I would like to note, that the problem of "expanding index files size during updating repositories" is solved. The reason of it was optimalization in Apache.Lucene. When optimalization crashed extended index files stayed.
Now indexer works fine.
Index, made from 70 repos. (about 20GB) gives 4GB index files.
And the second test one repo. which has 56000 only zip files (in them only text files) (0.9GB archive, 1,9GB unpacked)
gives index files 2.7GB size. Not that in the second test all files was indexed and all the text from archive was taken to index.
Please write your test result even if you indexed something on smaller repositories.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Version 1.2.1 SVNIndexer turn out not ideal solution for indexing huge repositories (tested one was about 20GB).
Now main problem is expanding index files size during updating repositories.
I will give a note when this and other problems will be solved.
Please write your opinion about using SVNIndexer and Searcher, even if you use it for small repositories.
I would like to note, that the problem of "expanding index files size during updating repositories" is solved. The reason of it was optimalization in Apache.Lucene. When optimalization crashed extended index files stayed.
Now indexer works fine.
Index, made from 70 repos. (about 20GB) gives 4GB index files.
And the second test one repo. which has 56000 only zip files (in them only text files) (0.9GB archive, 1,9GB unpacked)
gives index files 2.7GB size. Not that in the second test all files was indexed and all the text from archive was taken to index.
Please write your test result even if you indexed something on smaller repositories.