You are viewing a plain text version of this content. The canonical link for it is here.
Posted to slide-dev@jakarta.apache.org by Eirikur Hrafnsson <ei...@idega.is> on 2005/01/05 18:11:09 UTC
Question about a known issue in release notes for 2.2?
In the 2.2 release notes is says
KNOWN ISSUES
------------
WebDAV Layer:
- SEARCH doesn't work with WebdavServlet mapped to a path other than /
What path is this refering to?
The tomcat context or the servlet mapping?
I map the Slide servlet "/content/" on my tomcat server. Will a DASL
search fail for that setup?
Best Regards
Eirikur S. Hrafnsson, eiki@idega.is
Chief Software Engineer
Idega Software
http://www.idega.com
---------------------------------------------------------------------
To unsubscribe, e-mail: slide-dev-unsubscribe@jakarta.apache.org
For additional commands, e-mail: slide-dev-help@jakarta.apache.org
Re: Question about a known issue in release notes for 2.2?
Posted by Karl Øie <ka...@gan.no>.
On 7. jan. 2005, at 12.35, Eirikur Hrafnsson wrote:
> Wow thanks Karl, this really helpes :)
I'm glad, just started to dab into slide myself so i am happy for all
feedback too! :-)
> Really off topic here (could send me an email directly rather): So
> your working for the Norweigan government? We work a lot for Swedish
> and Icelandic municipalities, infact the Slide work I'm doing now is
> for a new version of our IdegaWeb eGov platform. Are you also doing a
> document management system for a municipality or the national
> government perhaps?
That s cool, i'm not governmental but a contractor. But yes, we're
about at the same place it seems. I'll send you en email later when
things have calmed down a bit. Got some bugs to squash and a delivery
to make first...
> This sounds very interesting I will take a look at the docs for
> extractors if there are any ;)
Docs are a bit lacking, but checking out HEAD and reading the source
for the interfaces and classes you find declared in the domain.xml is a
good start. I was a bit scared at first because its hard to know where
to start. But the layout and the logic in slide is very good, and once
you get it you can find a lot of helpful stuff by looking at the
available implementations.
> Heja Norge
> Eirikur, Idega.
Oh, "heja" is swedish, you're on my shit list now....
Mvh Karl Øie
>
>> Mvh karl Øie
>>
>> On 6. jan. 2005, at 16.41, Eirikur Hrafnsson wrote:
>>
>>> Hi guys!
>>>
>>> On 6.1.2005, at 13:35, Karl Øie wrote:
>>>
>>>> Yes, the index i am creating is not used by Slide at all but a
>>>> legacy webapp operating on a lucene directory. The IndexTrigger and
>>>> Index system is really clever for maintaining indexes that is not
>>>> integrated into Slide. As Stefan points out DASL search will not be
>>>> available by doing this and i would not advice you to index these
>>>> document into the same lucene index as Slide uses.
>>>>
>>>> The reason for this is that Lucene is a indexing toolkit, and
>>>> allows you to create any document->field strategy you want. By
>>>> using the IndexTrigger i can update a lucene index that has a
>>>> different strategy than DASL, and therefor not compatible with the
>>>> DASL searchers that requires a DASL strategy for searching. If you
>>>> got just a few days i would advice you to do this.
>>>>
>>>> Mvh Karl Øie
>>> First thanks a lot to both of you for taking the time to answer all
>>> my annoying questions, I hope this thread helps other developers
>>> also...
>>>
>>> I'm really thinking about taking your advice Karl and use a
>>> listener. What do I have to register to Domain.xml for that exactly?
>>> Just the listener itself or do I have to activate Lucene in some
>>> way?
>>> Any code example for a listener you might be willing to send me to
>>> get started?
>>>
>>> What do you do about security though? I think DASL ensures your
>>> search only returns allowed entries according to who is logged on
>>> (via webdavresource) but I would think you have to somehow filter
>>> the search results afterwards if you maintain your own indexer?
