- Zope-Dev - Zope lists

RE: [Zope-dev] ZFormulator review...
by Martijn Pieters 11 Jan '00

11 Jan '00

> -----Original Message----- > From: Chris McDonough [mailto:chrism@digicool.com] > > I'm working on a product that does a similar thing (except > the object is > a "font manager"). I'd be very interested to see how you're doing it > Martijn... > Here's what I do: def initialize(context): # Add the scheduler to the Zope root, and make it undeletable. # I know, this is an extreme hack. But supplying a patch for # OFS/Application.py is just not practical. # We also add a hook to the Product object, so we can make # the scheduler object removable again when this Product is deleted. app = context._ProductContext__app # Don't you love private name mangling if not hasattr(app, 'Scheduler'): oScheduler = ZScheduler.ZScheduler() app._setObject('Scheduler', oScheduler) prod = context._ProductContext__prod setattr(prod, 'manage_beforeDelete', beforeDeleteHook()) get_transaction().note('Added Scheduler') get_transaction().commit() if 'Scheduler' not in app._reserved_names: app._reserved_names = app._reserved_names + ('Scheduler',) # initialize ZEvent objects context.registerClass( ZEvent.OneTimeZEvent, permission = 'Add ZEvent', constructors = ( ZEvent.manage_addOneTimeZEventForm, ZEvent.manage_addOneTimeZEvent), icon = 'www/OneTimeZEvent.gif') class beforeDeleteHook(Persistent): """ Make sure that the Scheduler object can be deleted This should be persistent.... """ def __call__(self, prod, products): app = products.aq_parent.aq_parent rn = list(app._reserved_names) try: rn.remove('Scheduler') app._reserved_names = tuple(rn) except ValueError: pass The beforeDeleteHook object makes sure that if you _manually_ delete the ZScheduler Product in the management interface, the protection on the Scheduler object is removed. As this code is removed when you delete the actual Product subdir from the filesystem this does not get executed when you uninstall your Product that way. -- Martijn Pieters, Software Engineer | Digital Creations http://www.digicool.com | Creators of Zope http://www.zope.org | mailto:mj@digicool.com ICQ: 4532236 | PGP: http://wwwkeys.nl.pgp.net:11371/pks/lookup?op=get&search=0xA8A32149 -------------------------------------------

1 0

CORBA Integration Redux
by Tres Seaver 11 Jan '00

11 Jan '00

I have added a couple of documents (or a least starts at them) to the requirements document I posted about last week -- here is the whole set: * "Requirements":http://www.zope.org/Members/tseaver/CosNaming/Requirements * "Roadmap":http://www.zope.org/Members/tseaver/CosNaming/Roadmap * "CORBA @ 30,000 feet":http://www.zope.org/Members/tseaver/CosNaming/BirdsEyeCORBA Feedback welcome! Tres. -- ========================================================= Tres Seaver tseaver(a)palladion.com 713-523-6582 Palladion Software http://www.palladion.com

1 0

RE: [Zope-dev] How do index the contents of a File?
by Michel Pelletier 11 Jan '00

11 Jan '00

> -----Original Message----- > From: Evan Simpson [mailto:evan@tokenexchange.com] > Sent: Friday, January 07, 2000 1:50 PM > To: bauer(a)atlas.unisinos.br; zope-dev(a)zope.org > Subject: Re: [Zope-dev] How do index the contents of a File? > > > From: Carlos Henrique Bauer <bauer(a)atlas.unisinos.br> > > I created a ZCatalog to index all the contents of a site. > To index the > contents > > of the documents I added a 'raw' text index to the catalog, > but it seems > to be > > working just with DTML Documents. Do I need to add another > index to the > catalog > > to make it index File documents or is it a ZCatalog bug? > > File objects are not 'documents', although they may of course > contain files > which you consider to be documents. Since File objects can contain > *anything*, including binary executables or images, they don't define > 'contents' for ZCatalog purposes. Well stated Evan. File's do have an attribute called 'data' that ZCatalog will use like any other attribute (if you have an index called 'data'), however, as you pointed out, you may end up indexing quite a bit of garbage. > What you might want to do is create a sub-ZClass of File dedicated to > storing your document files, and define a > PrincipiaSearchSource method which > returns the contents of the object. You could even > pre-process the contents > in this method if you wanted to strip off headers or > formatting or any other > bits you don't really want in your full-text search. This is probably the best solution. -Michel

