September 2001 - Zope

Zope.org downtime
by Martijn Pieters 14 Sep '01

14 Sep '01

Zope.org will be restarted in about 5 minutes for a ZEO upgrade. We apologize for the inconvenience. -- Martijn Pieters | Software Engineer mailto:mj@zope.com | Zope Corporation http://www.zope.com/ | Creators of Zope http://www.zope.org/ ---------------------------------------------

1 1

Finding out what Operating System a site visitor is using
by Steven Grimes 14 Sep '01

14 Sep '01

I need to find out what operating system my clients are using. I know this information is available in the REQUEST object but I can not find a way to print that information on a page. For example: <html> <body> Hello <dtml-var AUTHENTICATED_USER>! You are running {Operating System}. </body> </html> I would like the Zope syntax to print the actual operating system used in the statement above. Thanks in advance, Steven Grimes Database Engineer sgrimes(a)atinucleus.com Accelerated Technology, Inc. 720 Oak Circle Drive East Mobile, AL 36609 TEL*: 251-661-5770 FAX*: 251-661-5788 www.acceleratedtechnology.com *Please note new area code "251" ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Over Ten Years of Providing Embedded Solutions ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

5 6

Redirecting to url with query information
by Michael Schulz 13 Sep '01

13 Sep '01

Hi, i am trying to redirect a request back to its origin-url after a user authentication using the response.redirect mechanism providing it with the url via http_referer. This works well unless the origin-url has a query section appended (e.g. http://www.xxx.yy/origin/url?id=007) Anybody has experienced the same and found a way how to solve that? No need to say that this query string is essential and it can't be omitted. TIA, Michael -- ----------------------------------------------------------- Michael Schulz in medias res Dipl.-Geologe Gesellschaft für Informationstechnologie mbH Sautierstr. 38, 79104 Freiburg 0761 55695-95 (Fax 96) mschulz(a)webgis.de www.webgis.de

4 3

Why can't I use a method of an objectmanger I am trying to add to ?
by Max Møller Rasmussen 13 Sep '01

13 Sep '01

I have an objectmanager that implements: def _getNewId(self): """ generates a new unique id for a message get new id and make a zero-padded string """ # 12 chars long, 999,999,999,999 messages max in one discussion! newID = '%012d' % self._newId self._newId += 1 self._p_changed = 1 return newID And it works nicely if I call it directly. But when I try to call the method from an object being added to the objectmanager like this: ------------------------------------------------------ def manage_addMessageAction(self, title='No title', comment='No comment', author='No author', email='', parent=None, REQUEST=None, RESPONSE=None): "Adds a message to the objectmanager" id = self._getNewId() .... etc. ------------------------------------------------------ I get an error: Error Type: AttributeError Error Value: _getNewId Traceback (innermost last): File C:\zope\zope240\lib\python\ZPublisher\Publish.py, line 223, in publish_module File C:\zope\zope240\lib\python\ZPublisher\Publish.py, line 187, in publish File C:\zope\zope240\lib\python\Zope\__init__.py, line 226, in zpublisher_exception_hook File C:\zope\zope240\lib\python\ZPublisher\Publish.py, line 171, in publish File C:\zope\zope240\lib\python\ZPublisher\mapply.py, line 160, in mapply (Object: manage_addMessageAction) File C:\zope\zope240\lib\python\ZPublisher\Publish.py, line 112, in call_object (Object: manage_addMessageAction) File C:\zope\zope240\lib\python\Products\mxm_Discuss\mxm_Message.py, line 106, in manage_addMessageAction AttributeError: (see above) ------------------------------------------------------ line 106 being "id = self._getNewId()" "manage_addMessageAction" is registered as a constructor. When it is being called from an ObjectManager, isn't the objectmanager passed to this unbound method? If so, why can't it find "_getNewId()" Regards Max M

3 2

Changing Zope COL ROW and WRAP in the TEXTAREA elements for Properties editing
by Ronald L. Chichester 13 Sep '01

13 Sep '01

I want to change TEXTAREA parameters, specifically the ROW (increase), COL (increase), and WRAP (=VIRTUAL) for editing purposes in the Properties tab. Is there any way that a ZCLass can do that? I have alot of text in a property and would like to have the same size TEXTAREA that is in the Contents tab for, say, DTML Method or DTML Document. Or is the only alternative to tweak zope specifically and then restart (and thus make it the same for every object)? Thanks in advance, Ron

2 1

Word wrapping the Text Editor in Zope
by Jan Smith 13 Sep '01

13 Sep '01

I know flash, PhotoShop, fireworks, illustrator, Dreamweaver and html - I am not a programmer. Occasionally I need to edit in the Zope text editing area. Is there anyway to word wrap the Zope code so you don't have to continually scroll sideways on some lines and loose your place. I find the Narrower, Wider, Taller, Shorter options do not fix this. I can widen the screen view by dragging on the edge of the text editing area but I can't I work out how to word wrap. Jan

2 1

Multiple ZCatalogs and automatic reindexing
by Milos Prudek 13 Sep '01

13 Sep '01

I'd like to have several searchable types in ZCatalog. Let's say I have ZClass Z1, Z2 and Z3. I need to catalog them SEPARATELY. 1. Do I need several ZCatalogs? 2. Do I need Zope 2.4.1? 3. For CatalogAware automatically reindexing, is it still neccessary to give the catalog the name "Catalog", and place it in root folder? The third question puzzles me the most, because it seems to prevent me from creating several ZCatalogs, if I need them reindexed automatically. -- Milos Prudek

2 1

Welcome to the Cmf-fr@factor-e.tv a new French mailing list about CMF!
by FR Chalaoux 13 Sep '01

