- Zope-Dev - Zope lists

RDBM Connections per User
by Christian Theune 05 Jun '01

05 Jun '01

Hello out there. I had following problem: In a software, using zope as appserver, we needed to have every user a single connection to the database (postgres 7.1) to have access- checks on database level. I hacked into the Connection code and found, that the interfaces are really hard to be used for this feature. The thing is, that many "people" are accessing the database connection handle directly whether I am using now a wrapper function "getConnection()" to get the current connection or open one. I am reading the current user out of the REQUEST["AUTHENTICATED_USER"] and am storing the handles in an _v_connections Dictionary, which is explicitly not stored in database, but lets every user have exactly one connection. Any comments on this? -- Christian Theune - ct(a)gocept.com gocept gmbh & co.kg - schalaunische strasse 6 - 06366 koethen/anhalt tel.+49 3496 3099112 - fax.+49 3496 3099118 mob. - 0178 48 33 981 reduce(lambda x,y:x+y,map(lambda x:chr(ord(x)^42),tuple('zS^BED\nX_FOY\x0b')))

3 2

manage_main modification time wrong
by Andrew Sydelko 05 Jun '01

05 Jun '01

Our Zope installation has this really strange problem associated with the displayed "Modification Time" shown for each file in the /manage_main of a folder. The time that is shown is exactly one-hour later than the system time. I think I've tracked down this to the bobobase_modification_time variable. Both reference to <dtml-var "ZopeTime()"> and <dtml-var "_.DateTime()"> show the system time. <dtml-var "_.DateTime().timezone()"> shows US/Eastern. This is really not our correct timezone, since we are in West Lafayette, Indiana, which has it's own timezone: US/East-Indiana, but where does Zope/python get this from? Any ideas why bobobase_modification_time is off by one hour? --andy. Purdue Engineering Computer Network

1 0

RE: [Zope-dev] SOAP again (or, xml-rpc client for MSIE)
by Adrian Hungate 05 Jun '01

05 Jun '01

What are you actually trying to achive? Adrian... Phil Harris wrote: > > Brad, > > There are a few COM/COM+ components around that do xml-rpc. > > I use them with MS Word to write Word files directly to Zope for instance. > > Some are better than others, but ymmv so I'll hesitate to recommend one. > > At least one of them is listed on xmlrpc.com. > > Having them you could script them via jscript, and basically open up the > connection easily.

3 2

RE: [Zope-dev] dtml-in batching improved
by Adrian Hungate 05 Jun '01

05 Jun '01

Back before the zope.org product directory became so easy to navigate I wrote an index that I used to use which has batch links top and bottom: http://www.zope.org/Members/haqa/productlist/view_source Dunno if this helps... Adrian... -----Original Message----- From: Chris Withers [mailto:chrisw@nipltd.com] Sent: Wednesday, 30 May 2001 22:03 To: Ivo van der Wijk; zope-dev(a)zope.org; andyd(a)nipltd.com; Simon Coles Subject: Re: [Zope-dev] dtml-in batching improved > When using batching in dtml-in, why is 'previous-sequence' only defined at > the first iteration of the current batch? And why is 'next-sequence' only > defined at the last iteration of the current batch? I know this problem ;-) I had it too, and stuck something in the collector about it. I did manage to get the layout you're after without modifying Zope though (see the search page on www.nipltd.com) If you want the code, maybe someone at NIP could fish it out, if you can wait until next Thursday, I'll grab it myself :-) cheers, Chris _______________________________________________ Zope-Dev maillist - Zope-Dev(a)zope.org http://lists.zope.org/mailman/listinfo/zope-dev ** No cross posts or HTML encoding! ** (Related lists - http://lists.zope.org/mailman/listinfo/zope-announce http://lists.zope.org/mailman/listinfo/zope )

1 0

aq_explicit bug?
by Chris Withers 05 Jun '01

05 Jun '01

This code from a python script gets called when iterating over either a list of objects or catalog brains: obj = _.getitem('sequence-item', 0).aq_explicit url = getattr(obj,'getURL',obj.absolute_url)() folderish = obj.isPrincipiaFolderish I would always expect to get an attribute error on the last line, since brains don't have that method, but I don't, folderish is always true since it appears to acquire the catalog's 'isPrincipiaFolderish'... What's going on here? cheers, Chris

