1 | ###########################################################################
|
---|
2 | #
|
---|
3 | # help.mg.src -- Source for the help command
|
---|
4 | # Copyright (C) 1994 Neil Sharman
|
---|
5 | #
|
---|
6 | # This program is free software; you can redistribute it and/or modify
|
---|
7 | # it under the terms of the GNU General Public License as published by
|
---|
8 | # the Free Software Foundation; either version 2 of the License, or
|
---|
9 | # (at your option) any later version.
|
---|
10 | #
|
---|
11 | # This program is distributed in the hope that it will be useful,
|
---|
12 | # but WITHOUT ANY WARRANTY; without even the implied warranty of
|
---|
13 | # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
---|
14 | # GNU General Public License for more details.
|
---|
15 | #
|
---|
16 | # You should have received a copy of the GNU General Public License
|
---|
17 | # along with this program; if not, write to the Free Software
|
---|
18 | # Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
|
---|
19 | #
|
---|
20 | # @(#)help.mg.src 1.8 21 Mar 1994
|
---|
21 | #
|
---|
22 | ###########################################################################
|
---|
23 | #
|
---|
24 | # The help file for mgquery.
|
---|
25 | # Lines starting with '#' are treated as comments and are discarded when
|
---|
26 | # the 'help.mg.h' file is produced
|
---|
27 | #
|
---|
28 | ###########################################################################
|
---|
29 |
|
---|
30 | HELP for mgquery
|
---|
31 | ================================
|
---|
32 |
|
---|
33 | The text is a summary of the information in the "mgquery" manual pages.
|
---|
34 |
|
---|
35 | The input to 'mgquery' consists of a series of input lines. The backslash
|
---|
36 | character ("\") is used at the end of lines to indicate that input
|
---|
37 | continues on the next line.
|
---|
38 |
|
---|
39 | Input lines on which the first character is a dot (".") are commands to
|
---|
40 | the mgquery program. Input lines that do not start with a dot are queries.
|
---|
41 |
|
---|
42 | A query consists of two parts. One part is a boolean or ranked query that
|
---|
43 | identifies documents. The second part is a post-processing pattern matching
|
---|
44 | operation. Any text between the first speech mark (") and the last speech
|
---|
45 | mark is considered to the a post-processing pattern.
|
---|
46 |
|
---|
47 |
|
---|
48 | The following command are available :-
|
---|
49 |
|
---|
50 | .help - displays this text.
|
---|
51 | .quit - quits the program.
|
---|
52 | .set name value - sets parameter "name" to "value" . If the parameter
|
---|
53 | is a boolean parameter and value is omitted the
|
---|
54 | parameter will be inverted (i.e. if it is true it
|
---|
55 | will change to false, if it is false it will change
|
---|
56 | to true).
|
---|
57 | .unset name - deletes parameter "name"
|
---|
58 | .reset - sets all the parameters to their initial state.
|
---|
59 | .display - displays the values of all the current parameters.
|
---|
60 | .push - pushes the current parameters on to a stack.
|
---|
61 | .pop - destroys the current parameters and pops a new set
|
---|
62 | of parameters off the stack.
|
---|
63 | .output arg - This is used to specify where to send the documents.
|
---|
64 | Arg may one be of the following:
|
---|
65 | > filename : Send output to the specified file.
|
---|
66 | >> filename : Append output to the specified file.
|
---|
67 | | command : The output is piped into command,
|
---|
68 | which is executed by sh.
|
---|
69 | .input arg - This is used to specify where input comes from.
|
---|
70 | Arg may one be of the following:
|
---|
71 | < filename : Get the input from the specified
|
---|
72 | file.
|
---|
73 | | command : The input comes from the standard
|
---|
74 | output of command, which is executed
|
---|
75 | by sh.
|
---|
76 |
|
---|
77 | On startup the mgquery program reads from the file .mgrc a sequence
|
---|
78 | of commands (NOTE: The .mgrc file may not contain any queries). mgquery
|
---|
79 | first looks for .mgrc in the current directory and then in the users home
|
---|
80 | directory. Lines starting with a '#' in the .mgrc file are considered to
|
---|
81 | be comments and are ignored.
|
---|
82 |
|
---|
83 | The following parameters (used in the .set and .unset commands) are
|
---|
84 | predefined and have special significance :-
|
---|
85 |
|
---|
86 | accumulator_method = `array'
|
---|
87 | This parameter is used during ranking, and specifies how the
|
---|
88 | weight for each document should be accumulated. The following
|
---|
89 | methods are available `array', `splay_tree', `hash_table', and
|
---|
90 | `list'.
