- Zope - Zope lists

- Zap-1.1.0pre
by Michel Pelletier 22 Jan '99

22 Jan '99

Hello, I'd like some of you linux folks who are interested to try out the new version of Zap (Zope + Apache). Extract the tarball in your top level Zope dir, and run 'zap/start'. This should fire up an Apache to server your Zope. I am working on a FAQ, Troubleshooting document, and a new home page, these things coming soon. Zap for other platforms is in the works. New features include: Apache 1.3.4 dynamic loaded modules (mod_rewrite and mod_proxy by default) binary now called 'zap' instead of 'httpd' for distinction in 'ps' fixed up startup and stop scripts to whack the Zope process properly http://www.zope.org/Download/Prereleases/Zap/Zap-1.1.0pre-linux2-x86.tar.gz Please report bugs to me (michel(a)digicool.com) Thanks! Michel Pelletier

1 0

RE: [Zope] - Website recommendation
by Paul Everitt 22 Jan '99

22 Jan '99

Kevin wrote: > I've seen a few Products that people have put together here > and posted on their own websites. I think it would be good to take > the contrib area on the Zope site seriously, so that it will grow > into a repository as useful as CPAN. Believe me, we are taking this particular area *very, very* seriously. This is part of the mandate for Ken Manheimer who starts on Monday. Ken's work operating python.org kept him firmly and painfully rooted in the problem of organizing the Python universe. Also, Ken and I are both librarians at heart :^) Michel Pelletier, of Trove/ZTrove fame, works here now and he has a lot of ideas from Trove. I had a lot of background in the mis-begotten Locator-SIG that I ran but failed at. The Package Archive support that we build into zope.org is going to be absolutely awesome. It is part of the larger "Zopespot" project, in that packages will have some relationship to authors (and thus the membership system), to bug reports (thus the Collector), and the sitewide catalog. > To get things going in that direction requires the cooperation > of everyone. If DC can keep the contrib area organized as the number > of Products increases, and if people make sure they submit their > Products to the site, we'll have the beginnings of CZAN :) Yes, there will be more than enough for all of us to do. Step one when Ken gets here will be to organize the requirements and manage the way the community reaches conclusion. Agreeing on standards for packages is very hard as there are so many: IAFA standards, RPM metadata, Trove, Dublin Core, etc. --Paul Paul Everitt Digital Creations paul(a)digicool.com 540.371.6909

1 0

RE: [Zope] - How can I get the column names from an SQL Method wi th DTML?
by Paul Everitt 22 Jan '99

22 Jan '99

Plamen wrote: > My question is: how can I get the column names from the SQL method? I > think I noticed that _searchable_result_columns() is used to get this > information when creating a Search Interface. Is it possible to get > it from DTML without using an External Method? The database adapter has a tab for a Browse view. Take a look at the source to see what methods it calls to produce the metadata. --Paul Paul Everitt Digital Creations paul(a)digicool.com 540.371.6909

1 0

- Search engine inside zope database (real)
by chalaouxf＠synbio.tpgnet.net 22 Jan '99

22 Jan '99

Hi All, How is it possible to search inside a zope database all the document objet containing a string occcurance (see regular expression) and construct the result in a document with an hypertexte link for each answer? I don't think about the administration tools present by default in zope, at least if I can use this tools. FR François-Régis Chalaoux Bioinformatics Group Synthélabo Biomoléculaire 67080 Strasbourg Cedex FRANCE Tél : 03 88 60 87 14 Fax : 03 88 45 90 70

1 0

- Search inside zope database
by chalaouxf＠synbio.tpgnet.net 22 Jan '99

22 Jan '99

François-Régis Chalaoux Bioinformatics Group Synthélabo Biomoléculaire 67080 Strasbourg Cedex FRANCE Tél : 03 88 60 87 14 Fax : 03 88 45 90 70