>>>
>>>>>> greeeeat...so Lucene only really works in 2.2 but the webdav
>>>>>> servlet doesn't so basically I can't search because I have to the
>>>>>> WebdavServlet mapped to "/content/"?
>>>>> ????
>>>>> The issue in 2 was: the DASL stuff only works if the WebdavServlet
>>>>> is mapped to "/" (Note: this does not mean the mapping of the
>>>>> slide web application). This is resolved in 2.2.
>>> Hmmm? Now your just confusing me on purpose hehehe ;)
>>> You say " only works if the WebdavServlet is mapped to "/" (Note:
>>> this does not mean the mapping of the slide web application)" what
>>> other mapping is there? As I understood your last reply it would
>>> only work if the servlet is mapped to "/" meaning the root of the
>>> context. Do you mean the root of the Slide repository e.g. where
>>> /roles, /users are? If so then I'm back in business!
>>>
>>> You also say "This is resolved in 2.2", so is it resolved in the
>>> HEAD version but just still mentioned as unresolved in the Release
>>> notes?
>>> Sorry for being completely stupid....
>>>
>>> Best Regards
>>>
>>> Eirikur S. Hrafnsson, eiki@idega.is
>>> Chief Software Engineer
>>> Idega Software
>>> http://www.idega.com
>>>
>>>
>>> ---------------------------------------------------------------------
>>> To unsubscribe, e-mail: slide-dev-unsubscribe@jakarta.apache.org
>>> For additional commands, e-mail: slide-dev-help@jakarta.apache.org
>>>
>>>
>> - you are what you eat. Avoid fruits and nuts...
>>
>>
>> ---------------------------------------------------------------------
>> To unsubscribe, e-mail: slide-dev-unsubscribe@jakarta.apache.org
>> For additional commands, e-mail: slide-dev-help@jakarta.apache.org
> Best Regards
>
> Eirikur S. Hrafnsson, eiki@idega.is
> Chief Software Engineer
> Idega Software
> http://www.idega.com
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: slide-dev-unsubscribe@jakarta.apache.org
> For additional commands, e-mail: slide-dev-help@jakarta.apache.org
>
>
Willow: Have you googled her yet? Xander: Willow, she's only seventeen.
---------------------------------------------------------------------
To unsubscribe, e-mail: slide-dev-unsubscribe@jakarta.apache.org
For additional commands, e-mail: slide-dev-help@jakarta.apache.org
Re: Question about a known issue in release notes for 2.2?
Posted by Eirikur Hrafnsson <ei...@idega.is>.
Wow thanks Karl, this really helpes :)
On 7.1.2005, at 11:11, Karl Øie wrote:
> Hi there Eirikur
>
> Here is the source for the indexer that i have worked on so far, the
> update() method is still not finished but it will give you an idea. As
> you can see i am using a Lucene xml indexer from another project here,
> and thats the reason why i have skipped the DASL route. Here is what
> my my domain.xml entry looks like:
>
> <listener classname="org.apache.slide.search.IndexTrigger">
> <configuration>
> <indexer classname="org.apache.slide.search.LoggingIndexer"
> synchronous="false" uri="/files/articles"/>
> <indexer classname="no.gan.slide.search.XMLIndexer"
> synchronous="false" uri="/files">
> <configuration>
> <!-- an existing directory to use as lucenes index -->
> <parameter name="lucene-path">${filespath}/lucene</parameter>
> <!-- the lucene Analyzer class to be used for indexing -->
> <parameter
> name="lucene-analyzer">no.gan.lucene.GanAlyzer</parameter>
> <!-- the subject that will perform the indexing -->
> <parameter name="subject"></parameter>
> </configuration>
> </indexer>
> </configuration>
> </listener>
>
> <XMLIndexer.java>
>
> The security has yet not been a issue for me as all the documents i am
> working with are public norwegian governmental publications available
> to all citizens. But yes, doing this you could a: only index after
> checking some authority, b: filter the search results against a
> authority. But that should be doable.