|
---|
91 |
|
---|
92 | briefstats = `off'
|
---|
93 | This is a boolean parameter that determines whether the the
|
---|
94 | totals for disk, memory and time usage statistics will be
|
---|
95 | displayed. at the end of each query.
|
---|
96 | NOTE: this takes precedence over the parameters "diskstats",
|
---|
97 | "memstats" and "timestats". This parameter may take the values
|
---|
98 | `yes', `no', `true', `false', `on' or `off'.
|
---|
99 |
|
---|
100 | buffer = `1048576'
|
---|
101 | When the documents are being read in they are read into a
|
---|
102 | buffer of this size and then displayed from this buffer. If
|
---|
103 | the documents are larger than this buffer the buffer is
|
---|
104 | expanded automatically. Having a large buffer gives a very
|
---|
105 | slight performance improvement because it allows the order of
|
---|
106 | disk operations to be optimised. The buffer size is measured
|
---|
107 | in bytes.
|
---|
108 |
|
---|
109 | diskstats = `off'
|
---|
110 | This is a boolean parameter that determines whether the disk
|
---|
111 | usage statistics for the preceding query will be displayed
|
---|
112 | after each query. This parameter may take the values `yes',
|
---|
113 | `no', `true', `false', `on' or `off'.
|
---|
114 |
|
---|
115 | doc_sepstr = `---------------------------------- %n\n'
|
---|
116 | This specifies the string that will be used to separate
|
---|
117 | documents when they are displayed for `boolean' or `docnums'
|
---|
118 | queries. The standard C escape character sequences (see the
|
---|
119 | man page) may be used to place special characters in the
|
---|
120 | string. For example, a newline would the `\n'. To include a `%'
|
---|
121 | use the sequence `%%'. To include the MG document number use
|
---|
122 | the sequence `%n'.
|
---|
123 |
|
---|
124 | expert = `false'
|
---|
125 | If this is true then a lot of the waffle that the program
|
---|
126 | spits out is suppressed. This parameter may take the values
|
---|
127 | `yes', `no', `true', `false', `on' or `off'.
|
---|
128 |
|
---|
129 | hash_tbl_size = `1000'
|
---|
130 | One of the options during ranking queries is to use a hash
|
---|
131 | table to accumulate the weights for each document. The hash
|
---|
132 | table is a simple chained type. This parameter specifies the
|
---|
133 | size of the hash table and may take any value between 8 and
|
---|
134 | 268435456.
|
---|
135 |
|
---|
136 | heads_length = `50'
|
---|
137 | When the mode is `heads' this specifies the number of
|
---|
138 | characters that will be output for each document.
|
---|
139 |
|
---|
140 | maxdocs = `all'
|
---|
141 | The maximum number of documents to display in response to a
|
---|
142 | query. This parameter may take on a numeric value between 1
|
---|
143 | and 429467295 or the word `all'.
|
---|
144 |
|
---|
145 | maxparas = `1000'
|
---|
146 | The maximum number of paragraphs to identify during a ranked
|
---|
147 | query with paragraph indexing. After the paragraphs have been
|
---|
148 | identified the paragraphs are converted into documents, and
|
---|
149 | because some of the paragraphs may refer to the same documents
|
---|
150 | the final number of answers may be less that maxparas. The
|
---|
151 | maxdocs parameter will then be applied. This parameter may
|
---|
152 | take on a numeric value between 1 and 429467295.
|
---|
153 |
|
---|
154 | max_accumulators = `50000'
|
---|
155 | This parameter limits the number of different paragraph/
|
---|
156 | document numbers to be accumulated during ranked queries when
|
---|
157 | the parameter `accumulator_method` is set to `splay_tree',
|
---|
158 | `hash_table', or `list'. This parameter may take any value
|
---|
159 | between 8 and 268435456.
|
---|
160 |
|
---|
161 | max_terms = `all'
|
---|
162 | This parameter limits the number of terms that will actually
|
---|
163 | be used during a ranked query. If more terms than the number
|
---|
164 | specified by max_terms are entered, then the extra terms will
|
---|
165 | be discarded. If `sorted_terms' is on then the limiting will
|
---|
166 | be done after the terms have been sorted. This parameter may
|
---|
167 | take any value between 1 and 429467295 or the word `all'.
|
---|
168 |
|
---|
169 | memstats = `off'
|
---|
170 | This is a boolean parameter that determines whether the memory
|
---|
171 | usage statistics for the preceding query will be displayed
|
---|
172 | after each query. This parameter may take the values `yes',
|
---|
173 | `no', `true', `false', `on' or `off'.
