- Zope - Zope lists

RE: [Zope] Zope performance too low ?
by Brian Lloyd 17 Jun '99

17 Jun '99

> I guess the subscribers to this list, more or less, like > working with Zope. > Developing dynamic pages is quick and easy, and code/design > integration is > very comfortable. > > However I'm starting to doubt if it's worth it. Some quick > benchmarking with > apache's ab utitlity gave the following results on my > server(50 requests, no > concurrency, ab ran on the server machine): > > Copy of the Zope welcome screen(served from disk): 204.08 > requests per > second > The Zope welcome screen(served from Zope): 14.38 > requests per second > A page with lots of DTML Code(about 50 lines): > 2.68 requests per > second I'm not that familiar with ab, but you don't give a lot of information on your setup here (processor, memory, load (discounting zope), publishing mechanism (pcgi? ZServer? mod_pcgi?), details of the DTML used), which would be helpful to put this kind of information in context. The numbers you give above seem _way_ low to me compared to some general benchmarking that has been done in the past. The only really reliable benchmark, of course, is a real live site - and Zope is currently running a number of rather high-traffic sites without problems (and with far better throughput than the numbers you got with ab). Note that I'm not discounting your results above, I just suspect that there is an underlying reason for them. Having done some benchmarking in the past, I'm a firm believer that you can prove anything you want with a benchmark - just ask the Apache group, who were totally hosed in a recent (laughable) NT/IIS vs. Linux/Apache benchmark :) That said, we do not think that benchmarking it useless - we are very receptive to constructive criticism, and having members of the Zope community develop credible benchmarks is a great way to help discover and solve performance problems. If nothing else, it would provide some comparative information to the community on how best to field Zope based on their performance requirements and server environment. For instance, I would bet that if you ran that same benchmark using Zope/ZServer instead of PCGI, you would get very different results, having removed the need to spawn a process on every hit. That could indicate that server load/memory/swap/process spawn time considerations are causing the hit seen above. If one were to investigate this and find that such-and-such seemed to cause a big hit on Zope response time, it would be very useful to make this data available to the community. Having this kind of background data would allow folks to better evaluate the kind of server environment they need, and allow us all to decide that Something Must Be Done if a problem seems apparent. Brian Lloyd brian(a)digicool.com Software Engineer 540.371.6909 Digital Creations http://www.digicool.com

2 1

Re: [Zope] Cool but...
by michel＠digicool.com 17 Jun '99

17 Jun '99

David J. Orme <orme(a)cs.columbia.edu> writes: > Okay, Zope looks *really*, *really* cool. The one thing I *really* wish > I could do is to NFS mount or SMB mount the object database onto my file > system. Then I wouldn't have to go through the nonsense of uploading > graphic files and Zope would automatically interoperate with all of my > HTML editing tools. > > Even so, I think I'm going to give Zope a whirl on a small project and see > what happens. I like PHP but wish it had exceptions to help create more > robust code. I think that Python is a better programming language, and > the way Zope integrates it into the web environment is very intriguing. > > > Regards, > > Dave Orme > Similar to what you want to do can be done with our FTP support and WebDAV support. With Emacs and FTP, you can move files around transparently, with ncftp you can also recursively FTP files into a server, and with MSIE5.0 you can drag and drop files from your windows machine into Zope using Web Folders. Also I believe kfm can do remote file managment with HTTP. The object database is not a filesystem, although I suppose it *could* be made to look like one to a filesystem transport protocol, like NFS or SMB, with vast amounts of effort. -Michel > > _______________________________________________ > 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 )

2 1

Re: [Zope] Revoking authentication (or: logging out)?
by Service Informatique CHSR 17 Jun '99

17 Jun '99

>>>> Anthony Baxter wrote >> >> If you use the "unsupported" UserDB system, you can store the auth info >> in cookies, instead of basic auth. You can then do a 'logout' by calling >> acl_users/docLogout. >> >> (or even modify the UserDB code to add an expiry date to the cookie) > >oops, almost forgot - we're using this here, and I've had _some_ reports >that IE won't properly expire the cookie I ran into this The problem is with IE 3.x The solution is to strip out the maxage attribute in the SetCookie. Don't know why, but with maxage, IE 3.x fails to expire the cookie. NS 3.x, 4.x, IE 4.x all expire the cookie, even if maxage is not specified. Regards, Jephte CLAIN Service Informatique CHSR

