SupoSE

Hi Christian,

the resultindex is intended to support scanning of multiple repositories and put the resulting index may be in different indexes...so usualy if you have only one repository you don't need to give the resultindex.
Details can be found Configuration descrition

I suggest to ask in the forum before you create an issue thanks in advance..
Kind regards
Karl Heinz Marbaise

Hi Karl,

I'm scanning multiple repositories at once, thats not the key. My problem is, that supose produces for each repository two index-folder. One named "index.#name#" and one named "#name#". Otherwise I wouldn't have raised an issue ;)

Kind regards
Christian Dehning

Hi Christian,

that's Ok (from my point of view) may be not from yours...

The basic idea is that you can define a result index (resultindex) which contains the result of different repositories.

So you can create a scan of repoA, repoB and repoC and put the result into resultA which contains repoA and repoB the last one can be put into a separate index resultC.

But on the other hand if you have multiple repositories which will be scanned but you don't like to have a summarized indexes (only one index per repository) than it might be good idea to think about a configuration to be aware of this kind of situations.

What I missed to say, that the index.de-svn will be created from the scratch every time the scanning job is running whereas the de-svn index is held intact which is intended to search in this index.
Kind regards
Karl Heinz Marbaise

Hi Christian,

i have to apologize, cause after some investigation i found out that the second scan of a repository will do a full scan instead scanning the changed revisions only.

So this is really a Bug ...

Thanks for your support....

Kind regards
Karl Heinz Marbaise

Applied in changeset r383.