1 |
#!/usr/bin/perl -w |
#!/usr/bin/perl -w |
2 |
# -*- Mode: Perl -*- |
# -*- Mode: Perl -*- |
3 |
# $Basename$ |
# $Basename$ |
4 |
# $Revision: 1.12 $ |
# $Revision: 1.14 $ |
5 |
# Author : Ulrich Pfeifer |
# Author : Ulrich Pfeifer |
6 |
# Created On : Mon Dec 31 13:57:11 2001 |
# Created On : Mon Dec 31 13:57:11 2001 |
7 |
# Last Modified By: Ulrich Pfeifer |
# Last Modified By: Ulrich Pfeifer |
48 |
'database=s', |
'database=s', |
49 |
'dir=s', |
'dir=s', |
50 |
'table=s', |
'table=s', |
51 |
|
'verbose!', |
52 |
|
'debug!', |
53 |
) || die "Usage: ...\n"; |
) || die "Usage: ...\n"; |
54 |
|
|
55 |
my @localtime = localtime; |
my @localtime = localtime; |
62 |
|
|
63 |
my $layout = new WAIT::Parse::Ora; |
my $layout = new WAIT::Parse::Ora; |
64 |
|
|
65 |
my $stem = ['OR_tr_20020124', 'OR_lc_20020125', 'split2', 'stop', 'Stem']; |
# my $stem = ['OR_tr_20020124', 'OR_lc_20020125', 'split2', 'stop', 'Stem']; |
66 |
# my $text = ['OR_tr_20020124', 'split2', 'OR_minus_20020311', 'OR_lc_20020125', 'split2', 'stop']; |
# my $text = ['OR_tr_20020124', 'split2', 'OR_minus_20020311', 'OR_lc_20020125', 'split2', 'stop']; |
67 |
my $text = ['OR_tr_20020124', 'split', 'OR_minus_20020311', 'OR_lc_20020125']; |
my $text = ['OR_tr_20020124', 'split', 'OR_minus_20020311', 'OR_lc_20020125']; |
68 |
my $wplus = ['split', 'OR_lc_20020125', 'OR_mixedonly_20020221']; |
my $wplus = ['OR_split_20020401', 'OR_lc_20020125', 'OR_mixedonly_20020221']; |
69 |
my $sound = ['OR_tr_20020124', 'OR_lc_20020125', 'split2', 'Soundex']; |
# my $sound = ['OR_tr_20020124', 'OR_lc_20020125', 'split2', 'Soundex']; |
70 |
my $trigr = ['OR_lc_20020125', 'OR_trigrams_20020125']; |
my $trigr = ['OR_lc_20020125', 'OR_trigrams_20020125']; |
71 |
|
|
72 |
# split6 is better than split13 or split10: it allows them to enter |
# split6 is better than split13 or split10: it allows them to enter |
73 |
# shorter sequences when searching. |
# shorter sequences when searching. |
74 |
my $isbn = ['split6', 'OR_isbn_20020127']; |
my $isbn = ['split6', 'OR_isbn_20020127']; |
93 |
'abstract' => $text, |
'abstract' => $text, |
94 |
'abstract' => $wplus, |
'abstract' => $wplus, |
95 |
'author' => $text, |
'author' => $text, |
96 |
|
'chapter' => $text, |
97 |
|
'chapter' => $wplus, |
98 |
'colophon' => $text, |
'colophon' => $text, |
99 |
'colophon' => $wplus, |
'colophon' => $wplus, |
100 |
'desc' => $text, |
'desc' => $text, |
119 |
my ($did, $value); |
my ($did, $value); |
120 |
binmode STDOUT, ":utf8"; |
binmode STDOUT, ":utf8"; |
121 |
my $ALL; |
my $ALL; |
122 |
my $traceALL = 0; # expensive |
my $traceALL = 0; |
123 |
|
$traceALL = 1 if $OPT{debug}; # expensive |
124 |
my $done = 0; |
my $done = 0; |
125 |
my $todo = keys %D; |
my $todo = keys %D; |
126 |
my $lasttimeround = my $starttime = time; |
my $lasttimeround = my $starttime = time; |
127 |
|
|
128 |
DOC: while (($did, $value) = each %D) { |
DOC: while (($did, $value) = each %D) { |
129 |
# next unless $did eq "jscook"; |
my $superdebug = 0; |
130 |
|
if ($superdebug && $OPT{debug}) { |
131 |
|
next unless $did =~ /perltb/; |
132 |
|
} |
133 |
|
printf "%15s...(%d/%d)\n", $did, $done, $todo if $OPT{verbose}; |
134 |
my $record = $layout->split($value); |
my $record = $layout->split($value); |
135 |
my $headline = $record->{title}; |
my $headline = $record->{title}; |
136 |
$headline =~ s/\s+/ /sg; |
$headline =~ s/\s+/ /sg; |
137 |
# printf "%15s %s\n", $record->{isbn}, substr($headline,0,60); |
# printf "%15s %s\n", $record->{isbn}, substr($headline,0,60); |
138 |
printf "%15s %s...\n", $did, substr($headline,0,60); |
printf "%15s %s...\n", $did, substr($headline,0,60) if $OPT{verbose}; |
139 |
|
if ($superdebug && $OPT{debug}) { |
140 |
|
# $record = { chapter => $record->{chapter}}; |
141 |
|
} |
142 |
|
$done++; |
143 |
|
next if oreilly_de_catalog::config::DONT_INDEX_JUST_PARSE() ; |
144 |
$tb->insert('docid' => $did, |
$tb->insert('docid' => $did, |
145 |
headline => $headline, |
headline => $headline, |
146 |
%{$record}); |
%{$record}); |
|
$done++; |
|
147 |
my $spenttime = time - $starttime; |
my $spenttime = time - $starttime; |
148 |
my $averagetime = $spenttime/$done; |
my $averagetime = $spenttime/$done; |
149 |
my $left = $todo-$done; |
my $left = $todo-$done; |
154 |
$spenttime, |
$spenttime, |
155 |
$averagetime, |
$averagetime, |
156 |
$left*$averagetime, |
$left*$averagetime, |
157 |
); |
) if $OPT{verbose}; |
158 |
$lasttimeround = time; |
$lasttimeround = time; |
159 |
if ($traceALL) { # costs a lot when reaching the 100th file or so |
if ($traceALL) { # costs a lot when reaching the 100th file or so |
160 |
$ALL->{$did} = $record; |
$ALL->{$did} = $record; |
213 |
unlink $sltmp; # may fail |
unlink $sltmp; # may fail |
214 |
symlink $dir, $sltmp or die "Couldn't symlink $dir, $sltmp: $!"; |
symlink $dir, $sltmp or die "Couldn't symlink $dir, $sltmp: $!"; |
215 |
rename $sltmp, $slwant or die "Couldn't rename $sltmp, $slwant: $!"; |
rename $sltmp, $slwant or die "Couldn't rename $sltmp, $slwant: $!"; |
216 |
warn "$slwant now points to $dir"; |
print "$slwant now points to $dir\n" if $OPT{verbose}; |
217 |
system("chmod 777 $slwant/*/read")==0 or die; |
system("chmod 777 $slwant/*/read")==0 or die; |
218 |
|
|
219 |
opendir DIR, "." or die "Could not opendir .: $!"; |
opendir DIR, "." or die "Could not opendir .: $!"; |