1 0

Re: [Zope] win32 binaries for Zope 2 ?
by michel＠digicool.com 17 Jun '99

17 Jun '99

Kent Tenney <ktenney(a)win.bright.net> writes: > are they available? if so where? > if not, when? > > TIA > Kent > There aren't any any alpha binaries for any platform. This is intentional. You must be able to compile your own to get an alpha working. Binaries are provided for betas. Unfortunately this means we seldom hear from the Windows crowd (other than a few hard core windows folks with compilers) due to the expensive nature of most windows compiler products. There is, however, a glimmer of hope with cygwin. If you can get Zope to compile with cygwin and gcc, then you're on. Please pass on your experiences if you do. If I had Windows, I'd give it a go myself. -Michel > _______________________________________________ > 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] Easily including external HTML file?
by michel＠digicool.com 17 Jun '99

17 Jun '99

Alexander Staubo <alex(a)mop.no> writes: > Afaik, Zope does not have any built-in document grabbing routines, but > it's real easy to do with httplib in an External Method. Something like: > > import httplib > > def GetLinuxToday(): > http = httplib.HTTP('linuxtoday.com') > http.putrequest('GET', '/backend/lthead.inc') > http.putheader('Accept', 'text/html') # we can handle text/html > http.putheader('Accept', 'text/plain') # we can handle text/plain > http.endheaders() > httpcode, httpmsg, headers = http.getreply() # get response headers > if httpcode != 200: > raise "Could not get document" # HTTP error > doc = http.getfile() > data = doc.read() # read file > doc.close() > return data > > Didn't test this inside Zope, but it works standalone. Just put it in a > file in Zope's extensions directory (eg., c:\zope\extensions), add an > External Method, enter GetLinuxToday as id and function name, and call > it up in a document: > >  > Alternatively you could genericize this method in the obvious way (pass the URL into the method) I'll speak to Jim about the possibility of getting something like this into the DTML _ namespace now that 2.0 is concurrent (The problem with this in 1.x is, of course, the getreply() call blocking the Zope process, unless you're *positive* that a blocking call will return almost immediately, you shouldn't use one in 1.x). -Michel PS My new mailer quotes properly! Forget Outlook! PPS (XEmacs with Gnus/nnimap) > Hope this helps. > > -- > Alexander Staubo http://www.mop.no/~alex/ > "`Ford, you're turning into a penguin. Stop it.'" > --Douglas Adams, _The Hitchhiker's Guide to the Galaxy_ > > >-----Original Message----- > >From: dyork(a)mercury.mv.net [mailto:dyork@mercury.mv.net]On > >Behalf Of Dan > >York > >Sent: 17. juni 1999 02:46 > >To: zope(a)zope.org > >Subject: [Zope] Easily including external HTML file? > > > > > >Greetings! I've been lurking for a bit and have appreciated > >the tips and ideas I've picked up on the list. But now I have > >a question. In one of my pages, I would like to pull in the > >latest stories from Linux Today. They explain what to do at: > > > > http://linuxtoday.com/linkus.html > > > >All I would like to do is pull the following file into the > >middle of a DTML document: > > > > http://linuxtoday.com/backend/lthead.inc > > > >Now I frequently did this before with Apache and server-side > >includes, but I can't figure out exactly how to do this using > >DTML. In looking around zope.org, I found the contributed > >external method for pulling in items from Slashdot.org: > > > > http://www.zope.org/Download/Contrib/slashdot.py > > > >And while that works... and I could sit and modify the python > >to have it go pull in the LinuxToday info... I just find myself > >sitting here thinking there's got to be a better way to do this! > >All I want to do is pull in the LT file... no formatting or > >anything else. > > > >Any ideas you have will be appreciated. > > > >Thanks, > >Dan > >-- > >-------------------------------------------------------- > >Dan York, Certification Program Manager, Linuxcare, Inc. > >dyork(a)linuxcare.com http://www.linuxcare.com/ > >415-740-4519 mobile, 603-268-0691 tel, 603-268-0103 fax > >Linuxcare. At the center of Linux. > >-------------------------------------------------------- > > > >_______________________________________________ > >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 ) > > > > > _______________________________________________ > 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] Revoking authentication (or: logging out)?
by michel＠digicool.com 17 Jun '99

