Raphael Parree | 1 Aug 2006 13:49
Picon

Powered By eXist

Hi,

 

Attached another entry for the Powered By eXist page on the site.

 

Cheers,

 

Raphael

-------------------------------------------------------------------------
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys -- and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
_______________________________________________
Exist-open mailing list
Exist-open <at> lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/exist-open
Chris Wallace | 1 Aug 2006 14:14
Picon
Favicon

web admin interface

Belatedly switching to 20060316 , I see that using the web admin client
uppercase characters in filenames (on Windows) are converted to lowercase.  This
is also the case in 1.0rc as running on the demo server. This is a regression
from 20051203.

-------------------------------------------------------------------------
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys -- and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
Jonas Lundberg | 1 Aug 2006 20:32
Picon

fn:substring test case difference: SAXON vs EXIST

I run the following query through saxon and exist (rev 3938):

xquery version "1.0";

<xml>
{substring("1234",1,3)}
</xml>

eXist returns:
<xml>234</xml>

Saxon returns:
<xml>123</xml>

I believe that the Saxon output is the correct one here (although I am
not sure).

Regards
Hans

-------------------------------------------------------------------------
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys -- and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
Chris Marasti-Georg | 1 Aug 2006 20:39
Picon

Re: automatically create subcollections when storing documents

Search around for java extensions.  Along the lines of:

declare namespace j-file="java:java.io.file";

let $parent := j-file:new("/")
for $child in j-file:list-files($parent)
return j-file:get-name($child)

On 7/29/06, Szekeres Laszlo <szlaca <at> gmail.com> wrote:
Hi,

sorry, I considered the terminology :)
I want this in the filesytem, not in the database, to store the directories and files into collections and resources.
Anyhow, thanks for the reply.

SzL

2006/7/29, Pierrick Brihaye <pierrick.brihaye <at> free.fr>:
Hi,

Szekeres Laszlo a écrit :

> I didn't find any xquery function which can return the subdirectories
> of a directory, or the full paths of the files in a directory, which
> is scanned recursively.

Reconsider your terminology : "children collections of a collection" and
"resources in a collection".

Then, look at the functions in the xmldb namespace :
http://demo.exist-db.org/xquery/functions.xq#N1054D

[snip the tail quotation]

Cheers,

p.b.



-------------------------------------------------------------------------
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys -- and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV

_______________________________________________
Exist-open mailing list
Exist-open <at> lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/exist-open



-------------------------------------------------------------------------
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys -- and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
_______________________________________________
Exist-open mailing list
Exist-open <at> lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/exist-open
Pierrick Brihaye | 1 Aug 2006 20:48
Picon
Favicon

Re: fn:substring test case difference: SAXON vs EXIST

Hi,

Jonas Lundberg a écrit :

 > I believe that the Saxon output is the correct one here (although I am
 > not sure).

Do you yhink the 100% XQTS compliant XQuery processor could be wrong ?

Well, my current version is patched with the enclosed patch. It works in 
your case. Not sure it does for every one though ; this function is 
trickier than it looks like at first glance.

Cheers,

p.b.
Index: C:/eclipse/workspace/eXist-svn/src/org/exist/xquery/functions/FunSubstring.java
===================================================================
---
C:/eclipse/workspace/eXist-svn/src/org/exist/xquery/functions/FunSubstring.java	(revision 3938)
+++
C:/eclipse/workspace/eXist-svn/src/org/exist/xquery/functions/FunSubstring.java	(working copy)
 <at>  <at>  -120,7 +120,7  <at>  <at> 
 	    			if (lengthValue.isNaN())
 	    				result = StringValue.EMPTY_STRING;
 	    			else if (((NumericValue)lengthValue.plus(startValue)).isInfinite() ||
-	    					((NumericValue)lengthValue.plus(startValue)).getInt() >= string.length()) {
+	    					((NumericValue)lengthValue.plus(startValue)).getInt() > string.length()) {
 	        			if (start < 0)
 	        				start = 0;        				
 	        			result = new StringValue(string.substring(start));
-------------------------------------------------------------------------
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys -- and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
_______________________________________________
Exist-open mailing list
Exist-open <at> lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/exist-open
Alessandro Vernet | 1 Aug 2006 20:52
Gravatar

Another process has locked a portion of the file

I am running into a situation where when my embedded eXist starts, I
get the message:

Another process has locked a portion of the file

Then eXist cannot access the database, and there is no other process
accessing the database. This happens only after VM that run eXist is
killed, I presume leaving the database files in an unstable state.
Obviously, I am not being gentle to eXist by killing the VM, but if it
happens, I would still like to be able to recover from it, instead of
having to uniquely rely on a previous backup of the database files.

