Project allows you to index files in a local directory and then query for them using a different API similar to a search engine.
First clone the repository in the directory to be indexed. Then install the below dependencies with the following commands.
boost: brew install boost
flask: pip install flask
First cd to the project directory. Then, run make sfs, followed by python run.py. This will open a web interface with the following two capabilities.
- Click index to generate an index of the selected directory
- Ente a search query (E.G: "term1" OR "term2" AND "term3")
The results of the commands are output on the page after processing.
Note that the commands can also be run through the command-line. This method is preferred since the above method is shown to provide slow performance.
- To index, run
project-dir/executables/sfs index - To query, run
project-dir/executables/sfs query "example_query"
There is support for PHRASE queries, AND queries, and OR queries. The grammar for the queries are as follows.
- PHRASE Queries are of the form: "term1 term2"
- AND Queries are of the form: (query1 AND query2)
- OR Queries are of the form: (query1 OR query2)
The list of subdirectories and documents that are indexed are specified in the direcotry project-dir/lib/utils/DirectoryFileUtils.h . This file can be modified to incorporate other kinds of documents into the index.