From isaac at structuredsystems.net Tue Oct 8 18:23:53 2002 From: isaac at structuredsystems.net (Isaac Levy) Date: Sun Aug 10 16:54:52 2008 Subject: [Zope-xml] zope <> flash In-Reply-To: Message-ID: Hi Ed, ike here. Well, it seems, there are a few ways I can see to do this: 1) sendAndLoad- this works fine, but for complex/involved data exchanges, it can become EXTREMELY tedious. 2) XML.sendAndLoad- this is better for more rich data, and one could conceivably build XML templates using Zope Page Templates (awesome!!!) that were dynamically generated server side from data. There's tons of other xml-related products for Zope too. 3) XMLRPC- well, we've just crossed paths on this one on the sourceforge side of the fence, (and this is in the works to finally work. The benefits here are the ability to exchange complex data types, and to call methods server-side, exchanging data-objects which each side understands in their respective languages. With that, Zope has a killer xmlrpc mechanism built in. 4) SWF Binary Data Exchanges- some of the open SWF generator products have Zope Components, Ming being one, (with php, python, and other interfaces)- SwiftGenerator being another, but there are some drawbacks here. The first is server-side binary generation takes CPU, the second is that the files are based on the open .swf format, and lacks a great deal of functionality of MX, (and even Flash5, from what I understand). Also, these are pretty steep to get into using. The plus here, is that if one were to be simply passing data back and forth, you'd be doing what MM is selling as remoting. The benefits of this are that the data passed is really tiny, all compressed as a binary, but this gets trumped by #5: 5) XMLRPC passed through an SSL tunnel- SSL, using various encryption mechanisms, compresses the data sent- considerably. The implict benefits of security aside, SSL can effectively trump the above method in file compression- as well as CPU load on a server- (there's a lot more involvment with OpenSSL development/optimization etc... than there could conceivably be on the MM side). Flash does SSL just fine when run in a web browser, and there are libraries for decrypting data using MD5 checksums within flash itself. #3, and #5, are what I'm working on. Hope this helps- Rocket- .ike On Wednesday, September 25, 2002, at 07:45 AM, Ed Colmar wrote: > > I'm a newbie to flash, so please pardon any silly questions. > > What is the most painless way to get data into and out of flash? > > I've found several xml-rpc documents (and I'm already a xml-rpc fan): > > http://www.zope.org/Members/ike/Flash-Zope/FLASH_xmlrpc_zope > http://sourceforge.net/projects/xmlrpcflash > > Is there an easier way? > > I'm hesitant to use the swiftgenerator, as that seems less than ideal. > > I know flash can POST directly into dtml/sql methods, but can it do > anything with the results? > > Thanks for any ideas! > > -ed- > > > > -- > Green Graphics ::: Print and Web Design ::: 510.923.0000 > > > _______________________________________________ > Zope-xml mailing list > Zope-xml@zope.org > http://lists.zope.org/mailman/listinfo/zope-xml > > From santee_sawant at yahoo.co.in Wed Oct 9 10:41:17 2002 From: santee_sawant at yahoo.co.in (=?iso-8859-1?q?santosh=20sawant?=) Date: Sun Aug 10 16:54:52 2008 Subject: [Zope-xml] need help in cataloging heavy xml data.... Message-ID: <20021009144117.76567.qmail@web8206.mail.in.yahoo.com> hi! I am facing problem with cataloging xml data. Firstly I added btree folder, inside that I uploaded xml data thru adding Parsed Xml. I also wrote 2 python scripts to indexing data and added 1 default.. "index_html" as textindex and 1 field index and 1 keyword index. My xml data is very heavy like 4.00 mb. I am cataloging data like... Find objects of type: All types ~~~~~~~~~ with ids: containing: expr: nodeName=='h2' ~~~~~~~~~~~~~~ modified: before ~~~~~~ where the roles: have permission: but cataloging could not complete, after hour or so, the zope server shows... 