1 0

memory eating Zope..
by cs＠comlounge.net 05 Jun '01

05 Jun '01

Hi there! I got another problem with my Zope server and that seems to be a memory hole.. I am using Zope 2.3.2 with Python 1.5.2 now on SuSE 7.0 (this seems to solve the hanging problem, which appeared on 7.1) and I use Python Script, ZSQL methods, ZMySQLDA and ZPatterns. Now using top I see over time how the memory consumption of the zope processes grows, which means I have to restart once a day as it then becomes too slow.. When looking at the refcount in the debug section I see the following: Acquisition.ImplicitAcquirerWrapper: 7509 DocumentTemplate.DT_Util.Eval: 719 Products.ZPatterns.Expressions.Expression: 716 ZPublisher.HTTPRequest.HTTPRequest: 457 ZServer.HTTPResponse.ZServerHTTPResponse: 450 ZPublisher.BaseRequest.RequestContainer: 447 So Acquisition.ImplicitAcquirerWrapper seems to be referenced quite often (this is in an early state..) So does anybody know what this might cause? Or maybe how to track that problem down? (seems to be a long way with that server to become it stable.. ;-) Thanks in advance! Christian -- COM.lounge http://comlounge.net/ communication & design info(a)comlounge.net

4 5

How do I create entries in another folder
by r.laarhoven＠aca.it 05 Jun '01

05 Jun '01

I am new with zope and i am trying to foundout something i found in the zope book on www.zope.org. >From the zope book i found this piece of code to create an entry in a guest book """ Create a guest book entry. """ # create a unique document id id='entry_%d' % len(context.objectIds()) # create the document context.manage_addProduct['OFSP'].manage_addDTMLDocument(id, title="", file=comments) # add a guest_name string property doc=getattr(context, id) doc.manage_addProperty('guest_name', guest_name, 'string') My questions are: what does OFSP means and how do I create the entry in another folder eg. GuestBookEntries? Please help me.

1 0

Boost.Python
by James Treleaven 05 Jun '01

05 Jun '01

Hi, I know that Zope has its own C++ binding mechanism for Python scripts, but are there any complications with respect to my using Boost.Python to bind any Zope Python scripts that I write to C++ code? cheers, James

3 2

SearchIndex: is this behaviour in 2.4?
by Erik Enge 04 Jun '01

04 Jun '01

Hi, sorry I can't test this myself. I get this behaviour in 2.3.2, and I find it strange. I'd classify it as a bug. If I pass this query to a TextIndex: (word1 OR word2) AND (word3) it is first translated to this: [['word1', 'or', 'word2'], 'and', ['word3']] which is fine. But then, after query_hook() in GlobbinLexicon.py: [['word1', 'or', 'word2'], ['word3']] and then the default_operator (Or) is used (by parse2()/query() in UnTextIndex.py, I think), and turns it into: [['word1', 'or', 'word2'], 'or', ['word3']] kinda not what I wanted. Since I always use parens around OR's I can just set the default_operator to And and this solves my problem, but it's still not working correctly. Is this fixed in Zope 2.4?

