Re: [VOTE] Todd Byrne as new Xindice committer

Vadim Gritsenko wrote: Hi, I'd like to propose Todd Byrne as new Xindice committer. He authored several patches and he is still around after that, even if he is less active than I'd like to :) So to nudge him to be more active with Xindice, I vote +1 to make him a committer. Please vote.

Re: Incompatible change in xalan-java [Fwd: [GUMP@brutus]: Project xml-xindice (in module xml-xindice) failed]

Hi, WRT the "new Compiler" problem, xalan developers removed static methods from the FunctionTable class. Now, when they instantiate a new Compiler, they pass an instance of a function table to the Compiler instead of having the Compiler use static methods of FunctionTable. Previously-static method

Re: Extending org.apache.xindice.core.indexer.Indexer

methods Object evalValComparison(int op, String owner, int pos) vs. funcContains(List args). Unfortunately, job responsibilities prevent me from working on this right now. :( When we are not close to the end of a product release cycle, my employer will allow me to donate company-paid time to the eff

Re: Extending org.apache.xindice.core.indexer.Indexer

Andy Armstrong wrote: > My current approach is to have a one-to-one relationship between Xindice > documents and Lucene's concept of a document but it's just occured > to me that I should maybe have a Lucene document for each chunk > of text in a Xindice document. One may want to implement a full t

Re: Registering the 'xmldb:' URL protocol in Java

I happened to be looking at an interesting article on the subject of protocol handlers the other day. You might be interested: /Tutorials & Code Camps/ /A New Era for Java Protocol Handlers/ http://java.sun.com/developer/onlineTraining/protocolhandlers/ -Terry Gianugo Rabellino wrote: Murray Althei

Re: XPath Query attribute/text/comment/etc results

+1 for (a). Like to see the attributes named. -Terry Vadim Gritsenko wrote: Hi all, Is there any opinion on how results of the XPath query returning attribute(s), text node(s), comment node(s), etc should look like? Up to the couple of days ago, it would just add those to the resulting document,

Re: Xindice full text indexing

I think one good way to make a full-text indexer useful in current XPath syntax would be to use it to build a substring index useful for assisting in evaluation of the "contains" and "ends-with" operators. Perhaps index all 3 character substrings and modify XPathQueryResolver and friends to look

Re: Xindice full text indexing

I assume your indexer only indexes "whole words" (perhps with stemming) and therefore would not be useful in assisting evaluation of the "contains" or "ends-with" functions (as a substring index would be useful for)? e.g. if the text was: "some words" contains("some words", "wor") would return true

Re: XML:DB XUpdate List

Jeroen Breedveld maintains the XUpdate reference implementation. I had him make a change to the implementation a few months ago to make it a bit more backward compatible. He was very responsive. Some of the changes he made in the most recent revision rendered the reference implementation incompati

Re: Top level tag attributes: src:col, src:key

a return XML set. Data is sacred. If you wish to have this information available it should be in the Resource object separate from the content. As it is for each and every returned item resource I must strip this out. Just my opinion. -Original Message----- From: Terry Rosenbaum [mailto:[EMAIL

Re: Top level tag attributes: src:col, src:key

These attributes should appear once in the result. That is not a bug. Currently, there is no way to turn off the source attribution information (src:col and src:key). The bug was that the xmlns:src and other namespace definitions were appearing in each element of the result thus providing useless r

Re: Bug in org.apache.xindice.core.xupdate.XObjectImpl

See below... David Kensche wrote: Hi, I found a bug. In org.apache.xindice.core.xupdate.XObjectImpl line 105 says: return (NodeList) _xobj.nodeset(); which causes a ClassCastException. This fixes: return _xobj.nodelist(); The Xalan API changed. Use early-model Xalan version with early model Xin

Re: Lexus as an Apache project?

s.html and look for "CVS Repositories" -Terry Alain Javier Guarnieri del Gesu wrote: * Terry Rosenbaum <[EMAIL PROTECTED]> [2003-10-24 02:03]: Michael Wechner wrote: I recently fixed a bug within Lexus, but it seems that the Lexus community at xmldb is kind of dead, or am I wr

Re: Lexus as an Apache project?