17 Jun '99

Jonathan Corbet <corbet(a)eklektix.com> writes: > I'm working on a system to make medical records available via a web > interface. It needs to make different levels of access available to > different sorts of people (doctors, nurses, clerical staff) - a perfect > match for Zope's roles. > > But I've encountered one rub: the web browser will be running on PC's > sitting in various spots in the clinic's offices: the doctor's office, work > areas, even examination rooms. There will be a different person sitting > down at it every few minutes. But, with "Basic" authentication, once the > web browser has your username/password in its clutches, it never lets go. > > We're dealing with medical records here, so it is a poor idea to leave a > "logged in" browser sitting around in a public place. What I am looking > for is a way to put in a "log out" option that stops short of killing and > restarting the browser. Has anybody else figured out a way to do this? >   Will raise a 401 to the browser, typicly causing (netscape at least) to prompt a box saying "Authorization Failed. Retry?" Clickin on yes will bring up a login box. Clicking cancel will show the Unauthorized Error. I guess as a quick hack, you can impliment a client pull in your browser to refetch itself, raising it's own unauthorization, forcing the user to click "Ok" then login again. Alternatively you can tell your users that they *must* as a policy click 'log out' (which takes you to a DTML Method with the above DTML code snippet). Cookies is probably the more elegant solution. -Michel > Thanks, > > jon > > Jonathan Corbet, Eklektix, Inc. > corbet(a)eklektix.com > > _______________________________________________ > 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

Zope does not find import
by steve＠hosys.com 17 Jun '99

17 Jun '99

PersistentExtra.py can't find Persistence, I don't see a file or directory named Persistence ? Using Zope-1.11.0pr1-win32-x86.exe >>> import Project Traceback (innermost last): File "<stdin>", line 1, in ? File "Project.py", line 2, in ? from Globals import HTMLFile, MessageDialog, Persistent File "C:\ZOPE\lib\python\Globals.py", line 90, in ? import Acquisition, ComputedAttribute, App.PersistentExtra, os File "C:\ZOPE\lib\python\App\PersistentExtra.py", line 87, in ? from Persistence import Persistent ImportError: cannot import name Persistent

