- Zope - Zope lists

Most stable database to use with zope, perl, jdbc and odbc?
by Robin Smidsrød 19 May '02

19 May '02

I need a database with transactions, stored procedures, foreign keys and common reqursive queries (generally SQL99-compliant) which is stable to use with zope. Does anybody have a good suggestion on a database engine (open-source databases preferred) which is usable? I've been using DB2 from IBM for a while, but the DA's available for it are either slow or buggy (tried ZDBI/Perl, SQLRelay and the one from those japanese dudes). None of them measure up. I also need a database who can handle load well and has good crash-recovery. Distributed database is not _that_ important, but would be fine. I also need to connect to my database from Java (JDBC or SQLJ), Windows ODBC and Linux-based Perl. So decent drivers for these need to exist as well. Hope someone has some good answers or pointers. The stability and the crash-recovery is the most crucial features I need, the other can be discussed. TIA, -- Robin

2 1

serve files and browse directories from the local file system ?
by Marcio Marchini 19 May '02

19 May '02

Is this what people use when they want to serve files and browse directories from the local file system: http://www.zope.org/Members/jfarr/Products/LocalFS ? I am concerned about the age/status 0.10.1 (Development) 2001/02/10 thanks, marcio

1 0

Newbie question on running external method
by Leo 19 May '02

19 May '02

Hello, I'm still working on my Zope site, and I have another problem again. I have a MySQL table called Photos and with column names: filename, foldername, PhotoID(primary key). ReadPhotoID is the ZSQL method by passing folder data and get back filename, PhotoID. photorename is the external method to pass the foldername, filename and PhotoID data, and rename the file name from "filename" to "PhotoID", eg. from "abcd.jpg" to "123.jpg". photorename code is : import os def photorename(folder, filename, photoID): os.renames("/Finish/"+folder+"/"+filename,"/Finish/"+folder+"/"+photoID+".jpg") So, I use the following code: <dtml-in "ReadPhotoID(foname=REQUEST.get('folder'))"> <dtml-var "photorename(REQUEST.get('folder'),filename, PhotoID)"> </dtml-in> It shows a site error with "_mysql_exceptions.NotSupportedError". However, it runs ok when I just show the data: <dtml-in "ReadPhotoID(foname=REQUEST.get('folder'))"> <dtml-var filename> <dtml-var PhotoID> </dtml-in> or put the data manually in the external method: <dtml-var "photorename(REQUEST.get('folder'), 'abcd.jpg', '123')"> Did I miss something? I worked on this problem 2 days and still cannot find out which part is wrong. Please tell me anything wrong if you found it. Thanks in advance. Leo

3 2

(no subject)
by Derek Basch 19 May '02

19 May '02

Message: 1 Date: Fri, 17 May 2002 13:44:58 +0400 From: Oleg Broytmann <phd(a)phd.pp.ru> To: zope(a)zope.org Subject: Re: [Zope] Re: ZFireBirdDA connection string Q's Reply-To: Oleg Broytmann <phd(a)phd.pp.ru> Oleg, When entering the ZFirebirdDA connection string in window$ it will fail if you include any white space. Do you Unders~1? :) Peace, Derek Basch On Thu, May 16, 2002 at 11:57:26PM -0700, Derek Basch wrote: > The ZFireBirdDA connection string works > fine if you convert long file names to ******~1 > format. (ie. localhost:c:\progra~1\test.gdb). My Windows unders~1 long filena~1, and yours? :) Oleg. -- Oleg Broytmann http://www.zope.org/Members/phd/ phd(a)phd.pp.ru Programmers don't die, they just GOSUB without RETURN. __________________________________________________ Do You Yahoo!? LAUNCH - Your Yahoo! Music Experience http://launch.yahoo.com

2 1

How About Storing Python Data Structures Directly in ZODB?
by Dan Shafer 19 May '02

19 May '02

Is there a way to store, e.g., a Python dictionary object *directly* in the ZODB either as a DTML Document/Method or as a File Object, e.g.? I *think* it would be possible to use a string to store the dictionary in a DTML Method and then do an eval() or exec() on the string depending on the context, but I'm not really sure. Still learning.

