December 2003 - Zope

Zope community + Zope corp. IRC meeting Tues. Dec. 16th
by Paul Winkler 15 Dec '03

15 Dec '03

I noticed this in the news on zopezen.org ... looks like it was posted to zope-annce but not the main zope list, so it's possible that many people haven't seen it: >From Rob Page's announcement at http://mail.zope.org/pipermail/zope-announce/2003-December/001310.html """In addition, we are going to experiment with a few mechanisms at improving the regularity, predictability and quality of communication we have with the Zope community. The first such experiment will be an IRC session as follows: - Tuesday, Dec 16, 2003, 11a-12p (US/EST GMT-5) on irc.zope.com in #zope-community Zope reps on the IRC session will include Rob Page, Jim Fulton, Brian Lloyd, Jens Vagelpohl and Shane Hathaway. We hope you can join us! Agenda for Dec 16, 2003 IRC Session ------------------------------------ 15 mins: Zope.org operations 15 mins: Trademark management 15 mins: Zope 2.x issues 15 mins: Zope 3 issues """ There is other interesting stuff in Rob's original message, about the new zope.org community involvement process and about the recent trademark issues. I strongly recommend a careful reading of all of the material linked from Rob's message before joining the IRC session, because the first half hour will be about those topics. Big IRC meetings can get out of hand, let's keep this one on-topic and productive! -- Paul Winkler http://www.slinkp.com Look! Up in the sky! It's THE AMAZING BALEFUL RHINOCEROUS BALLBOY! (random hero from isometric.spaceninja.com)

1 0

Re[2]: [Zope] Traversing question
by fowlertrainer＠anonym.hu 15 Dec '03

15 Dec '03

Hello Andreas, Monday, December 15, 2003, 5:14:24 PM, you wrote: I try with this in my access_rule: rq=context.REQUEST s=str(rq.PATH_INFO) rq.set("MainPath",s) if (s<>'/manage') and (s<>'/'): from string import split l=split(s,'/') rq.set("MainPath",l) s='here/index_html' context.unrestrictedTraverse(s) But it is don't working. Have anybody an idea ? How to redirect the user from these nonexisting virtual objects to the existing main handler ? Thanx: KK AJ> What are you trying to accomplish? Ever heard of restrictedTraverse()? AJ> -aj AJ> --On Montag, 15. Dezember 2003 17:09 Uhr +0100 fowlertrainer(a)anonym.hu AJ> wrote: >> Hello zope, >> >> I have to all requests are handled by only one method/zpt. >> >> I set the access rule like this: >> >> rq=context.REQUEST >> s=str(rq.PATH_INFO) >> rq.set("MainPath",s) # debug >> if (s<>'/manage') and (s<>'/'): >> from string import split >> l=split(s,'/') >> rq.set("MainPath",l) # debug >> TraversalRequestNameStack=['index_html'] >> <--- I try to the Zope get >> this object, and not the original URL >> >> but it have failed. >> >> When I try with: >> >> http://mysite/o >> >> I want to this request are stored into MainPath ("/o"), and the Zope >> don't drop an error, but it is get the default index_html, and process >> that. >> >> But it is drop the error: >> >> Site Error >> An error was encountered while publishing this resource. >> Resource not found >> Sorry, the requested resource does not exist. >> Check the URL and try again. >> Resource: o GET >> >> What I can do to prevent the error, and redirect the process to the >> main index_html ??? >> >> Thanx for any advance !!! >> >> KK -- Best regards, fowlertrainer mailto:fowlertrainer@anonym.hu

1 0

Re[2]: [Zope] Traversing question
by fowlertrainer＠anonym.hu 15 Dec '03

15 Dec '03

Hello Andreas, Monday, December 15, 2003, 5:14:24 PM, you wrote: AJ> What are you trying to accomplish? Ever heard of restrictedTraverse()? Sorry, but I don't heard from that. How to work ? Have anybody with little working example ? Thanx for any info what more that RTFM... :-) AJ> -aj AJ> --On Montag, 15. Dezember 2003 17:09 Uhr +0100 fowlertrainer(a)anonym.hu AJ> wrote: >> Hello zope, >> >> I have to all requests are handled by only one method/zpt. >> >> I set the access rule like this: >> >> rq=context.REQUEST >> s=str(rq.PATH_INFO) >> rq.set("MainPath",s) # debug >> if (s<>'/manage') and (s<>'/'): >> from string import split >> l=split(s,'/') >> rq.set("MainPath",l) # debug >> TraversalRequestNameStack=['index_html'] >> <--- I try to the Zope get >> this object, and not the original URL >> >> but it have failed. >> >> When I try with: >> >> http://mysite/o >> >> I want to this request are stored into MainPath ("/o"), and the Zope >> don't drop an error, but it is get the default index_html, and process >> that. >> >> But it is drop the error: >> >> Site Error >> An error was encountered while publishing this resource. >> Resource not found >> Sorry, the requested resource does not exist. >> Check the URL and try again. >> Resource: o GET >> >> What I can do to prevent the error, and redirect the process to the >> main index_html ??? >> >> Thanx for any advance !!! >> >> KK

