myisam_ftdump displays information about
FULLTEXT indexes in
tables. It reads the
MyISAM index file
directly, so it must be run on the server host where the table
is located. Before using myisam_ftdump, be
sure to issue a
FLUSH TABLES statement first
if the server is running.
myisam_ftdump scans and dumps the entire index, which is not particularly fast. On the other hand, the distribution of words changes infrequently, so it need not be run often.
Invoke myisam_ftdump like this:
tbl_name argument should be the
name of a
MyISAM table. You can also specify
a table by naming its index file (the file with the
.MYI suffix). If you do not invoke
myisam_ftdump in the directory where the
table files are located, the table or index file name must be
preceded by the path name to the table's database directory.
Index numbers begin with 0.
Example: Suppose that the
contains a table named
mytexttable that has
the following definition:
CREATE TABLE mytexttable ( id INT NOT NULL, txt TEXT NOT NULL, PRIMARY KEY (id), FULLTEXT (txt) ) ENGINE=MyISAM;
The index on
id is index 0 and the
FULLTEXT index on
index 1. If your working directory is the
test database directory, invoke
myisam_ftdump as follows:
myisam_ftdump mytexttable 1
If the path name to the
you can also specify the table name argument using that path
name. This is useful if you do not invoke
myisam_ftdump in the database directory:
myisam_ftdump /usr/local/mysql/data/test/mytexttable 1
You can use myisam_ftdump to generate a list of index entries in order of frequency of occurrence like this:
myisam_ftdump -c mytexttable 1 | sort -r
myisam_ftdump supports the following options:
Display a help message and exit.
Calculate per-word statistics (counts and global weights).
Dump the index, including data offsets and word weights.
Report the length distribution.
Report global index statistics. This is the default operation if no other operation is specified.
Verbose mode. Print more output about what the program does.