| 1 |
#!/usr/bin/perl -w |
| 2 |
######################### -*- Mode: Perl -*- ######################### |
| 3 |
## |
| 4 |
## $Basename: cpanwait $ |
| 5 |
## $Revision: 1.7 $ |
| 6 |
## |
| 7 |
## Author : Ulrich Pfeifer |
| 8 |
## Created On : Sat Jan 4 18:09:28 1997 |
| 9 |
## |
| 10 |
## Last Modified By : Ulrich Pfeifer |
| 11 |
## Last Modified On : Sun Nov 22 18:44:36 1998 |
| 12 |
## |
| 13 |
## Copyright (c) 1996-1997, Ulrich Pfeifer |
| 14 |
## |
| 15 |
## |
| 16 |
###################################################################### |
| 17 |
|
| 18 |
use strict; |
| 19 |
|
| 20 |
use File::Path; |
| 21 |
use DB_File; |
| 22 |
use Getopt::Long; |
| 23 |
use File::Find; |
| 24 |
use File::Basename; |
| 25 |
use IO::File; |
| 26 |
|
| 27 |
use lib '/data/wait/lib'; |
| 28 |
|
| 29 |
require WAIT::Config; |
| 30 |
require WAIT::Database; |
| 31 |
require WAIT::Parse::Pod; |
| 32 |
require WAIT::Document::Tar; |
| 33 |
|
| 34 |
sub fname($); |
| 35 |
|
| 36 |
# maximum number of archives to index (set to -1 for unlimited) |
| 37 |
my $max = -1; |
| 38 |
|
| 39 |
my %OPT = (database => 'DB', |
| 40 |
dir => $WAIT::Config->{WAIT_home} || '/tmp', |
| 41 |
table => 'cpan', |
| 42 |
clean => 0, |
| 43 |
remove => [], |
| 44 |
force => 0, |
| 45 |
# cpan => '/usr/src/perl/CPAN/sources', |
| 46 |
cpan => '/rest/cpan/CPAN/', |
| 47 |
trust_mtime => 1, |
| 48 |
match => 'authors/id/', |
| 49 |
test => 0, |
| 50 |
# cpan => 'ftp://ftp.rz.ruhr-uni-bochum.de/pub/CPAN', |
| 51 |
# cpan => 'ftp://ftp.uni-hamburg.de:/pub/soft/lang/perl/CPAN', |
| 52 |
keep => '/tmp/CPAN/', |
| 53 |
); |
| 54 |
|
| 55 |
GetOptions(\%OPT, |
| 56 |
'database=s', |
| 57 |
'dir=s', |
| 58 |
'cpan=s', |
| 59 |
'table=s', |
| 60 |
'keep=s', |
| 61 |
'match=s', |
| 62 |
'clean!', |
| 63 |
'test=i', # test level 0: normal |
| 64 |
# 1: don't change db |
| 65 |
# 2: don't look at archives even |
| 66 |
|
| 67 |
'remove=s@', |
| 68 |
'force!', # force indexing even if seen |
| 69 |
'trust_mtime!', # use mtime instead of version number |
| 70 |
'max=i', |
| 71 |
'reorg!', |
| 72 |
) || die "Usage: ...\n"; |
| 73 |
|
| 74 |
$max ||= $OPT{max}; |
| 75 |
|
| 76 |
clean_database( |
| 77 |
database => $OPT{database}, |
| 78 |
dir => $OPT{dir}, |
| 79 |
table => $OPT{table}, |
| 80 |
) if $OPT{clean}; |
| 81 |
|
| 82 |
my $db = WAIT::Database->open( |
| 83 |
name => $OPT{database}, |
| 84 |
'directory' => $OPT{dir}, |
| 85 |
) |
| 86 |
|| WAIT::Database->create( |
| 87 |
name => $OPT{database}, |
| 88 |
'directory' => $OPT{dir}, |
| 89 |
) |
| 90 |
or die "Could not open/create database '$OPT{dir}/$OPT{database}': $@"; |
| 91 |
|
| 92 |
my $layout= new WAIT::Parse::Pod; |
| 93 |
|
| 94 |
my $tb = $db->table(name => $OPT{table}) |
| 95 |
|| create_table(db => $db, table => $OPT{table}, layout => $layout); |
| 96 |
|
| 97 |
# Map e.g. '.../latest' to 'perl'. Used in wanted(). Effects version |
| 98 |
