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: mgmerge.1 3745 2003-02-20 21:20:24Z mdewsnip $
|
---|
8 | .\"------------------------------------------------------------
|
---|
9 | .TH mgmerge 1 \*(Dt CITRI
|
---|
10 | .SH NAME
|
---|
11 | mgmerge \- update an mg system database with new documents
|
---|
12 | .SH SYNOPSIS
|
---|
13 | .B mgmerge
|
---|
14 | [
|
---|
15 | .B \-c
|
---|
16 | ]
|
---|
17 | [
|
---|
18 | .BI \-g " get"
|
---|
19 | ]
|
---|
20 | [
|
---|
21 | .BI \-s " source"
|
---|
22 | ]
|
---|
23 | [
|
---|
24 | .B \-S
|
---|
25 | ]
|
---|
26 | [
|
---|
27 | .B \-w
|
---|
28 | ]
|
---|
29 | .I collection-name
|
---|
30 | .SH DESCRIPTION
|
---|
31 | .B mgmerge
|
---|
32 | is a
|
---|
33 | .B csh
|
---|
34 | script that executes all the appropriate programs in the correct order
|
---|
35 | to completely merge a current
|
---|
36 | .BR mg (1)
|
---|
37 | system database ready with some new documents, saving the need for a complete
|
---|
38 | database rebuild with
|
---|
39 | .B mgbuild.
|
---|
40 | It does this by building a second
|
---|
41 | .BR mg (1)
|
---|
42 | system database from the new documents and then merging this database
|
---|
43 | with the old one.
|
---|
44 | This program makes use of the
|
---|
45 | .BR mg_get_merge (1)
|
---|
46 | script to obtain the text of the collection.
|
---|
47 | .B mgmerge
|
---|
48 | cannot edit or delete documents already in a database; only new documents
|
---|
49 | can be added.
|
---|
50 | .SH OPTIONS
|
---|
51 | Options can occur in any order, but the collection name must be last.
|
---|
52 | .TP "\w'\fIcollection-name\fP'u+2n"
|
---|
53 | .BI \-c
|
---|
54 | This specifies whether the
|
---|
55 | .I get
|
---|
56 | program is \*(lqcomplex\*(rq. If a
|
---|
57 | .I get
|
---|
58 | program is \*(lqcomplex\*(rq, then it requires initialisation and
|
---|
59 | cleanup with the
|
---|
60 | .B \-i
|
---|
61 | and
|
---|
62 | .B \-c
|
---|
63 | options.
|
---|
64 | .TP
|
---|
65 | .BI \-g " get"
|
---|
66 | This specifies the program to use for getting the source text for the
|
---|
67 | build. If no
|
---|
68 | .B \-g
|
---|
69 | option is given, the default program
|
---|
70 | .BR mg_get_merge (1)
|
---|
71 | is used.
|
---|
72 | .TP
|
---|
73 | .BI \-s " source"
|
---|
74 | The
|
---|
75 | .B mgmerge
|
---|
76 | program consists of two parts. The first part initializes variables
|
---|
77 | to default values. The second part uses these variables to control
|
---|
78 | how the
|
---|
79 | .BR mg (1)
|
---|
80 | database is built. This option specifies a program to execute between
|
---|
81 | the first and second parts. The details of what the variables are, and
|
---|
82 | how they may be changed, are in comments in the
|
---|
83 | .B mgmerge
|
---|
84 | program.
|
---|
85 | If this option is used, the program should ideally be the same one that
|
---|
86 | was called by
|
---|
87 | .BR mgbuild (1)
|
---|
88 | to build the database since some parameters need to be consistent between
|
---|
89 | .B mgbuild
|
---|
90 | and
|
---|
91 | .B mgmerge.
|
---|
92 | .TP
|
---|
93 | .B \-S
|
---|
94 | This option will cause a slow merge to be performed on the inverted files,
|
---|
95 | where each inverted file entry is decoded and recoded.
|
---|
96 | The default is a fast merge. Accumulated fast merges slowly degrade
|
---|
97 | compression performance on the resulting inverted file so
|
---|
98 | a periodic slow merge is recommended.
|
---|
99 | .TP
|
---|
100 | .B \-w
|
---|
101 | Adding new documents can have an effect on the weight of the previous ones.
|
---|
102 | By default the weights for documents already in the collection are not
|
---|
103 | recomputed since the change in their values is usually small.
|
---|
104 | This option forces new weights to be recomputed.
|
---|
105 | Periodic use of this option, as for the "-S" option, is recomended,
|
---|
106 | otherwise query rankings may become inaccurate.
|
---|
107 | .TP
|
---|
108 | .I collection-name
|
---|
109 | This is the collection name, as required by the
|
---|
110 | .BR mg_get_merge (1)
|
---|
111 | program. It serves both as a
|
---|
112 | .I case
|
---|
113 | statement selector, and as the name of a subdirectory that holds the
|
---|
114 | indexing files.
|
---|
115 | .SH ENVIRONMENT
|
---|
116 | .TP "\w'\fBMGDATA\fP'u+2n"
|
---|
117 | .SB MGDATA
|
---|
118 | If this environment variable exists, then its value is used as the
|
---|
119 | default directory where the
|
---|
120 | .BR mg (1)
|
---|
121 | collection files are. If this variable does not exist, then the
|
---|
122 | directory \*(lq\fB.\fP\*(rq is used by default. The command line
|
---|
123 | option
|
---|
124 | .BI \-d " directory"
|
---|
125 | overrides the directory in
|
---|
126 | .BR MGDATA .
|
---|
127 | Note that a temporary directory under the
|
---|
128 | .B MGDATA
|
---|
129 | directory is used to perform the merge.
|
---|
130 | The default name for this directory is
|
---|
131 | .BR MERGE .
|
---|
132 | .SH FILES
|
---|
133 | .TP 20
|
---|
134 | .B *.invf
|
---|
135 | Inverted file.
|
---|
136 | .TP
|
---|
137 | .B *.invf.dict
|
---|
138 | Compressed stemmed dictionary.
|
---|
139 | .TP
|
---|
140 | .B *.invf.dict.blocked
|
---|
141 | The `on-disk' stemmed dictionary.
|
---|
142 | .TP
|
---|
143 | .B *.invf.idx
|
---|
144 | The index into the inverted file.
|
---|
145 | .TP
|
---|
146 | .B *.weight
|
---|
147 | The exact weights file.
|
---|
148 | .TP
|
---|
149 | .B *.text
|
---|
150 | Compressed documents.
|
---|
151 | .TP
|
---|
152 | .B *.text.stats
|
---|
153 | Text statistics.
|
---|
154 | .TP
|
---|
155 | .B *.text.dict
|
---|
156 | Compressed compression dictionary.
|
---|
157 | .TP
|
---|
158 | .B *.text.idx
|
---|
159 | Index into the compressed documents.
|
---|
160 | .TP
|
---|
161 | .B *.text.idx.wgt
|
---|
162 | Interleaved index into the compressed documents and document weights.
|
---|
163 | .TP
|
---|
164 | .B *.weight.approx
|
---|
165 | Approximate document weights.
|
---|
166 | .SH "SEE ALSO"
|
---|
167 | .na
|
---|
168 | .BR mg (1),
|
---|
169 | .BR mgbuild (1),
|
---|
170 | .BR mg_get_merge (1),
|
---|
171 | .BR mg_get (1),
|
---|
172 | .BR mg_invf_merge (1),
|
---|
173 | .BR mg_text_merge (1),
|
---|
174 | .BR mg_query (1),
|
---|
175 | .BR mg_weights_build (1).
|
---|