1 0

dtml-in batching improved
by Ivo van der Wijk 03 Jun '01

03 Jun '01

Hi all, When using batching in dtml-in, why is 'previous-sequence' only defined at the first iteration of the current batch? And why is 'next-sequence' only defined at the last iteration of the current batch? This behaviour makes it difficult to display a batch like this: -- begin sample -- Item N Item N+1 Item N+2 (link to previous items) (link to next items) -- end sample -- Because the code would be something like: -- begin sample -- <dtml-in something start=query_start size=3> [do something with sequence-item] <dtml-if sequence-end>  <dtml-if previous-sequence> (link to previous items) </dtml-if> <dtml-if next-sequence> (link to next items) </dtml-if> </dtml-if> </dtml-in> -- end sample -- However, if your batchsize is larger than 1, 'previous-sequence' will always be untrue at the last item (when sequence-item is true). The only solution would be implement a second dtml-in after the first one purely for displaying the previous link. Also, dtml-in would re-execute your "expr" (which it does not very efficiently), so if you want to avoid doing the "expr" twice, you'd have to store it temporarily using dtml-let (or my dtml-set tag ;). The end result would be: -- start sample -- <dtml-let foo="expr"> <dtml-in foo start=start_query size=3> [do something with sequence-item] </dtml-in> <dtml-in foo start=start_query size=3>  <dtml-if sequence-end> <dtml-if previous-sequence> (link to previous items) </dtml-if> <dtml-if next-sequence> (link to next items) </dtml-if> </dtml-if> </dtml-in> </dtml-let> -- end sample -- And at this point I really would start to implement the batching myself in python :) Of course, basically the same applies if you want both the previous/next link at the top of the batch display. (Imagine you want both! :) Either I'm missing something really obvious here, or the patches below make sense. There are two versions: - patch #1: This version fixes the issue above by enabling the values of previous/next-sequence throughout the entire batch (and other related variables as well: X-sequence, X-sequence-start-index, X-sequence-end-index, X-sequence-size) (which may break existing sites), and - patch #2: This version introduces new variablenames, leaving the behaviour of previous/next-sequence. A third alternative would be to only define the variables at the top and bottom of the batch. Both versions fix (I think) a grammatical error. It's up to the DC guys if they feel like incorporating either of these patches in a next release :) patch #1: define variables everywhere in the batch This patch makes sure that the following variables are available through each iteration of dtml-in, for each item. !!!!! This patch may break alot of dtml code !!!!!! next-sequence next-sequence-start-index next-sequence-end-index next-sequence-size previous-sequence previous-sequence-start-index previous-sequence-end-index previous-sequence-size -- patch begins here -- *** Zope-2.3.2-orig/lib/python/DocumentTemplate/DT_In.py Fri Apr 13 21:30:38 2001 --- Zope-2.3.2-src/lib/python/DocumentTemplate/DT_In.py Tue May 29 19:52:36 2001*************** *** 583,611 **** for index in range(first,end): # preset kw['previous-sequence']= 0 ! kw['next-sequence']= 0 # now more often defined then previously # ! if index==first or index==last: ! # provide batching information ! if first > 0: ! pstart,pend,psize=opt(0,first+overlap, ! sz,orphan,sequence) ! if index==first: kw['previous-sequence']=1 ! kw['previous-sequence-start-index']=pstart-1 ! kw['previous-sequence-end-index']=pend-1 ! kw['previous-sequence-size']=pend+1-pstart ! try: ! # The following line is a sneaky way to ! # test whether there are more items, ! # without actually computing a length: ! sequence[end] ! pstart,pend,psize=opt(end+1-overlap,0, ! sz,orphan,sequence) ! if index==last: kw['next-sequence']=1 ! kw['next-sequence-start-index']=pstart-1 ! kw['next-sequence-end-index']=pend-1 ! kw['next-sequence-size']=pend+1-pstart ! except: pass if index==last: kw['sequence-end']=1 --- 583,610 ---- for index in range(first,end): # preset kw['previous-sequence']= 0 ! kw['next-sequence']= 0 # now more often defined than previously # ! # provide batching information ! if first > 0: ! pstart,pend,psize=opt(0,first+overlap, ! sz,orphan,sequence) ! kw['previous-sequence']=1 ! kw['previous-sequence-start-index']=pstart-1 ! kw['previous-sequence-end-index']=pend-1 ! kw['previous-sequence-size']=pend+1-pstart ! try: ! # The following line is a sneaky way to ! # test whether there are more items, ! # without actually computing a length: ! sequence[end] ! pstart,pend,psize=opt(end+1-overlap,0, ! sz,orphan,sequence) ! kw['next-sequence']=1 ! kw['next-sequence-start-index']=pstart-1 ! kw['next-sequence-end-index']=pend-1 ! kw['next-sequence-size']=pend+1-pstart ! except: pass if index==last: kw['sequence-end']=1 -- end -- patch #2: introduce new variables This patch introduces the variables batch-next-sequence batch-next-sequence-start-index batch-next-sequence-end-index batch-next-sequence-size batch-previous-sequence batch-previous-sequence-start-index batch-previous-sequence-end-index batch-previous-sequence-size Which behave exactly the same as the variables withouth 'batch-' in front, but these are available with each iteration/item of dtml-in, not just the first/last. This patch should not break any existing dtml code. -- patch begins here -- *** Zope-2.3.2-orig/lib/python/DocumentTemplate/DT_In.py Fri Apr 13 21:30:38 2001 --- Zope-2.3.2-src/lib/python/DocumentTemplate/DT_In.py Tue May 29 20:23:24 2001 *************** *** 583,589 **** for index in range(first,end): # preset kw['previous-sequence']= 0 ! kw['next-sequence']= 0 # now more often defined then previously # if index==first or index==last: # provide batching information --- 583,609 ---- for index in range(first,end): # preset kw['previous-sequence']= 0 ! kw['next-sequence']= 0 # now more often defined than previously ! # provide batching information ! if first > 0: ! pstart,pend,psize=opt(0,first+overlap, ! sz,orphan,sequence) ! kw['batch-previous-sequence']=1 ! kw['batch-previous-sequence-start-index']=pstart-1 ! kw['batch-previous-sequence-end-index']=pend-1 ! kw['batch-previous-sequence-size']=pend+1-pstart ! try: ! # The following line is a sneaky way to ! # test whether there are more items, ! # without actually computing a length: ! sequence[end] ! pstart,pend,psize=opt(end+1-overlap,0, ! sz,orphan,sequence) ! kw['batch-next-sequence']=1 ! kw['batch-next-sequence-start-index']=pstart-1 ! kw['batch-next-sequence-end-index']=pend-1 ! kw['batch-next-sequence-size']=pend+1-pstart ! except: pass # if index==first or index==last: # provide batching information -- To apply either of them, cd to your SOFTWARE_HOME (i.e. Zope-2.3.2-src) and patch -p1 < /location/of/patch NOTE! It only makes sense to apply one of these patches, not both! Cheers, Ivo -- Drs. I.R. van der Wijk -=- Brouwersgracht 132 Amaze Internet Services V.O.F. 1013 HA Amsterdam -=- Tel: +31-20-4688336 Linux/Web/Zope/SQL Fax: +31-20-4688337 Network Solutions Web: http://www.amaze.nl/ Consultancy Email: ivo(a)amaze.nl -=-

3 7