'The page cannot be displayed' error. Is there any other way to catalog so heavy data? Please let me know about this ASAP. Thank you very much for ur kind support. With best regards, Santosh Sawant ________________________________________________________________________ Missed your favourite TV serial last night? Try the new, Yahoo! TV. visit http://in.tv.yahoo.com From paul at eurozope.org Wed Oct 9 12:47:51 2002 From: paul at eurozope.org (Paul Everitt) Date: Sun Aug 10 16:54:52 2008 Subject: [Zope-xml] need help in cataloging heavy xml data.... In-Reply-To: <20021009144117.76567.qmail@web8206.mail.in.yahoo.com> Message-ID: Couple of questions... 1) What version of Zope are you using? 2) Have you tried greatly increasing the size of your Zope object cache (Control Panel -> Database -> Cache) -- perhaps to 4000? 3) How is the data getting loaded, how many data files, how big is the data, etc.? 4) If you are indexing via a Python script, you might try using subtransactions. --Paul On mercredi, oct 9, 2002, at 16:41 Europe/Paris, santosh sawant wrote: > hi! > > I am facing problem with cataloging xml data. > Firstly I added btree folder, inside that I uploaded > xml data thru adding Parsed Xml. > I also wrote 2 python scripts to indexing data and > added 1 default.. "index_html" as textindex and 1 > field index and 1 keyword index. > > My xml data is very heavy like 4.00 mb. > > I am cataloging data like... > > Find objects of type: All types > ~~~~~~~~~ > with ids: > > containing: > > expr: nodeName=='h2' > ~~~~~~~~~~~~~~ > modified: before > ~~~~~~ > where the roles: > > have permission: > > > but cataloging could not complete, after hour or so, > the zope server shows... > > 'The page cannot be displayed' error. > > Is there any other way to catalog so heavy data? > > Please let me know about this ASAP. > > Thank you very much for ur kind support. > > With best regards, > Santosh Sawant > > _______________________________________________________________________ > _ > Missed your favourite TV serial last night? Try the new, Yahoo! TV. > visit http://in.tv.yahoo.com > > _______________________________________________ > Zope-xml mailing list > Zope-xml@zope.org > http://lists.zope.org/mailman/listinfo/zope-xml > From santee_sawant at yahoo.co.in Fri Oct 11 03:18:13 2002 From: santee_sawant at yahoo.co.in (=?iso-8859-1?q?santosh=20sawant?=) Date: Sun Aug 10 16:54:53 2008 Subject: [Zope-xml] page cannot be displayed error while cataloging xml data Message-ID: <20021011071813.13895.qmail@web8206.mail.in.yahoo.com> Hi! Let me explain me whole with new changes made. My Xml file structure is ::

The h2 element data is more important which needs to show as search results. I have uploaded all 52 xml files of diff. sections thru ParsedXML, which I could have make 1 xml file and upload it. But keeping in mind, the future maintenance, updation of xml data and better performance while showing search results content, I made it section wise. Total h2 elements inside 52 xml files are 9750. The words inside 52 xml files are 639451. All of 52 files are uploaded inside btree folder. I am creating vocabulary while creating ZCatalog. There is only one TextIndex defined, no metadata or other indexes defined. I gave "index_html" name to TextIndex which is method of ParsedXML, which reads the whole xml data like document_src method in XMLDocument. *** While cataloging, "Find Objects"-- I am secifying..... Find objects of type: All Types with ids: containing: expr: nodeName=='h2' modified: before where the roles: have permission: Advanced: Subtransactions are Enabled Subtransaction threshold 10000 *** Database Management at /Control_Panel/Database Total number of objects in the database 763352 Total number of objects in all caches 155 Target Size 100000 Target max time between accesses 60 I could not catalog so heavy xml data. And I am getting error after hour or so like.... "Page cannot be