2 2

RE: [Zope] HTTP Cache & Zope (Relative/Absolute URL's, seperate s ervers)
by sean.upton＠uniontrib.com 18 May '02

18 May '02

As someone previously said, use a VHM or something similar to Virtual Host with Squid+ZServer. That said, if you use a redirector with Squid to get VHM interaction working (you would have to, actually - you don't have a choice, I suggest searching freshmeat.net for pyredir), you should be aware that what gets cached is actually based on the back-end URL. I think that most well-behaving CMF tools using the getURL() function for an object will work with a VHM just fine, as long as your templates or scripts use this rather than the physical path to lookup URLs... If, by chance you are using ZEO, you want to keep your back-end Zope servers using the same back-end URLs if you have more than one (for example, with ZEO); compile Squid with --disable-internal-dns and use Squid's external dnsserver process to resove from /etc/hosts to manage the names of your back-end servers (Squid will round-robin among all IPs with the same hostname in /etc/hosts) this way. Sean -----Original Message----- From: Jean-Francois.Doyon(a)CCRS.NRCan.gc.ca [mailto:Jean-Francois.Doyon@CCRS.NRCan.gc.ca] Sent: Tuesday, May 14, 2002 10:50 AM To: zope(a)zope.org Subject: [Zope] HTTP Cache & Zope (Relative/Absolute URL's, seperate servers) Hello, I am working onn adding Squid to my Zope setup to speed things up, but am wondering abour the URL's. In many parts of the site, and even in the default setup of CMF for example, absolute URL functions are used. Now, this becomes a problem if I want to run Squid on a separate machine, since I'm guessing Zope will guess it's URL as being different from what is being used by the user. If I have http://mydomain.com setup as the Squid cache, but the Zope site actually lives at http://nocache.mydomain.com , Zope will give URL outputs as the "nocache" version, which is a problem, since I want Zope to *think* it's actually on mydomain.com ... Has anybody run into this problem ? How do I get around it ? For my own stuff it's easy enough, but in the case of CMF content, it's a problem since a lot of the skin stuff uses absolute paths .... Help?! Thanks, Jean-François Doyon Internet Service Development and Systems Support GeoAccess Division Canadian Center for Remote Sensing Natural Resources Canada http://atlas.gc.ca Phone: (613) 992-4902 Fax: (613) 947-2410 _______________________________________________ Zope maillist - Zope(a)zope.org http://lists.zope.org/mailman/listinfo/zope ** No cross posts or HTML encoding! ** (Related lists - http://lists.zope.org/mailman/listinfo/zope-announce http://lists.zope.org/mailman/listinfo/zope-dev )