# considerations. Value *must* match common prefix. Aliasing should be |
| 99 |
# used if CPAN contains serveral distributions with different name but |
| 100 |
# same root directory. |
| 101 |
# We still have a problem if there are different root directories! |
| 102 |
|
| 103 |
my %ALIAS = (# tar name real (root) name |
| 104 |
'latest' => 'perl', |
| 105 |
'perl5db-kit' => 'DB', |
| 106 |
'SGI-FM' => 'FM', |
| 107 |
'net-ext' => 'Net', |
| 108 |
'VelocisSQL' => 'Velocis', |
| 109 |
'Net-ext' => 'Net', |
| 110 |
'Curses-DevKit' => 'Cdk', |
| 111 |
'PostgresPerl' => 'Postgres', |
| 112 |
'perlpdf' => 'PERLPDF', |
| 113 |
'Des-perl' => 'Des', |
| 114 |
'SGI-GL' => 'GL', |
| 115 |
'DBD-DB2' => 'DB2', |
| 116 |
); |
| 117 |
my %NEW_ALIAS; # found in this pass |
| 118 |
|
| 119 |
# Map module names to pathes. Generated by wanted() doing alisaing. |
| 120 |
my %ARCHIVE; |
| 121 |
|
| 122 |
# Map module names to latest version. Generated by wanted() |
| 123 |
my %VERSION; |
| 124 |
|
| 125 |
|
| 126 |
# Mapping for modules with common root not matching modules name that |
| 127 |
# are not aliased. This is just for prefix stripping and not strictly |
| 128 |
# necessary. Takes effect after version considerations. |
| 129 |
my %TR = (# tar name root to strip |
| 130 |
'Net_SSLeay.pm' => 'SSLeay/', |
| 131 |
'EventDrivenServer' => 'Server/', |
| 132 |
'bio_lib.pl.' => '', |
| 133 |
'AlarmCall' => 'Sys/', |
| 134 |
'Cdk-ext' => 'Cdk/', |
| 135 |
'Sx' => '\d.\d/', |
| 136 |
'DumpStack' => 'Devel/', |
| 137 |
'StatisticsDescriptive'=> 'Statistics/', |
| 138 |
'Term-Gnuplot' => 'Gnuplot/', |
| 139 |
'iodbc_ext' => 'iodbc-ext-\d.\d/', |
| 140 |
'UNIVERSAL' => '', |
| 141 |
'Term-Query' => 'Query/', |
| 142 |
'SelfStubber' => 'Devel/', |
| 143 |
'CallerItem' => 'Devel/', |
| 144 |
); |
| 145 |
|
| 146 |
my $DIR = $tb->dir; |
| 147 |
my $DATA = $tb->dir . "/data"; |
| 148 |
my $LWP; |
| 149 |
|
| 150 |
|
| 151 |
if (@{$OPT{remove}}) { |
| 152 |
my $pod; |
| 153 |
for $pod (@{$OPT{remove}}) { |
| 154 |
unless (-e $pod) { |
| 155 |
$pod = "$DIR/$pod"; |
| 156 |
} |
| 157 |
index_pod(file => $pod, remove => 1) if -f $pod; |
| 158 |
unlink $pod or warn "Could not unlink '$pod': $!\n"; |
| 159 |
#$tb->sync; |
| 160 |
} |
| 161 |
$tb->close; |
| 162 |
$db->close; |
| 163 |
exit; |
| 164 |
} |
| 165 |
|
| 166 |
# Now get the beef |
| 167 |
if ($OPT{cpan} =~ /^(http|ftp):/) { |
| 168 |
$LWP = 1; |
| 169 |
require LWP::Simple; |
| 170 |
LWP::Simple->import(); |
| 171 |
|
| 172 |
mkpath($DATA,1,0755) or |
| 173 |
die "Could not generate '$DATA/': $!" |
| 174 |
unless -d $DATA; |
| 175 |
|
| 176 |
if (! -f "$DATA/find-ls.gz" or -M "$DATA/find-ls.gz" > 0.5) { |
| 177 |
my $status = mirror("$OPT{cpan}/indices/find-ls.gz", "$DATA/find-ls.gz"); |
| 178 |
if ($status != &RC_OK and $status != &RC_NOT_MODIFIED) { |
| 179 |
