May 1999 - Zope - Zope lists

RE: [Zope] DTML question: URLn for n > 9?
by Michel Pelletier 26 May '99

26 May '99

> -----Original Message----- > From: Anthony Baxter [mailto:anthony@interlink.com.au] > Sent: Wednesday, May 26, 1999 7:11 AM > To: zope(a)zope.org > Subject: [Zope] DTML question: URLn for n > 9? > > > DTML questions, probably really stupid ones (I'm tired, and I > can't see the answer in the source, or in the DTML ref...) > > Is there an equivalent for URL1, URL2, &c for when you > go past a 9 deep hierarchy? Uhm. Nope. Hard coded from 0 to 9. You're welcome to take a stab at extending it (ZPublisher/HTTPRequest.py 'round line 455 or so). I belive it's small and tight like this for speed and that the URLns are sort of convienence. alternatively, you could have your application keep track of the URLs itself. -Michel > > (yes, it's autogenerated - I don't have anything _that_ evil built > by hand :) > > Anthony > > _______________________________________________ > Zope maillist - Zope(a)zope.org > http://www.zope.org/mailman/listinfo/zope > > (For developer-specific issues, use the companion list, > zope-dev(a)zope.org - http://www.zope.org/mailman/listinfo/zope-dev ) >

1 0

RE: [Zope] Problem with Zope & Apache
by Michel Pelletier 26 May '99

26 May '99

> -----Original Message----- > From: Corey Ralph [mailto:corey@net-tech.com.au] > Sent: Wednesday, May 26, 1999 1:15 PM > To: zope(a)zope.org > Subject: [Zope] Problem with Zope & Apache > > > I am having a problem with zope & apache. > > Everything works OK except when I try to make any changes in > the manage, I get > a window saying "the document contains no data", and it > according to the logs, > the apache child process caused a segmentation fault. > > With the built in zope http server I have no problems at all. > Then I suspect the problem is not Zope, but your Apache. What version? Can your upgrade and recompile? I haven't seen this behavior before, so you either have a quirky config that is causing Apache to segfault or it's just plain broke and you need to replace it with a recent binary. If you can't do any of these things I would check the error and access logs to look fora pattern that may be avoidable in a configuration setting, or a strange module that is loaded that may be causing the violation. Try stripping your Apache out of all but the rewrite module (if you have dynamic loading) and all of the extra configuration options, and then put them back in one by one until it fails. -Michel > Any ideas? > > _______________________________________________ > Zope maillist - Zope(a)zope.org > http://www.zope.org/mailman/listinfo/zope > > (For developer-specific issues, use the companion list, > zope-dev(a)zope.org - http://www.zope.org/mailman/listinfo/zope-dev ) >

1 0

Weird crashes with  on ZServer 1.11.0pr1
by Alexander Staubo 26 May '99

26 May '99

This used to work before. However, when I put this into some document and views it, Zope starts behaving _extremely weird_ until I remove the #var. It starts premuturely disconnecting, giving garbage text, etc. And it never ever gives me the contents of REQUEST. Same thing with some other variables such as RESPONSE, by the way, but not, say, PATH_INFO. I've tried this on three platforms: NTW, NTS/SBS and Linux 2.0.2x. -- Alexander Staubo http://www.mop.no/~alex/ "In the end, we all assume room temperature." --John Maynard Keynes

1 0

1.x to 2.x query
by Randy Edwards 26 May '99

26 May '99

Well, now that I have the stable version of Zope running, I was curious if some of you experienced Zope users could give me some advice on the wisdom of instead simply running the Zope 2.x beta. From what I gather reading about it, there's a backward incompatability (I'm not sure how large though) which I figure I could avoid if I went the 2.x route now. Any advice on the wisdom of this? Thanks in advance. -- Regards, | <redwards(a)golgotha.net> | Linux: superior operating . | <http://www.golgotha.net> | system tools for those Randy | | who know how to use them.