1 0

RE: [Zope] Easily including external HTML file?
by Alexander Staubo 17 Jun '99

17 Jun '99

>-----Original Message----- >From: michel(a)digicool.com [mailto:michel@digicool.com] >Sent: 17. juni 1999 04:16 >To: Alexander Staubo >Cc: Zope Mailing List (E-mail) >Subject: Re: [Zope] Easily including external HTML file? > > >Alexander Staubo <alex(a)mop.no> writes: > >> Afaik, Zope does not have any built-in document grabbing >routines, but >> it's real easy to do with httplib in an External Method. >Something like: >> >> import httplib >> >> def GetLinuxToday(): >> http = httplib.HTTP('linuxtoday.com') >> http.putrequest('GET', '/backend/lthead.inc') >> http.putheader('Accept', 'text/html') # we can handle text/html >> http.putheader('Accept', 'text/plain') # we can handle text/plain >> http.endheaders() >> httpcode, httpmsg, headers = http.getreply() # get >response headers >> if httpcode != 200: >> raise "Could not get document" # HTTP error >> doc = http.getfile() >> data = doc.read() # read file >> doc.close() >> return data >> >> Didn't test this inside Zope, but it works standalone. Just >put it in a >> file in Zope's extensions directory (eg., c:\zope\extensions), add an >> External Method, enter GetLinuxToday as id and function >name, and call >> it up in a document: >> >>  >> > >Alternatively you could genericize this method in the obvious way >(pass the URL into the method) Thought I'd leave that as an exercise for the reader. >I'll speak to Jim about the >possibility of getting something like this into the DTML _ namespace >now that 2.0 is concurrent (The problem with this in 1.x is, of >course, the getreply() call blocking the Zope process, unless you're >*positive* that a blocking call will return almost immediately, you >shouldn't use one in 1.x). Of course, as with anything else -- lord knows posting rows to an SQL database can take time, too. :-) Which is the reason I (and everybody else, I guess) is looking hungrily towards Zope 2.0 and tearing my hair out in frustration over its unfinished state. :-) >PS My new mailer quotes properly! Forget Outlook! Your new mailer has calendar scheduling, contact management, hierchical folder storage, scheduled offline archiving, custom folder views, and support for Exchange Server? If so, I'll _gladly_ use it. (Okay, I could do without the Exchange part. What a piece of feces. Phewey!) >PPS (XEmacs with Gnus/nnimap) Emacs for mail. Now _that_ sounds fun. >> Hope this helps. >> >> -- >> Alexander Staubo http://www.mop.no/~alex/ >> "`Ford, you're turning into a penguin. Stop it.'" >> --Douglas Adams, _The Hitchhiker's Guide to the Galaxy_ >> >> >-----Original Message----- >> >From: dyork(a)mercury.mv.net [mailto:dyork@mercury.mv.net]On >> >Behalf Of Dan >> >York >> >Sent: 17. juni 1999 02:46 >> >To: zope(a)zope.org >> >Subject: [Zope] Easily including external HTML file? >> > >> > >> >Greetings! I've been lurking for a bit and have appreciated >> >the tips and ideas I've picked up on the list. But now I have >> >a question. In one of my pages, I would like to pull in the >> >latest stories from Linux Today. They explain what to do at: >> > >> > http://linuxtoday.com/linkus.html >> > >> >All I would like to do is pull the following file into the >> >middle of a DTML document: >> > >> > http://linuxtoday.com/backend/lthead.inc >> > >> >Now I frequently did this before with Apache and server-side >> >includes, but I can't figure out exactly how to do this using >> >DTML. In looking around zope.org, I found the contributed >> >external method for pulling in items from Slashdot.org: >> > >> > http://www.zope.org/Download/Contrib/slashdot.py >> > >> >And while that works... and I could sit and modify the python >> >to have it go pull in the LinuxToday info... I just find myself >> >sitting here thinking there's got to be a better way to do this! >> >All I want to do is pull in the LT file... no formatting or >> >anything else. >> > >> >Any ideas you have will be appreciated. >> > >> >Thanks, >> >Dan >> >-- >> >-------------------------------------------------------- >> >Dan York, Certification Program Manager, Linuxcare, Inc. >> >dyork(a)linuxcare.com http://www.linuxcare.com/ >> >415-740-4519 mobile, 603-268-0691 tel, 603-268-0103 fax >> >Linuxcare. At the center of Linux. >> >-------------------------------------------------------- >> > >> >_______________________________________________ >> >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 ) >> > >> >> >> _______________________________________________ >> 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