Xindice does not even use the latest available revision of Lexus. The latest Lexus implementation in the xmldb project's CVS is not quite compatible with Xindice. I have been picking away at that, working toward integrating the latest Lexus version into Xindice. In the latest version, there have be

[PATCH] 4 Patches Added to bugzilla

4 new items added in bugzilla: 23595 - [PATCH] Unit Test Enhancements Indexed search tests (tests that searches use available indexes). Tests for the following 3 bugs. 23956 - [PATCH] XPathQueryRseolver doesn't evaluate functions Due to a bug XPath functions were n

Re: Patches and Unit tests

Kevin, I supplied a unit test with my patch. However, as noted in the bug report, the fix exposes lots of Xincide code that has not been executed/tested yet. This afternoon I uncovered a bug in some of that untested function evaluation code. A fix for that is underway. Please hold off on applying t

[PATCH] Bug 23468 (starts-with and other functions not evaluated using index)

Bug 23468 starts-with and other functions not evaluated using index Actually, functions have not been evaluated at all by Xindice since the beginning (due to the bug)! Noticed that queries using starts-with (e.g. //person[starts-with(last, 'Smi')]) were taking the same length of time whether or no

Re: BUG 22156 - Attributes can't be selected

See below... Kevin O'Neill wrote: On Fri, 26 Sep 2003 11:46:05 -0400, Terry Rosenbaum wrote: Hi, I'm currently working on this issue. From the bugzilla report: - 22156 - /XPath queries in the form //foo/@bar fail due to the way that resul

Re: [PATCH] BUG 22122 (src prefix added to search...) REOPENED

eill wrote: On Thu, 25 Sep 2003 13:32:32 -0400, Terry Rosenbaum wrote: Hi, Re: BUG 22122 (src prefix added to search results overwrites existing src prefix) I've reopened this bug and attached a patch. Could someone review/commit please? The descriptive comment I made in bugzilla follows: T

BUG 22156 - Attributes can't be selected

Hi, I'm currently working on this issue. From the bugzilla report: - 22156 - /XPath queries in the form //foo/@bar fail due to the way that result sets are/ /created. / /I've added unit tests that expose the issue./ /--- Additional Comments From Kevin O

[PATCH] BUG 22122 (src prefix added to search...) REOPENED

Hi, Re: BUG 22122 (src prefix added to search results overwrites existing src prefix) I've reopened this bug and attached a patch. Could someone review/commit please? The descriptive comment I made in bugzilla follows: The fix for this bug can produce undesirable results. Sometimes the "src" pr

Re: 1.1 - The bug list is growing ... sorry

I think a list of known issues would be very useful. Should not prevent release of the current version. Here's one of my pet issues that I would add to such a list: XPathQuery/ResourceSet Design Limitations No bug report on this, but I consider it a major outstanding issue. See Thread "XPat

Re: XMLDB API

Have you been in touch with "Jeroen Breedveld" <[EMAIL PROTECTED]> ? I think he is the last remaining active commiter to the project. I believe the company he works for (X-Hive) has an interest in XMLDB API. -Terry Kevin O'Neill wrote: The founders of the XMLDB API seem to have drifted off onto ot

Re: [CROSS POST] RE: current state of XMLDB initiative?

t which does appear in the archive (forwarded by Tom Bradford). The message received no replies: ------- From: Terry Rosenbaum <[EMAIL PROTECTED]> Date: Thu Sep 19, 2002 9:09:33 PM America/Phoenix To: xindice-dev <[EMAIL PROTECTED]>, [EMAIL PROTECTE

saxon-aelfred.jar ??

Hi, I noticed that saxon-aelfred.jar was added to build.xml as one of the "core.jars" about 5 weeks ago (1.40 by Gary Shea via Kevin Ross) But, saxon-aelfred.jar is not in the java/lib

Re: minimum jvm requirements for xindice 1.1?

Users of the embedded version may care more about this choice. As for me, stuck on 1.3 for now due to reasons beyond my control. So, I request JVM >= 1.3. -Terry Kevin Ross wrote: My vote, 1.4+ BUT, that's just me. Everyone else? I'm afraid some people might be stuck on older vm(s) bas

Re: [VOTE] Release Manager for 1.1

I believe that only commiters are allowed to vote. Since I am not a commiter, I will merely express my opinion as osspoed to "voting". +1 -Terry Kevin Ross wrote: Complying with Apache conventions, I would like to volunteer as release manager for 1.1. A release plan

Re: 1.1 Release Plan

I too agree that a release of 1.1 is long overdue. I think it's a bit like getting married or having children -- if you wait for everything to be perfect, you'll be single and/or childless your whole life. Take the plunge. Just do it! -Terry Kimbro Staken wrote: This is fine with me. I held