3 2

RE: [Zope] RE: What method do I use to check access?
by Jay, Dylan 26 May '99

26 May '99

> -----Original Message----- > From: Michel Pelletier [mailto:michel@digicool.com] > Sent: Thursday, May 13, 1999 15:51 > To: Jay, Dylan; 'Martijn Pieters'; 'Jason Spisak'; zope(a)zope.org > Subject: RE: [Zope] RE: What method do I use to check access? > > > I want a method such that > >  > > or something similar. > > > > I've never seen or used anything like this, so perhaps you > may want to > try to refactor your problem, otherwise: > > Users don't have permissions, Users have roles. Using the security > screen in the managment interface, you map which roles have which > permissions, and when you create Users, you give them roles. If your > user is in a Role which has the 'View' permission associated with it, > then you don't need to do the #if check. I think maybe what your > looking for is "Do any of my User's *Roles* map to the permission x?". > (Not using 'View' as an example because anonymous maps to 'View' by > default) > > This would require a bit of tinkering, but it could be done with > AUTHENTICATED_USER.getRoles() and rolesOfPermission(x). You may also > want to talk a look at AccessControl/Roles.py, there are > several methods > defined in there to exampine roles and permissions, like > permissionsOfRole (the inverse of rolesOfPermision) and > acquiredRolesAreUsedBy. I've looked at rolesOfPermission etc and they don't seem to give what I want in that they don't recursivly check parent permissions if permissions are aquired. What I really want is an easy way conditionally include something based on weather the user can view it (or maybe another permission). I don't want to have to hard code the roles as this is duplicating code. What I want is  <a href="">Goto SomeObject</a>  The link will only appear if the user can actually go to it. This must be possible as somewhere in the code this validation must happen already. I've tried looking but really need some help.

3 2

DTML question: URLn for n > 9?
by Anthony Baxter 26 May '99

26 May '99

DTML questions, probably really stupid ones (I'm tired, and I can't see the answer in the source, or in the DTML ref...) Is there an equivalent for URL1, URL2, &c for when you go past a 9 deep hierarchy? (yes, it's autogenerated - I don't have anything _that_ evil built by hand :) Anthony

2 1

RE: [Zope] RE: What method do I use to check access?
by Rob Page 26 May '99

26 May '99

> What I want is >  > <a href="">Goto > SomeObject</a> >  How about:  <LI> <A HREF="Requester/index_html">Requester menu</A>  Reading between the lines of your original post it sounds like this approach might work but require you to add more role definitions... --Rob

1 0

Re: [Zope] FYI: Zope slides from WWW8 on-line.
by Martijn Pieters 26 May '99

26 May '99

At 19:58 25/05/99 , Amos Latteier wrote: >Hi, > >I've just finished getting Paul Everitt's WWW8 presentation on Zope >on-line. > > http://www.digicool.com/Library/FTPB/index.html > >It's a fairly interesting discussion of open source from a business >perspective. And of course, it features Zope. Is there any information on the 10 minute demo's you have been giving? I am not that good at demoing Zope, If I could use your's as a base, would be a great help. -- Martijn Pieters, Web Developer | Antraciet http://www.antraciet.nl | Tel: +31-35-6254545 Fax: +31-35-6254555 | mailto:mj@antraciet.nl http://www.antraciet.nl/~mj | PGP: http://wwwkeys.nl.pgp.net:11371/pks/lookup?op=get&search=0xA8A32149 ------------------------------------------

1 0

ANN: Zope Training Full
by Rob Page 26 May '99

26 May '99

Hello everybody: Just wanted to thank the Zope Community for a _great_ response to Digital Creations' first public presentation of an Introduction to Zope course. All twelve seats are filled. Thanks again! --Rob -- Rob Page V: 540 371 6909 x103 Digital Creations http://www.digicool.com ---- Publishers of Zope The Open Source Application Server @ http://www.zope.org

1 0

nasty datetime bug: _nearsec() sometimes says '60' (!)
by Anthony Baxter 26 May '99

26 May '99

Inside the DateTime object, it uses the attr '_nearsec()' for printing times which are rounded to the nearest second. This is: self._nearsec=round(sc) This is all well and good, except when it's 59.7 seconds past the minute, in which case you get times like '07:34:60'. Then SQL Server and Oracle both get sad at me, and foul things up. My quick fix: --- /opt/zope/inst/lib/python/DateTime/DateTime.py Mon Feb 22 22:10:48 1999 +++ DateTime/DateTime.py Wed May 26 10:33:49 1999 @@ -91,6 +91,7 @@ from string import strip,split,upper,lower,atoi,atof,find,join from time import time,gmtime,localtime,asctime,tzname,strftime,mktime from types import InstanceType,IntType,FloatType,StringType +import math @@ -580,7 +581,7 @@ self._months[mo],self._months_a[mo],self._months_p[mo] self._fday,self._aday,self._pday= \ self._days[dx],self._days_a[dx],self._days_p[dx] - self._nearsec=round(sc) + self._nearsec=math.floor(sc) self._year,self._month,self._day =yr,mo,dy self._hour,self._minute,self._second =hr,mn,sc self.time,self._d,self._t,self._tz =s,d,t,tz (so always truncate the seconds) (yes, it's posted to Collector, but in case anyone else hits this...) Anthony

1 0