December 2003 - Zope

grup and ldapuserfolder
by Bin Zhang 06 Dec '03

06 Dec '03

I am trying to set up GRUP working with LDAPUserFolder. The LDAP is actually an Exchange 5.5 box. LDAPFolder works fine, but I have not been able to see any users in GRUP. Can someone help? Thanks bz

2 1

RedHat Alternative...
by Charlie Jones 05 Dec '03

05 Dec '03

With the recent departure of Redhat Linux from the "free" distros, can somebody recommend a version that is stable with the latest release of production releases of Zope/Plone. charlie

9 8

print
by Andre Meyer 05 Dec '03

05 Dec '03

Hi All In a file-system-based product, when I print some output in a Python method, where does the output go? There is no command line output as far as I can see. It would be nice to at least print some results when things go wrong. Any ideas? thanks André

2 1

Zope URLs and spiders' behavior
by Sean Lee 05 Dec '03

05 Dec '03

I wonder if anyone has experienced this: a) Google displays links to Zope site with a space in them For example, the search result has the right Title (and click on it works), but underneath the page summary the URL (some, not all) may be displayed like this http://domain.com/dir/ subdir/index_html b) Spiders download non-existing directories For example, the site has two directories, site.com/dir1, site.com/dir2 The weird thing is that dir1 can be accessed in these (and more) ways: http://site.com/dir2/dir1 http://site.com/dir1/dir2/dir1 Unnecessary to mention, the spider keeps downloading all the time. The worst of all is, the directories really can be accessed that way. Why is this possible? I noticed some spiders (Altavista, for example) are more likely to do that, some less. I blocked some that I can, the others are too important to block. Thanks Zoper 2.5.x

3 4

Problems with mime types
by Juho Vuori 05 Dec '03

05 Dec '03

Hello. I ended up working with zope version 2.6.3 without knowing much about it. One of the services it is running is a message board that allows file attatchments in the messages. It works fine most of the time, but say open office .sxw files end up being octet streams and become difficult for users to open and they complain. I've been peeking around zope and found some references to mime types and tried adding the mime types somewhere as well, but my modifications don't seem to have any effect. That's propably because I didn't know at all what I was doing and where zope mime types are stored in the first place. Any ideas? Our messageboard attachments work briefly like this: The message is read in from a form: <form method="post" action="<dtml-var URL0>" enctype="multipart/form-data"> ... <input type="file" name="attached_file"><br><br> ... </form> Values from the form are then passed to a python script that gets the mime type from file headers(?): ... filetype = match('Content-Type: ([\w/]+)', attached_file.headers) ... Who sets these headers and how to change its behaviour? Cheers, Juho

2 1

General Best Practices Setup
by Michael Havard 05 Dec '03

05 Dec '03

I'm looking for some observations on a best practices for setting up ZOPE, which products to use and how to use them. We've been using ZOPE/CMF for about a year now and it's great but we have run into a few problems which I think are setup issues and so we are looking at the architecture again. The architecture I'm thinking of going with right now is: Apache (rewrite rule) -----> ZEO Client -----> ZEO Storage Server (using ZODB) I think Plone/CMF is definitely in as well as Archetypes, External Editor, BTreeFolder2, DCWorkflow, and EPOZ. We need a DEV and QA area on separate machines from the Production instances but an easy means of migrating templates, python, and infrastructure code from one stage to the next. In the past we've used ZSyncer with decent success. But is there a better way, maybe using something via CVS or some other product I'm not familiar with just yet? We have about 3000+ documents (policies, procedures, how-tos, bulletins), daily updates by the content managers (time sensitive information), and probably about 19gb of bandwidth a month consumed (2 million hits a month) with those documents served from the old IIS server. We've previously used DCOracle2 and DCOracleStorage to house field level content and may continue down that path for some simple feeds using ZSQL, but I think the majority will remain in the ZODB. Keeping it all in Oracle seemed too problematic with our setup. We continually have problems packing because of allocation of rollback segments and the database bloating exponentially (300 meg of uploaded code somehow consumed 1.8gb or Oracle table space). My main concern is that the system needs to be responsive. So if using another storage mechanism gains 500ms or more per transaction I'd be willing to look into it. But it also needs to be fairly easy to maintain because we're working with EDS. Our content developers have typically used Frontpage but are NOT good at creating HTML documents, script, CSS or anything that would be considered "CODE". Essentially they use Frontpage like you'd use Word or Powerpoint. So we'd like to have as many tools as possible to ease their content creation and formatting efforts. I'd thought about teaching them structured text and/or using EPOZ. But our goal is really to keep the content being entered as clean as possible from ye ol' legacy code (font, b, small, i, and such) and to take advantage of some existing CSS files that are shared between many of our sites/systems. Does anyone have any success stories, documentation, suggestions, or even failures so that I don't go down the wrong path. _________________________________________________________________ Get holiday tips for festive fun. http://special.msn.com/network/happyholidays.armx

