1 |
#!/usr/bin/perl -w |
2 |
# -*- Mode: Perl -*- |
3 |
# $Basename: sman $ |
4 |
# $Revision: 1.14 $ |
5 |
# Author : Ulrich Pfeifer |
6 |
# Created On : Fri Aug 30 15:52:25 1996 |
7 |
# Last Modified By: Ulrich Pfeifer |
8 |
# Last Modified On: Mon May 8 11:03:46 2000 |
9 |
# Language : CPerl |
10 |
# |
11 |
# (C) Copyright 1996-2000, Ulrich Pfeifer |
12 |
# |
13 |
|
14 |
use strict; |
15 |
|
16 |
use Term::ReadLine; |
17 |
use Getopt::Long; |
18 |
use Fcntl qw(O_RDONLY); |
19 |
use Config; |
20 |
|
21 |
use lib '/data/wait/lib'; |
22 |
|
23 |
require WAIT::Config; |
24 |
require WAIT::Database; |
25 |
require WAIT::Query::Base; |
26 |
require WAIT::Query::Wais; |
27 |
|
28 |
|
29 |
$SIG{PIPE} = 'IGNORE'; |
30 |
my %OPT = (database => 'DB', |
31 |
dir => $WAIT::Config->{WAIT_home} || '/tmp', |
32 |
table => 'man', |
33 |
pager => $WAIT::Config->{'pager'} || 'more', |
34 |
filter => 0, |
35 |
max => 15, |
36 |
); |
37 |
|
38 |
GetOptions(\%OPT, |
39 |
'database=s', |
40 |
'dir=s', |
41 |
'table=s', |
42 |
'filter=i', |
43 |
'max=i', |
44 |
'pager:s') || die " |
45 |
Usage: $0 |
46 |
[--database database] |
47 |
[--dir dir ] |
48 |
[--table table ] |
49 |
[--filter integer ] |
50 |
[--max integer ] |
51 |
[--pager pager ] |
52 |
"; |
53 |
|
54 |
my $db = WAIT::Database->open(name => $OPT{database}, |
55 |
mode => O_RDONLY, |
56 |
directory => $OPT{dir}) |
57 |
or die "Could not open database $OPT{database}: $@"; |
58 |
|
59 |
my $tb = $db->table(name => $OPT{table}) |
60 |
or die "Could not open table $OPT{table}: $@"; |
61 |
|
62 |
# not used: my $layout = $tb->layout; # a WAIT::Parse::Nroff object |
63 |
|
64 |
my $term = new Term::ReadLine 'Simple Query Interface'; |
65 |
|
66 |
require WAIT::Format::Term; |
67 |
my $format; |
68 |
if ($Config::Config{'archname'} eq 'i586-linux') { |
69 |
# for color xterm |
70 |
$format = new WAIT::Format::Term query_s => "[01;31m", query_e => "[00m"; |
71 |
} else { |
72 |
$format = new WAIT::Format::Term; |
73 |
} |
74 |
|
75 |
my $pager = ($OPT{pager}) ? \&pager : \&less; |
76 |
$pager = \&less; # force pager usage |
77 |
my $OUT = $term->OUT; |
78 |
|
79 |
my $st = 1; |
80 |
print $OUT "Enter 'h' for help.\n"; |
81 |
|
82 |
# sman is basically offering three services: find the hits and show |
83 |
# them (a query), show metadata for a hit (a view), show a hot (display) |
84 |
|
85 |
my($query, @did); |
86 |
|
87 |
while (defined ($_ = &myreadline("$st> "))) { |
88 |
chomp; $st++; |
89 |
|
90 |
my(%hits, $query_text); |
91 |
if (/^$/) { |
92 |
next; |
93 |
} elsif (/^m (\d+)$/) { |
94 |
$OPT{max} = $1; |
95 |
} elsif (/^f\s*(\d+)?$/) { |
96 |
$OPT{filter} = $1; |
97 |
next; |
98 |
} elsif (/^t$/i) { |
99 |
if ($pager eq \&less) { |
100 |
$pager = \&pager; |
101 |
} else { |
102 |
$pager = \&less; |
103 |
} |
104 |
next; |
105 |
} elsif (/^(\d+)$/) { |
106 |
if (defined $did[$1]) { |
107 |
display($did[$1]); # <----------- display (full doc) |
108 |
next; |
109 |
} |
110 |
} elsif (/^d\s*(\d+)/) { |
111 |
if (defined $did[$1]) { |
112 |
view($did[$1]); # <----------- view (metadata from WAIT) |
113 |
next; |
114 |
} |
115 |
} elsif (/^q$/i) { |
116 |
last; |
117 |
} elsif (/^l$/i) { |
118 |
# fall through |
119 |
} elsif (/^[h?]$/i) { |
120 |
help(); |
121 |
next; |
122 |
} elsif (/^hh$/i) { |
123 |
extended_help(); |
124 |
next; |
125 |
} else { # <----------- A query (Display a list) |
126 |
$query_text = $_; |
127 |
eval {$query = WAIT::Query::Wais::query($tb, $_)}; |
128 |
if ($@ ne '') { |
129 |
print $OUT "$_ => $query\n\$\@='$@'\n"; |