1 0

Apache URhell
by mh 15 Dec '03

15 Dec '03

(Apologies for a potential duplicate posting of this message) Thanks to those that responded off-list to my original query, but I'm still having no joy whatsoever in getting Zope to properly handle requests proxied to it from Apache. My Apache configuration is: <VirtualHost 10.0.0.2> ServerName www.test.com ServerAlias test.com RewriteEngine On RewriteLog "/var/log/httpd/rewrite_log" RewriteLogLevel 9 RewriteRule ^/(.*) http://localhost:8080/VirtualHostBase/http/test.com:80/VirtualHostRoot/$1 [L,P] </VirtualHost> (RewriteRule directive shown wrapped over three lines.) test.com is defined as a zone on my local name server pointing to 10.0.0.2. I have a VirtualHostMonster in the Zope's root folder with no mappings defined there. Also in the root folder is a folder "test.com" with the default DTML document index_html. When a request is made to www.test.com, Zope intercepts it but displays the (acquired?) content from the root folder's index_html. Snippets from the error_log show this: TraversalRequestNameStack [] VirtualRootPhysicalPath ('',) URL 'http://test.com:8080/%241' SERVER_URL 'http://test.com:8080' URL0 http://test.com:8080/%241 URL1 http://test.com:8080 BASE0 http://test.com:8080 BASE1 http://test.com:8080 BASE2 http://test.com:8080/%241 (No idea what the %241 implies, but it's there in the error log) PATH_INFO '/VirtualHostBase/http/test.com:8080/VirtualHostRoot/$1' HTTP_HOST '127.0.0.1:8080' REQUEST_METHOD 'GET' PATH_TRANSLATED '/VirtualHostBase/http/test.com:8080/VirtualHostRoot/$1' The "/$1" on the end of PATH_INFO and PATH_TRANSLATED would suggest to my naive eyes that Zope isn't properly mapping the URL to a local resource, but that's just conjecture on my part. VirtualRootPhysicalPath also looks a little suspect. If I make an explicit request for http://10.0.0.2:8080/VirtualHostBase/http/test.com:80/VirtualHostRoot/ (as suggested by an off-list response), then again, the root folder's index_html is returned, not that of test.com. Requests to 127.0.0.1 as opposed to 10.0.0.2 generate the same result. Muchos kudos to anyone who can shed some light! mh

3 3

Problem with run Zope at win2k
by DeHa 15 Dec '03

15 Dec '03

Hi, I want to say hello to everybody - I'm new to zope(:. At the begining - I am really sorry for my english. I have a problem with starting zope at my machine: MS Windows 2k Python 2.3.1 Zope 2.6.2 (binary install) I found at google that if I want to run zope with my existing Python installation I have to change file start.bat in zope directory to point to my python install. I did this, but en error occurs: The dynamic link library python21.dll could not be found in the specified path. I get python21.dll from zope directory and copy to one of the directory of specified path and an error stop to occur, but zope doesn't start. Please help(: Best regards ---------- Dawid Hulisz also known as DeHa Dobrze widzi się tylko sercem. Najważniejsze jest niewidoczne dla oczu. ----------

2 2

RE: [Zope] Help on identification
by Tiller, Michael (M.M.) 15 Dec '03

15 Dec '03

> From: J Cameron Cooper [mailto:jccooper@jcameroncooper.com] > Subject: Re: [Zope] Help on identification > > > Tiller, Michael (M.M.) wrote: > > >I'm trying to setup a Zope server on our corporate intRAnet > set. We have a > >formal system of authenticating users. It is centralized. > I don't want to > >go into too much detail, but suffice it to say that I can > invoke a program > >and it will tell me whether the person requesting my page has a valid > >certificate and, if so, who the person is. > > > >What is the best scheme for integrating such a program into the Zope > >authentication and identification process. Key things are: > > > >1) The user does not provide a password, only the validity of the > >certificate is checked. > > This detail I cannot really advise on. You're trying to get a > certificate? If it comes in REQUEST it's easy to get. It comes in REQUEST. There is no issue here, I just added this as information. The important point is that no password is required from the user. > >2) How do I invoke an external program from Zope?!? > > > > > Any filesystem-based code can do it, using the usual Python > calls. (This means External Methods and Products.) Great. I just wanted to confirm that. Unfortunately, I don't have easy access to the file system (although I'm not completely blocked) so I wanted to know if there was a way around this. In a sense, it is reassuring that there isn't. :-) > > 3) Are there any "out-of-the-box" user folders that can > > handle this? I couldn't find any. If not, what is the best > > open ended product and what would I need to do? > SimpleUserFolder I looked into this, but it wants to feedback into the user source. It wants to be able to create new users and edit the roles and password of an existing user. I'm not in a position to provide any of this. > ExtensibleUserFolder exUserFolder says that the one method that is mandatory is "listOneUser" which has to return the username, password and a list of roles for the user. In our system, I already know the identity of the person. I don't need to authenticate them. > PluggableUserFolder This folder is quite generic. Based on my reading of the "INTERNALS.txt" file, I would need to provide my own identification method (which would fit pretty simply into this), the authentication would really do nothing and the authorization should be based on local user data (information about roles) that would be maintained over and above our intranet indentification/authentication system. So this one is a possibility but a possibility with a perhaps steep learning curve and lots of coding. > all provide some simplified API for arbitrary user sources, > authorization, and authentication. The closest thing I have found is "RemoteUser" which identifies a user based on the REMOTE_USER environment variable. This would work great if I had access to REMOTE_USER, but it isn't clear that I will. In any case, it seems to function the way I want (it doesn't care about passwords, it just grabs the user's identity and adds them to the user folder if they don't exist already). This is along the lines of what I want, but it is going to require me to hack it a little bit in order to support our specific system. I was hoping for something that I might be able to configure "through the web". I'll also look into how authentication is done with SSL since that is very similar to what we do. Thanks. > --jcc -- Mike

