- Zope - Zope lists

[ANN] TextIndexNG 3.0.2 released
by Andreas Jung 02 Jun '05

02 Jun '05

F I am pleased to announce the release of TextIndexNG V 3.0.2. TextIndexNG V 3 is a complete new implementation based on Zope 3 technologies and can be used both in Zope 2.8 or 2.7 (with Five) or in Zope 3. What's new? - multi-field indexing and query support - multi-lingual support - configurable converters (through ZCML) - new indexing API (allowing you to hook your custom content types with TextIndexNG through Zope 3 adapters). Changes since V 3.0.2: - multi-field query support added: the query parser now accepts queries where you can search within multiple fields using a single query e.g. 'title::phrase(The Zope Book) and author::or(michel amos)' which searches for all documents with the exact title 'The Zope Book' and where the author field contains 'michel' or 'amos'. See doc/README.txt for details on the query syntax. - new extension module 'txngstemmer' providing language dependant stemmer support based on the Snowball stemmer library. This is a reimplementation of the older PyStemmer extension module based on the new Snowball reimplementation allowing a much cleaner and easier integration. Outstanding compilation problems and problems with unicode should be finally resolved. Stemmer support will be added to TextIndexNG in a later version. Requirements: - Zope 2.8 or Zope 2.7 (+ Five) or Zope 3 (Zope 3 has nothing like catalogs yet) TextIndexNG V3 will *not* run with bare Zope 2.7 installations without Five (and this will not be changed in the future). Download: http://sf.net/projects/textindexng For installation and documentation issues refer to doc/README.txt from the archive. It's basically the same procedure as with former versions except you *need* to recompile the extension modules. Windows binaries of the required extension modules are currently not available (any volunteers?). TextIndexNG V 3 is published under the GPL. Andreas Jung ----------------------------------------------------------------------- - Andreas Jung ZOPYX Software Development and Consulting - - E-mail: info(a)zopyx.com Web: www.zopyx.com - -----------------------------------------------------------------------

1 0

DTML in Zope 2 vs Zope 3
by David H 01 Jun '05

01 Jun '05

Hi, I understand that DTML will be in Zope 3. My question is, do "namespaces" and "context" also migrate to Zope 3? If not, could break existing code? David

11 17

Visual Reporting tool.
by Fernando Lujan 01 Jun '05

01 Jun '05

Hi, I want to know, if is there a visual reporting tool? Like JasperReport, for example, wich works with reportlab's libraries and generate a .py. Thanks and regards, Fernando Lujan

1 0

Problem with ZPsyopg
by David 01 Jun '05

01 Jun '05

Zope 2.7 / ZPsyopgDA / PostgreSQL7.3.4 My problem is that we're running a lot of little websites on a single server. Each site has its own zope/zpsyopg instance and its own postgresql database. But all databases are served by a single postmaster (postgresql instance). The issue arising is that I'm reaching max_connections (and as this ties into memory, there are limits to how high I can set this without affecting the performance of the machine), while most connections sit around being idle most of the time. I had a look at pgpool, but if I understand it correctly, it won't help (or will any other pooling solution), as it requires as many connections to the backend, as it's serving to clients. (Or did I get this wrong?) I do understand how I can reduce the number of threads within Zope to reduce the number of connections, but that only goes so far. Is there a way how I can configure my system, so that idle connections expire when not used for a certain time? Or is there another solution to the problem? Thanks DR

2 1

Saving the html-code
by Flemming Bjerke 01 Jun '05

01 Jun '05

With a dtml-script I renders data from a mysql-database. But, sometimes I should like to save the html-code generated by the script. When I try to do this with a pythonscript I run into problems with acquiring the html-code generated by the dtml-script. I looks like the dtml-script does not get the proper namespace. How do I solve this? Flemming

3 3

Namespace for PythonScript
by Hong Yuan 01 Jun '05

01 Jun '05

Hello, If a PythonScript is called from a Page Template, how can it access the variables defined using: <tal:define="global ..." /> in the calling script? Of course, one can pass the variables to the PythonScript explicitly as parameters, but is there any other way to let the PythonScript to easily access the name space of its calling page template? Hong Yuan

