- Zope - Zope lists

Open Source -- Advantages
by Chris Keyes 21 Aug '01

21 Aug '01

Hi Robb I was reading an article in Linux User last night, an interview with the chairman of RedHat. He was talking about the problems of companies getting tied to a proprietry system and then building their business model around them... (See page 3, 'On Customer Control')... There were some arguments that seem obvious but aren't the ones that always spring to mind... http://www.linuxuser.co.uk/articles/issue12/lu12-Cover_feature-Bob_the_Build er.pdf 150k PDF... Well you can't expect everything from a publisher can you ;-) Not sure if that helps, but, its something that made sence to me :-) Good luck! Chris

1 0

RE: [Zope] Zope vs. the Big Boys
by sean.upton＠uniontrib.com 21 Aug '01

21 Aug '01

What kind of business is your company in? Perhaps you could describe the problem domain you wish to tackle? For certain problem domains, certain tools may or may not be more or less appropriate. Zope is a very good general purpose toolkit, but I can think of quite a few applications where it would be leaps and bounds better than other choices. I guess the answer really depends on what you want to do. Sean -----Original Message----- From: Robb Allen [mailto:myself@robballen.com] Sent: Monday, August 20, 2001 6:17 PM To: zope(a)zope.org Subject: [Zope] Zope vs. the Big Boys Honestly, how does Zope compare to some of the big names such as BEA and WebSphere? My company is looking at spending a lot of money in the Application Server realm, and I wondered how Zope stacks up against them. We're two shakes of a pen away from purchasing something, and I'd like to at least come back with an open-source solution just so that we can compare something. Fire away. I need as much ammo as I can get (although, honestly, I'm leaning towards BEA) R _______________________________________________ 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 )

4 3

Timetable Tag for Zope 2.4.0
by Chris Withers 21 Aug '01

21 Aug '01

Hi, Does anyone know where I can get a working Timetable Tag for Zope 2.4.0? cheers, Chris

2 1

RE: [Zope] New Zope-powered site
by Max Møller Rasmussen 21 Aug '01

21 Aug '01

> From: Kevin Lewandowski [mailto:kevinsl@yahoo.com] > I thought I'd mention my site that I have recently converted > from PHP to Zope. > > http://www.discogs.com Great job ... looks nice. Regards Max M

1 0

problems with MailHost
by Robert Rottermann 21 Aug '01

21 Aug '01

Hi there, the ISP where my Zope Server is situated is using an MTA (Mercury 1.47 for Novell) that seams not to accept the mails generated by Zope's MailHost. It complains that the FROM: field is not valid. Now I have two questions: - Does anyone of you know a solution. - is there a mailserver (preferably in python) I could install on my NT box with which I can test what the difference between a Zope generated and a "normal" email is. Thanks for your help Robert www.redcor.ch don't mind my typos I see them neither

2 1

Compiling errors on 2.4
by Stephan Goeldi 21 Aug '01

21 Aug '01

When running the install script of Zope 2.4 I get this error message. Anyway, Zope runs just fine after this: Compiling python modules File "/some/path_to/zope/new/lib/python2.1/test/nocaret.py", line 2 [x for x in x] = x SyntaxError: can't assign to list comprehension SyntaxError: from __future__ imports must occur at the beginning of the file (te st_future3.py, line 3) SyntaxError: from __future__ imports must occur at the beginning of the file (te st_future4.py, line 3) SyntaxError: from __future__ imports must occur at the beginning of the file (te st_future5.py, line 4) SyntaxError: from __future__ imports must occur at the beginning of the file (te st_future6.py, line 3) SyntaxError: from __future__ imports must occur at the beginning of the file (te st_future7.py, line 3) -------------------------- _________________________________________________________________ Get your FREE download of MSN Explorer at http://explorer.msn.com/intl.asp

1 0

Warning: Extreme Newbie Tutorial Question
by Robb Allen 21 Aug '01

21 Aug '01

I've installed Zope on my Win2K Advanced Server over the weekend to test it out. I want to do the tutorial, but every time I try it tells me the tutorial isn't installed. So I followed the directions by going to http://www:8080/manager and logging in as the only user there is, me. I then add from the list Zope Tutorial. I give it an ID of "Zope Tutorial" and it brings me to a button that says launch. I click it, and I still get the same error. I reeeeealy want to learn this and play with the tutorial, but I must be doing something wrong. Thanks, R

4 7

Strip-o-Gram 1.0 Released!
by Chris Withers 21 Aug '01

21 Aug '01

Strip-o-Gram is an HTML Conversion Library that can convert HTML to Plain Text and strip specified tags from HTML. See http://www.zope.org/Members/chrisw/StripOGram for more details. cheers, Chris

