| 1 | #!/usr/bin/perl -w
|
|---|
| 2 |
|
|---|
| 3 |
|
|---|
| 4 | # Both this script and its associated process_html.pl were written by
|
|---|
| 5 | # Marcel ?, while a student at Waikato University. Unfortunately he
|
|---|
| 6 | # was very new to perl at the time so the code is neither as clean nor
|
|---|
| 7 | # as fast as it could be (I've cleaned up a few of the more serious
|
|---|
| 8 | # bottlenecks -- it could do with alot more work though). The code
|
|---|
| 9 | # does work though, if a little slowly. It's not ready for primetime
|
|---|
| 10 | # however and is included in the Greenstone source tree mostly so that
|
|---|
| 11 | # I don't lose it. -- Stefan - 24 Jul 2001
|
|---|
| 12 |
|
|---|
| 13 |
|
|---|
| 14 | # This script will download an entire collection (and its associated pictures and files)
|
|---|
| 15 | # and store them in a temporary directory ($outputdir).
|
|---|
| 16 | # A second script (process_html.pl) can be then be used to 'rebuild' the collection and link all the
|
|---|
| 17 | # downloaded pages and pictures together into a usable static collection.
|
|---|
| 18 |
|
|---|
| 19 | # This script will generate a number of text files required by the second script
|
|---|
| 20 | # and for possible recovery, namely:
|
|---|
| 21 | # - links.txt - contains all the http links that were downloaded
|
|---|
| 22 | # - images.txt - contains a list of all the images that were downloaded
|
|---|
| 23 | # - image_dirs.txt - contains a list of all the image-prefixes that need to be wiped from the html files
|
|---|
| 24 | # (otherwise you won't get to see any pictures)
|
|---|
| 25 |
|
|---|
| 26 | # Both this script and the html processing script have a recovery feature built in, they can continue from wherever
|
|---|
| 27 | # they left off, but this only works if $outputdir and $finaldir are set to different values.
|
|---|
| 28 |
|
|---|
| 29 | # This is where all the downloaded html files end up, eg. 'temp_html/'
|
|---|
| 30 | my $outputdir = 'temp_html/';
|
|---|
| 31 | # This is where all the processed html files (and the pictures) will end up eg. 'my_static_collection/'
|
|---|
| 32 | my $finaldir = 'envl_collection/';
|
|---|
| 33 |
|
|---|
| 34 | # This is where we start our mirroring
|
|---|
| 35 | $address = 'http://nowhere.com/cgi-bin/library?a=p&p=about&c=demo&u=1';
|
|---|
| 36 |
|
|---|
| 37 |
|
|---|
| 38 | # whatever is specified in $option will be attached to the end of each html link before it is downloaded.
|
|---|
| 39 | # eg. "&u=1" to disable various features in the greenstone collections that are not needed with a static
|
|---|
| 40 | # collection.
|
|---|
| 41 | # another example : "&l=nl" to set the entire collection to dutch (NetherLands :)
|
|---|
| 42 | my $option = "&u=1";
|
|---|
| 43 |
|
|---|
| 44 | # Most OS have a limit on the maximum amount of files per directory (or folder).
|
|---|
| 45 | # A static collection can easily contain >3000 html files. Putting all those files
|
|---|
| 46 | # into one single directory is just asking for trouble. It's also very unwieldy ;^)
|
|---|
| 47 | # Hence...this value will set how much html files will be stored in one directory.
|
|---|
| 48 | # These directories themselves will be numbered,
|
|---|
| 49 | # so if $dir_entries = 500 then the directories will be "0/", "500/", "1000/", "1500/", "2000/", etc.
|
|---|
| 50 | my $dir_entries = 250;
|
|---|
| 51 |
|
|---|
| 52 | # Occasionally a page occurs which contains no data (because &cl is not set) This option fixes that.
|
|---|
| 53 | my $fix_empty_pages = "&cl=CL1";
|
|---|
| 54 |
|
|---|
| 55 | # These are the files that wget will download.
|
|---|
| 56 | # more can be added if necessary.
