- Zope - Zope lists

export Z_DEBUG_MODE=no -- broken?
by Michael Hamilton 24 Feb '00

24 Feb '00

Inside lib/python/ZPublisher/HTTPResponse.py there is the following code: def _traceback(self,t,v,tb): tb=self.format_exception(t,v,tb,200) tb=join(tb,'\n') tb=self.quoteHTML(tb) if self.debug_mode: _tbopen, _tbclose = '<PRE>', '</PRE>' else: _tbopen, _tbclose = '' return "\n%s\n%s\n%s" % (_tbopen, tb, _tbclose) which seems to be a way of switching off traceback dumps in the browser window and have them wind up in an html comment instead. How do I enable this feature. I tried setting Z_DEBUG_MODE to something other that 1 or yes, but it doesn't seem to do anything. If I manually hack the above code to force debug_mode off, then it does what I want. --- Michael Hamilton ( http://home.clear.net.nz/pages/michael.hamilton )

1 0

RE: [Zope] [SiteAccess/GUF -- was (no subject)]
by Marcus Collins 24 Feb '00

24 Feb '00

We're using SiteRoot (1.0.0) and Generic User Folder (1.1.0) on Zope 2.1.3 src on our site, without any hassles. When you have a SiteRoot, you need to use http://server_name/zope_path/__no_before_traverse__/manage to bypass SiteRoot's rewriting of the URL and get to the management interface. hth, -- Marcus > -----Original Message----- > From: Curtis Maloney [mailto:curtis@umd.com.au] > Sent: 24 February 2000 07:58 > To: zope(a)zope.org > Subject: Re: [Zope] (no subject) > > > On Thu, 24 Feb 2000, you wrote: > > ----- Original Message ----- > > From: Curtis Maloney <curtis(a)umd.com.au> > > > SiteAccess doesn't seem to like working with GUF all that much. > > > > Could you give me a series of steps to reproduce the > problem? Also, what > > version of GUF and SiteAccess are you using? > > > > Cheers, > > > > Sure thing... I'll retell them as accurately as i can.. > > to my knowledge, all programs are the latest versions. > > 1) Installed Zope 2.1.4 from src. > 2) Added a couple of neccesary products (XML Document, ZMySQLdb, ...) > 3) Developed site > 4) Added SiteAccess object > 5) created SiteRoot object in project folder: > Base = <company url>/foldername > Path = / > 6) Added GUF product > 7) Deleted acl_users from project folder > 8) Created GUF in project folder. > > From that point on, even though i was logged in as superuser, > i could not get > to ANY management screens in the project folder. > > > Evan @ 4-am & digicool > > Anything else you want to know? > > Have a better one, > Curtis. > > _______________________________________________ > Zope maillist - Zope(a)zope.org > http://lists.zope.org/mailman/listinfo/zope > ** No cross posts or HTML encoding! ** > (Related lists - > http://lists.zope.org/mailman/listinfo/zope-announce > http://lists.zope.org/mailman/listinfo/zope-dev ) >

1 0

Re: [Zope] (no subject)
by Curtis Maloney 24 Feb '00

24 Feb '00

On Thu, 24 Feb 2000, you wrote: > ----- Original Message ----- > From: Curtis Maloney <curtis(a)umd.com.au> > > SiteAccess doesn't seem to like working with GUF all that much. > > Could you give me a series of steps to reproduce the problem? Also, what > version of GUF and SiteAccess are you using? > > Cheers, > Sure thing... I'll retell them as accurately as i can.. to my knowledge, all programs are the latest versions. 1) Installed Zope 2.1.4 from src. 2) Added a couple of neccesary products (XML Document, ZMySQLdb, ...) 3) Developed site 4) Added SiteAccess object 5) created SiteRoot object in project folder: Base = <company url>/foldername Path = / 6) Added GUF product 7) Deleted acl_users from project folder 8) Created GUF in project folder. >From that point on, even though i was logged in as superuser, i could not get to ANY management screens in the project folder. > Evan @ 4-am & digicool Anything else you want to know? Have a better one, Curtis.

1 0

ZODB 7 thread limit?
by Brian Takashi Hooper 24 Feb '00

24 Feb '00

Hi there, I was just wondering if any one can explain the reasons for the 'pool_size=7' and 'version_pool_size=3' numbers which are part of the defaults for the ZODB (lib/python/ZODB/DB.py)? Is there something behind the magic numbers, or should I change them only at great risk to myself and my Zope installation? I'd like to raise the number of threads active on my site above the 7 limit... Searching for ZODB wisdom-ly yours, --Brian Hooper

1 0

cataloging nodes of an XML Document?
by Anthony Baxter 24 Feb '00

24 Feb '00

Has anyone tried to use ZCatalog to catalog the nodes of an XML Document? Trying the simple Find to ZCatalog approach fails... -- Anthony Baxter <anthony(a)ekit-inc.com> ekit.com, Level 3, 28 Clarendon St, Sth Melbourne Australia 3205.