Really off topic here (could send me an email directly rather): So your
working for the Norweigan government? We work a lot for Swedish and
Icelandic municipalities, infact the Slide work I'm doing now is for a
new version of our IdegaWeb eGov platform. Are you also doing a
document management system for a municipality or the national
government perhaps?
>
> A quick route to index into slides DASL is to configure a extractor:
>
> <extractors>
> <extractor classname="org.apache.slide.extractor.SimpleXmlExtractor"
> uri="/files/articles/test.xml">
> <configuration>
> <instruction property="title" xpath="/article/title/text()"/>
> <instruction property="summary" xpath="/article/summary/text()"/>
> </configuration>
> </extractor>
> </extractors>
>
> Create a class that extends SimpleXmlExtractor if neccessary, or just
> tweak the configuration to suite your documents. If these documents
> are not XML take a look at how to implement a extractor. This should
> just plug into Slides own Lucene indexing and be available in DASL
> with security to go. But yeah, i havent tried this yet, only looked at
> the sources.
This sounds very interesting I will take a look at the docs for
extractors if there are any ;)
Heja Norge
Eirikur, Idega.
> Mvh karl Øie
>
>
>
> On 6. jan. 2005, at 16.41, Eirikur Hrafnsson wrote:
>
>> Hi guys!
>>
>> On 6.1.2005, at 13:35, Karl Øie wrote:
>>
>>> Yes, the index i am creating is not used by Slide at all but a
>>> legacy webapp operating on a lucene directory. The IndexTrigger and
>>> Index system is really clever for maintaining indexes that is not
>>> integrated into Slide. As Stefan points out DASL search will not be
>>> available by doing this and i would not advice you to index these
>>> document into the same lucene index as Slide uses.
>>>
>>> The reason for this is that Lucene is a indexing toolkit, and allows
>>> you to create any document->field strategy you want. By using the
>>> IndexTrigger i can update a lucene index that has a different
>>> strategy than DASL, and therefor not compatible with the DASL
>>> searchers that requires a DASL strategy for searching. If you got
>>> just a few days i would advice you to do this.
>>>
>>> Mvh Karl Øie
>> First thanks a lot to both of you for taking the time to answer all
>> my annoying questions, I hope this thread helps other developers
>> also...
>>
>> I'm really thinking about taking your advice Karl and use a listener.
>> What do I have to register to Domain.xml for that exactly? Just the
>> listener itself or do I have to activate Lucene in some way?
>> Any code example for a listener you might be willing to send me to
>> get started?
>>
>> What do you do about security though? I think DASL ensures your
>> search only returns allowed entries according to who is logged on
>> (via webdavresource) but I would think you have to somehow filter the
>> search results afterwards if you maintain your own indexer?
>>
>>>>> greeeeat...so Lucene only really works in 2.2 but the webdav
>>>>> servlet doesn't so basically I can't search because I have to the
>>>>> WebdavServlet mapped to "/content/"?
>>>> ????
>>>> The issue in 2 was: the DASL stuff only works if the WebdavServlet
>>>> is mapped to "/" (Note: this does not mean the mapping of the slide
>>>> web application). This is resolved in 2.2.
>> Hmmm? Now your just confusing me on purpose hehehe ;)
>> You say " only works if the WebdavServlet is mapped to "/" (Note:
>> this does not mean the mapping of the slide web application)" what
>> other mapping is there? As I understood your last reply it would only
>> work if the servlet is mapped to "/" meaning the root of the context.
>> Do you mean the root of the Slide repository e.g. where /roles,
>> /users are? If so then I'm back in business!
>>
>> You also say "This is resolved in 2.2", so is it resolved in the HEAD
>> version but just still mentioned as unresolved in the Release notes?
>> Sorry for being completely stupid....
