June 2001 - Zope-Dev

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

ANNOUNCE: Zope 2.4.0 alpha 1 released
by Brian Lloyd 03 Jun '01

03 Jun '01

Hello all, Zope 2.4.0 alpha 1 has been released - you can download it from Zope.org: http://www.zope.org/Products/Zope/2.4.0a1/ Highlights of what is new in Zope 2.4: - Zope 2.4 supports (and requires) Python 2.1 - New restricted execution architecture - WebDAV Level 2 locking support - New "WebDAV Access" permission - A new built-in 'Authenticated' role - Product refresh without restart during development - New "Drop-in indexes" support for ZCatalog Be sure to see the Zope 2.4 Migration Guide: http://www.zope.org/Products/Zope/2.4.0/Zope24MigrationGuide.html which outlines the new features and provides other important information for users of Zope 2.4. For more information on what is new in this release, see the CHANGES.txt and HISTORY.txt files for the release: http://www.zope.org/Products/Zope/2.4.0a1/CHANGES.txt http://www.zope.org/Products/Zope/2.4.0a1/HISTORY.txt For more information on the available Zope releases, guidance for selecting the right distribution and installation instructions, please see: http://www.zope.org/Documentation/Misc/InstallingZope.html Note that this is an alpha release, so it is available as a source distribution only. We will make binary releases available with the first beta release. Brian Lloyd brian(a)digicool.com Software Engineer 540.371.6909 Digital Creations http://www.digicool.com

10 9

WEBDav - ZClass Problems
by Eric Roby 03 Jun '01

03 Jun '01

I have two ZClasses, one with base class of CatalogAwareBase and ZDTMLDocument, the other with base class of _ZClass_for_Renderable and CatalogAwareBase. Manager role has the full set of permissions for the classes. ZInstances are created as expected, TTW. However, when I try to manage these ZInstances via a WEBDav client, the WEBDav Server simply doesn't show that the objects are there. Is the WebDav server setup to ONLY manage DTMLDocuments and DTMLMethods? I have had no problems with my WEBDav client when managing these traditional Zope objects. Is this a side effect of ZClasses or of CatalogAware? As a side, the FTP client has no problems with these ZInstances. What gives??? Eric

1 0

Re: [Zope-dev] Does creating REAL over-the-web Python code becomerealistic with Zope 2.4?
by Tres Seaver 02 Jun '01

02 Jun '01

"Joachim Werner" <joe(a)iuveno-net.de> wrote: >> > > this distinction so that all components can be edited via the >> > > filesystem and live in the object database. > >> >> This does worry me. I think I understand at least one motivation (ZEO), >> but... >> >> Well, as long as it doesn't compromise my ability to write Zope python >> code using filesystem tools, it should be OK. It just makes me nervous <grin>. We want to give th programmer the freedom to use comfortable tools, while still preserving the "distributability" of having the objects in the ZODB (replicating filesystem-based code is a hairy problem). > I guess the crucial point is that everything should be editable from BOTH > the filesystem (or a mountable Zope storage ;-)) and the web front end. > > Actually our motivation is a bit different from the one at DC. They mainly > see the programmer as the target. For us, it's the power user, who wants to > creat new objects, but only within a limited framework. For adding > completely new functionality, these people would rely on professional help. > But adding a new attribute to an object or changing a form should all be > possible over the web. We see FTP and especially WebDAV as the "best fit" technoligies for the "content owners"; we spent a good deal of effoert on cleaning up FTP/WebDAV for yesterday's CMF release, for instance. Tres. -- =============================================================== Tres Seaver tseaver(a)digicool.com Digital Creations "Zope Dealers" http://www.zope.org

2 1

Does creating REAL over-the-web Python code become realistic with Zope 2.4?
by Joachim Werner 02 Jun '01

02 Jun '01

Hi! When I read about the reload functionality of Zope 2.4, an old vision came back to my mind: If we can reload Python code at runtime, and we can also edit, generate and store files (containing code) via the web (as LocalFS or extFile prove), then it should be perfectly possible to have something like ZClasses (it terms of the over-the-web front-end) that spits out and dynamically loads REAL Python classes instead of the rather experimental simulation of classes we get with ZClasses, right? I'll skip the part about why we should not be able to manipulate Zope code over the web, which can be discussed separately ... What do you think about this idea? If it finds some friends, I'll set up a formal proposal for something like "ZClass-like features to build REAL Zope Python classes over the web". Joachim

6 8

RE: [Zope-dev] ANNOUNCE: Zope 2.4.0 alpha 1 released
by sean.upton＠uniontrib.com 01 Jun '01