# we could use Net:FTP here ... |
| 180 |
die "Was unable to mirror '$OPT{cpan}/indices/find-ls.gz'\n"; |
| 181 |
} |
| 182 |
} |
| 183 |
my $fh = new IO::File "gzip -cd $DATA/find-ls.gz |"; |
| 184 |
die "Could not open 'gzip -cd $DATA/find-ls.gz': !$\n" unless $fh; |
| 185 |
|
| 186 |
my $line; |
| 187 |
while (defined ($line = <$fh>)) { |
| 188 |
chomp($line); |
| 189 |
my ($mon, $mday, $time, $file, $is_link) = (split ' ', $line)[7..11]; |
| 190 |
|
| 191 |
next if defined $is_link; |
| 192 |
my $mtime = mtime($mon, $mday, $time); |
| 193 |
|
| 194 |
$file =~ s:^\./::; |
| 195 |
($_) = fileparse($file); |
| 196 |
$File::Find::name = $file; |
| 197 |
wanted($mtime); |
| 198 |
} |
| 199 |
} else { |
| 200 |
find(sub {&wanted((stat($_))[9])}, $OPT{cpan}); |
| 201 |
} |
| 202 |
|
| 203 |
ARCHIVE: |
| 204 |
for my $tar (sort keys %ARCHIVE) { |
| 205 |
next if $OPT{match} and $ARCHIVE{$tar} !~ /$OPT{match}/o; |
| 206 |
my $base = (split /\//, $ARCHIVE{$tar})[-1]; |
| 207 |
my $parent; |
| 208 |
|
| 209 |
# logging |
| 210 |
if ($OPT{trust_mtime}) { |
| 211 |
printf "%-20s %10s %s\t", $tar, |
| 212 |
substr(scalar(localtime($VERSION{$tar})),0,10), $base; |
| 213 |
} else { |
| 214 |
printf "%-20s %10.5f %s\t", $tar, $VERSION{$tar}, $base; |
| 215 |
} |
| 216 |
|
| 217 |
# Remember the archive |
| 218 |
# We should have an extra table for the tar file data ... |
| 219 |
if (!$OPT{force} and $tb->have(docid => $base)) { |
| 220 |
print "skipping\n"; |
| 221 |
next ARCHIVE; |
| 222 |
} else { |
| 223 |
$parent = $tb->insert(docid => $base, |
| 224 |
headline => $ARCHIVE{$tar}) unless $OPT{test}; |
| 225 |
print "indexing\n"; |
| 226 |
} |
| 227 |
|
| 228 |
next ARCHIVE if $OPT{test} > 1; |
| 229 |
|
| 230 |
my $TAR = myget($tar); |
| 231 |
|
| 232 |
next ARCHIVE unless $TAR; # not able to fetch it |
| 233 |
|
| 234 |
my %tar; |
| 235 |
tie (%tar, |
| 236 |
'WAIT::Document::Tar', |
| 237 |
sub { $_[0] =~ /\.(pm|pod|PL)$/i or $_[0] =~ /readme/i}, |
| 238 |
#sub { $_[0] !~ m:/$: }, |
| 239 |
$TAR) |
| 240 |
or warn "Could not tie '$TAR'\n"; |
| 241 |
|
| 242 |
my $sloppy; |
| 243 |
my ($key, $val); |
| 244 |
|
| 245 |
FILE: |
| 246 |
while (($key, $val) = each %tar) { |
| 247 |
my $file = fname($key); |
| 248 |
|
| 249 |
# don't index directories |
| 250 |
next if $file =~ /\/$/; |
| 251 |
|
| 252 |
# is it a POD file? |
| 253 |
next FILE unless $file =~ /readme/i or $val =~ /\n=head/; |
| 254 |
|
| 255 |
# remove directory prefix |
| 256 |
unless ($sloppy # no common root |
| 257 |
or $file =~ s:^\Q$tar\E[^/]*/:: # common root, maybe alias |
| 258 |
or ($TR{$tar} # common root, not aliased |
| 259 |
and $file =~ s:^\Q$TR{$tar}\E::) |
| 260 |
) { |
| 261 |
# try to determine an alias |
| 262 |
warn "Bad directory prefix: '$file'\n"; |
| 263 |
my ($prefix) = split /\//, $file; |
| 264 |
|
| 265 |
while ($key = (tied %tar)->NEXTKEY) { |
| 266 |