>>
>> Best Regards
>>
>> Eirikur S. Hrafnsson, eiki@idega.is
>> Chief Software Engineer
>> Idega Software
>> http://www.idega.com
>>
>>
>> ---------------------------------------------------------------------
>> To unsubscribe, e-mail: slide-dev-unsubscribe@jakarta.apache.org
>> For additional commands, e-mail: slide-dev-help@jakarta.apache.org
>>
>>
> - you are what you eat. Avoid fruits and nuts...
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: slide-dev-unsubscribe@jakarta.apache.org
> For additional commands, e-mail: slide-dev-help@jakarta.apache.org
Best Regards
Eirikur S. Hrafnsson, eiki@idega.is
Chief Software Engineer
Idega Software
http://www.idega.com
---------------------------------------------------------------------
To unsubscribe, e-mail: slide-dev-unsubscribe@jakarta.apache.org
For additional commands, e-mail: slide-dev-help@jakarta.apache.org
Re: Question about a known issue in release notes for 2.2?
Posted by Karl Øie <ka...@gan.no>.
Hi there Eirikur
Here is the source for the indexer that i have worked on so far, the
update() method is still not finished but it will give you an idea. As
you can see i am using a Lucene xml indexer from another project here,
and thats the reason why i have skipped the DASL route. Here is what my
my domain.xml entry looks like:
<listener classname="org.apache.slide.search.IndexTrigger">
<configuration>
<indexer classname="org.apache.slide.search.LoggingIndexer"
synchronous="false" uri="/files/articles"/>
<indexer classname="no.gan.slide.search.XMLIndexer"
synchronous="false" uri="/files">
<configuration>
<!-- an existing directory to use as lucenes index -->
<parameter name="lucene-path">${filespath}/lucene</parameter>
<!-- the lucene Analyzer class to be used for indexing -->
<parameter
name="lucene-analyzer">no.gan.lucene.GanAlyzer</parameter>
<!-- the subject that will perform the indexing -->
<parameter name="subject"></parameter>
</configuration>
</indexer>
</configuration>
</listener>
Re: Question about a known issue in release notes for 2.2?
Posted by Eirikur Hrafnsson <ei...@idega.is>.
Hi guys!
On 6.1.2005, at 13:35, Karl Øie wrote:
> Yes, the index i am creating is not used by Slide at all but a legacy
> webapp operating on a lucene directory. The IndexTrigger and Index
> system is really clever for maintaining indexes that is not integrated
> into Slide. As Stefan points out DASL search will not be available by
> doing this and i would not advice you to index these document into the
> same lucene index as Slide uses.
>
> The reason for this is that Lucene is a indexing toolkit, and allows
> you to create any document->field strategy you want. By using the
> IndexTrigger i can update a lucene index that has a different strategy
> than DASL, and therefor not compatible with the DASL searchers that
> requires a DASL strategy for searching. If you got just a few days i
> would advice you to do this.
>
> Mvh Karl Øie
First thanks a lot to both of you for taking the time to answer all my
annoying questions, I hope this thread helps other developers also...
I'm really thinking about taking your advice Karl and use a listener.
What do I have to register to Domain.xml for that exactly? Just the
listener itself or do I have to activate Lucene in some way?
Any code example for a listener you might be willing to send me to get
started?
What do you do about security though? I think DASL ensures your search
only returns allowed entries according to who is logged on (via
webdavresource) but I would think you have to somehow filter the search
results afterwards if you maintain your own indexer?
>>> greeeeat...so Lucene only really works in 2.2 but the webdav servlet
>>> doesn't so basically I can't search because I have to the
>>> WebdavServlet mapped to "/content/"?
>> ????
>> The issue in 2 was: the DASL stuff only works if the WebdavServlet is
>> mapped to "/" (Note: this does not mean the mapping of the slide web
>> application). This is resolved in 2.2.
Hmmm? Now your just confusing me on purpose hehehe ;)
You say " only works if the WebdavServlet is mapped to "/" (Note: this
does not mean the mapping of the slide web application)" what other
mapping is there? As I understood your last reply it would only work if
the servlet is mapped to "/" meaning the root of the context. Do you
mean the root of the Slide repository e.g. where /roles, /users are? If
so then I'm back in business!