2 1

put python script in other directory.
by Jianping Zhu 05 Dec '03

05 Dec '03

Diretory stucture is root --subdir1 --ssubdir1 --ssubdir2 --dev --mydev I need to search some Zobject in subdir1 from diretory mydev. following python scirpt works if i put it in sundir1 [code] from Products.PythonScripts.standard import html_quote request = container.REQUEST RESPONSE = request.RESPONSE catalog =container if not type: type = ['Units', 'UnitHeadquarter', 'SupervisoryUnit'] query = {} query['TemplateType'] = type if Unit: query['id'] = Unit query['sort_on'] = 'Title' results = catalog.searchResults(query) return results [/code] The problem is that i have not right to put scirpt in subdir1 i one can put script in dev->mydev [code] from Products.PythonScripts.standard import html_quote request = root.subdir1.REQUEST RESPONSE = request.RESPONSE catalog =root.subdir1 if not type: type = ['Units', 'UnitHeadquarter','SupervisoryUnit'] query = {} query['TemplateType'] = type if Units query['id'] = Unit query['sort_on'] ='Title' results =catalog.searchResults(query) return results [/code] But this way does not work, how can i change the script so i can put in in mydev and be able to search Zobjects in subdir1? Thanks

3 2

view rendered zsql query without execution?
by Elena Schulz 05 Dec '03

05 Dec '03

Hi, as the rendered zsql query source is the key for the cached results, it should be possible to view the source even if the query is failing due to db-problems. Has anybody done the trick yet or can give a hint how to do it? -- Thanks, Elena PS.: I know the one with <dtml-var "myZsqlMethod(REQUEST, src__=1)"> but this requires an errorfree execution of the method.

2 1

passing context using expr="dtml_method()"?
by John Withers 05 Dec '03

05 Dec '03

Folks, Newbieish question. When I am calling a dtml method using a dtml tag such as <dtml-var mymethod> it appears to pass the namespace of the calling document. However, when I am calling a dtml method using the expr syntax, such as <dtml-var expr="mymethod()"> it doesn't. For instance, a <dtml-var title_or_id> in the tag calling version will pick up the title of the calling document, and in the expr syntax version will not unless I explicitly pass the title_or_id variable as a parameter to the call. I am trying to organize a project into a semblance of order with subfolders at this point, and it would be very handy to be able to pass the calling document namespace wholesale to my dtml-methods, instead of having to pass each variable the method is going to use explicitly. Is there some simple way to do this that I am missing in the documentation? Thanks. -- John Withers <jwithers(a)nautilus.org>

3 2

python Script and download file
by Hervé Richard 05 Dec '03

05 Dec '03

Hi, It's probably a newbie question : I want to write a python script form action which return a file to download: In a folder i have 2 files ("A.tgz" and "B.tgz") and a Page Template Object (download.html). This page is a form with multiple choice in order to seclect a file to download (if the user select "A" button, he get the A.tgz file, if select "B" he get A.tgz ...) I want to write a python script method to render the file but i don't know how create the command to pass the file to downlown. I don't find any example, and i'm very glad if anybody can help me ! Thanks _____________________________________________________________________________ Herve RICHARD Herve.Richard(a)avignon.inra.fr INRA - Biometrie

2 3