- Zope - Zope lists

RE: [Zope] sample site
by Geoff Nordli 29 Oct '99

29 Oct '99

Thanks. I downloaded it, and I am going to try to modify for the PygreSQL database. It would be great to create an area on the page with sample sites. so people can upload them to a common area. Or even have links pointing to sites like Zgold's. http://d031.ml.uwcm.ac.uk/z/zGold geoff > -----Original Message----- > From: Tres Seaver [mailto:tseaver@palladion.com] > Sent: Friday, October 29, 1999 11:44 AM > To: Geoff Nordli > Subject: Re: [Zope] sample site > > > Check the list over the last couple of days for a reference > to "ZGold". > > it was put together as a demo, and has an export file which > can be downloaded > and installed directly (I think). > > Tres > -- > ========================================================= > Tres Seaver tseaver(a)palladion.com 713-523-6582 > Palladion Software http://www.palladion.com >

2 1

Re: NotMail
by J Kinsley 29 Oct '99

29 Oct '99

I saw your post in the Zope archive regarding the imort error with NotMail. I had the exact same error on a Linux box and managed to fix it by doing the following: In NotMail.py at line 199-200, change: def __init__(self, id, title='', default_domain, smpt_server, smtp_port, imap_server, imap_port, REQUEST=None): to: def __init__(self, id, title='', default_domain='', smpt_server='', smtp_port=25, imap_server='', imap_port=143, REQUEST=None): Although I am not yet certain whether or not this modification causes any adverse effects somewhere else (to the best of my knowledge it should not as long as everything is configured properly), it does fix the SyntaxError: non-default argument follows default argument (line 199) problem. However, I have discovered more problems when I try to access NotMail, but I have not yet looked at them. I also made a modification to Connection.py similar to the one referenced in http://www.zope.org/pipermail/zope/1999-October/011505.html (knotmail) for IMAPAdapter.py (if 'NotMail' not in roles).... Anyway, hope this helps.... Regards Jarrod

1 0

Zope.org site text should wrap
by Thomas B. Passin 29 Oct '99

29 Oct '99

Many pages on the zope.org site are annoying to read because they don't word-wrap. This is probably caused by a width="100%" attribute in the main <table> tag. On most pages with this problem, getting rid of the widh="100%" lets the text wrap like it should. Whoever is the template custodian, please let our text wrap! Thomas Passin

3 2

Re: [Zope] Re: New Version of Zope RPMS Available
by Karl Anderson 29 Oct '99

29 Oct '99

"Jeff Rush" <jrush(a)timecastle.net> writes: > I think the problem is that I made the change to omit the ZSYSLOG > env var in the zserver.sh script for running PCGI+ZServer and you > are running the ZServer-standalone version, right? (hope, hope) yep. > I'm not sure why just running "python z2.py" still fails for you but > would like to hear of your results with the new RPMs. Sorry for > the hassle. Again, it was only minor - using the ZSYSLOG var requirs a listening logger & a ZSYSLOG_SERVER var. I commented out the ZSYSLOG var & both calling the script direct & using the init.d script works fine without the ZSYSLOG_SERVER var, so the new RPM should be fine for everyone without making them mess with the script. Thx for making the RPM. -- Karl Anderson kra(a)monkey.org <URL:http://www.pobox.com/~kra/>

1 0

Internal links?
by Ethan Fremen 29 Oct '99

29 Oct '99

Zopistas, I just noticed this making a howto for css- Zope automatically sets the <BASE> tag as the current URL. This is great, but in NS 4.7 this means that href="#foobar" comes out as href="www.zope.org/currenturl/#foobar". Which means when I click on it, it seems to have to load the page. I tried experimenting with ../currenturl#foobar, but it gives the same results. Is this a bug in NS or what? -- Ethan "mindlace" Fremen you cannot abdicate responsibility for your ideology.