displayed." Is there any other way to catalog xml data? -- santosh sawant ============================================================================================= On jeudi, oct 10, 2002, at 14:12 Europe/Paris, santosh sawant wrote: Hi Paul, 1) What version of Zope are you using? -- Zope 2.3.3 (binary release, python 1.5.2, win32-x86) 2) Have you tried greatly increasing the size of your Zope object cache (Control Panel -> Database -> Cache) -- perhaps to 4000? -- No Unfortunately, changing the cache size might not be as helpful under 2.3.x as needed. Under Zope 2.6b1 (or current checkouts) the usefulness of the cache settings is improved. Also, there is useful report that shows cache activity. I tried with cache setting - 40000, but it didn't work. 3) How is the data getting loaded, how many data files, how big is the data, etc.? -- I am uploading xml data thru adding ParsedXML into BTreeFolder. Total data files are 46, total data size is around 4.00 MB. Are you loading the xml data files one at a time using a web form? :: yes Just to make sure I am correct: uploading the files is not a problem, but running the Python script is the problem, right? How many elements are in each xml data file? :: total elements are 9975 which I trying to index thru ZCatalog. Hmm, this doesn't seem like enough to cause such problems. 4) If you are indexing via a Python script, you might try using subtransactions. -- Subtransactions are Enabled. Subtransaction threshold - 10000 Hmm, you have your subtransaction threshold set much higher than your object cache size. You might be in cache thrash. :: I tried with threshold - 100000, but it didn't work. Ok. On the other hand, if you are loading your files one-at-a-time, you might never be reaching 10000 objects. :: But the ZCatalog creates object as per the expression given in "find objects", irrespective the file nos and size. They are marginally related. The catalog creates objects relative to the number of words (or in your case, elements) in the original data. And I am giving expr= nodeName=='h2', so h2 elements are total 9975. Did you say that the total of all elements in all files was 9,975? So the indexed objects/ catalog contents should be 9975. Do you have any statistics on cpu/memory/disk usage to indicate what is happening on the machine? :: While cataloging zope server occupies 100% CPU usage and memory usage is between 30000 to 100000 bytes. That sounds normal. I apologize, I didn't read your original note carefully enough. Here are some more points: 1) You get the "Page could not be completed" message, but that's simply because the browser times out and gives up. You could move your script to an external method and run the task as a thread, in the background. 2) What kind of indexes do you have defined on the elements -- full-text, keyword, ??? --Paul ============================================================================================= On jeudi, oct 10, 2002, at 07:04 Europe/Paris, santosh sawant wrote: 1) What version of Zope are you using? -- Zope 2.3.3 (binary release, python 1.5.2, win32-x86) By the way, Jim Fulton (CTO at Zope Corp) is in India right now. (I'm guessing by your email address that you're there also.) 2) Have you tried greatly increasing the size of your Zope object cache (Control Panel -> Database -> Cache) -- perhaps to 4000? -- No Unfortunately, changing the cache size might not be as helpful under 2.3.x as needed. Under Zope 2.6b1 (or current checkouts) the usefulness of the cache settings is improved. Also, there is useful report that shows cache activity. 