You also say "This is resolved in 2.2", so is it resolved in the HEAD
version but just still mentioned as unresolved in the Release notes?
Sorry for being completely stupid....
Best Regards
Eirikur S. Hrafnsson, eiki@idega.is
Chief Software Engineer
Idega Software
http://www.idega.com
---------------------------------------------------------------------
To unsubscribe, e-mail: slide-dev-unsubscribe@jakarta.apache.org
For additional commands, e-mail: slide-dev-help@jakarta.apache.org
Re: Question about a known issue in release notes for 2.2?
Posted by Karl Øie <ka...@gan.no>.
Yes, the index i am creating is not used by Slide at all but a legacy
webapp operating on a lucene directory. The IndexTrigger and Index
system is really clever for maintaining indexes that is not integrated
into Slide. As Stefan points out DASL search will not be available by
doing this and i would not advice you to index these document into the
same lucene index as Slide uses.
The reason for this is that Lucene is a indexing toolkit, and allows
you to create any document->field strategy you want. By using the
IndexTrigger i can update a lucene index that has a different strategy
than DASL, and therefor not compatible with the DASL searchers that
requires a DASL strategy for searching. If you got just a few days i
would advice you to do this.
Mvh Karl Øie
On 6. jan. 2005, at 14.22, Stefan Lützkendorf wrote:
> Eirikur Hrafnsson wrote:
>
>> greeeeat...so Lucene only really works in 2.2 but the webdav servlet
>> doesn't so basically I can't search because I have to the
>> WebdavServlet mapped to "/content/"?
> ????
> The issue in 2 was: the DASL stuff only works if the WebdavServlet is
> mapped to "/" (Note: this does not mean the mapping of the slide web
> application). This is resolved in 2.2.
>
>> Any ideas? Do I have to wait for 2.2, I only have a few days to
>> finish the search part of my project. Or should I maybe write an
>> external indexer like Karl did and add an indexer listener?
> If you use our own indexer, you will also have to implement your own
> search too, I think. I have no much knowledge about this way. May be
> Karl knows better.
>
> Stefan
>
> --
> Stefan Lützkendorf -- luetzkendorf@apache.org
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: slide-dev-unsubscribe@jakarta.apache.org
> For additional commands, e-mail: slide-dev-help@jakarta.apache.org
>
>
Remember kids, given half a chance a cow would eat you and everyone you
care about. (Pan and zoom to a cow with a crazed look.) MOO-OOO
---------------------------------------------------------------------
To unsubscribe, e-mail: slide-dev-unsubscribe@jakarta.apache.org
For additional commands, e-mail: slide-dev-help@jakarta.apache.org
Re: Question about a known issue in release notes for 2.2?
Posted by Stefan Lützkendorf <lu...@apache.org>.
Eirikur Hrafnsson wrote:
> greeeeat...so Lucene only really works in 2.2 but the webdav servlet
> doesn't so basically I can't search because I have to the WebdavServlet
> mapped to "/content/"?
????
The issue in 2 was: the DASL stuff only works if the WebdavServlet is
mapped to "/" (Note: this does not mean the mapping of the slide web
application). This is resolved in 2.2.
> Any ideas? Do I have to wait for 2.2, I only have a few days to finish
> the search part of my project. Or should I maybe write an external
> indexer like Karl did and add an indexer listener?
If you use our own indexer, you will also have to implement your own
search too, I think. I have no much knowledge about this way. May be
Karl knows better.
>
Stefan
--
Stefan Lützkendorf -- luetzkendorf@apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: slide-dev-unsubscribe@jakarta.apache.org
For additional commands, e-mail: slide-dev-help@jakarta.apache.org
Re: Question about a known issue in release notes for 2.2?
Posted by Eirikur Hrafnsson <ei...@idega.is>.
greeeeat...so Lucene only really works in 2.2 but the webdav servlet
doesn't so basically I can't search because I have to the WebdavServlet
mapped to "/content/"?