|
---|
174 |
|
---|
175 | mgdir = `.'
|
---|
176 | This specifies the directory where the MG files may be found.
|
---|
177 | If the environment variable `MGDATA' is set then `mgdir' is
|
---|
178 | initialised to the value in `MGDATA'.
|
---|
179 |
|
---|
180 | mgname = `'
|
---|
181 | This specifies the name of the MG database to process.
|
---|
182 |
|
---|
183 | mode = `text'
|
---|
184 | This specifies how documents should be displayed when they
|
---|
185 | are retrieved it may take four different values `text',
|
---|
186 | `docnums', `silent', `heads' or `count'. `text' displays
|
---|
187 | the contents of the document. `docnums' displays only the
|
---|
188 | document numbers. `Silent' retrieves all the documents but
|
---|
189 | displays nothing except how many documents were retrieved.
|
---|
190 | This mode is intended to be used in timing experiments.
|
---|
191 | `Heads` is used to print out the head of each document.
|
---|
192 | `Count' does the minimum amount of work required to determine
|
---|
193 | how many documents would be retrieved, but does not retrieve
|
---|
194 | them.
|
---|
195 |
|
---|
196 | pager = `more'
|
---|
197 | This is the name of the program that will be used to display
|
---|
198 | the help and the retrieved documents. If the environment
|
---|
199 | variable "PAGER" is defined then `pager' takes on that value.
|
---|
200 |
|
---|
201 | para_sepstr = `\n######## PARAGRAPH %n ########\n'
|
---|
202 | This specifies the string that will be used to separate
|
---|
203 | paragraphs. The standard C escape character sequences (see the
|
---|
204 | man page) may be used to place special characters in the
|
---|
205 | string. For example, a newline would the `\n'. To include a `%'
|
---|
206 | use the sequence `%%'. To include the paragraph number within
|
---|
207 | the document use the sequence `%n'.
|
---|
208 |
|
---|
209 | para_start = `***** Weight = %w *****\n'
|
---|
210 | This specifies the string that will be used at the head of
|
---|
211 | paragraphs for a paraghaph level index following a ranked query.
|
---|
212 | The standard C escape character sequences (see the man page)
|
---|
213 | may be used to place special characters in the string. For
|
---|
214 | example, a newline would the `\n'. To include a `%' use the
|
---|
215 | sequence `%%'. To include the paragraph weight use the
|
---|
216 | sequence `%w'.
|
---|
217 |
|
---|
218 | qfreq = `true'
|
---|
219 | This determine whether the ranked queries will take into
|
---|
220 | account the number of times each query term is specified.
|
---|
221 | When this is `true' the number of times a term appears in
|
---|
222 | the query is used in the ranking. When this is `false' all
|
---|
223 | query term are assumed to occur only once. This parameter
|
---|
224 | may take the values `yes', `no', `true', `false', `on' or
|
---|
225 | `off'.
|
---|
226 |
|
---|
227 | query = `boolean'
|
---|
228 | This specifies the type of queries that are to be specified.
|
---|
229 | It can take four different values `boolean', `ranked',
|
---|
230 | `docnums' or `approx-ranked'.
|
---|
231 |
|
---|
232 | `boolean' is for boolean queries.
|
---|
233 | The yacc grammar for boolean queries is as follows :-
|
---|
234 |
|
---|
235 | query : or;
|
---|
236 |
|
---|
237 | or : or '|' and
|
---|
238 | | and ;
|
---|
239 |
|
---|
240 | and : and '&' not
|
---|
241 | | and not
|
---|
242 | | not ;
|
---|
243 |
|
---|
244 | not : term
|
---|
245 | | '!' not ;
|
---|
246 |
|
---|
247 | term : TERM
|
---|
248 | | '(' or ')' ;
|
---|
249 |
|
---|
250 |
|
---|
251 |
|
---|
252 |
|
---|
253 | `ranked' and `approx-ranked' are for queries ranked by the
|
---|
254 | cosine measure. `approx-ranked' uses only the low
|
---|
255 | precision document lengths, and therefore only
|
---|
256 | produces an approximation to full cosine ranking.
|
---|
257 |
|
---|
258 | query : TERM
|
---|
259 | | query TERM ;
|
---|
260 |
|
---|
261 | `docnums' allows the entry of document numbers. Multiple
|
---|
262 | numbers separated by spaces may be specified
|
---|
263 | or ranges separated by hyphens.