1 0

RE: [Zope] - ZMySQLDA, broken product
by Brian Lloyd 22 Jan '99

22 Jan '99

> Hi all, > > After struggling with the NT version with IIS, I decided to > bite the bullet > and installed on a Linux box, it is now working fine 8^). > > I'm now getting into the DB side of things and decided to try > the ZMySQLDA, > installed, compiled ok 8^), but Zope tells me that the > product is broken, > this is the traceback: > > <traceback> > > Traceback (innermost last): > File > "/home/brian/temp/build/tmp_build/Zope-1.9.0-linux2-x86/lib/py > thon/OFS/Appli > cation.py", line 378, in import_products > File "lib/python/Products/ZMySQLDA/__init__.py", line 100, in ? > File "lib/python/Products/ZMySQLDA/DA.py", line 102, in ? > File "lib/python/Products/ZMySQLDA/db.py", line 100, in ? > ImportError: /lib/libm.so.5: undefined symbol: __getfpucw > > </traceback> > > Any ideas? Try replacing the "-lm" in your Setup file with "-lgcc" and recomplile the MySQL module. Brian Lloyd brian(a)digicool.com Software Engineer 540.371.6909 Digital Creations http://www.digicool.com

1 0

- The beginning of darkness...
by Evan Gibson 22 Jan '99

22 Jan '99

I have started down the path to evil... Mr Bailey made a small dtml object that let us input dates and to call the dtml while passing in the names we wanted on the objects we went:    This returns us start_year, start_month and start_day. This was very good. The requirement of using the #with tag was very bad. It looks ugly and messy. What I _wanted_ was to type:  and then later on add default values:  With any further keywords being passed into the namespace so the DTML can find them. This looks neat and makes DTML components useable more like functions. There is namespace contamination in that you can't use anything your tag would normally (so don't expect "fmt" to be anything other than the format command of dtml!) but I think it's safe enough. So I started to delve into the not so hidden, but not so looked at code of DocumentTemplate with the aim of implementing a var2 tag as a test. I copied DT_Var.py to DT_Var2.py, had DT_Strings.py import it, no problem. Looking through DT_Var2 I noticed that the main problem is parse_params from DT_Utils.py which raises errors on non-valid keywords. I changed it to, instead, collect and return the invalid keywords if there are any. If not it returns the normal result. This shouldn't break anything unless someone was relying on the failure mode. It doesn't work.  returns an "Invalid Attribute Name" error on date_input itself. I put some debug info in to test it, but just kept getting the same error and not my modified error message, even though I stopped and started the server. It was as if it was using the old parse_params even though var2 was being noticed as a valid tag and var2 doesn't even _import_ parse_params and was calling parse_params2. Very strange. So, anyone have any ideas why? Smile. It's been fun messing around in the guts of it all... -------------------------- def parse_params2(text, result=None, otherparams=None, tag='', unparmre=regex.compile( '\([\0- ]*\([^\0- =\"]+\)\)'), qunparmre=regex.compile( '\([\0- ]*\("[^"]*"\)\)'), parmre=regex.compile( '\([\0- ]*\([^\0- =\"]+\)=\([^\0- =\"]+\)\)'), qparmre=regex.compile( '\([\0- ]*\([^\0- =\"]+\)="\([^"]*\)\"\)'), **parms): result=result or {} otherparams=otherparams or {} if parmre.match(text) >= 0: name=lower(parmre.group(2)) value=parmre.group(3) l=len(parmre.group(1)) elif qparmre.match(text) >= 0: name=lower(qparmre.group(2)) value=qparmre.group(3) l=len(qparmre.group(1)) elif unparmre.match(text) >= 0: name=unparmre.group(2) l=len(unparmre.group(1)) if result: if parms.has_key(name): if parms[name] is None: raise ParseError, ( 'Attribute %s requires a value' % name, tag) result[name]=parms[name] else: raise ParseError, ( 'Invalid attribute name, "%s" result:%s' % (name,`result`) , tag) else: result['']=name return apply(parse_params,(text[l:],result),parms) elif qunparmre.match(text) >= 0: name=qunparmre.group(2) l=len(qunparmre.group(1)) if result: raise ParseError, ( 'Invalid attribute name, "%s" result:%s' % (name,`result`) , tag) else: result['']=name return apply(parse_params,(text[l:],result),parms) else: if not text or not strip(text): return result raise ParseError, ('invalid parameter: "%s"' % text, tag) if result.has_key(name): p=parms[name] if type(p) is not ListType or p: raise ParseError, ( 'Duplicate values for attribute "%s"' % name, tag) if not parms.has_key(name): otherparams[name]=value #raise ParseError, ( # 'Invalid attribute name, "%s"' % name, tag) else: result[name]=value text=strip(text[l:]) if text: return apply(parse_params,(text,result,otherparams),parms) else: if otherparams: return (result,otherparams) else: return result -------------------------- And these are the changes to class Var in DT_Var to make it Var2: def __init__(self, args, fmt='s'): args = parse_params2(args, name='', lower=1, upper=1, expr='', capitalize=1, spacify=1, null='', fmt='s', size=0, etc='...', thousands_commas=1, html_quote=1, url_quote=1, sql_quote=1, newline_to_br=1) self.newnamespace = {} if type(args) == type(()) and length(args) == 2: self.newnamespace = args[1] args = args[0] self.args=args self.modifiers=tuple( map(lambda t: t[1], filter(lambda m, args=args, used=args.has_key: used(m[0]) and args[m[0]], modifiers))) name, expr = name_param(args,'var',1) self.__name__, self.expr = name, expr self.fmt = fmt if len(args)==1 and fmt=='s': if expr is None: expr=name else: expr=expr.eval self.simple_form=expr, def render(self, md): md.update(self.newnamespace) { and then the rest of the old render... } -- Evan ~ThunderFoot~ Gibson ~ nihil mutatem, omni deletum ~ May the machines watch over you with loving grace.