1 0

RE: [Zope] zope + squid
by sean.upton＠uniontrib.com 18 May '02

18 May '02

Squid will cache HTML even without expires headers if Last-modified is present in the headers, albeit with refresh patterns configured in squid.conf, which really are not well suited to dynamic content. However, it may also be easier to just use (and understand the implications of doing so) refresh patterns in squid.conf for images (instead of worrying about expires headers), and use cache managers (or add expires to the RESPONSE object) for dynamic stuff. The downside to this is that an image that may have changed a long time ago may take a while before doing freshness checks when updated; on a high-traffic site, though, it might just be best to keep your refresh patterns low, since it won't really do too much to your HIT/MISS ratio. Also, if you are using ZEO, consider just running an NTP server on your proxy; this seems like the proper place for it, since it is likely you would have one (or two) of those boxes to many Zope servers that could sync to it. Sean -----Original Message----- From: Andreas Jung [mailto:andreas@andreas-jung.com] Sent: Thursday, May 16, 2002 3:53 AM To: Daniel Meier; zope(a)zope.org Subject: Re: [Zope] zope + squid - check the access.log of Squid (HITS vs. MISS) - check that your Zope objects set the 'expires' HTTP header and other caching related headers - check that the clocks of your Squid and your Zope machine are synchronized (when running on different machines). -aj ----- Original Message ----- From: "Daniel Meier" <krams(a)krims.ch> To: <zope(a)zope.org> Sent: Thursday, May 16, 2002 06:21 Subject: [Zope] zope + squid > hi there > > i just installed squid on my zopemachine to use > it as a reverse proxy (speed up things ;-). > i can access the site through squid, but I think > it's not caching. is there something i should put > in my zpt and dtml-objetcs and if yes how (I read > something about http-header). can somebody explain? > > thx > > daniel > > > > > _______________________________________________ > Zope maillist - Zope(a)zope.org > http://lists.zope.org/mailman/listinfo/zope > ** No cross posts or HTML encoding! ** > (Related lists - > http://lists.zope.org/mailman/listinfo/zope-announce > http://lists.zope.org/mailman/listinfo/zope-dev ) > _______________________________________________ Zope maillist - Zope(a)zope.org http://lists.zope.org/mailman/listinfo/zope ** No cross posts or HTML encoding! ** (Related lists - http://lists.zope.org/mailman/listinfo/zope-announce http://lists.zope.org/mailman/listinfo/zope-dev )

1 0

How a WebSite works using Zope?
by Raphael Ribeiro 18 May '02

18 May '02

I don't know , then i'm asking this, but is it the way i'm thinking? In your index and other pages you call your zope objects using DTML , it's like that, you create an object in zope that takes some data of your SQL Server DB, and then in your index page, you call that object that takes some data? And then the DATA appears... Another question , is: Is there any DTML editor , and how can i separarate the business logic and the front-end in ZOPE? It's like, i want to make a design , and then i want to integrate it , with my Zope website, and i don't want the codes to be a mass , how can i do it?

1 0

atexit.register() doesn't work
by Jens Wolk 18 May '02

18 May '02

Hi everybody, in a python based product (running on a Zope 2.5.0 installation) I try to implement a class which hands out unique ids. (*) The idea is to save the last id in the server's file system. To make sure that this works even if the Zope server crashed, the object should save an 'earmark' in the file system if it is destroyed in a regular way. I tried to do this with the atexit module provided by Python, but it doesn't work. (nor does __del__). Does anybody know why this is so? Or can anybody suggest a working alternative? (One working solution provided by alex.khan (*) is to patch the source code of Zope itself. But I would like to find another solution, so that others who might want to use the product, have not to fiddle with the Zope source code.) Here are the relevant parts of the implementation: class ManageGlobalZypId: def __init__(self, bucketSize, FileName): import os, thread, string, App.FindHomes, atexit ... self.opened = 1 atexit.register(self.close) def close(self): if self.opened: # save 'earmark' self.opened = 0 ... # GlobalIdHandout is the one and only object from which other objects get # their ids GlobalIdHandout = ManageGlobalZypId(256, 'GSID.txt') Thank you. Jens (*) based on an idea by alex.khan: http://www.zope.org/Members/alex.khan/GlobalSystemId?pp=1

1 0

ZOPE and ActivePython
by José Orlando Ribeiro 18 May '02

18 May '02

I have recently installed ZOPE 2.5 and before that i had installed ActivePython 2.2 , but at zope info page it is like that: Zope Version (Zope 2.5.0 (binary release, python 2.1, win32-x86), python 2.1.2, win32) Python Version 2.1.2 (#31, Jan 15 2002, 17:28:11) [MSC 32 bit (Intel)] System Platform win32 And then , can't i compile zope with ActivePython , or can't i compile extensions to zope using the active state version? I want explanations with that , for i'm very confused here , don't know what to do...

3 5