my $file = fname($key); |
| 267 |
|
| 268 |
next if $file =~ /\/$/; |
| 269 |
unless ($file =~ m:^$prefix/: or $file eq $prefix) { |
| 270 |
warn "Archive contains different prefixes: $prefix,$file\n"; |
| 271 |
$prefix = ''; |
| 272 |
last; |
| 273 |
} |
| 274 |
} |
| 275 |
if ($prefix) { |
| 276 |
print "Please alias '$tar' to '$prefix' next time!\n"; |
| 277 |
print "See alias table later.\n"; |
| 278 |
$NEW_ALIAS{$tar} = $prefix; |
| 279 |
$tb->delete_by_key($parent); |
| 280 |
next ARCHIVE; |
| 281 |
} else { |
| 282 |
print "Assuming that tar file name $tar is a valid prefix\n"; |
| 283 |
$sloppy = 1; |
| 284 |
|
| 285 |
# We may reset too much here! But that this is not exact |
| 286 |
# science anyway. Maybe we should ignore using 'next ARCHIVE'. |
| 287 |
|
| 288 |
$key = (tied %tar)->FIRSTKEY; |
| 289 |
redo FILE; |
| 290 |
} |
| 291 |
} |
| 292 |
|
| 293 |
# remove /lib prefix |
| 294 |
$file =~ s:^lib/::; |
| 295 |
|
| 296 |
# generate new path |
| 297 |
my $path = "$DATA/$tar/$file"; |
| 298 |
|
| 299 |
my ($sbase, $sdir) = fileparse($path); |
| 300 |
my $fh; |
| 301 |
|
| 302 |
unless ($OPT{test}) { |
| 303 |
if (-f $path) { |
| 304 |
index_pod(file => $path, remove => 1); |
| 305 |
unlink $path or warn "Could not unlink '$path' $!\n"; |
| 306 |
} elsif (!-d $sdir) { |
| 307 |
mkpath($sdir,1,0755) or die "Could not mkpath($sdir): $!\n"; |
| 308 |
} |
| 309 |
$fh = new IO::File "> $path"; |
| 310 |
die "Could not write '$path': $!\n" unless $fh; |
| 311 |
} |
| 312 |
|
| 313 |
if ($file =~ /readme|install/i) { # make READMEs verbatim pods |
| 314 |
$val =~ s/\n/\n /g; |
| 315 |
$val = "=head1 NAME\n\n$tar $file\n\n=head1 DESCRIPTION\n\n $val" |
| 316 |
unless $val =~ /^=head/m; |
| 317 |
} else { # remove non-pod stuff |
| 318 |
my $nval = $val; $val = ''; |
| 319 |
my $cutting = 1; |
| 320 |
|
| 321 |
for (split /\n/, $nval) { |
| 322 |
if (/^=cut|!NO!SUBS!/) { |
| 323 |
$cutting = 1; |
| 324 |
} elsif ($cutting and /^=head/) { |
| 325 |
$cutting = 0; |
| 326 |
} |
| 327 |
unless ($cutting) { |
| 328 |
$val .= $_ . "\n"; |
| 329 |
} |
| 330 |
} |
| 331 |
} |
| 332 |
unless ($OPT{test}) { |
| 333 |
$fh->print($val); |
| 334 |
index_pod(file => $path, parent => $parent, |
| 335 |
text => $val, source => $ARCHIVE{$tar}); |
| 336 |
} |
| 337 |
} |
| 338 |
|
| 339 |
if ($LWP and !$OPT{keep}) { |
| 340 |
unlink $TAR or warn |
| 341 |
"Could not unlink '$TAR': $!\n"; |
| 342 |
} |
| 343 |
} |
| 344 |
|
| 345 |
if (%NEW_ALIAS) { |
| 346 |
print "\%ALIAS = (\n"; |
| 347 |
for (keys %NEW_ALIAS) { |
| 348 |
print "\t'$_'\t=> '$NEW_ALIAS{$_}',\n"; |
| 349 |
} |
| 350 |
print "\t);\n"; |
| 351 |
} |
| 352 |
|
| 353 |
if ($OPT{reorg}) { |
| 354 |
my $now = time; |
| 355 |
warn "Starting reorg\n"; |
| 356 |
$tb->set(top=>1); |
| 357 |
warn sprintf "Finished reorg %d seconds\n", time - $now; |
| 358 |
} |
| 359 |
|
| 360 |
# we are done |
| 361 |
