1 | .\"------------------------------------------------------------
|
---|
2 | .\" Id - set Rv,revision, and Dt, Date using rcs-Id tag.
|
---|
3 | .de Id
|
---|
4 | .ds Rv \\$3
|
---|
5 | .ds Dt \\$4
|
---|
6 | ..
|
---|
7 | .Id $Id: mg_perf_hash_build.1 856 2000-01-14 02:26:25Z sjboddie $
|
---|
8 | .\"------------------------------------------------------------
|
---|
9 | .TH mg_perf_hash_build 1 \*(Dt CITRI
|
---|
10 | .SH NAME
|
---|
11 | mg_perf_hash_build \- generate an order-preserving hash function for the stemmed dictionary
|
---|
12 | .SH SYNOPSIS
|
---|
13 | .B mg_perf_hash_build
|
---|
14 | [
|
---|
15 | .B \-h
|
---|
16 | ]
|
---|
17 | [
|
---|
18 | .BI \-r " rndseed"
|
---|
19 | ]
|
---|
20 | [
|
---|
21 | .BI \-d " directory"
|
---|
22 | ]
|
---|
23 | .if n .ti +9n
|
---|
24 | .BI \-f " name"
|
---|
25 | .SH DESCRIPTION
|
---|
26 | .B mg_perf_hash_build
|
---|
27 | generates an order-preserving hashing function from the compressed
|
---|
28 | stemmed dictionary and writes it out to disk.
|
---|
29 | .BR mg_passes (1)
|
---|
30 | will make use of the hash function when it builds the inverted file.
|
---|
31 | .SH OPTIONS
|
---|
32 | Options may appear in any order.
|
---|
33 | .TP "\w'\fB\-d\fP \fIdirectory\fP'u+2n"
|
---|
34 | .B \-h
|
---|
35 | This displays a usage line on
|
---|
36 | .IR stderr .
|
---|
37 | .TP
|
---|
38 | .BI \-r " rndseed"
|
---|
39 | This specifies the random seed to be used in the generation of the
|
---|
40 | hash function; it must be an integer. If this is not specified, the
|
---|
41 | current time will be used as the random seed.
|
---|
42 | .TP
|
---|
43 | .BI \-d " directory"
|
---|
44 | This specifies the directory where the document collection can be found.
|
---|
45 | .TP
|
---|
46 | .BI \-f " name"
|
---|
47 | This specifies the base name of the document collection.
|
---|
48 | .SH ENVIRONMENT
|
---|
49 | .TP "\w'\fBMGDATA\fP'u+2n"
|
---|
50 | .SB MGDATA
|
---|
51 | If this environment variable exists, then its value is used as the
|
---|
52 | default directory where the mg
|
---|
53 | collection files are. If this variable does not exist, then the
|
---|
54 | directory \*(lq\fB.\fP\*(rq is used by default. The command line
|
---|
55 | option
|
---|
56 | .BI \-d " directory"
|
---|
57 | overrides the directory in
|
---|
58 | .BR MGDATA .
|
---|
59 | .SH FILES
|
---|
60 | .TP 20
|
---|
61 | .B *.invf.dict
|
---|
62 | Stemmed dictionary.
|
---|
63 | .TP
|
---|
64 | .B *.invf.dict.hash
|
---|
65 | Data for a order-preserving perfect hash function.
|
---|
66 | .SH "SEE ALSO"
|
---|
67 | .na
|
---|
68 | .BR mg_compression_dict (1),
|
---|
69 | .BR mg_fast_comp_dict (1),
|
---|
70 | .BR mg_invf_dict (1),
|
---|
71 | .BR mg_passes (1),
|
---|
72 | .BR mg_stem_idx (1),
|
---|
73 | .BR mg_weights_build (1)
|
---|