2 1

Uncatalog absent id bug - one possible cause
by Itamar Shtull-Trauring 11 Jan '00

11 Jan '00

Good morning Michael, I'm also encountering the "Uncatalog absent id" in a bad way (users can't enter data cause reindex fails), and I found at least one reason why, in addition to the slash problem (which I deal with by trying to uncatalog both with and without a slash.) I have a folder with a space in it - "Do Ahead". When I index using the "Find objects to index", they get indexed as "/Do Ahead/object". When I try to unindex them it tries to unindex "/Do%20Ahead/object", and then that error comes up. I added it to the Collector. -- Itamar S.T. itamars(a)ibm.net

1 0

ZDP ANNOUNCEMENT
by Tom Deprez 11 Jan '00

11 Jan '00

Weekly news at ZDP updated.

1 0

Re: Zope source code version control
by Andrey V Khavryutchenko 11 Jan '00

11 Jan '00

Hi, Kenneth! >>>>> "KYKY" == Kenneth Y K YOUNG writes: KYKY> If anyone can point out what's needed to make that work, I will KYKY> seriously _try to_ write patches, though I don't know how difficult KYKY> that would be. Is it of interest to ye developers out there? I'm interested, but at the moment can be the devt version tester at best :( Now, as a quick solution I use scripts, that I've posted week ago to Zope list. The basic idea is the following: - each object is stored in the file (index.dtml, etc) - Zope folder is the directory - when I want to build the site, I run the script, which traverses the directory tree, importing it and calling build() function, which does all the stuff. -- SY, Andrey V Khavryutchenko http://www.kbi.kiev.ua/~akhavr Software & SPI Engineer Visit my site Shick's Law: There is no problem a good miracle can't solve.