130 |
} elsif (ref($query)) { |
131 |
%hits = $query->execute(top => $OPT{max}, picky => 1); |
132 |
# the hash %hits has as keys document numbers and as values |
133 |
# quality figures. The doc numbers are not what we have as docid |
134 |
# to find the item in the access class, they are WAIT's private |
135 |
# numbers. |
136 |
} else { |
137 |
next; |
138 |
} |
139 |
} |
140 |
|
141 |
next unless %hits; |
142 |
my $no = 1; # numbering the hits for the result table that is |
143 |
# presented to the user |
144 |
|
145 |
@did = (); # store the internal numbers (keys of %hits). The user |
146 |
# will use $no in sman's interface to select a hit. |
147 |
|
148 |
# the following loop uses the values of %hits to sort the results |
149 |
# according to the quality and cut after a number of rows. After |
150 |
# that %hits isn't needed anymore. |
151 |
print "Query: $query_text\n"; |
152 |
for my $did (sort {$hits{$b} <=> $hits{$a}} keys %hits) { |
153 |
|
154 |
my %tattr = $tb->fetch($did); |
155 |
# the hash %tattr contains several attributes of the item we are |
156 |
# referring to, namely the attributes that we named in the "attr" |
157 |
# argument of the create_table statement in smakewhatis |
158 |
|
159 |
printf $OUT "%2d %6.3f %s\n", $no, $hits{$did}, |
160 |
substr($tattr{headline} ||'',0,68); |
161 |
$did[$no] = $did; |
162 |
last if $no++ >= $OPT{max}; |
163 |
|
164 |
} |
165 |
|
166 |
} continue { |
167 |
# we don't do this since Andreas Koenig does not think of it as feature |
168 |
# $term->SetHistory(grep length($_)>4, $term->GetHistory) |
169 |
} |
170 |
warn "Thank you for using sman\n"; |
171 |
|
172 |
$tb->close; |
173 |
$db->close; |
174 |
|
175 |
sub myreadline { |
176 |
if (@ARGV) { |
177 |
return shift @ARGV; |
178 |
} else { |
179 |
$term->readline(@_); |
180 |
} |
181 |
} |
182 |
sub help { |
183 |
my $idb = "\n\t'". join(q[', '], $tb->fields()) . "'"; |
184 |
print $OUT qq[Available commands: |
185 |
|
186 |
<num> Show the document <num> |
187 |
d <num> Show the db entry of document <num> |
188 |
f <num> Display only <num> lines context |
189 |
h,? Display this help message |
190 |
hh Display query examples |
191 |
m <num> Set maxhits to <num> |
192 |
t Toggle display mode (term/less) |
193 |
q Exit from $0 |
194 |
l redisplay last ranking |
195 |
Other input is tried as wais query. |
196 |
The following fields for table '$OPT{table}' are known: $idb |
197 |
] ; |
198 |
} |
199 |
|
200 |
sub extended_help { |
201 |
print q{ |
202 |
Here are some query examples: |
203 |
|
204 |
information retrieval free text query |
205 |
information or retrieval same as above |
206 |
des=information retrieval `information' must be in the description |
207 |
des=(information retrieval) one of them in description |
208 |
des=(information or retrieval) same as above |
209 |
des=(information and retrieval) both of them in description |
210 |
des=(information not retrieval) `information' in description and |
211 |
`retrieval' not in description |
212 |
des=(information system*) wild-card search |
213 |
au=ilia author names may be misspelled |
214 |
|
215 |
You can build arbitary boolean combination of the above examples. |
216 |
Field names may be abbreviated. |
217 |
} |
218 |
} |
219 |
|
220 |
sub view { |
221 |
my $did = shift; |
222 |
my %tattr = $tb->fetch($did); |
223 |
for (keys %tattr) { |
224 |
print $OUT "$_ $tattr{$_}\n"; |
225 |
} |
226 |
} |
227 |
|
228 |
sub display { |
229 |
my $did = shift; |
230 |
|
231 |
return unless defined $query and defined $did; |