FYI: Zope Weekly News
by Amos Latteier 17 Jun '99

17 Jun '99

Hello there, Two new Zope releases were announced this week, 2.0a2 and 1.10.3b1. In addition, there were announcements on other exciting Zope projects, and the volume on the Zope mailing lists continued at high levels. There's so much activity in Zope land that people are starting to talk about splitting the Zope lists again. * Zope 2.0 a2 was released. This release is the second alpha source release of Zope 2. Many bugs were fixed, including an important memory leak, and a few new features were added including an experimental DTML syntax. We're getting closer to a Zope 2 beta release, but there's still a ways to go. http://www.zope.org/Download/Releases/Zope-2.0.0a2 * Zope 1.10.3b1 was released. This is a maintenance release for Zope 1.10.2 and mainly plugs a memory leak. http://www.zope.org/Download/Releases/Zope-1.10.3b1/ http://www.zope.org/pipermail/zope/1999-June/005361.html * Butch Landingin announced his "Squishdot" Slashdot clone Zope product. There's also a Squishdot demo site. http://members.tripod.com/squishdot/index.html http://www.libc.org/Community http://www.zope.org/pipermail/zope/1999-June/005241.html * Alexander Staubo announced a client ISAPI extension for Zope which allows efficient communication between IIS 4.0 and Zope. http://www.mop.no/~alex/download/zisapi-0.1.zip http://www.zope.org/pipermail/zope/1999-June/005370.html * Another Zope job posting appeared. http://www.zope.org/pipermail/zope/1999-June/005494.html * Python luminary Jody Winston announced that he is available for Zope contracting work. http://www.zope.org/pipermail/zope/1999-June/005577.html * Alex Rice announced that a New Mexico Zope User's Group is forming. http://www.zope.org/pipermail/zope/1999-June/005427.html * Jules Allen posted a draft of a short Zope tutorial for review. http://zdp.zope.org/test/jules/navbar_howto_html http://www.zope.org/pipermail/zope/1999-June/005290.html * Pavlos Christoforou announced an HTML to PDF Zope product. http://starship.python.net/crew/pavlos/ http://www.zope.org/pipermail/zope/1999-June/005535.html * There was a lot of discussion about DTML syntax and how to improve it. The tone was particularly level-headed this time round. One upshot of the discussion was an announcement of an experimental new DTML syntax for Zope 2.0a2, and the goal of creating some kind of Python Method object which is editable through the web. http://www.zope.org/pipermail/zope/1999-June/005350.html http://www.zope.org/pipermail/zope/1999-June/005357.html http://www.zope.org/pipermail/zope/1999-June/005621.html http://www.zope.org/pipermail/zope/1999-June/005574.html See you next week. -Amos -- Amos Latteier mailto:amos@digicool.com Digital Creations http://www.digicool.com

1 0

RE: [Zope] Revoking authentication (or: logging out)?
by Alexander Staubo 17 Jun '99

17 Jun '99

Cookies or URL embedding are two possible solutions. Cookies store store login state and expiry information, your pages validate this information and emits a 401 if incorrect. This is to augment, not replace, Zope's security system, although most likely you'll be forced to do your own HTTP authentication (which iirc means writing your own UserFolder descendant wrapped in a Zope Product). URL embedding is just a variation that embeds the information in URLs rather than inside cookies (you should obfuscate or hexbin the information to avoid problems); this is typically more relevant on public sites where some users might not have cookies enabled, or is running a non-cookie-enabled browser. Otoh, the downside is you have to put this stuff in every single secure URL. I hear Chris Petrilli is the resident security dude at DC. Perhaps he can shed more light on this subject; I just drained the bulk of my knowledge on this topic. :-) Imho a solution to this problem should be built into Zope, since it's so common. -- Alexander Staubo http://www.mop.no/~alex/ "`Ford, you're turning into a penguin. Stop it.'" --Douglas Adams, _The Hitchhiker's Guide to the Galaxy_ >-----Original Message----- >From: corbet(a)eklektix.com [mailto:corbet@eklektix.com] >Sent: 17. juni 1999 02:41 >To: zope(a)zope.org >Subject: [Zope] Revoking authentication (or: logging out)? > > >I'm working on a system to make medical records available via a web >interface. It needs to make different levels of access available to >different sorts of people (doctors, nurses, clerical staff) - a perfect >match for Zope's roles. > >But I've encountered one rub: the web browser will be running on PC's >sitting in various spots in the clinic's offices: the doctor's >office, work >areas, even examination rooms. There will be a different >person sitting >down at it every few minutes. But, with "Basic" >authentication, once the >web browser has your username/password in its clutches, it >never lets go. > >We're dealing with medical records here, so it is a poor idea >to leave a >"logged in" browser sitting around in a public place. What I >am looking >for is a way to put in a "log out" option that stops short of >killing and >restarting the browser. Has anybody else figured out a way to do this? > >Thanks, > >jon > >Jonathan Corbet, Eklektix, Inc. >corbet(a)eklektix.com > >_______________________________________________ >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