01 Jun '01

My impression is that FileStorage implements a 32-bit id-type-thingy somewhere (look at ZODB docs, I think there is something about this somewhere), which limits it (in addition to the Linux kernel ext2 fs limit), to < 2GB. With 7.5 GB, I'd use a more advanced storage like the new - well, not quite released ;) - Berkeley (libdb3 based) storages, though I haven't tried anything like that myself... Does anyone have any comments on the reliability of PartitionedFileStorage in this regard as a temporary solution? Is anyone using PartitionedFileStorage in a production environment? Sean -----Original Message----- From: Holger Lehmann [mailto:lehmann@catworkx.de] Sent: Friday, June 01, 2001 2:33 AM To: Brian Lloyd; zope-dev(a)zope.org Subject: Re: [Zope-dev] ANNOUNCE: Zope 2.4.0 alpha 1 released Hi Brian, hi all, great to hear this. Now comes a tricky question: Suppose we are running a Linux Kernel 2.4.x and using the latest Python 2.x distribution. And supposing we can recompile all code if neccessary: How large can a Data.fs grow until it breaks ? We need to be able to store about 7.5 GBytes (note: Gigabytes) in the ZODB. I would rather use the ZODB than a LocalFS .... Thanks in advance - Holger Am Donnerstag, 31. Mai 2001 22:55 schrieb Brian Lloyd: > Hello all, > > Zope 2.4.0 alpha 1 has been released - you can download it from > Zope.org: > > http://www.zope.org/Products/Zope/2.4.0a1/ > > > Highlights of what is new in Zope 2.4: > > - Zope 2.4 supports (and requires) Python 2.1 > > - New restricted execution architecture > > - WebDAV Level 2 locking support > > - New "WebDAV Access" permission > > - A new built-in 'Authenticated' role > > - Product refresh without restart during development > > - New "Drop-in indexes" support for ZCatalog > > > Be sure to see the Zope 2.4 Migration Guide: > > http://www.zope.org/Products/Zope/2.4.0/Zope24MigrationGuide.html > > which outlines the new features and provides other important > information for users of Zope 2.4. > > For more information on what is new in this release, see the > CHANGES.txt and HISTORY.txt files for the release: > > http://www.zope.org/Products/Zope/2.4.0a1/CHANGES.txt > > http://www.zope.org/Products/Zope/2.4.0a1/HISTORY.txt > > For more information on the available Zope releases, guidance for > selecting the right distribution and installation instructions, > please see: > > http://www.zope.org/Documentation/Misc/InstallingZope.html > > Note that this is an alpha release, so it is available as a source > distribution only. We will make binary releases available with the > first beta release. > > > Brian Lloyd brian(a)digicool.com > Software Engineer 540.371.6909 > Digital Creations http://www.digicool.com > > > > > _______________________________________________ > 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 ) -- ----------------------------------- catWorkX GmbH Dipl.-Ing. Holger Lehmann Stresemannstr. 364 22761 Hamburg Tel: +49 (0700) catWorkX Tel: +49 (40) 890 646-0 Fax: +49 (40) 890 646-66 mailto:lehmann@catworkx.de http://www.catworkx.de http://www.catbridge.de _______________________________________________ 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 )

6 7

RE: [Zope-dev] ANNOUNCE: Zope 2.4.0 alpha 1 released
by sean.upton＠uniontrib.com 01 Jun '01

01 Jun '01

Perhaps I was looking at old docs, as the docs had indicated that filestorage uses 32-bit position pointers; I'm guessing that this has since changed... Sean -----Original Message----- From: R. David Murray [mailto:bitz@bitdance.com] Sent: Friday, June 01, 2001 9:54 AM To: sean.upton(a)uniontrib.com Cc: lehmann(a)catworkx.de; brian(a)digicool.com; zope-dev(a)zope.org Subject: RE: [Zope-dev] ANNOUNCE: Zope 2.4.0 alpha 1 released On Fri, 1 Jun 2001 sean.upton(a)uniontrib.com wrote: > My impression is that FileStorage implements a 32-bit id-type-thingy > somewhere (look at ZODB docs, I think there is something about this > somewhere), which limits it (in addition to the Linux kernel ext2 fs limit), > to < 2GB. With 7.5 GB, I'd use a more advanced storage like the new - well, > not quite released ;) - Berkeley (libdb3 based) storages, though I haven't > tried anything like that myself... FileStorage does *not* have a 2GB limit. The only 2GB limit is the old Linux filesystem limit. I know this, because I've had >2GB Data.fs files on FreeBSD. --RDM

1 0

SOAP again (or, xml-rpc client for MSIE)
by Brad Clements 01 Jun '01

01 Jun '01

Hate to bring this up again. 1999 Press release from DC http://www.prnewswire.com/cgi-bin/stories.pl?ACCT=104&STORY=/www/story/12-0… > An early 2000 Zope > release will include support for two protocols heavily supported by Microsoft, > the Simple Object Access Protocol (SOAP) and Web Distributed Authoring and > Versioning (WebDAV) protocol We have WebDav.. How's SOAP look? I'm looking for a decent RPC mechanism for MSIE 5.5 to talk to Zope. MS has their webService behavior that supports SOAP. But without SDL or WSDL in Zope it's gonna take a lot of code to get working. Otherwise, I could use an XML-RPC client for IE5.5, but I can't find one! Brad Clements, bkc(a)murkworks.com (315)268-1000 http://www.murkworks.com (315)268-9812 Fax netmeeting: ils://ils.murkworks.com AOL-IM: BKClements

3 2