Is there anything I can do to "rebuild" the database files?

Alex
--

-- 
Blog (XML, Web apps, Open Source):
http://www.orbeon.com/blog/

-------------------------------------------------------------------------
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys -- and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
Pierrick Brihaye | 1 Aug 2006 20:59
Picon
Favicon

Re: Another process has locked a portion of the file

Hi,

Alessandro Vernet a écrit :

> Obviously, I am not being gentle to eXist by killing the VM, but if it
> happens, I would still like to be able to recover from it

I think the JVM implementation is in cause : I also sometimes experience 
the same behaviour on Windows when I have to kill my JVM, mainly during 
debugging sessions though...

Cheers,

p.b.

-------------------------------------------------------------------------
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys -- and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
Wolfgang Meier | 1 Aug 2006 21:03
Picon

Re: Another process has locked a portion of the file

> Another process has locked a portion of the file
>
> Then eXist cannot access the database, and there is no other process
> accessing the database. This happens only after VM that run eXist is
> killed, I presume leaving the database files in an unstable state.

As the message says, there's a lock on one of the database files or
the journal lock. You would need to manually release this lock.
However, I have no idea how to do that or if it is possible at all. I
guess it depends on the JVM version and the operating system. eXist
relies entirely on Java NIO file locks. On windows it sometimes help
to wait a few seconds.

Anyway, if this is a critical issue for other users as well I would
rather prefer to leave the files unlocked.

Wolfgang

-------------------------------------------------------------------------
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys -- and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
Jonas Lundberg | 1 Aug 2006 21:40
Picon

Re: fn:substring test case difference: SAXON vs EXIST

Thanks! That seems to fix it.

Regards
Hans

On 8/1/06, Pierrick Brihaye <pierrick.brihaye <at> free.fr> wrote:
> Hi,
>
> Jonas Lundberg a écrit :
>
>  > I believe that the Saxon output is the correct one here (although I am
>  > not sure).
>
> Do you yhink the 100% XQTS compliant XQuery processor could be wrong ?
>
> Well, my current version is patched with the enclosed patch. It works in
> your case. Not sure it does for every one though ; this function is
> trickier than it looks like at first glance.
>
> Cheers,
>
> p.b.
>
>
> Index: C:/eclipse/workspace/eXist-svn/src/org/exist/xquery/functions/FunSubstring.java
> ===================================================================
> --- C:/eclipse/workspace/eXist-svn/src/org/exist/xquery/functions/FunSubstring.java
(revision 3938)
> +++ C:/eclipse/workspace/eXist-svn/src/org/exist/xquery/functions/FunSubstring.java
(working copy)
>  <at>  <at>  -120,7 +120,7  <at>  <at> 
>  if (lengthValue.isNaN())
>  result = StringValue.EMPTY_STRING;
>  else if (((NumericValue)lengthValue.plus(startValue)).isInfinite() ||
> - ((NumericValue)lengthValue.plus(startValue)).getInt() >= string.length()) {
> + ((NumericValue)lengthValue.plus(startValue)).getInt() > string.length()) {
>  if (start < 0)
>  start = 0;
>  result = new StringValue(string.substring(start));
>
>
> -------------------------------------------------------------------------
> Take Surveys. Earn Cash. Influence the Future of IT
> Join SourceForge.net's Techsay panel and you'll get the chance to share your
> opinions on IT & business topics through brief surveys -- and earn cash
> http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
>
> _______________________________________________
> Exist-open mailing list
> Exist-open <at> lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/exist-open
>
>
>

-------------------------------------------------------------------------
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys -- and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
Pierrick Brihaye | 1 Aug 2006 21:50
Picon
Favicon

Re: fn:substring test case difference: SAXON vs EXIST

Hi,

Jonas Lundberg a écrit :

 > Thanks! That seems to fix it.

I'm committing my patch since I'm getting this excellent XQTS result :
fn:substring [21/0/0]

[would you snipping the tail quotation next time ?]

Cheers,

p.b.

-------------------------------------------------------------------------
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys -- and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV

Gmane