|
|---|
| 57 | my @graphic_formats = ('.gif','.jpg','.bmp','.png','.pdf','.mov','.mpeg','.jpeg','.rm');
|
|---|
| 58 |
|
|---|
| 59 | # ---------------------------------------[ System specific options ]----------------------------------------------------
|
|---|
| 60 |
|
|---|
| 61 | # The lynx variable specifies the command line for the lynx web browser
|
|---|
| 62 | # -- This is what I use under dos/win32
|
|---|
| 63 | # my $lynx = 'e:\lynx_w32\lynx -cfg=e:\lynx_w32\lynx.cfg';
|
|---|
| 64 |
|
|---|
| 65 | # -- This is what I use under linux
|
|---|
| 66 | my $lynx = 'lynx';
|
|---|
| 67 |
|
|---|
| 68 | # and the same for the wget utility
|
|---|
| 69 | my $wget = 'wget';
|
|---|
| 70 |
|
|---|
| 71 | # NB: There is one other linux specific command all the way at the end of this script, where I've used 'cp' to copy a file.
|
|---|
| 72 |
|
|---|
| 73 | # Another NB: When saving the dl-ed html files to disk, I've set lynx to dump the html-source to the standard output,
|
|---|
| 74 | # which I then simply redirect to a target file, BUT
|
|---|
| 75 | # this does not work under DOS/win32. Redirecting standard output in a script causes it to be displayed on
|
|---|
| 76 | # the screen instead. The easiest way to get around this I found was by doing the actual redirection in a simple
|
|---|
| 77 | # batch file (say grab.bat), which contains the following line:
|
|---|
| 78 | # @e:\lynx_w32\lynx -cfg=e:\lynx_w32\lynx.cfg -dump -source "%1" > %2
|
|---|
| 79 | #
|
|---|
| 80 | # Then replace line nr 326 -> 'system ("$kommand > $target");' with 'system("grab.bat $address $target");'
|
|---|
| 81 | # Not a very elegant solution, but it works :)
|
|---|
| 82 |
|
|---|
| 83 | #------------------------------------------------------------------------------------------------------------------------
|
|---|
| 84 |
|
|---|
| 85 | my %image_list;
|
|---|
| 86 | my $image_pointer = 0;
|
|---|
| 87 |
|
|---|
| 88 | my %linkz_list;
|
|---|
| 89 | my %short_linkz_list;
|
|---|
| 90 | my $linkz_pointer = 0;
|
|---|
| 91 |
|
|---|
| 92 | my %image_dirs_list;
|
|---|
| 93 | my $image_dirs_pointer = 0;
|
|---|
| 94 |
|
|---|
| 95 | my $numberdir = 0;
|
|---|
| 96 |
|
|---|
| 97 | my $start_time = (times)[0];
|
|---|
| 98 |
|
|---|
| 99 | # check if directories exist and create them if necessary..
|
|---|
| 100 | if ((-e $outputdir)&&(-d $outputdir))
|
|---|
| 101 | {
|
|---|
| 102 | print " ** ",$outputdir," directory already exists..\n";
|
|---|
| 103 | }
|
|---|
| 104 | else
|
|---|
| 105 | {
|
|---|
| 106 | print " ** Creating ",$outputdir," directory..\n";
|
|---|
| 107 | mkdir($outputdir, 0777) or die " Cannot create output directory: $!\n";
|
|---|
| 108 | }
|
|---|
| 109 |
|
|---|
| 110 | if ((-e $finaldir)&&(-d $finaldir))
|
|---|
| 111 | {
|
|---|
| 112 | print " ** ",$finaldir," directory already exists..\n";
|
|---|
| 113 | }
|
|---|
| 114 | else
|
|---|
| 115 | {
|
|---|
| 116 | print " ** Creating ",$finaldir," directory..\n";
|
|---|
| 117 | mkdir($finaldir, 0777) or die " Cannot create final directory: $!\n";
|
|---|
| 118 | }
|
|---|
| 119 |
|
|---|
| 120 | #-----------------------------------------------------------------------------------------------
|
|---|
| 121 | # No need to start from scratch everytime, we can recover/continue from wherever we left off
|
|---|
| 122 | # simply by checking which html files have been created
|
|---|
| 123 | #-----------------------------------------------------------------------------------------------