13 Sep '01

Welcome to the Cmf-fr(a)factor-e.tv a new French mailing list about CMF! To post to this list, send your email to: cmf-fr(a)factor-e.tv General information about the mailing list is at: http://www.factor-e.tv/mailman/listinfo/cmf-fr by, FRC

1 0

Problems creating auto-navigation links (LONG)
by Dave Lehman 13 Sep '01

13 Sep '01

>I am having a lot of problems trying to get a breadcrumb or tree or >other page-by-page, automatically generated navigational scheme put >into place on my Zope site. I have a sense that I may be trying too >hard or that I've created some complexity here that I didn't need to >(or that my ISP has done so). I have just finished wrestling with exactly this problem! Coincidentally enough, also on a Zope site at Hurrah. (And I would strongly concur that David Elkins is very helpful and knowledgeable. I'm sure if you ask him, he'll shake his head and tell you all kinds of stories about me...) Here are the problems I encountered, and how I overcame them. In this example, I have a Zope CMF site inside a folder called "portal". My domain name points directly to the portal folder, so when users go to "http://www.mysite.com" they are already at the root of my CMF site, instead of having to go into the portal folder. That is: http://www.mysite.com is mapped to http://mysite.hurrah.com/cgi-bin/Zope.cgi/portal/ 1. If possible, breadcrumbs should display the friendly-reading title for each folder instead of the usually cryptic Id Most breadcrumb implementations that i had found just iterated back through the id's of all the parent folders. But instead of displaying: home/my_folder/folder125/some_story I wanted to display something like: Home / My Folder / Stories / The Quest for the Holy Grail However, it *should* use the Id if you forgot to assign a title to a certain folder. After going a couple of rounds with DTML and Python, and pleading my ignorance to the list, Tres Seaver finally did things The Right Way(tm) and supplied a Python script (see below) which returns a data structure that could be used to create the actual breadcrumbs. 2. Each breadcrumb should be hyperlinked to it's folder, except for the "leaf" breadcrumb, which is the page you're currently on. OK, now we're getting somewhere. This is minor, but took a bit of thinking on my part. I wanted each step of the breadcrumb "chain" to be a hyperlink to that actual folder (so that the user could step back to any part of the folder hierarchy) except for the leaf folder-- which is the user's current location. Once I had the right python script, this turned out to be just a TAL condition statement: tal:condition="repeat/crumb/end" 3. The hyperlinks returned by a breadcrumb should only start at the "root" of the CMF site, which isn't (necessarily) the same as the root of your Zope installation So I had it *almost* working, but the hyperlinks for each breadcrumb returned something like "http://mysite.com/portal/folder1" when what I wanted was "http://mysite.com/folder1". Initially, I wrote thought I would have to augment my python script to manually strip out the "portal" section of the URL, but that seemed unnecessarily ugly (and thank goodness it was). The key here was to create a "SiteRoot" object in my /portal folder to tell Zope that it should make this folder look like the root of the site to the website visitor. Now the "crumb/url" returned by the page template is correct. The code I used is shown below. It looks simple once you have all the pieces, but for a Zope-Newbie like myself, it took a bit of work to figure out how to get exactly what I wanted. But at least I had confidence all along that Zope *could* do what I wanted, if I just could figure out how... I'm sure all this is as clear as mud. Please let me know if you want more clarification. Regards, Dave -------------------- (This page template is pretty of ugly-- i'm obviously no ZPT guru-- so anyone should feel free to clean this up!) but does the job. It creates the breadcrumbs string, hyperlinked at each level except the leaf, and displays the Title for that folder, when possible.) "breadcrumb_zpt" (PageTemplate) * put in portal_skins/custom folder <div tal:define="crumbs here/breadcrumbs"> / title id <a href="url" tal:attributes="href crumb/url"> title id </a> </div> -------------------- "Breadcrumbs" (Python script) * put in root CMF folder ## Script (Python) "breadcrumbs.py $Revision: 1.1 $" ##bind context=context ##bind namespace= ##bind script=script ##bind subpath=traverse_subpath ##parameters=include_root=1 ##title=Return breadcrumbs ## from string import join result = [] portal_url = context.portal_url() if include_root: result.append( { 'id' : 'root' , 'title' : context.portal_properties.title() , 'url' : portal_url } ) relative = context.portal_url.getRelativeContentPath( context ) portal = context.portal_url.getPortalObject() for i in range( len( relative ) ): now = relative[ :i+1 ] obj = portal.restrictedTraverse( now ) result.append( { 'id' : now[ -1 ] , 'title' : obj.Title() , 'url' : portal_url + '/' + join( now, '/' ) } ) return result -------------------- SiteRoot object * if necessary, install the "SiteRoot" product * go to your CMF folder (or wherever you want the root to be) and create a SiteRoot object title: whatever base: http://www.mysite.com path: /

1 0

Re: [Zope] Retrieving Rows, Columns, and ??Column Headings?? from RDB?
by Oliver Bleutgen 13 Sep '01

13 Sep '01

> Using an external database such as MySQL, does Zope(2-3-2) > (or Python-in-Zope) provide any mechanism to retrieve the > names of the columns queried, rather than simply the DATA > within those columns? > Please allow me to obfuscate this question with an example > of what I need.. :-) several ways to achieve what you want: <dtml-with "user_query(fname='Bob')"> <dtml-in dictionaries> <dtml-var sequence-item> </dtml-in> </dtml-with> returns for each sequence item a dictionary of name:value pairs. Equivalent: <dtml-in "user_query(fname='Bob').dictionaries()"> "user_query(fname='Bob').names()" gives a list of the column names "user_query(fname='Bob').tuples()" returns a list(tuple?) of the result rows in a tuple (better try it, words lost me). So you could also combine names() and tuples() cheers, oliver

1 0