1 | ###########################################################################
|
---|
2 | #
|
---|
3 | # phproc.pm -- the Phind document processor
|
---|
4 | #
|
---|
5 | # Copyright (C) 2000 Gordon Paynter
|
---|
6 | #
|
---|
7 | # A component of the Greenstone digital library software
|
---|
8 | # from the New Zealand Digital Library Project at the
|
---|
9 | # University of Waikato, New Zealand.
|
---|
10 | #
|
---|
11 | # This program is free software; you can redistribute it and/or modify
|
---|
12 | # it under the terms of the GNU General Public License as published by
|
---|
13 | # the Free Software Foundation; either version 2 of the License, or
|
---|
14 | # (at your option) any later version.
|
---|
15 | #
|
---|
16 | # This program is distributed in the hope that it will be useful,
|
---|
17 | # but WITHOUT ANY WARRANTY; without even the implied warranty of
|
---|
18 | # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
---|
19 | # GNU General Public License for more details.
|
---|
20 | #
|
---|
21 | # You should have received a copy of the GNU General Public License
|
---|
22 | # along with this program; if not, write to the Free Software
|
---|
23 | # Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
|
---|
24 | #
|
---|
25 | ###########################################################################
|
---|
26 |
|
---|
27 | # This is the phind document processor object. It is used by the document
|
---|
28 | # reader plugins to extract the clauses from each document.
|
---|
29 |
|
---|
30 | package phproc;
|
---|
31 |
|
---|
32 | use docproc;
|
---|
33 | use util;
|
---|
34 |
|
---|
35 | sub BEGIN {
|
---|
36 | @ISA = ('docproc');
|
---|
37 | }
|
---|
38 |
|
---|
39 | sub new {
|
---|
40 | my ($class, $archive_dir, $phindex_dir,
|
---|
41 | $language, $delimiter, $verbosity, $outhandle) = @_;
|
---|
42 |
|
---|
43 | my $self = new docproc ();
|
---|
44 |
|
---|
45 | # $self->{'collection'} = $collection;
|
---|
46 | $self->{'archive_dir'} = $archive_dir;
|
---|
47 | $self->{'phindex_dir'} = $phindex_dir;
|
---|
48 |
|
---|
49 | $language =~ s/,/\|/g;
|
---|
50 | $self->{'language_exp'} = $language;
|
---|
51 | $self->{'delimiter'} = $delimiter;
|
---|
52 |
|
---|
53 | $self->{'verbosity'} = $verbosity;
|
---|
54 | $self->{'outhandle'} = STDERR;
|
---|
55 | $self->{'outhandle'} = $outhandle if defined $outhandle;
|
---|
56 |
|
---|
57 | &util::rm("$phindex_dir/clauses") if (-e "$phindex_dir/clauses");
|
---|
58 | open(TEXT, ">$phindex_dir/clauses")
|
---|
59 | || die "Cannot open $phindex_dir/clauses: $!";
|
---|
60 | $self->{'txthandle'} = TEXT;
|
---|
61 |
|
---|
62 | &util::rm("$phindex_dir/mg-d.txt") if (-e "$phindex_dir/mg-d.txt");
|
---|
63 | open(DOCS, ">$phindex_dir/mg-d.txt")
|
---|
64 | || die "Cannot open $phindex_dir/mg-d.txt: $!";
|
---|
65 | $self->{'dochandle'} = DOCS;
|
---|
66 |
|
---|
67 | return bless $self, $class;
|
---|
68 |
|
---|
69 | }
|
---|
70 |
|
---|
71 | sub process {
|
---|
72 | my $self = shift (@_);
|
---|
73 | my ($doc_obj) = @_;
|
---|
74 |
|
---|
75 | my $verbosity = $self->{'verbosity'};
|
---|
76 | my $top_section = $doc_obj->get_top_section();
|
---|
77 |
|
---|
78 | my $title = $doc_obj->get_metadata_element ($top_section, "Title");
|
---|
79 | print "process: $title\n" if ($verbosity > 1);
|
---|
80 |
|
---|
81 |
|
---|
82 | # only consider english-language files
|
---|
83 | my $doclanguage = $doc_obj->get_metadata_element ($top_section, "Language");
|
---|
84 | my $phrlanguage = $self->{'language_exp'};
|
---|
85 | return if ($doclanguage && ($doclanguage !~ /$phrlanguage/i));
|
---|
86 |
|
---|
87 | # record this file
|
---|
88 | my $total++;
|
---|
89 | print "file $total: $file\n" if ($self->{'$verbosity'});
|
---|
90 |
|
---|
91 | # store object ID & title in document index file
|
---|
92 | my $OID = $doc_obj->get_OID();
|
---|
93 | $OID = "NULL" unless defined $OID;
|
---|
94 |
|
---|
95 | my $dochandle = $self->{'dochandle'};
|
---|
96 | # print "dochandle: =$dochandle=\n";
|
---|
97 | print $dochandle "$OID\t$title\n";
|
---|
98 |
|
---|
99 | # store the text
|
---|
100 | $text = convert_gml_to_tokens($doc_obj->get_text());
|
---|
101 |
|
---|
102 | my $txthandle = $self->{'txthandle'};
|
---|
103 | print $txthandle $self->{'delimiter'}, "\n$text\n";
|
---|
104 | }
|
---|
105 |
|
---|
106 |
|
---|
107 |
|
---|
108 | sub convert_gml_to_tokens {
|
---|
109 |
|
---|
110 | $_ = shift @_;
|
---|
111 |
|
---|
112 | # FIRST, remove GML tags
|
---|
113 |
|
---|
114 | # Replace all whitespace with a simple space
|
---|
115 | s/\s+/ /gs;
|
---|
116 |
|
---|
117 | # Remove everything that is in a tag
|
---|
118 | s/\s*<p>\s*/ PARAGRAPHBREAK /isg;
|
---|
119 | s/\s*<br>\s*/ LINEBREAK /isg;
|
---|
120 | s/<[^>]*>/ /sg;
|
---|
121 |
|
---|
122 | # Now we have the text, but it may contain HTML
|
---|
123 | # elements coded as > etc. Remove these tags.