|
|---|
| 124 |
|
|---|
| 125 | $linknumber = 0; # used to name/number the dl-ed html files
|
|---|
| 126 |
|
|---|
| 127 | my $failed = 0;
|
|---|
| 128 | while ($failed == 0)
|
|---|
| 129 | {
|
|---|
| 130 | if ($linknumber % $dir_entries == 0)
|
|---|
| 131 | {
|
|---|
| 132 | if (!((-e $outputdir.$linknumber)&&(-d $outputdir.$linknumber)))
|
|---|
| 133 | {
|
|---|
| 134 | $failed++;
|
|---|
| 135 | mkdir($outputdir.$linknumber, 0777) or print " ** Cannot create ",$outputdir.$linknumber, "!: $!\n";
|
|---|
| 136 | }
|
|---|
| 137 | $numberdir = $linknumber;
|
|---|
| 138 | }
|
|---|
| 139 |
|
|---|
| 140 | $check_file = $outputdir.$numberdir."/".$linknumber.".html";
|
|---|
| 141 | if ((-e $check_file)&&($failed == 0))
|
|---|
| 142 | {
|
|---|
| 143 | $linknumber++;
|
|---|
| 144 | }
|
|---|
| 145 | else
|
|---|
| 146 | {
|
|---|
| 147 | $failed++;
|
|---|
| 148 | # I'm subtracting 1 from the starting link,
|
|---|
| 149 | # just in case it only loaded half the page ;^)
|
|---|
| 150 | if($linknumber>0)
|
|---|
| 151 | {
|
|---|
| 152 | $linknumber--;
|
|---|
| 153 | }
|
|---|
| 154 | print " Will start downloading at number $linknumber \n";
|
|---|
| 155 | }
|
|---|
| 156 | }
|
|---|
| 157 |
|
|---|
| 158 | # if we're starting from scratch, then we might as well nuke the links file
|
|---|
| 159 | #if ($linknumber == 0)
|
|---|
| 160 | #{
|
|---|
| 161 | # print " Starting from scratch - clobbering the old text files...\n";
|
|---|
| 162 | # if (-e 'links.txt')
|
|---|
| 163 | # {
|
|---|
| 164 | # print " Removing links.txt...\n";
|
|---|
| 165 | # unlink <links.txt> or print " ** Cannot delete links textfile: $!\n";
|
|---|
| 166 | # }
|
|---|
| 167 | # if (-e 'images.txt')
|
|---|
| 168 | # {
|
|---|
| 169 | # print " Removing images.txt...\n";
|
|---|
| 170 | # unlink <images.txt> or print " ** Cannot delete images textfile: $!\n";
|
|---|
| 171 | # }
|
|---|
| 172 | # if (-e 'image_dirs.txt')
|
|---|
| 173 | # {
|
|---|
| 174 | # print " Removing image_dirs.txt...\n";
|
|---|
| 175 | # unlink <image_dirs.txt> or print " ** Cannot delete image_dirs textfile: $!\n";
|
|---|
| 176 | # }
|
|---|
| 177 | #}
|
|---|
| 178 |
|
|---|
| 179 | # if we're NOT starting from scratch, then read in old links from links text file
|
|---|
| 180 | # and grab the old image-links as well...
|
|---|
| 181 | if ($linknumber != 0)
|
|---|
| 182 | {
|
|---|
| 183 | # load the old links from links.txt, if it doesn't exist, then give up :(
|
|---|
| 184 | my $this = "";
|
|---|
| 185 | my $that = "";
|
|---|
| 186 | open (CHECK, "links.txt") or die " ** Cannot find/open links.txt file!: $! **\n";
|
|---|
| 187 | while(eof CHECK == 0)
|
|---|
| 188 | {
|
|---|
| 189 | while($this ne "\n")
|
|---|
| 190 | {
|
|---|
| 191 | read CHECK, $this ,1;
|
|---|
| 192 | $that = $that.$this;
|
|---|
| 193 | }
|
|---|
| 194 | $linkz_list[$linkz_pointer] = $that;
|
|---|
| 195 |
|
|---|
| 196 | for my $search(0 .. (length($that) - 3))
|
|---|
| 197 | {
|
|---|
| 198 | if((substr($that, $search, 3) eq '?a=')||(substr($that, $search, 3) eq '&a='))
|
|---|
| 199 | {
|
|---|
| 200 | $short_linkz_list[$linkz_pointer] = substr($that, $search);
|
|---|
| 201 | last;
|
|---|
| 202 | }
|
|---|
| 203 | }
|
|---|
| 204 | $linkz_pointer++;
|
|---|
| 205 | $that = ""; $this = "";
|
|---|
| 206 | }
|
|---|
| 207 | close(CHECK);
|
|---|
| 208 | print "- I found ",($#linkz_list + 1)," links in links.txt -\n";
|
|---|