2 1

Re[2]: [Zope] Zope Requests - POST vs. GET
by fowlertrainer＠anonym.hu 15 Dec '03

15 Dec '03

Hello Andreas, Monday, December 15, 2003, 2:46:53 PM, you wrote: AJ> --On Montag, 15. Dezember 2003 14:36 Uhr +0100 Tino Wildenhain AJ> <tino(a)wildenhain.de> wrote: >> Hi Andreas, >> >> Andreas Jung schrieb: >>> This a single REQUEST is either a POST or GET request there is no need >>> for this separation IMO. Everything's in the REQUEST. >> >> That's not entirely true. Sadly there is a design flaw in cgi.py >> which ignores GET variables still present when doing POST. >> >> The whole lib is a bit ugly so its not easy to fix it. >> And its even worser to fix if you want to achieve compatibility >> with current implementations. >> >> Should variables be merget? Get variables before POST >> or vice versa? >> AJ> big hmmmmmmm :-) I don't have the CGI specs under my pillow and I am too AJ> lazy AJ> to read them but POST usually transmits the data in the body and GET in the AJ> QUERY AJ> strings. I have never (and needed) a usecase where both were mixed (if AJ> allowed). AJ> -aj I want to mix them before, but I see the chance to replace my code with virtual urls. (url?thissite -> url/thissite) But how ? I want to do this: My site has only 1 control file. This named index_html. I want to all of the requests are redirected to this file. Pseudo: If url is not base_url: delete base_url from url set my variable to url traversedobject: index_html In index html I process the datas, and others. I create a test script to the main/root folder: # access_rule from string import split rq = container.REQUEST urllist = split(rq.HTTP_URL, '/') if (len(urllist)>1): urlold=urllist rq.set("BHAPath",urlold) RequestTraversalNamePath=[rq.SERVER_URL] I set rule, but it is nothing happened. What I doing wrong ??? Thanx for any advance: KK -- Best regards, fowlertrainer mailto:fowlertrainer@anonym.hu

1 0

Job: Python/Zope developer
by Ezra Wolfe 15 Dec '03

15 Dec '03

DLC Solutions is seeking a senior-level software engineer with experience in application development using Zope, Python, and ActionScript. Five or more years experience in systems architecture, database design, and legacy systems integration is required. Experience in medical education or e-learning field is a plus. This hire will be the programming lead for DLC Solution's product development for web-based educational technologies. We are looking for someone to develop our core learning management system built using Zope/Python/DTML. This position requires dedication to and a love of building elegant, modular products that are optimized for administrative simplicity, top-notch user experiences and rich reporting capabilities. The successful applicant will also be able to understand and participate in the business process of our product development effort. We are looking for a team player who is willing to contribute both lines of code and design leadership; therefore, strong communication skills are necessary. A strong applicant would also have experience working in a small, fast-growing software development department and be willing to embrace the challenge of developing processes and mentoring junior level programmers. This position does not require managing employees. Two or more years of experience are required in the following areas: Zope Python ActionScript DTML SQL Server and database architecture Windows 2000 Server ASP/VBScript Familiarity with the following technologies is a plus: HTML, CSS, JavaScript Streaming audio Flash Communications Server PHP Learning management systems Content management systems To apply for this position, email your resume (in Word format) and salary requirements to info(a)dlc-solutions.com.

1 0

Howto prevent deletion of acl_users
by Florian Schulze 15 Dec '03

15 Dec '03

Hi! How can I prevent people with FTP access from deleting their acl_users folder? Is there a simple access_rule I can use or some special security setting? I tried to remove the right to delete things inside the acl_users folder, but the folder itself can still be deleted. Thanks and regards, Florian Schulze

3 3

What tool to test Zope performance with Apache proxy cache
by Patrick Gelin 15 Dec '03

15 Dec '03

Hi, I'm looking for a tool to test performances with my Zope server. I don't use Zeo, just 1 serveur with Apache proxy cache activated. So I'm looking for a tool taking into acount cache infromation (this is not the case of ApacheBench). Thanks.

1 0