Contents of this chapter:

What are preference files for?
a preference file example
an output file example

What are preference files for?

Preference files are a way for the CorpusSearch user to set custom default values. If you find yourself continually copying the same information into your query files, you would most likely benefit from a preference file.

CorpusSearch sets its own default values for certain variables but if a preference file exists, the commands contained in it override the defaults set within CorpusSearch. If the current query file contains commands of the same type as the preference file, the commands in the query file override those in the preference file. So, for instance, CorpusSearch sets the command line comment delimiter "//", your preference file might set it to "/!/", and your query file might set it to "/>". It is the last value that CorpusSearch will use as the comment delimiter.

a preference file example

Here's an example of a basic preference file, designed for searching the Penn Korean TreeBank (Han, Han, and Ko (2001)). The name of the file is "korean.prf". Preference file names must always have the extension .prf and must be stored in the same directory as the query file.

// a preference file for the Korean corpus.
corpus_file_extension: fid
corpus_comment_begin: <
corpus_comment_end: >
corpus_line_comment: ;;
node: S

an output file example

Here's the preface from an output file using "korean.prf". Notice the line

preference file:  korean.prf

This line is how you know that CorpusSearch accessed your preference file. If you don't see this line, your preference file was not found. Check that your preference file name ends with ".prf", and that it's stored in the same directory as your query file.

CorpusSearch copyright Beth Randall 2000.
Date:  Mon Feb 26 09:28:55 EST 2001

command file:     kor.q
preference file:  korean.prf
input file:       add.out
output file:      kor.out

node:   S
query:  (NNC iPrecedes NNC)