| 209 |
|
|---|
| 210 | #make sure that we start dl-ing the correct first page
|
|---|
| 211 | $address = $linkz_list[$linknumber];
|
|---|
| 212 |
|
|---|
| 213 | # load the old image links from image.txt (if it doesn't exist, no big deal ;)
|
|---|
| 214 | my $im_this = "";
|
|---|
| 215 | my $im_that = "";
|
|---|
| 216 | open (IMAGES, "images.txt") || print " ** Cannot find/open images.txt file! : $! **\n";
|
|---|
| 217 | while(eof IMAGES == 0)
|
|---|
| 218 | {
|
|---|
| 219 | while($im_this ne "\n")
|
|---|
| 220 | {
|
|---|
| 221 | read IMAGES, $im_this ,1;
|
|---|
| 222 | $im_that = $im_that.$im_this;
|
|---|
| 223 | }
|
|---|
| 224 | $image_list[$image_pointer] = $im_that;
|
|---|
| 225 | $image_pointer++;
|
|---|
| 226 | $im_that = ""; $im_this = "";
|
|---|
| 227 | }
|
|---|
| 228 | close(IMAGES);
|
|---|
| 229 | print "- I found ",($#image_list + 1)," picture-links in images.txt -\n";
|
|---|
| 230 |
|
|---|
| 231 | #..and last but not least, load any image_dirs from image_dirs.txt
|
|---|
| 232 | # again, if its not there, no big deal :)
|
|---|
| 233 | my $imd_this = "";
|
|---|
| 234 | my $imd_that = "";
|
|---|
| 235 | open (IMAGE_DIR, "image_dirs.txt") || print " ** Cannot find/open image_dirs.txt file!: $! **\n";
|
|---|
| 236 | while(eof IMAGE_DIR == 0)
|
|---|
| 237 | {
|
|---|
| 238 | while($imd_this ne "\n")
|
|---|
| 239 | {
|
|---|
| 240 | read IMAGE_DIR, $imd_this ,1;
|
|---|
| 241 | $imd_that = $imd_that.$imd_this;
|
|---|
| 242 | }
|
|---|
| 243 | $image_dirs_list[$image_dirs_pointer] = $imd_that;
|
|---|
| 244 | $image_dirs_pointer++;
|
|---|
| 245 | $imd_that = ""; $imd_this = "";
|
|---|
| 246 | }
|
|---|
| 247 | close(IMAGE_DIR);
|
|---|
| 248 | print "- I found ",($#image_dirs_list + 1)," picture directories in image_dirs.txt -\n";
|
|---|
| 249 | }
|
|---|
| 250 |
|
|---|
| 251 | # Just keep going till we can find no more new links
|
|---|
| 252 | while(($#linkz_list < 0)||($#linkz_list+1 > $linknumber))
|
|---|
| 253 | {
|
|---|
| 254 |
|
|---|
| 255 | # This line specifies the command line for the lynx web browser
|
|---|
| 256 | my $kommand = $lynx.' -dump -image_links "'.$address.'"';
|
|---|
| 257 |
|
|---|
| 258 | # dump page into text-array and find starting-point of the references/links
|
|---|
| 259 | chomp(@data=`$kommand`);
|
|---|
| 260 | for my $i(0 .. $#data)
|
|---|
| 261 | {
|
|---|
| 262 | if ($data[$i] eq "References") {
|
|---|
| 263 | $here = $i;}
|
|---|
| 264 | }
|
|---|
| 265 | $here = $here+2;
|
|---|
| 266 |
|
|---|
| 267 | # process references/links
|
|---|
| 268 | for $i($here .. $#data){
|
|---|
| 269 |
|
|---|
| 270 | $its_an_image = 0;
|
|---|
| 271 |
|
|---|
| 272 | #chop-off refs leading number&spaces (eg. '1. http://www.cs.waikato.ac.nz')
|
|---|
| 273 | # ^^^
|
|---|
| 274 |
|
|---|
| 275 | # $temp = substr($data[$i],3);
|
|---|
| 276 | # @temp = split(/ /, $temp, 2);
|
|---|
| 277 |
|
|---|
| 278 | #check if the last 4 characters of the link equal .gif .jpg .png .bmp .pdf .mov .mpeg etc etc
|
|---|
| 279 |
|
|---|
| 280 | # for my $g(0 .. $#graphic_formats)
|
|---|
| 281 | # {
|
|---|
| 282 | # if(substr($temp[1],(length($graphic_formats[$g]) * -1)) eq $graphic_formats[$g])
|
|---|
| 283 | # {
|
|---|
| 284 | # $its_an_image = 1;
|
|---|
| 285 | # }
|
|---|
| 286 | # }
|
|---|
| 287 |
|
|---|
| 288 | $data[$i] =~ s/^\s*\d+\.\s+//;
|
|---|
| 289 | if ($data[$i] =~ /\.(gif|jpe?g|png|bmp|pdf|mov|mpe?g|rm)$/i) {
|
|---|
| 290 | $its_an_image = 1;