$db->close(); |
| 362 |
exit; |
| 363 |
|
| 364 |
sub fname ($) { |
| 365 |
my $key = shift; |
| 366 |
my ($ntar, $file) = split $;, $key; |
| 367 |
|
| 368 |
# remove leading './' - shudder |
| 369 |
$file =~ s/^\.\///; |
| 370 |
|
| 371 |
return($file); |
| 372 |
} |
| 373 |
|
| 374 |
sub myget { |
| 375 |
my $tar = shift; |
| 376 |
my $TAR; |
| 377 |
|
| 378 |
if ($LWP) { # fetch the archive |
| 379 |
if ($OPT{keep}) { |
| 380 |
$TAR = "$OPT{keep}/$ARCHIVE{$tar}"; |
| 381 |
print "Keeping in '$TAR'\n" unless -e $TAR; |
| 382 |
my ($base, $path) = fileparse($TAR); |
| 383 |
unless (-d $path) { |
| 384 |
mkpath($path,1,0755) or |
| 385 |
die "Could not mkpath($path)\n"; |
| 386 |
} |
| 387 |
} else { |
| 388 |
$TAR = "/tmp/$tar.tar.gz"; |
| 389 |
} |
| 390 |
unless (-e $TAR) { # lwp mirror seems to fetch ftp: in any case? |
| 391 |
print "Fetching $OPT{cpan}/$ARCHIVE{$tar}\n"; |
| 392 |
my $status = mirror("$OPT{cpan}/$ARCHIVE{$tar}", $TAR); |
| 393 |
if ($status != &RC_OK and $status != &RC_NOT_MODIFIED) { |
| 394 |
warn "Was unable to mirror '$ARCHIVE{$tar}, skipping'\n"; |
| 395 |
return; |
| 396 |
} |
| 397 |
} |
| 398 |
} else { |
| 399 |
$TAR = $ARCHIVE{$tar}; |
| 400 |
} |
| 401 |
$TAR; |
| 402 |
} |
| 403 |
|
| 404 |
sub index_pod { |
| 405 |
my %parm = @_; |
| 406 |
my $did = $parm{file}; |
| 407 |
my $rel_did = $did; |
| 408 |
my $abs_did = $did; |
| 409 |
|
| 410 |
if ($rel_did =~ s:$DIR/::) { |
| 411 |
$abs_did = "$DIR/$rel_did"; |
| 412 |
} |
| 413 |
|
| 414 |
undef $did; |
| 415 |
|
| 416 |
# check for both variants |
| 417 |
if ($tb->have('docid' => $rel_did)) { |
| 418 |
$did = $rel_did; |
| 419 |
} elsif ($tb->have('docid' => $abs_did)) { |
| 420 |
$did = $abs_did; |
| 421 |
} |
| 422 |
if ($did) { # have it version |
| 423 |
if (!$parm{remove}) { |
| 424 |
warn "duplicate: $did\n"; |
| 425 |
return; |
| 426 |
} |
| 427 |
} else { # not seen yet |
| 428 |
$did = $rel_did; |
| 429 |
if ($parm{remove}) { |
| 430 |
print "missing: $did\n"; |
| 431 |
return; |
| 432 |
} |
| 433 |
} |
| 434 |
|
| 435 |
$parm{'text'} ||= WAIT::Document::Find->FETCH($abs_did); |
| 436 |
|
| 437 |
unless (defined $parm{'text'}) { |
| 438 |
print "unavailable: $did\n"; |
| 439 |
return; |
| 440 |
} |
| 441 |
|
| 442 |
my $record = $layout->split($parm{'text'}); |
| 443 |
|
| 444 |
if (! $record) { |
| 445 |
print "empty pod: $did\n"; |
| 446 |
return; |
| 447 |
} |
| 448 |
|
| 449 |
$record->{size} = length($parm{'text'}); |
| 450 |
my $headline = $record->{name} || $did; |
| 451 |
|
| 452 |
$headline =~ s/^$DATA//o; # $did |
| 453 |
$headline =~ s/\s+/ /g; $headline =~ s/^\s+//; |
| 454 |
|
| 455 |
printf "%s %s\n", ($parm{remove})?'-':'+', substr($headline,0,70); |
| 456 |
if ($parm{remove}) { |
| 457 |
$tb->delete('docid' => $did, |
| 458 |
headline => $headline, |
| 459 |
%{$record}); |
| 460 |
} else { |
| 461 |