|
---|
264 |
|
---|
265 |
|
---|
266 | query : range
|
---|
267 | | query range ;
|
---|
268 |
|
---|
269 |
|
---|
270 | range : num
|
---|
271 | | num '-' num ;
|
---|
272 |
|
---|
273 |
|
---|
274 | ranked_doc_sepstr = `---------------------------------- %n %w\n'
|
---|
275 | This specifies the string that will be used to separate
|
---|
276 | documents when they are displayed for `ranked' or
|
---|
277 | `approx-ranked' queries. The standard C escape character
|
---|
278 | sequences (see the man page) may be used to place special
|
---|
279 | characters in the string. For example, a newline would the
|
---|
280 | `\n'. To include a `%' use the sequence `%%'. To include the
|
---|
281 | MG document number use the sequence `%n'. To include the
|
---|
282 | document weight use the sequence `%w'.
|
---|
283 |
|
---|
284 | sizestats = `false'
|
---|
285 | If this is true then various numbers are output at the end
|
---|
286 | of each query indicating what went on during the query.
|
---|
287 | This parameter may take the values `yes', `no', `true',
|
---|
288 | `false', `on' or `off'.
|
---|
289 |
|
---|
290 | skip_dump = `skips.%d'
|
---|
291 | If this parameter is set then during ranked queries on skipped
|
---|
292 | inverted files when `accumulator_method' is set to `splay_tree',
|
---|
293 | `hash_table', or `list` a file will be produced in the current
|
---|
294 | directory. The name of the file is the value of this parameter,
|
---|
295 | a `%d' in the file name will be replaced with the process id of
|
---|
296 | mgquery. This file will contain information about the usage of
|
---|
297 | skips during the query processing. This option is expensive;
|
---|
298 | use `.unset skip_dump' to obtain optimal performance.
|
---|
299 |
|
---|
300 | sorted_terms = `on'
|
---|
301 | This specifies whether of not the terms should be sorted into
|
---|
302 | decreasing occurrence in documents so that the least often
|
---|
303 | occurring terms are processed first when ranked queries are
|
---|
304 | being done. When this is true the terms are sorted. When this
|
---|
305 | is false the terms are not sorted and are instead processed in
|
---|
306 | order of occurrence. This parameter may take the values `yes',
|
---|
307 | `no', `true', `false', `on' or `off'.
|
---|
308 |
|
---|
309 |
|
---|
310 | stop_at_max_accum = `on'
|
---|
311 | This specifies what should happen when the maximum number of
|
---|
312 | accumulators set by `max_accumulators' is reached. When this
|
---|
313 | is true the the processing of terms is stopped at the completion
|
---|
314 | of the current term. When this is false processing continues but
|
---|
315 | no new accumulators are created. This parameter may take the
|
---|
316 | values `yes', `no', `true', `false', `on' or `off'.
|
---|
317 |
|
---|
318 | terminator = `'
|
---|
319 | This specifies the string that will be output after the last
|
---|
320 | document from the previous query has been output. The standard
|
---|
321 | C escape character sequences (see the man page) may be used to
|
---|
322 | place special characters in the string. For example, a newline
|
---|
323 | would the `\n'. To include a `%' use the sequence `%%'.
|
---|
324 |
|
---|
325 |
|
---|
326 | timestats = `false'
|
---|
327 | If this is true then the time to process a query is displayed
|
---|
328 | in both real time and CPU time. This parameter may take the
|
---|
329 | values `yes', `no', `true', `false', `on' or `off'.
|
---|
330 |
|
---|
331 | verbatim = `off'
|
---|
332 | This is a boolean parameter that determines whether the program
|
---|
333 | should attempt to do a regular expression match on the retrieved
|
---|
334 | text. If verbatim is `on' and a post-processing strng is specified
|
---|
335 | with the query then the post-processing string will be searched for
|
---|
336 | in the documents just before they are displayed. If the string is
|
---|
337 | found the document will be displayed, if not the document will not
|
---|
338 | be displayed. If verbatim is `off' the post-processing string will
|
---|
339 | be considered a regular expression like in `vi' or `egrep'.
|
---|
340 | E.G. If verbatim is `on', "and.*the" will look for the 8 character
|
---|
341 | sequence "and.*the". If verbatim is `off', "and.*the" will
|
---|
342 | look for the sequence "and" followed somewhere later in the
|
---|
343 | document by the sequence "the".
|
---|
344 | This parameter may take the values `yes', `no', `true', `false',
|
---|
345 | `on' or `off'.
|
---|
346 |
|
---|
347 |
|
---|