2 1

- MySQLDA, connection string
by Phil Harris 22 Jan '99

22 Jan '99

Hi all, I've got Zope working at last, at least on Linux. I've now installed the MySQLDA product and was wondering if someone could give me an example of a connection string for this scenario: DB Aspire user test password test host localhost TIA Phil wmlph(a)d032.ml.uwcm.ac.uk

1 0

- ZopeHTTPServer
by Pavlos Christoforou 22 Jan '99

22 Jan '99

Hello Zopistas I have been running a Zope site for a couple of months using the ZopeHTTPServer. Unfortunately the server crashes occasionaly with no informative comments in the var/server.log. Has anyone else encountered a similar situation? I am waiting for the medusa integration. A bobo app I was running over medusa proved to be very reliable. A month ago I was contemplating implementing Userland's XML-RPC protocol in python, and yesterday I noticed Fredrik Lundh posted a module for it. It is a very simple RPC protocol different from bci (oops ZClient) and it might be useful to integrate it into the forthcoming medusa server. Skip who seems to have taken note of it already might provide some useful info. Pavlos

3 2

- ZMySQLDA, broken product
by Phil Harris 22 Jan '99

22 Jan '99

Hi all, After struggling with the NT version with IIS, I decided to bite the bullet and installed on a Linux box, it is now working fine 8^). I'm now getting into the DB side of things and decided to try the ZMySQLDA, installed, compiled ok 8^), but Zope tells me that the product is broken, this is the traceback: <traceback> Traceback (innermost last): File "/home/brian/temp/build/tmp_build/Zope-1.9.0-linux2-x86/lib/python/OFS/Appli cation.py", line 378, in import_products File "lib/python/Products/ZMySQLDA/__init__.py", line 100, in ? File "lib/python/Products/ZMySQLDA/DA.py", line 102, in ? File "lib/python/Products/ZMySQLDA/db.py", line 100, in ? ImportError: /lib/libm.so.5: undefined symbol: __getfpucw </traceback> Any ideas? TIA Phil wmlph(a)d032.ml.uwcm.ac.uk

1 0