CompleteSearch

Old Wiki from the MPII (lots of detailed / internal information)

Quick Intro

Follow these steps to checkout the CompleteSearch code from our SVN, build it, build an index, run a server on that index, and ask queries to that server via HTTP. Don't be afraid, it's easy. If you have questions, send an email to <bast@informatik.uni-freiburg.de>.

0. Get source code

svn checkout http://vulcano.informatik.uni-freiburg.de/svn/completesearch/codebase Username: [ask us] Password: [ask us]

1. Compile

make all

This will build three binaries:

buildIndex buildDocsDB startCompletionServer

If you call any of these binaries without parameters you will get usage info with all the available options.

2. Input (to be produced by a suitable parser)

a <name>.words file, with lines of the form

Must be sorted so that sort -c -k1,1 -k2,2n -k4,4n does not complain.

And a <name>.docs file, with lines of the form

Must be sorted so that sort -c -k1,1n does not complain.

You find a very simple example under http://www.mpi-inf.mpg.de/~bast/topsecret/example.tgz

3. Build the word index

buildIndex HYB <name>.words

This produces the main index file <name>.hybrid needed for prefix search (this is a binary file). It also produces the file <name>.vocabulary, that provides the mapping from word ids to words (it's an ascii file, you can just look at it).

Note that by default, HYB is built with block of fixed sizes. It is more efficient though to pass it an explicit list of block boundaries (-B option). Let's talk about this more when efficiency becomes an issue for you.

4. Build the doc index

buildDocsDB <name>.docs

This produces the file <name>.docs.DB which provides efficient mapping from doc ids to documents. Needed if you want to show excerpts/snippets from documents matching the query.

5. Start server

startCompletionServer -Z <name>.hybrid

This starts the server. If you run it without argument, it prints usage information. The -Z argument lets the server run in the foreground, and output everything to the console, which is convenient for testing.

6. Queries

The server listens on the port you specified in step 6 (8888 by default), and speaks HTTP. For example:

curl "http://localhost:8888/?q=die*&h=1&c=3"

This will return the result as an XML, which should be self-explanatory.

Here is the list of parameters which you may pass along with the query (q=...)

h : number of hits c : number of completions (of last query word, if you put a * behind it) f : send hits starting from this one (default: 0) en : number of excerpts per hit er : size of excerpt rd : how to rank the documents (0 = by score, 1 = by doc id, 2 = by word id, append a or d for ascending or descending) rw : how to rank the words (0 = by score, 1 = by doc count, 2 = by occurrence count, 3 = by word id, 4 = by doc id, append a or d as above) s : how to aggregate scores (expert option, ignore for the moment)

Jede Menge Doku auf http://search.mpi-inf.mpg.de/wiki/CompleteSearch