1 0

RE: [Zope-dev] searching/viewing Zope source?
by Michel Pelletier 11 Jan '00

11 Jan '00

> -----Original Message----- > From: Jon Udell [mailto:udell@monad.net] > Sent: Saturday, January 08, 2000 9:08 PM > To: zope-dev(a)zope.org > Subject: [Zope-dev] searching/viewing Zope source? > > > On the "eat your own dogfood" principle, I decided today that since -- > let's face it -- the Zope Python, DTML, and C source files *are* the > definitive documentation -- I ought to index them, and why not with > ZCatalog? Sounds reasonable. > First, I imported /zope/lib/python into Zope. Which took longer than I > expected, mainly because although I remembered having used fsimport.py > in Zope 1.x, I could not find it in 2.x, nor in any searchable Zope > place I could think of (zope.org, egroups.com, altavista, deja.com, > etc.). I could not find it either, it's not distributed with Zope. I eventually found it by hunting it down on the old site through the managment 'find'. > Fortunately, I found a backup of Zope 1 that had fsimport.py, so I set > it up as an external method, made a Zopedir called src, and ran this: > > http://host/src/fsimport?fsdir=/usr/zope2/lib/python > > Cool. Now the source tree is in Zope. I added a ZCatalog to the src > folder, and told it to add everything. I'd have liked to have > added just > *.py, *.dtml, and *.c, but none of the strings I could think of typing > into the > > with ids: > containing: > expr: > > boxes of the "Find Items to ZCatalog" > > seemed to find anything. I'm guessing that regexps can be used here > somewhere, but none of the usual suspects (zope.org, egroups.com, > altavista, deja.com, etc.) yielded any clues. (This, of > course, is *why* > I was trying to index the source in the first place!) Not regexprs, but python exprs. 'with ids' does an exact string comparison to the object's id, 'containing' does a grep style search through the object's 'PrincipiaSearchSource' value (slow!), and 'expr' does a python expression evaluation against the object. It's the identical interface to the regular Zope find, which I believe is documented in one of the guides. Note that Zope's 'find mechanism' is pre-catalog, it does not use the catalog in any way, it's alot like UNIX 'find' and 'grep' rolled into one tool. > Whatever, I just took the whole lot, 800 and some files. Then tried > searching, and realized -- oops, no text index, only a > meta_data index. > After searching the usual suspects again (zope.org, > egroups.....) I came > up with the idea of indexing the 'data' attribute of the File objects. That sounds right. > But no joy. It's looking like either: You created a text index called 'data'? That should do it, I'm trying to do it myself but fsimport doesn't apear to work with the new help system in the CVS version of Zope, argh. Did you create a new search interface to query the text index also? > a) wrap those files, in some way, so that they have a catalog-able > attribute (which I guess means changing fsimport.py?) Yes, you would need fsimport to create instances of this new object, instead of File object, but File objects should work on their 'data' attribute, I can't explain why it doesn't. > or > > b) use a conventional search engine > > I guess b) sounds easier :-) Not after I fix whatever bug is causing this... It's a short hill up, but it's a long ride down. ;) Obviously, one of the big problems here is documentation, as you mention. There is little to no ZCatalog docs, other than the managment interface, and such a complex beast should have more documentation. Bad me! -Michel

2 1

RE: More on the Absent id bug
by Michel Pelletier 11 Jan '00

11 Jan '00

> -----Original Message----- > From: Jason Spisak [mailto:webmaster@mtear.com] > Sent: Thursday, January 06, 2000 3:07 PM > To: zope-dev(a)zope.org; michel(a)digicool.com > Subject: More on the Absent id bug > > > More findings on the extra backslash. The url stored in the Catalog > when I ask for: > > <dtml-in "Catalog(id=id)"> > <dtml-var url> > </dtml-in> > > is correct. It has no leading slash. However under the 'Cataloged > Items' tab as well as when I use getpath() I get a leading > slash. This > screws up unindex_object because it's looking for an object > without the > forward slash, and I guess it doesn't see one. Michel, is > this fixed in > 2.1.2? Is it related to having old Catalaog Aware ZClasses? It's > really drving me crazy. An app that was working yeasterday, is pretty > useless today. :( Can you open a collector ticket? I'm exploring catalog issues right now... -Michel

3 2

property types
by Lalo Martins 11 Jan '00

11 Jan '00

I have some stuff I'd like to do with ZClasses - I'm trying to avoid writing Python classes at all when possible. But I'd have to store some stuff sequentially, and some stuff in dictionaries. Is there any way to have list or dictionary properties? Is there any way to add new property types? (I believe list/dict props could be really useful if the general goal is to deprecate python code in favour of ZClasses and Python Methods). []s, |alo +---- -- I am Lalo of deB-org. You will be freed. Resistance is futile. http://www.webcom.com/lalo mailto:lalo@webcom.com pgp key in the web page Debian GNU/Linux --- http://www.debian.org Brazil of Darkness -- http://zope.gf.com.br/BroDar

1 0

XML-RPC, SOAP: xml-dist-app w3c mailing list
by Edd Dumbill 11 Jan '00

11 Jan '00

Those interested in the XML-RPC and SOAP specifications may find this mailing list of use: http://lists.w3.org/Archives/Public/xml-dist-app/1999Dec/0000.html -- Edd -- Edd Dumbill ----/ voice: +44 702-093-6870 fax: +44 870-164-0230 /---- | Managing Editor, XML.com <http://xml.com> | Publishing Editor, XMLhack.com <http://xmlhack.com>

1 0