Is these any way to set up a custom Extractor in Jackrabbit 2.

[taha ben salah <ta...@gmail.com>]

Hi all,

Is these any way please to set up a custom Extractor in Jackrabbit 2.

What i ended up to guess is that JR2 uses Tika for all text extracting but
it does not give a way to specify textFilterClasses as in previous versions.
When looking into JackrabbitParser.java (in JR implementation), i found a
fuzzy :
new AutoDetectParser(new
TikaConfig(JackrabbitParser.class.getResourceAsStream("tika-config.xml")))
which closes all possibilities to handle custom extractors.
further more only for backwork compatibility textFilterClasses values (in
workspace.xml) are handled for solely for "APACHE implemeted classes" and
does a simple
logger.warn("Ignoring unknown text extractor class: {}", name);
for all the rest.

Thanks for the help.
Taha Ben Salah

Re: Is these any way to set up a custom Extractor in Jackrabbit 2.

[Justin Edelson <ju...@gmail.com>]

Perhaps you should vote for
https://issues.apache.org/jira/browse/JCR-2642 and/or provide a patch
for that issue.

Justin

On 10/14/10 1:14 PM, taha ben salah wrote:
> Hi all,
> 
> Is these any way please to set up a custom Extractor in Jackrabbit 2.
> 
> What i ended up to guess is that JR2 uses Tika for all text extracting but
> it does not give a way to specify textFilterClasses as in previous versions.
> When looking into JackrabbitParser.java (in JR implementation), i found a
> fuzzy :
> new AutoDetectParser(new
> TikaConfig(JackrabbitParser.class.getResourceAsStream("tika-config.xml")))
> which closes all possibilities to handle custom extractors.
> further more only for backwork compatibility textFilterClasses values (in
> workspace.xml) are handled for solely for "APACHE implemeted classes" and
> does a simple
> logger.warn("Ignoring unknown text extractor class: {}", name);
> for all the rest.
> 
> Thanks for the help.
> Taha Ben Salah
>