232 |
|
233 |
print $OUT "Wais display document $did\n"; |
234 |
my %tattr = $tb->fetch($did); |
235 |
my $tdid = $tattr{docid}; |
236 |
# WHAT DOES HE DO HERE? ULI??? |
237 |
# Re: some indexing scripts did use pathnames relative to the table directory |
238 |
# especially the cpanwait script does this. uli |
239 |
if ($tdid !~ m(^/)) { |
240 |
$tdid = $tb->dir . '/' . $tdid; |
241 |
} |
242 |
|
243 |
# The main task of all that follows from here is highlighting. WAIT |
244 |
# is designed to make it possible to show the user why a certain |
245 |
# document was chosen by the indexer. |
246 |
|
247 |
my $buf = $tb->fetch_extern($tdid); |
248 |
# This $buf can be an object that can have enough information to do |
249 |
# highlighting without WAIT's help. If you prefer to implement your |
250 |
# own highlighting, you can do so now with e.g. print |
251 |
# $buf->highlight(query => $query) |
252 |
|
253 |
# All you need to know to implement highlighting is how a |
254 |
# WAIT::Query::Base object looks like (left as an exercise for the |
255 |
# reader). |
256 |
|
257 |
# The impatient reader may want to implement something without |
258 |
# highlighting, in which case he does not need any info about the |
259 |
# query object and can rightaway run e.g. |
260 |
# print $buf->as_string |
261 |
|
262 |
# Thus the impatient reader does not necessarily need the following |
263 |
# heavy wizardry. Just to give you an idea what's going on: every |
264 |
# word in the text must be compared to every word in the query if it |
265 |
# is worth highlighting, and which part of the word is worth |
266 |
# highlighting. This must be done differently for every field in the |
267 |
# table and for every index defined for that field. Try to run a |
268 |
# query with 100 words and you'll be amazed to see it really works. |
269 |
# Or maybe it doesn't. You should be aware that the hilighting code |
270 |
# is to be regarded as alpha. It is certainly the least tested part |
271 |
# of WAIT so far. |
272 |
|
273 |
if ($buf) { |
274 |
my @txt = $query->hilight($buf); |
275 |
# In this operation the following things melt into one piece: |
276 |
# $query: The query entered by the user (Class isa WAIT::Query::Base) |
277 |
# $tb: The table we queried (Class WAIT::Table) |
278 |
# $buf: The document to display (User defined class or string) |
279 |
# The steps taken are: |
280 |
# 1.) $query calls "hilight" on $tb and passes |
281 |
# filtered and raw search terms ($query->{Plain} and $query->{Raw}). |
282 |
# 2.) $tb asks the layout object to tag the object which results |
283 |
# in an array with alternating elements of tags (anon HASHes) and |
284 |
# strings. |
285 |
# 3.) $tb adds some markup on its own: {qt=>1} or some such |
286 |
|
287 |
# The result of that process can optionally be sent through a |
288 |
# filter, just to impress your friends with yet more heavy |
289 |
# wizardry |
290 |
if ($OPT{filter}) { |
291 |
@txt = &filter(@txt); |
292 |
} |
293 |
|
294 |
# And then a formatter (in our case a terminal formatter) turns |
295 |
# all the markup into escape sequences and strings that can in |
296 |
# turn be sent through a pager for instance |
297 |
&$pager($format->as_string(\@txt)); |
298 |
} |
299 |
|
300 |
# Hey, that's it. The user out there is deeply impressed now. You |
301 |
# can lean back again:-) He got a document that has some words |
302 |
# hilighted and will probably read and enjoy it. Maybe he'll send |
303 |
# you an email. |
304 |
} |
305 |
|
306 |
sub filter { |
307 |
