actually, given that the streaming interface for indexing has been removed, we have a couple options:
1) re-create a streaming interface for indexing (but note that the URI must map to a file at query time in order to produce results)
2) change the filter to write to a temporary file and index that (same problem as #1, the URI must map to a file at query time, although if the actual DAV storage is not on disk the filter could create a copy for XDB to use.)
3) change the filter to be an output filter to catch the output side of the processing, not manipulating the output in any way but knowing at that point the file has been created and can, therefore be indexed
4) fall back to "dav_mon" for indexing
4.1) create an XDB-specific dav_mon module that calls API hooks rather than command-line commands
Thoughts?
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
actually, given that the streaming interface for indexing has been removed, we have a couple options:
1) re-create a streaming interface for indexing (but note that the URI must map to a file at query time in order to produce results)
2) change the filter to write to a temporary file and index that (same problem as #1, the URI must map to a file at query time, although if the actual DAV storage is not on disk the filter could create a copy for XDB to use.)
3) change the filter to be an output filter to catch the output side of the processing, not manipulating the output in any way but knowing at that point the file has been created and can, therefore be indexed
4) fall back to "dav_mon" for indexing
4.1) create an XDB-specific dav_mon module that calls API hooks rather than command-line commands
Thoughts?