|
|---|
| 291 | }
|
|---|
| 292 |
|
|---|
| 293 | # ignore mailto urls
|
|---|
| 294 | if ($data[$i] !~ /mailto:/i) {
|
|---|
| 295 |
|
|---|
| 296 | #----------- the link is NOT an image ----------------
|
|---|
| 297 | if ($its_an_image == 0)
|
|---|
| 298 | {
|
|---|
| 299 | &its_a_link($data[$i], $outputdir);
|
|---|
| 300 | }
|
|---|
| 301 |
|
|---|
| 302 | #----------- the link IS an image ----------------
|
|---|
| 303 | if ($its_an_image != 0)
|
|---|
| 304 | {
|
|---|
| 305 | &its_an_image($data[$i], $finaldir);
|
|---|
| 306 | }
|
|---|
| 307 | }
|
|---|
| 308 | }
|
|---|
| 309 |
|
|---|
| 310 | # save the web page to disk (in the appropriate numbered directory)
|
|---|
| 311 | $kommand = $lynx.' -dump -source "'.$address.'"';
|
|---|
| 312 |
|
|---|
| 313 | if ($linknumber % $dir_entries == 0)
|
|---|
| 314 | {
|
|---|
| 315 | if ((-e $outputdir.$linknumber)&&(-d $outputdir.$linknumber))
|
|---|
| 316 | {
|
|---|
| 317 | print " ** ",$outputdir.$linknumber, " - Directory allready exists.\n";
|
|---|
| 318 | }
|
|---|
| 319 | else
|
|---|
| 320 | {
|
|---|
| 321 | mkdir($outputdir.$linknumber, 0777) or print " ** Cannot create ",$outputdir.$linknumber, "!: $!\n";
|
|---|
| 322 | mkdir($finaldir.$linknumber, 0777) or print " ** Cannot create ",$outputdir.$linknumber, "!: $!\n";
|
|---|
| 323 | }
|
|---|
| 324 | $numberdir = $linknumber;
|
|---|
| 325 | }
|
|---|
| 326 | my $target = $outputdir.$numberdir."/".$linknumber.".html";
|
|---|
| 327 |
|
|---|
| 328 | #---------------------------------------------------------------------------------------------------------------
|
|---|
| 329 | # NOTE: This next command will NOT work under win32/dos, as redirecting standard output in a script causes it to
|
|---|
| 330 | # be dumped straight to the screen as opposed to into the target file.
|
|---|
| 331 | #---------------------------------------------------------------------------------------------------------------
|
|---|
| 332 | system ("$kommand > $target");
|
|---|
| 333 | #---------------------------------------------------------------------------------------------------------------
|
|---|
| 334 |
|
|---|
| 335 | print " Saved $target\n";
|
|---|
| 336 |
|
|---|
| 337 | $linknumber++;
|
|---|
| 338 |
|
|---|
| 339 | $address = $linkz_list[$linknumber];
|
|---|
| 340 | }
|
|---|
| 341 |
|
|---|
| 342 | my $end_time = (times)[0];
|
|---|
| 343 |
|
|---|
| 344 | print "\n\n\n *----------------------------*\n";
|
|---|
| 345 | print " | Whew! Task completed! :-D |\n";
|
|---|
| 346 | print " *----------------------------*\n";
|
|---|
| 347 | printf" Script took %.2f CPU seconds to complete ;^)\n", $end_time - $start_time;
|
|---|
| 348 | print "\n\n";
|
|---|
| 349 | print " Now execute the process_html.pl script to link the downloaded collection together.\n";
|
|---|
| 350 | print " Please do make sure that it is executed with the same options as this script ;-)\n";
|
|---|
| 351 |
|
|---|
| 352 | sub its_a_link
|
|---|
| 353 | {
|
|---|
| 354 | local($found) = @_;
|
|---|
| 355 | # local($ok = 0, $kommand);
|
|---|
| 356 | local($kommand);
|
|---|
| 357 | local $short_link = "";
|
|---|
| 358 |
|
|---|
| 359 | return if ($found =~ /\#.*$/);
|
|---|
| 360 |
|
|---|
| 361 | # attach the custom options
|
|---|
| 362 | $found .= $option;
|
|---|
| 363 |
|
|---|
| 364 | #little bit of trickery here - check if there is a &d= option present in the link
|
|---|
| 365 | #if there is, then wipe the &cl= option!