3) How is the data getting loaded, how many data files, how big is the data, etc.? -- I am uploading xml data thru adding ParsedXML into BTreeFolder. Total data files are 46, total data size is around 4.00 MB. Are you loading the xml data files one at a time using a web form? How many elements are in each xml data file? 4) If you are indexing via a Python script, you might try using subtransactions. -- Subtransactions are Enabled. Subtransaction threshold - 10000 Hmm, you have your subtransaction threshold set much higher than your object cache size. You might be in cache thrash. On the other hand, if you are loading your files one-at-a-time, you might never be reaching 10000 objects. Do you have any statistics on cpu/memory/disk usage to indicate what is happening on the machine? --Paul ============================================================================================= Couple of questions... 1) What version of Zope are you using? 2) Have you tried greatly increasing the size of your Zope object cache (Control Panel -> Database -> Cache) -- perhaps to 4000? 3) How is the data getting loaded, how many data files, how big is the data, etc.? 4) If you are indexing via a Python script, you might try using subtransactions. --Paul On mercredi, oct 9, 2002, at 16:41 Europe/Paris, santosh sawant wrote: > hi! > > I am facing problem with cataloging xml data. > Firstly I added btree folder, inside that I uploaded > xml data thru adding Parsed Xml. > I also wrote 2 python scripts to indexing data and > added 1 default.. "index_html" as textindex and 1 > field index and 1 keyword index. > > My xml data is very heavy like 4.00 mb. > > I am cataloging data like... > > Find objects of type: All types > ~~~~~~~~~ > with ids: > > containing: > > expr: nodeName=='h2' > ~~~~~~~~~~~~~~ > modified: before > ~~~~~~ > where the roles: > > have permission: > > > but cataloging could not complete, after hour or so, > the zope server shows... > > 'The page cannot be displayed' error. > > Is there any other way to catalog so heavy data? > > Please let me know about this ASAP. > > Thank you very much for ur kind support. > > With best regards, > Santosh Sawant ============================================================================================= ________________________________________________________________________ Missed your favourite TV serial last night? Try the new, Yahoo! TV. visit http://in.tv.yahoo.com From toman at marge.cyber-dyne.com Tue Oct 15 14:53:00 2002 From: toman at marge.cyber-dyne.com (toman) Date: Sun Aug 10 16:54:53 2008 Subject: [Zope-xml] Windows install? Message-ID: <3DAC640C.60508@marge.cyber-dyne.com> How do I get PyXML running on Windows? Specifically, the windows installer doesn't work because (guessing) the version of python in the Zope heirarchy has no registry settings associated with it, so it doesn't know I have a python 2.1 installed. Trying to build from source doesn't seem to work either. There aren't any includes with the Zope install and my regular python install is v2.2 . Finally, replacing pyexpat.pyd with the one I built against python v2.2 doesn't work either. Ideas? I guess I could find a box and install python 2.1 and VC++ and build there, though that seems somewhat painful. Given I have a good build, where does it get installed in the Zope heirarchy and how much of it do I need? J. Toman From rik.hoekstra at inghist.nl Wed Oct 16 03:48:57 2002 From: rik.hoekstra at inghist.nl (Rik Hoekstra) Date: Sun Aug 10 16:54:53 2008 Subject: [Zope-xml] Windows install? References: <3DAC640C.60508@marge.cyber-dyne.com> Message-ID: <3DAD19E9.8010904@inghist.nl> toman wrote: > > How do I get PyXML running on Windows? Specifically, > the windows installer doesn't work because (guessing) > the version of python in the Zope heirarchy has no registry > settings associated with it, so it doesn't know I have a python 2.1 > installed. Have your tried unzipping the exe manually with winzip to your zope installation (to the /bin directory?). This worked for me. Rik Hoekstra From toman at marge.cyber-dyne.com Wed Oct 16 11:35:39 2002 From: toman at marge.cyber-dyne.com (toman) Date: Sun Aug 10 16:54:53 2008 Subject: [Zope-xml] Windows install? References: <3DAC640C.60508@marge.cyber-dyne.com> <3DAD19E9.8010904@inghist.nl> Message-ID: <3DAD874B.5050209@marge.cyber-dyne.com> Rik