4 3

using DOM from within Zope
by karl.eklund＠amnesty.se 01 Jun '05

01 Jun '05

Maybe this is the wrong way or something, but how can I use xml.dom.minidom (or other module in its entirety) from within Zope? I've tried various stuff in the "MyScriptModules" Product. This same approach seems to work with the "re" module. Example: Products/MyScriptModules/__init__.py: from Products.PythonScripts.Utility import allow_module ... allow_module("xml") allow_module("xml.dom") allow_module("xml.dom.minidom") Python Script "inside" Zope: import xml.dom.minidom xmlstring="<a><b>texttextext</b><c>hejhejhej</c></a>" domtree=xml.dom.minidom.parseString(xmlstring) return domtree.toxml() Result: Username/Password question. "Error Type: Unauthorized Error Value: You are not allowed to access 'toxml' in this context"

4 3

Creating links dynamically
by John Poltorak 01 Jun '05

01 Jun '05

I want to provide a set of links on a Zope site dynamically. ie don't want to hard code a number of links, but would like to load them from an external source which I can add to independently, but I have no idea as to how to set about it. Any ideas? I'd be quite happy to have the links in a text file with one link per line. -- John

7 17

XML Content handling questions.
by Nikko Wolf 01 Jun '05

01 Jun '05

Three questions, in order of importance (to me): 1) Is it possible to validate an (uploaded) XML file against a schema (currently on the Zope server)? What, if any, extra packages are needed? I want to use this in my upload forms to alert the user if their files are invalid. 2) I also need to serve up web-based forms and allow the user to create XML content based on a user-selected schema. Something like what Chiba (http://chiba.sf.net/) is targeting. Of course, though, that isn't a Python/Zope solution. Has anyone seen anything that works? At this point, installing Tomcat and Chiba is what I presume I'll be doing in a couple of weeks.... 3) Is there a function built-in to Zope that allows merging XML files? (ie: all the <abc> within a list of user-selected files). I've got an Extension that works correctly (using the Amara/4Suite packages), so an answer here is mostly for me to learn Zope better. Thanks, Nikko. PS: I'm pretty new to Python & Zope, but not web servers or other programming. Server: Fedora Core 3 Python 2.3.4 Zope 2.7.6-final Plone 2.0.5 4Suite 1.0b1 Amara 1.0b2

3 3

ANNOUNCE: Pound - reverse proxy and load balancer - v1.9
by Robert Segall 01 Jun '05

01 Jun '05

This is to announce the release of Pound v1.9. Changes in this version: - Added the VerifyList configuration flag (CA root certs + CRL) - CRL checking code - RewriteRedirect 2 - ignores port value for host matching - Added -c flag (check-only mode) - Added -v flag (verbose mode) - Added -p flag for pid file name Problems fixed: - fixed a potential buffer overflow problem (in checking the Host header) - added call to SSL_library_init - added a check for MSIE before forcing SSL shutdown - X-SSL-Cipher header is added only if HTTPSHeaders is non-zero - added code for shorter linger on badly closed connections (IE work-around) - fixed the locking for session checking (mutex_lock/unlock) The software is at version 1.9 (production quality). Further testing (especially under heavy loads), improvements and suggestions are welcome. What is Pound (from the Web page) - a reverse-proxy: it passes requests from client browsers to one or more back-end servers. - a load balancer: it will distribute the requests from the client browsers among several back-end servers, while keeping session information. - an SSL wrapper: Pound will decrypt HTTPS requests from client browsers and pass them as plain HTTP to the back-end browsers. - an HTTP/HTTPS sanitizer: Pound will verify requests for correctness and accept only well-formed ones. - an HTTP/1.1 to 1.0 proxy. - a fail-over-server: should a back-end server fail, Pound will take note of the fact and stop passing requests to it until it recovers. - a request redirector - it will pass client requests to separate groups of servers, based on required URL and the presence or absence of headers (pattern matching) Pound was specifically developed to serve as a front-end for multiple instances of Zope running on top of a common ZEO storage, but can be used with any other web server. A patch for the Python source of z2.py is available as part of the distribution to allow using Pound's SSL capabilities. Pound is currently in use in several medium to large volume sites and seems to be holding up quite well: largest reported site does about 5.5 million requests per day, peaking at over 450 requests per second. The speed is well superior to other, equivalent solutions, the configuration is much simpler and the necessary resources (CPU, RAM, disk) are much, much lower. Pound is distributed under the GPL. For further information, download, etc. please see http://www.apsis.ch/pound Pound has its own mailing-list - just send a message to pound-subscribe(a)apsis.ch in order to join. Enjoy and let me know how it works for you. -- Robert Segall Apsis GmbH Postfach, Uetikon am See, CH-8707 Tel: +41-44-920 4904

1 0