2 1

Re: cataloging nodes of an XML Document?
by Anthony Baxter 24 Feb '00

24 Feb '00

>>> Anthony Baxter wrote > Has anyone tried to use ZCatalog to catalog the nodes of an XML Document? > > Trying the simple Find to ZCatalog approach fails... Ah - the problem here is my patch for making ZCatalog catalog acquired objects is breaking things. XMLDocument.objectIds() returns values that you can't access with XMLDocument.foo Fixed version of my patch attached... (line numbers might be slightly wonky) Anthony

1 0

urgent UserDB question
by Sam Gendler 24 Feb '00

24 Feb '00

Has anyone seen the following behaviour? If I set the UserDb to be in cookie mode and I access a page that I am unauthorized to view, I get the docLogin page. The docLogin page has a built in html header, and when it gets delivered the standard_html_header also gets included, so there are 2 headers. However, if a user is trying to access a document inside a folder that needs authorization, the user is unable to retrieve the standard_html_header, so they get an ugly page, lacking all of the header content I thought I would resolve this by adding the contents of the standard_html_header and footer to the docLogin, and then putting a special hack in standard_error_message that checks for a LoginRequired Exception, and skips the inclusion of the standard_html_header if serving the docLogin. However, as soon as I replace the top of the docLogin property of the UserDb, instead of sending the docLogin form, it raises the Unauthorized exception, which results in a username/password dialog box from netscape. With the header, by docLogin ends up being about 4200 bytes. Is there maybe a 4096bytes limit on the docLogin property? Has anyone else solved this problem? If so how. Please cc my own email account with your response. --sam -- -------------------------- Sam Gendler CTO, Impossible, Inc. 1222 State St. Suite 250 Santa Barbara, CA. 93101 p: 805-560-0508 f: 805-560-0608 c: 805-689-1191 e: sgendler(a)impossible.com

1 0

(no subject)
by Curtis Maloney 24 Feb '00

24 Feb '00

Greetings, Well... looks like i've found out the problem... SiteAccess doesn't seem to like working with GUF all that much. After removing the SiteAccess product, i had 0 (cont them, ZERO) problems getting back into my directory... so.... all those with their brains in D (for Damage) will now be expecting me to ask something like "How do I stop this, because I need SiteAccess!??!!?!" And so I don't need to ask... just wait for the reply. Have a better one, Curtis.

1 0

XMLDocument "WrongDocument" error and object comparision...
by Anthony Baxter 24 Feb '00

24 Feb '00

I'm playing with XML Document, and I'm finding that I'm getting a bogus WrongDocument error from the sample file. If I open up the 'slides' document and hit 'Change' I'm getting a WrongDocument exception at line 433 of Node.py. if document and ownerDocument and document != ownerDocument: raise ZDOM.WrongDocumentException() (I also see this when I try to create a new document) Putting a print "DOC", document, ownerDocument print "DOC2", id(document), id(ownerDocument) on the line before it gives back DOC <Document instance at 863bf90> <Document instance at 863bf90> DOC2 139736152 139811552 so they _look_ like the same object, but comparision is failing. On a hunch, I tried changing the code to compare document.aq_base and ownerDocument.aq_base, and this works. I have _no_ idea if this is correct, though. I'm running the latest CVS, including Jim's changes to Acquisition.c (r1.36) for numeric object comparision - could this be the problem? anyone? Anthony (patch to make it work for me: ) --- Node.py 2000/02/05 00:33:16 1.18 +++ Node.py 2000/02/24 04:17:55 @@ -430,7 +430,7 @@ raise ZDOM.HierarchyRequestException('Node type does not support children.') document=newChild.getOwnerDocument() ownerDocument=self.getOwnerDocument() - if document and ownerDocument and document != ownerDocument: + if document and ownerDocument and document.aq_base != ownerDocument.aq_base: raise ZDOM.WrongDocumentException() if newChild.getNodeType() == DOCUMENT_FRAGMENT_NODE: newChildren=newChild

1 0

FW: getPreviousSibling (PLEASE HELP)
by RC Compaan 24 Feb '00

24 Feb '00

-----Original Message----- From: RC Compaan [mailto:sparroy@adept.co.za] Sent: 20 February 2000 10:51 To: Zope Subject: FW: getPreviousSibling I still haven't found a solution for the following problem. Help would be greatly appreciated: -----Original Message----- From: RC Compaan [mailto:roche@up-front.co.za] Sent: 17 February 2000 01:57 To: Zope Subject: getPreviousSibling I have a list of folders containing thumbnail images. In each folder I make a link to the previous and next folder with "getPreviousSibling" and "getNextSibling". This works perfectly in all folders except for the first one. This is my code: <dtml-if getPreviousSibling> <dtml-with getPreviousSibling> <a href="<dtml-var absolute_url>"><< <dtml-var title_or_id></a> </dtml-with> </dtml-if> And Zope says: Error Type: SystemError Error Value: infinite recursion in document template My logic says <dtml-if getPreviousSibling> should return "none" for the first folder like <dtml-if getNextSibling> returns none for the last folder in the list. What could it be? roché

1 0