Hoekstra wrote: > toman wrote: > >> >> How do I get PyXML running on Windows? Specifically, >> the windows installer doesn't work because (guessing) >> the version of python in the Zope heirarchy has no registry >> settings associated with it, so it doesn't know I have a python 2.1 >> installed. > > > Have your tried unzipping the exe manually with winzip to your zope > installation (to the /bin directory?). This worked for me. > > Rik Hoekstra Actually I followed my own advice: installed VC++ and python 2.1 on a different server, compiled it and installed. The install put it in the wrong place (probably my fault) so I had to move the installed directories manually. Anyhow, the ParsedXML Product showed up so I guess it's working. Is there any one-stop shopping for API documentation or am I back to crawling through code like I did for Zope? J. Toman From dev at projekt01.ch Thu Oct 24 08:15:01 2002 From: dev at projekt01.ch (Projekt01, Roger Ineichen) Date: Sun Aug 10 16:54:53 2008 Subject: [Zope-xml] ParsedXML pyexpat other parser? Message-ID: <000401c27b56$fa9a1100$0b01a8c0@projekt01.local> Hello together I tried the product ParsedXML. This product tries to import the class pyexpat. Both the CVS and 1.3 version. I read somthing that this product schould work with 4XPath. Whats schould I do for to use the product? Wich version of pyexpat is ok? I run it on Win XP for development. Mit freundlichen Gr?ssen Roger Ineichen ___________________________ Projekt01 GmbH www.projekt01.ch Langackerstrasse 8 6330 Cham phone +41 (0)41 781 01 78 mobile +41 (0)79 340 52 32 fax +41 (0)41 781 00 78 email r.ineichen@projekt01.ch ___________________________ END OF MESSAGE From r.ineichen at projekt01.ch Thu Oct 24 11:52:07 2002 From: r.ineichen at projekt01.ch (Projekt01, Roger Ineichen) Date: Sun Aug 10 16:54:53 2008 Subject: [Zope-xml] ParsedXML won't work Message-ID: <000a01c27b75$4e63f940$0b01a8c0@projekt01.local> Hello It's not possible to install ParsedXML on zope 2.6 win. Can't import pyexpat. Any idea or hints? Mit freundlichen Gr?ssen Roger Ineichen ___________________________ Projekt01 GmbH www.projekt01.ch Langackerstrasse 8 6330 Cham phone +41 (0)41 781 01 78 mobile +41 (0)79 340 52 32 fax +41 (0)41 781 00 78 email r.ineichen@projekt01.ch ___________________________ END OF MESSAGE From fred at zope.com Thu Oct 24 11:56:41 2002 From: fred at zope.com (Fred L. Drake, Jr.) Date: Sun Aug 10 16:54:53 2008 Subject: [Zope-xml] ParsedXML won't work In-Reply-To: <000a01c27b75$4e63f940$0b01a8c0@projekt01.local> References: <000a01c27b75$4e63f940$0b01a8c0@projekt01.local> Message-ID: <15800.6201.136529.82198@grendel.zope.com> Projekt01, Roger Ineichen writes: > It's not possible to install ParsedXML on zope 2.6 win. > Can't import pyexpat. If your Python does not include the pyexpat module, install the PyXML package: http://pyxml.sourceforge.net/ Python on Windows bundles the required module, but it is not always part of the default build on Unix systems since it requires a 3rd-party library. PyXML 0.8 and newer bundles the library, as will Python 2.3 when released. -Fred -- Fred L. Drake, Jr. PythonLabs at Zope Corporation From huxley1 at westnet.com.au Thu Oct 24 21:53:47 2002 From: huxley1 at westnet.com.au (Jeremy Rew) Date: Sun Aug 10 16:54:53 2008 Subject: [Zope-xml] ParsedXML issue In-Reply-To: <20021002031002.2774.38442.Mailman@mail.python.org> Message-ID: <5.1.1.6.0.20021025094341.00b45d10@localhost> I have installed ParsedXML into Zope on RedHat 7.1 with python 2.1, and it all appears to work with the examples provided - eg: the slides - given with Examples.zexp. The problem is that any time I change any part of the xml file, or try to use my own, I get the following error from zope: Error Type: AttributeError Error Value: meta_type The same error occurs when i change the XML file back to exactly how it was when it was working. Has anyone had this problem/know how to fix it. Any help will be great. Thanks Jeremy