$tb->insert('docid' => $did, |
| 462 |
headline => $headline, |
| 463 |
source => $parm{source}, |
| 464 |
parent => $parm{parent}, |
| 465 |
%{$record}); |
| 466 |
} |
| 467 |
} |
| 468 |
|
| 469 |
# This *must* remove the version in *any* case. It should compute a |
| 470 |
# resonable version number - but usually mtimes should be used. |
| 471 |
sub version { |
| 472 |
local ($_) = @_; |
| 473 |
|
| 474 |
# remove alpha/beta postfix |
| 475 |
s/([-_\d])(a|b|alpha|beta|src)$/$1/; |
| 476 |
|
| 477 |
# jperl1.3@4.019.tar.gz |
| 478 |
s/@\d.\d+//; |
| 479 |
|
| 480 |
# oraperl-v2.4-gk.tar.gz |
| 481 |
s/-v(\d)/$1/; |
| 482 |
|
| 483 |
# lettered versions - shudder |
| 484 |
s/([-_\d\.])([a-z])([\d\._])/sprintf "$1%02d$3", ord(lc $2) - ord('a') /ei; |
| 485 |
s/([-_\d\.])([a-z])$/sprintf "$1%02d", ord(lc $2) - ord('a') /ei; |
| 486 |
|
| 487 |
# thanks libwww-5b12 ;-) |
| 488 |
s/(\d+)b/($1-1).'.'/e; |
| 489 |
s/(\d+)a/($1-2).'.'/e; |
| 490 |
|
| 491 |
# replace '-pre' by '0.' |
| 492 |
s/-pre([\.\d])/-0.$1/; |
| 493 |
s/\.\././g; |
| 494 |
s/(\d)_(\d)/$1$2/g; |
| 495 |
|
| 496 |
# chop '[-.]' and thelike |
| 497 |
s/\W$//; |
| 498 |
|
| 499 |
# ram's versions Storable-0.4@p |
| 500 |
s/\@/./; |
| 501 |
|
| 502 |
if (s/[-_]?(\d+)\.(0\d+)\.(\d+)$//) { |
| 503 |
return($_, $1 + "0.$2" + $3 / 1000000); |
| 504 |
} elsif (s/[-_]?(\d+)\.(\d+)\.(\d+)$//) { |
| 505 |
return($_, $1 + $2/1000 + $3 / 1000000); |
| 506 |
} elsif (s/[-_]?(\d+\.[\d_]+)$//) { |
| 507 |
return($_, $1); |
| 508 |
} elsif (s/[-_]?([\d_]+)$//) { |
| 509 |
return($_, $1); |
| 510 |
} elsif (s/-(\d+.\d+)-/-/) { # perl-4.019-ref-guide |
| 511 |
return($_, $1); |
| 512 |
} else { |
| 513 |
if ($_ =~ /\d/) { # smells like an unknown scheme |
| 514 |
warn "Odd version Numbering: '$File::Find::name'\n"; |
| 515 |
return($_, undef); |
| 516 |
} else { # assume version 0 |
| 517 |
warn "No version Numbering: '$File::Find::name'\n"; |
| 518 |
return($_, 0); |
| 519 |
} |
| 520 |
|
| 521 |
} |
| 522 |
} |
| 523 |
|
| 524 |
sub wanted { |
| 525 |
my $mtime = shift; # called by parse_file_ls(); |
| 526 |
|
| 527 |
return unless /^(.*)\.tar(\.gz|\.Z)$/; |
| 528 |
|
| 529 |
return if (! $max); |
| 530 |
$max--; |
| 531 |
|
| 532 |
my ($archive, $version) = version($1); |
| 533 |
|
| 534 |
unless (defined $version) { |
| 535 |
warn "Skipping $1\n"; |
| 536 |
return; |
| 537 |
} |
| 538 |
|
| 539 |
# Check for file alias |
| 540 |
$archive = $ALIAS{$archive} if $ALIAS{$archive}; |
| 541 |
|
| 542 |
# Check for path alias. |
| 543 |
if ($File::Find::name =~ m(/CPAN/(?:source/)?(.*\Q$archive\E))) { |
| 544 |
if ($ALIAS{$1}) { |
| 545 |
$archive = $ALIAS{$1}; |
| 546 |
} |
| 547 |
} |
| 548 |
|
| 549 |
if ($OPT{trust_mtime}) { |
| 550 |
$version = $mtime; |
| 551 |
} else { |
| 552 |
$version =~ s/(\d)_/$1/; |
| 553 |
$version ||= $mtime; # mtime |
| 554 |
} |
| 555 |
|
| 556 |
if (!exists $ARCHIVE{$archive} |