5 5

RE: [Zope] ZServer ACL
by sean.upton＠uniontrib.com 21 Aug '01

21 Aug '01

Some suggestions, from practical experience: 1 - You want a firewall or router rule to block public access to 8080 on that box. Why not use NetFilter/IPTables to set this up on your Linux box, or for that matter, just set up a Cisco router ACL, which is even easier if you can? 2 - Bind Zope to an destination interface/IP with the -a option for z2.py a. If squid/apache runs on the same box, bind it only to loopback (very secure, but limits some remote admin functionality if you are not careful). b. If you need to, bind ZServer to a management network that is only accessed by the proxy and any private administrative clients. c. If you are really paranoid, segment proxy traffic onto its own network, and bind ZServer onto that network; then force your managment/admin traffic to go through the proxy or from localhost. 3 - Use squid; you will get much more powerful acls that will allow you to block certain pages as well as use acls to determine what is cached and not cached. You can do all sorts of ACLs, for that matter. You can also use a redirector both to rewrite URLs for VHM, and also rewrite blocked management URLs to internal URLs that only can be used on a private network (that is, if you are a bit paranoid). Squid is going to be faster anyhow. Squid also does a nice job round-robin load distribution to ZEO clients. 4 - If you are running ZEO, protect the storage server by keeping it on its own VLAN with a private IP. 5 - use squid ACLs for everything we need in the ACL department, including blocking ZMI access on the public interface(s) of the caching proxy. 6 - Keep our Zope/ZEO nodes on a private network behind Squid proxies to keep them out of 'public' internet view, and give them internet access via NAT on a firewall. Squid serves both a private and a public network, and the private network is less restricted, and you rely on host/application/Zope security beyond that. Sean -----Original Message----- From: Todd Hepler [mailto:thepler@caspiannetworks.com] Sent: Monday, August 20, 2001 5:32 PM To: zope(a)zope.org Subject: [Zope] ZServer ACL I'm relatively new to Zope and python. I have Apache with ProxyPass (port 80) set up on the same box as Zope with a VirtualHostMonster (port 8080). I want to make it so that clients cannot "go around" Apache and talk directly to port 8080, so I'm looking for a way to make Zope only accept requests that come from localhost (or a specified IP address). I can't find anything on zope.org related to ACLs of this nature or blocking access to ZServer based on IP address. Any ideas? I dug through the medusa and ZServer code, but the answer isn't jumping out at me. I'm running Zope 2.4.0 with python 2.1.1 on solaris 2.7. Why would I want this? (you might ask) To scale, I want to be able to turn on caching in Apache, or replace Apache with squid. This won't improve performance if the pages that come out of Zope refer directly to port 8080. The clients would end up "going around" the cache. I know there are ways to make Zope generate appropriate URLs and thus not go around port 80, but I'm not interested in those solutions here. Even those solutions won't stop someone from hardcoding a reference to port 8080 in one of their pages. I want to simply shut off access to it in the first place so that if someone even tries it, they fail. Thanks, -todd _______________________________________________ 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

Zope 2.4 / DTML problem: With looping over two sets
by spex66＠yahoo.com 21 Aug '01

21 Aug '01

Hi there, maybe a newbie question... I try to dtml a cursor-object of an own python-database, swigged from a C-project (Gigabase). Actual the cursor-class has __getitem__ with throwing of IndexError. BUT: it has no real random access via index-number! its behavior is like this (a little bit simplifying): ... cursor.first() #get first position cursor[anyNumber] #if number == 0: return cursor.get() #else: cursor.next().get() #if next() has no result or number > len (cursor), or len(cursor)==0 the exception is thrown and a loop will stop ... so, a simple code-snippet cursor.first() for c in cursor: for r in c.rows: ...process something... works perfect for me! Now following DTML-code makes a unaccepted error (I've had very few accepted errors :-))... obviously the render-code access the cursor.__getitem__ method in second loop (-->c_item) again with static index-number! Bad for my stupid implementation, and not pythonlike where a loop in a loop works with an own loop-variable. I doesn't expect a continous __getitem__ with 'same' index-number, and I don't know how to workaround it on C-language-site first steps to program this page (and this is only the beginning!) the "classic" way, leads me directly to your mailinggroup and the urgent wink for help!!! I like the DTML way, and I want to use it! Anybody a tip or patch, workaround, whatever??? thanks a lot Peter (=PA=) : t = DocumentTemplate.HTML(""" <html><head><title><dtml-var expr="cursor.entityName"></title></head> <body> <dtml-if expr="cursor.__len__()"> <p>Die Suche ergab <dtml-var expr="cursor.__len__()"> Treffer! </p> <table> <tr> <dtml-in expr="cursor.rowNames"> <----- OUTER LOOP <th><dtml-var sequence-item></th> </dtml-in> <tr> <dtml-in cursor prefix="c"> <------ INNER LOOP <tr> <dtml-in expr="cursor.rowNames" prefix="r"> <td><dtml-var expr="c_item.__getattr__ (r_item)"></td> <---- c_item access my cursor again with __getitem__! instead of a loop-variable </dtml-in> </tr> </dtml-in> </table> <dtml-else> <p>Die Suche ergab keine Treffer!</p> </dtml-if> <p>Generated by Miracle</p> </body> </html> """)

1 0