previous-sequence' stuck around until 'sequence-end'. Or for that matter, if there was a way, within an if block, to set variables in the surrounding namespace as opposed to an internal #with grouping. Then I could do like this:  <tr> <td>Blah </td> <td>Blah </td> <td>Blah </td> <td>Blah </td> </tr>  Set 'my_previous_blah' equal to 'previous_blah'   Previous href (using my_sequence-prev) Next href (using sequence-next)   which would still be a kludge, and yet still be easier to understand.

1 0