|
---|
124 | s/</</sg;
|
---|
125 | s/>/>/sg;
|
---|
126 |
|
---|
127 | s/\s+/ /sg;
|
---|
128 | s/\s*<p>\s*/ PARAGRAPHBREAK /isg;
|
---|
129 | s/\s*<br>\s*/ LINEBREAK /isg;
|
---|
130 | s/<[^>]*>/ /sg;
|
---|
131 |
|
---|
132 | # remove & and other miscellaneous markup tags
|
---|
133 | s/&/&/sg;
|
---|
134 | s/</</sg;
|
---|
135 | s/>/>/sg;
|
---|
136 | s/&/&/sg;
|
---|
137 |
|
---|
138 | # replace<p> and <br> placeholders with carriage returns
|
---|
139 | s/PARAGRAPHBREAK/\n/sg;
|
---|
140 | s/LINEBREAK/\n/sg;
|
---|
141 |
|
---|
142 |
|
---|
143 | # Exceptional punctuation
|
---|
144 | #
|
---|
145 | # We make special cases of some punctuation
|
---|
146 |
|
---|
147 | # remove any apostrophe that indicates omitted letters
|
---|
148 | s/(\w+)\'(\w*\s)/ $1$2 /g;
|
---|
149 |
|
---|
150 | # remove period that appears in a person's initals
|
---|
151 | s/\s([A-Z])\./ $1 /g;
|
---|
152 |
|
---|
153 | # replace hyphens in hypheanted words and names with a space
|
---|
154 | s/([A-Za-z])-\s*([A-Za-z])/$1 $2/g;
|
---|
155 |
|
---|
156 |
|
---|
157 | # Convert the remaining text to "clause format",
|
---|
158 | # This means removing all excess punctuation and garbage text,
|
---|
159 | # normalising valid punctuation to fullstops and commas,
|
---|
160 | # then putting one cluse on each line.
|
---|
161 |
|
---|
162 | # Insert newline when the end of a sentence is detected
|
---|
163 | # (delimter is: "[\.\?\!]\s")
|
---|
164 | s/\s*[\.\?\!]\s+/\n/g;
|
---|
165 |
|
---|
166 | # split numbers after four digits
|
---|
167 | s/(\d\d\d\d)/$1 /g;
|
---|
168 |
|
---|
169 | # split words after 32 characters
|
---|
170 |
|
---|
171 | # squash repeated punctuation
|
---|
172 | tr/A-Za-z0-9 //cs;
|
---|
173 |
|
---|
174 | # save email addresses
|
---|
175 | # s/\w+@\w+\.[\w\.]+/EMAIL/g;
|
---|
176 |
|
---|
177 | # normalise clause breaks (mostly punctuation symbols) to commas
|
---|
178 | s/[^A-Za-z0-9 \n]+/ , /g;
|
---|
179 |
|
---|
180 | # Remove repeated commas, and replace with newline
|
---|
181 | s/\s*,[, ]+/\n/g;
|
---|
182 |
|
---|
183 | # remove extra whitespace
|
---|
184 | s/ +/ /sg;
|
---|
185 | s/^\s+//mg;
|
---|
186 | s/\s*$/\n/mg;
|
---|
187 |
|
---|
188 | # remove lines that contain one word or less
|
---|
189 | s/^\w*$//mg;
|
---|
190 | s/^\s*$//mg;
|
---|
191 | tr/\n//s;
|
---|
192 |
|
---|
193 | return $_;
|
---|
194 | }
|
---|
195 |
|
---|
196 |
|
---|
197 |
|
---|
198 | 1;
|
---|
199 |
|
---|