| 557 |
or $VERSION{$archive} < $version) { |
| 558 |
$ARCHIVE{$archive} = $File::Find::name; |
| 559 |
$VERSION{$archive} = $version; |
| 560 |
} |
| 561 |
} |
| 562 |
|
| 563 |
sub clean_database { |
| 564 |
my %parm = @_; |
| 565 |
|
| 566 |
my $db = WAIT::Database->open( |
| 567 |
name => $parm{database}, |
| 568 |
'directory' => $parm{dir}, |
| 569 |
) |
| 570 |
or die "Could not open database '$parm{dir}/$parm{database}': $@"; |
| 571 |
my $tbl = $db->table(name => $parm{table}); |
| 572 |
if ($tbl) { |
| 573 |
$tbl->drop or |
| 574 |
die "Could not open table '$parm{tabel}': $@"; |
| 575 |
} |
| 576 |
|
| 577 |
$db->close; |
| 578 |
} |
| 579 |
|
| 580 |
sub create_table { |
| 581 |
my %parm = @_; |
| 582 |
|
| 583 |
my $access = bless {}, 'WAIT::Document::Find'; |
| 584 |
|
| 585 |
my $stem = [{ |
| 586 |
'prefix' => ['isotr', 'isolc'], |
| 587 |
'intervall' => ['isotr', 'isolc'], |
| 588 |
}, 'isotr', 'isolc', 'split2', 'stop', 'Stem']; |
| 589 |
my $text = [{ |
| 590 |
'prefix' => ['isotr', 'isolc'], |
| 591 |
'intervall' => ['isotr', 'isolc'], |
| 592 |
}, |
| 593 |
'isotr', 'isolc', 'split2', 'stop']; |
| 594 |
my $sound = ['isotr', 'isolc', 'split2', 'Soundex'],; |
| 595 |
|
| 596 |
my $tb = |
| 597 |
$parm{db}->create_table |
| 598 |
(name => $parm{table}, |
| 599 |
attr => ['docid', 'headline', 'source', 'size', 'parent'], |
| 600 |
keyset => [['docid']], |
| 601 |
layout => $parm{layout}, |
| 602 |
access => $access, |
| 603 |
invindex => |
| 604 |
[ |
| 605 |
'name' => $stem, |
| 606 |
'synopsis' => $stem, |
| 607 |
'bugs' => $stem, |
| 608 |
'description' => $stem, |
| 609 |
'text' => $stem, |
| 610 |
'environment' => $text, |
| 611 |
'example' => $text, 'example' => $stem, |
| 612 |
'author' => $sound, 'author' => $stem, |
| 613 |
] |
| 614 |
); |
| 615 |
die "Could not create table '$parm{table}'" unless $tb; |
| 616 |
$tb; |
| 617 |
} |
| 618 |
|
| 619 |
my %MON; |
| 620 |
my $YEAR; |
| 621 |
|
| 622 |
BEGIN { |
| 623 |
my $i = 1; |
| 624 |
for (qw(Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec)) { |
| 625 |
$MON{$_} = $i++; |
| 626 |
} |
| 627 |
$YEAR = (localtime(time))[5]; |
| 628 |
} |
| 629 |
|
| 630 |
# We could/should use Date::GetDate here |
| 631 |
use Time::Local; |
| 632 |
sub mtime { |
| 633 |
my ($mon, $mday, $time) = @_; |
| 634 |
my ($hour, $min, $year, $monn) = (0,0); |
| 635 |
|
| 636 |
if ($time =~ /(\d+):(\d+)/) { |
| 637 |
($hour, $min) = ($1, $2); |
| 638 |
$year = $YEAR; |
| 639 |
} else { |
| 640 |
$year = $time; |
| 641 |
} |
| 642 |
$monn = $MON{$mon} || $MON{ucfirst lc $mon} || warn "Unknown month: '$mon'"; |
| 643 |
my $guess = timelocal(0,$min,$hour,$mday,$monn-1,$year); |
| 644 |
if ($guess > time) { |
| 645 |
$guess = timelocal(0,$min,$hour,$mday,$monn-1,$year-1); |
| 646 |
} |
| 647 |
$guess; |
| 648 |
} |
| 649 |
|
| 650 |
|
| 651 |
__END__ |
| 652 |