my @result; |
308 |
my @context; |
309 |
my $lines = 0; |
310 |
my $clines = 0; |
311 |
my $elipsis = 0; |
312 |
|
313 |
print STDERR "Filter ..."; |
314 |
while (@_) { |
315 |
my %tag = %{shift @_}; |
316 |
my $txt = shift @_; |
317 |
|
318 |
for (split /(\n)/, $txt) { |
319 |
if ($_ eq "\n") { |
320 |
if (exists $tag{_qt}) { |
321 |
#die "Weird!"; |
322 |
push @result, {_i=>1}, "[WEIRD]"; |
323 |
} elsif ($lines) { |
324 |
push @result, {}, $_; |
325 |
$lines--; |
326 |
} else { |
327 |
push @context, {}, $_; |
328 |
$clines++; |
329 |
} |
330 |
} else { |
331 |
if (exists $tag{_qt}) { |
332 |
push @result, {_i=>1}, "\n[ $elipsis linesĀ ]\n" if $elipsis; |
333 |
push @result, @context, {%tag}, $_; |
334 |
delete $tag{_qt}; |
335 |
@context = (); $clines = 0; $elipsis=0; |
336 |
$lines = $OPT{filter}+1; |
337 |
} elsif ($lines) { |
338 |
push @result, \%tag, $_; |
339 |
} else { |
340 |
push @context, \%tag, $_; |
341 |
} |
342 |
} |
343 |
if ($clines>$OPT{filter}) { |
344 |
my (%tag, $txt); |
345 |
while ($clines>$OPT{filter}) { |
346 |
%tag = %{shift @context}; |
347 |
$txt = shift @context; |
348 |
if ($txt =~ /\n/) { |
349 |
$clines--; |
350 |
$elipsis++; |
351 |
} |
352 |
} |
353 |
} |
354 |
} |
355 |
} |
356 |
print STDERR " done\n"; |
357 |
@result; |
358 |
} |
359 |
|
360 |
sub less { |
361 |
my $flags; |
362 |
if ($WAIT::Config->{pager} =~ /less/) { |
363 |
$flags = '-r'; |
364 |
} elsif ($WAIT::Config->{pager} =~ /more/) { |
365 |
$flags = '-c'; |
366 |
} |
367 |
open(PAGER, "|$WAIT::Config->{pager} $flags") or die; |
368 |
print PAGER @_; |
369 |
close PAGER; |
370 |
} |
371 |
|
372 |
sub pager { |
373 |
my @lines = split /\n/, $_[0]; |
374 |
my $line = 0; |
375 |
for (@lines) { |
376 |
print "$_\n"; $line++; |
377 |
if ($line % 24 == 0) { |
378 |
my $key = $term->readline("[return]"); |
379 |
return if $key =~ /^q/i; |
380 |
} |
381 |
} |
382 |
} |
383 |
|
384 |
|
385 |
__END__ |
386 |
## ################################################################### |
387 |
## pod |
388 |
## ################################################################### |
389 |
|
390 |
=head1 NAME |
391 |
|
392 |
sman - Search and disply manuals interactive |
393 |
|
394 |
=head1 SYNOPSIS |
395 |
|
396 |
B<sman> |
397 |
[B<-database> I<database name>] |
398 |
[B<-dir> I<database directory>] |
399 |
[B<-table> I<name>] |
400 |
[B<-less>] |
401 |
[B<-filter> I<num>] |
402 |
[B<-max> I<num>] |
403 |
|
404 |
=head1 DESCRIPTION |
405 |
|
406 |
B<Sman> is an interactive search interface to your systems manual pages. |
407 |
|
408 |
=head2 OPTIONS |
409 |
|
410 |
=over 10 |
411 |
|
412 |
=item B<-database> I<database name> |
413 |
|
414 |
Change the default database name to I<database name>. |
415 |
|
416 |
=item B<-dir> I<database directory> |
417 |
|
418 |
Change the default database directory to I<database directory>. |
419 |
|
420 |
=item B<-table> I<name> |
421 |
|
422 |
Use I<name> instead of C<man> as table name. |
423 |
|
424 |
=item B<-pager> I<name> |
425 |
|
426 |
Use I<name> instead of the default pager. If no I<name> is supplied a |
427 |
buildin pager is used. |
428 |
|
429 |
=item B<-filter> I<num> |
430 |
|
431 |
Display only I<num> lines above and below an occurance of a search |
432 |
term in the manual. |
433 |
|
434 |
=item B<-max> I<num> |
435 |
|
436 |
Display only I<num> hits. Default is to 10. |
437 |
|
438 |
=head1 SEE ALSO |
439 |
|
440 |
L<smakewhatis>. |
441 |
|
442 |
=head1 AUTHOR |
443 |
|
444 |
Ulrich Pfeifer E<lt>F<pfeifer@ls6.informatik.uni-dortmund.de>E<gt> |