Re: MemValueIndexer

Original Message- From: Terry Rosenbaum [mailto:[EMAIL PROTECTED] Sent: Thursday, June 05, 2003 1:26 AM To: [EMAIL PROTECTED] Subject: MemValueIndexer Attached are a couple of classes that you may find useful. Not a major piece of work, but I'm donating these to the cause if anyone

MemValueIndexer

ice.core.query.QueryEngine; /** * Provides an in-memory implementation of Xindice Indexer * based upon the Java TreeSet class. This implementation * is not persistent. A MemValueIndexer is useful for indexing * of temporary Collections (e.g. those based on MemFiler). It is also * useful for

Re: XPathQueryResolver Always Collection Scans On UNION [fix]

XPathQueryResolver-new.java > diffs.txt The key thing to recognize is that a union is easily evaluated by or'ing keysets. Evaluate the left into a keyset, evaluate the right into a keyset, or the left and right keysets together. -Terry Gianugo Rabellino wrote: Terry Rosenbaum wrote: Whenever a UNION

XPathQueryResolver Always Collection Scans On UNION [fix]

Hi, Whenever a UNION operator ("|") is encountered in an XPath, the XPathQueryResolver gives up on using indexes to resolve the query and reverts to a collection scan. This can have horrible performance consequences for a large database even if the proper indexes are in place. A suggested fix is to

ValueIndexer for Multi-Byte Numeric Types Broken

Hi, I've discovered that defining a ValueIndexer with multi-byte numeric types (e.g. long, int, short, etc.) does not work correctly when resolving relational comparisons other than equals (e.g. can produce incorrect results for gretater than, less than, etc.). The reason is that all data is handle

Re: Proposed Patch

If you want to eliminate the wrapper, do not forget that you must also deal with the embedded driver code (which cannot return results as an XMLRPC array). -Terry Christian Gross wrote: At 23:43 1/13/2003 +1100, you wrote: I must confess I am not well versed in Java, but is there much performanc

Re: Concurrency

There is non-thread-safe code lurking in the XMLDB XUpdate implementation. See bug 13745 . This code is executed only if/when you use XUpdate. -Terry Mark J. Stang wrote: I believe it is thread safe. However, there isn't any "locking" avai

Re: New security code in scratchpad

Gianugo Rabellino wrote: OK, until now you are making a technical point from a convenience functionality. Which is fair, but IMHO doesn't really pay off from what we are losing from an architectural and developer POV: we have to maintain two drivers, keeping them in sync and evolving them at the

Re: Embed driver: getId returns null... bug?

Vladimir R. Bossicard wrote: Very good explanation. So now the next (and last) step will be to put this into the documentation. Terry, thanks for volunteering. :-) Done. Attached is the updated guide-developer.xml file containing a new section below the section entitled "Querying with XPath". C

Re: Embed driver: getId returns null... bug?

e to expect the Java API to behave as it would when you call Collection.getResource(id). Of course it is understood that if you query a resource without an id, there is no way to populate with one. I believe that *this* is the execptional case where the getID() shoud be null. -Kevin Terry Rosen

Re: Embed driver: getId returns null... bug?

The problem is that depending on the XPath, a query can return all or part of a document. Documents have a resource ID (the document ID) and as such can be retrieved (in their entirity) from the database using that ID. A part of a document has no ID. The part can only be retrieved using an XPath th

Re: Bugzilla

Go to the query page. (http://nagoya.apache.org/bugzilla/query.cgi) Use multi-select in Status: field to highlight UNCONFIRMED through REOPENED inclusive. (To see all bugs in any state, unselect all items in the Status: field). Select Xindice in Program: field. Hit Submit Query button. Enjoy. -Te

Re: speeding up XPATH

> Is there any way to speed ... [xpath] up ? One thing that came up recently on this list was a defficiency in xerces causing slow XPath searching. A workaround for the problem is to supply the following command line arguments to your JVM: -Dorg.apache.xml.dtm.DTMManager=org.apache.xml.dtm.ref.DTMM

Re: servlet issues

> you already have multiple interfaces, so thread safety must already be taken into account at the core level Xindice only supports one Database instance at a time. You can't use 2 different servlets accessing the same DB simultaneously. You can't use a servlet and a process using the embedded in

Re: XPath query broken in CVS (get's an Exception)?

> I used be able to perform XPath queries as follows (I am working with the > CVS sources, refreshed earlier today): It would also be useful if you posted a sample of the XML document you were querying when the error occurred. It's often difficult to debug XPath problems without looking at the XML