## ################################################################### |
| 653 |
## pod |
| 654 |
## ################################################################### |
| 655 |
|
| 656 |
=head1 NAME |
| 657 |
|
| 658 |
cpanwait - generate an WAIT index for CPAN |
| 659 |
|
| 660 |
=head1 SYNOPSIS |
| 661 |
|
| 662 |
B<cpanwait> |
| 663 |
[B<-clean>] [B<-noclean>] |
| 664 |
[B<-cpan> I<url or directory>] |
| 665 |
[B<-database> I<dbname>] |
| 666 |
[B<-dir> I<directory>] |
| 667 |
[B<-force>] [B<-noforce>] |
| 668 |
[B<-keep> I<directory>] |
| 669 |
[B<-match> I<regexp>] |
| 670 |
[B<-table> I<table name>] |
| 671 |
[B<-test> I<level>] |
| 672 |
[B<-trust_mtime>] [B<-notrust_mtime>] |
| 673 |
|
| 674 |
=head1 DESCRIPTION |
| 675 |
|
| 676 |
TBS |
| 677 |
|
| 678 |
=head1 OPTIONS |
| 679 |
|
| 680 |
=over 5 |
| 681 |
|
| 682 |
=item B<-clean> / B<-noclean> |
| 683 |
|
| 684 |
Clean the table befor indexing. Default is B<off>. |
| 685 |
|
| 686 |
=item B<-cpan> I<url or directory> |
| 687 |
|
| 688 |
Default directory or URL for indexing. If an URL is given, there |
| 689 |
currently must be a file F<indices/find-ls.gz> relative to it which |
| 690 |
contains the output of C<find . -ls | gzip>. |
| 691 |
Default is F<ftp://ftp.rz.ruhr-uni-bochum.de/pub/CPAN>. |
| 692 |
|
| 693 |
|
| 694 |
=item B<-database> I<dbname> |
| 695 |
|
| 696 |
Specify database name. Default is F<DB>. |
| 697 |
|
| 698 |
=item B<-dir> I<directory> |
| 699 |
|
| 700 |
Alternate directory were databases are located. Default is the |
| 701 |
directory specified during configuration of WAIT. |
| 702 |
|
| 703 |
=item B<-force> |
| 704 |
|
| 705 |
Force reindexing, even if B<cpan> thinks files are up to date. |
| 706 |
Default is B<off> |
| 707 |
|
| 708 |
=item B<-keep> I<directory> |
| 709 |
|
| 710 |
If fetching from a remote server, keep files in I<directory>. Default is |
| 711 |
F</app/unido-i06/src/share/lang/perl/96a/CPAN/sources>. |
| 712 |
|
| 713 |
=item B<-match> I<regexp> |
| 714 |
|
| 715 |
Limit to patches matching I<regexp>. Default is F<authors/id/>. |
| 716 |
|
| 717 |
=item B<-table> I<table name> |
| 718 |
|
| 719 |
Specify an alternate table name. Default is C<cpan>. |
| 720 |
|
| 721 |
=item B<-test> I<level> |
| 722 |
|
| 723 |
Set test level, were B<0> means normal operation, B<1> means, don't |
| 724 |
really index and B<2> means, don't even get archives and examine them. |
| 725 |
|
| 726 |
=item B<-trust_mtime> / B<-notrust_mtime> |
| 727 |
|
| 728 |
If B<on>, the files mtimes are used to decide, which version of an |
| 729 |
archive is the newest. If b<off>, the version extracted is used |
| 730 |
(beware, there are far more version numbering schemes than B<cpan> can |
| 731 |
parse). |
| 732 |
|
| 733 |
=head1 AUTHOR |
| 734 |
|
| 735 |
Ulrich Pfeifer E<lt>F<pfeifer@ls6.informatik.uni-dortumund.de>E<gt> |
Parent Directory
| 
Revision Log