Any ideas? Do I have to wait for 2.2, I only have a few days to finish
the search part of my project. Or should I maybe write an external
indexer like Karl did and add an indexer listener?
>> <listener classname="org.apache.slide.search.IndexTrigger">
>> <configuration>
>> <indexer classname="YourIndexer" synchronous="false"
>> uri="/files/your/path"/>
>> </configuration>
>> </listener>
best regards
Eirikur, idega.
On 6.1.2005, at 09:30, Stefan Lützkendorf wrote:
> This issue will be resolved in 2.2.
>
> This issue means that there where problems map the WebdavServlet in
> your web.xml to an other path than /.
>
> e.g.
> <servlet-mapping>
> <servlet-name>webdav</servlet-name>
> <url-pattern>/webdav/*</url-pattern>
> </servlet-mapping>
>
> instead of
> <servlet-mapping>
> <servlet-name>webdav</servlet-name>
> <url-pattern>/</url-pattern>
> </servlet-mapping>
>
> Stefan
>
> Eirikur Hrafnsson wrote:
>> In the 2.2 release notes is says
>> KNOWN ISSUES
>> ------------
>> WebDAV Layer:
>> - SEARCH doesn't work with WebdavServlet mapped to a path other than /
>> What path is this refering to?
>> The tomcat context or the servlet mapping?
>> I map the Slide servlet "/content/" on my tomcat server. Will a DASL
>> search fail for that setup?
>> Best Regards
>> Eirikur S. Hrafnsson, eiki@idega.is
>> Chief Software Engineer
>> Idega Software
>> http://www.idega.com
>> ---------------------------------------------------------------------
>> To unsubscribe, e-mail: slide-dev-unsubscribe@jakarta.apache.org
>> For additional commands, e-mail: slide-dev-help@jakarta.apache.org
>
> --
> Stefan Lützkendorf -- luetzkendorf@apache.org
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: slide-dev-unsubscribe@jakarta.apache.org
> For additional commands, e-mail: slide-dev-help@jakarta.apache.org
>
>
>
Best Regards
Eirikur S. Hrafnsson, eiki@idega.is
Chief Software Engineer
Idega Software
http://www.idega.com
---------------------------------------------------------------------
To unsubscribe, e-mail: slide-dev-unsubscribe@jakarta.apache.org
For additional commands, e-mail: slide-dev-help@jakarta.apache.org
Re: Question about a known issue in release notes for 2.2?
Posted by Stefan Lützkendorf <lu...@apache.org>.
This issue will be resolved in 2.2.
This issue means that there where problems map the WebdavServlet in your
web.xml to an other path than /.
e.g.
<servlet-mapping>
<servlet-name>webdav</servlet-name>
<url-pattern>/webdav/*</url-pattern>
</servlet-mapping>
instead of
<servlet-mapping>
<servlet-name>webdav</servlet-name>
<url-pattern>/</url-pattern>
</servlet-mapping>
Stefan
Eirikur Hrafnsson wrote:
> In the 2.2 release notes is says
>
> KNOWN ISSUES
> ------------
> WebDAV Layer:
> - SEARCH doesn't work with WebdavServlet mapped to a path other than /
>
> What path is this refering to?
> The tomcat context or the servlet mapping?
> I map the Slide servlet "/content/" on my tomcat server. Will a DASL
> search fail for that setup?
>
> Best Regards
>
> Eirikur S. Hrafnsson, eiki@idega.is
> Chief Software Engineer
> Idega Software
> http://www.idega.com
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: slide-dev-unsubscribe@jakarta.apache.org
> For additional commands, e-mail: slide-dev-help@jakarta.apache.org
>
--
Stefan Lützkendorf -- luetzkendorf@apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: slide-dev-unsubscribe@jakarta.apache.org
For additional commands, e-mail: slide-dev-help@jakarta.apache.org