1 0

Assorted questions (long).
by Otto Hammersmith 29 Oct '99

29 Oct '99

Hello all, FWIW, I've copied the UserDB product (1.3.0) from the old Zope.org site and created a package. It's cataloged now, but the URL is http://www.zope.org/Members/otto/userdb. I've downloaded the beta ZDiscussions product (very nice, BTW) and it has brought up a question about ZCatalog and classes that inherit from it. If an object inherits from ZCatalog does that mean it's sub-objects are only cataloged there? For example, if I create a ZDiscussion topic how can I have a site wide search interface that will also search the ZDiscussion items in that topic. I've done some experimenting and I've found out I can have a ZCatalog catalog the discussion itself, which is somewhat useful... but there doesn't seem to be a way to catalog the items outside the discussion. Wouldn't it make more sense for ZDiscussion to not inherit from ZCatalog but allow provide a search interface limited to searching discussion? It should be easy to limit the search based on meta-type, but I don't see a way to limit it to a single discussion. Also, to allow the possibility of keeping the discussion out of the default catalog it should be possible to create a new catalog as a sub-object of the ZDiscussion. On the other hand would it make sense (or is is possible already) to "chain" ZCatalogs? Just a few thoughts... :) My second question is on AUTHENTICATED_USER. What properties does it have? The DTML User's Guide says "This object currently provides no public attributes" but the source for ZDiscussions clearly uses some functions and attributes like getName() and 'email'. Is this something that UserDB gives you? (I haven't had the chance to play around with that much, yet.) TIA. -Otto.

1 0

Obtaining URL of Object w/Non-Var Name
by Jeff Rush 29 Oct '99

29 Oct '99

Alright, what's the trick to be able to retrieve the URL of a Zope object via DTML whose id has blanks or punctuation in it? For example, in a normally-named object called 'site_css', I'd do: Ex. 1 <LINK REL=stylesheet HREF=<dtml-var "site_css.absolute_url()"> TYPE="text/css"> But if the object is named 'site.css', I can use it w/o another layer/call or escape code. I tried the following (and a lot of others) and the _[] expression is returning a string, which then fails when it lacks a 'absolute_url' attribute: Ex. 2 <LINK REL=stylesheet HREF=<dtml-var "_['site.css'].absolute_url()"> TYPE="text/css"> Any solution must work with acquisition, as example one does, so I can keep the stylesheet in the site root folder. -Jeff Rush

3 3

RE: [Zope] sample site
by Dom Mitchell 29 Oct '99

29 Oct '99

> -----Original Message----- > From: Geoff Nordli [mailto:geoff@gnaa.net] > Sent: Friday, October 29, 1999 12:08 AM > To: 'Pavlos Christoforou' > Cc: 'Zope List (E-mail)' > Subject: RE: [Zope] sample site > > > That is great, but I was thinking of a full-blown site. Something > that you could import that has all the objects. That way you > can see the different objects that were created, and how they were > used. You get a sense of the whole picture. Maybe we need another option under Products, "Demo Sites". We could start by listing zGold, as well as Gerard's site. -Dom -- ********************************************************************** This email and any files transmitted with it are intended solely for the use of the individual or entity to whom they are addressed. The email has been swept by MIMEsweeper for the presence of computer viruses. www.palmerharvey.co.uk **********************************************************************

3 2

ANN: SiteAccess v0.1.3 (looking good)
by Evan Simpson 29 Oct '99

29 Oct '99

For anyone who hasn't seen the news item at zope.org, just thought I'd let you know that SiteAccess v0.1.3 is up. It's looking pretty solid at this point, and is in production use at several sites. The new version is at http://zope.org/Members/4am/SiteAccess as usual, and here are the new CHANGES.txt entries since v0.1.1: * Using DTML Docs/Methods as Access Rules should work better, and allow normal acquisition. You won't have permissions for anything unless you give the rule a Proxy Role, though. * __no_before_traverse__ at the start of a URL path now persists in generated URLs, making debugging easier. * Fixed absolute_url() of objects acquired from above the point at which setURL was called (usually by a SiteRoot). (thanks again to Bruce Perens and technocrat.net) * Added Base and Path properties to SiteRoot, with Path defaulting to '/', the most commonly used value. If these are blank, it will search for SiteRootBASE and SiteRootPATH. * REQUEST.setURL now accepts either a string or sequence of strings for its path argument. Evan @ 4-am

1 0

installing ZPyGreSQLDA
by Geoff Nordli 29 Oct '99

29 Oct '99

I am trying to get "ZPyGreSQLDA 0.03" connection working, and have a few questions about installation. I can connect to the database, but am unable to run any queries against it. I was reading some of the archived mail, and Bruce Elrick states that you just have to install the product in the /lib/python/products directory, and restart Zope. This is supposed to work. In the README.txt file you have to do what Bruce says, and compile some code: make -f Makefile.pre.in boot make then copy the file _pgmodule.so from the src to the ../src Which is right? I have RH 6.1, Zope RPMs from Jeff, + all the Postgres RPMs from RedHat. thanks, Geoff Nordli

3 3