|
|---|
| 366 | #This should cut down multiple copies by 75%!!
|
|---|
| 367 |
|
|---|
| 368 | #but, if there is no &d option, and the &cl option is not set, then we have to set the &cl option to something
|
|---|
| 369 | #otherwise we get pages which contain no data :\
|
|---|
| 370 |
|
|---|
| 371 | if ($found =~ /[&\?]a=d/) {
|
|---|
| 372 | if ($found =~ /[&\?]d=/) {
|
|---|
| 373 | $found =~ s/[&\?]cl=[^&]*//;
|
|---|
| 374 | } elsif ($found !~ /[&\?]cl=/) {
|
|---|
| 375 | $found .= $fix_empty_pages;
|
|---|
| 376 | }
|
|---|
| 377 | }
|
|---|
| 378 |
|
|---|
| 379 | # we also want to sort out any xxx.pr OIDs that we come across
|
|---|
| 380 | $found =~ s/([&\?](cl|d)=.*?)\.\d+\.pr/$1/g;
|
|---|
| 381 |
|
|---|
| 382 | # attach the EOL character.
|
|---|
| 383 | $found = $found."\n";
|
|---|
| 384 |
|
|---|
| 385 |
|
|---|
| 386 | # the hard way !!!
|
|---|
| 387 | # for my $search(0 .. (length($found) - 3))
|
|---|
| 388 | # {
|
|---|
| 389 | # if((substr($found, $search, 3) eq '?d=')||(substr($found, $search, 3) eq '&d='))
|
|---|
| 390 | # {
|
|---|
| 391 | # for my $second_search(0 .. (length($found) - 4))
|
|---|
| 392 | # {
|
|---|
| 393 | # if((substr($found, $second_search, 4) eq '?cl=')||(substr($found, $second_search, 4) eq '&cl='))
|
|---|
| 394 | # {
|
|---|
| 395 | # for my $third_search(($second_search + 3) .. (length($found) - 1))
|
|---|
| 396 | # {
|
|---|
| 397 | # if((substr($found, $third_search, 1)) eq '&')
|
|---|
| 398 | # {
|
|---|
| 399 | # substr($found, $second_search, $third_search - $second_search) = "";
|
|---|
| 400 | # last;
|
|---|
| 401 | # }
|
|---|
| 402 | # }
|
|---|
| 403 | # last;
|
|---|
| 404 | # }
|
|---|
| 405 | # }
|
|---|
| 406 | # last;
|
|---|
| 407 | # }
|
|---|
| 408 | # else
|
|---|
| 409 | # {
|
|---|
| 410 | # if( $search == (length($found) - 3))
|
|---|
| 411 | # {
|
|---|
| 412 | # for my $second_search(0 .. (length($found) - 4))
|
|---|
| 413 | # {
|
|---|
| 414 | # if((substr($found, $second_search, 4) eq '?cl=')||(substr($found, $second_search, 4) eq '&cl='))
|
|---|
| 415 | # {
|
|---|
| 416 | # for my $third_search(($second_search + 3) .. (length($found) - 1))
|
|---|
| 417 | # {
|
|---|
| 418 | # if((substr($found, $third_search, 1)) eq '&')
|
|---|
| 419 | # {
|
|---|
| 420 | # if (substr($found, $second_search, $third_search - $second_search) eq '&cl=')
|
|---|
| 421 | # {
|
|---|
| 422 | # substr($found, $second_search, $third_search - $second_search) = $fix_empty_pages;
|
|---|
| 423 | # }
|
|---|
| 424 | # last;
|
|---|
| 425 | # }
|
|---|
| 426 | # }
|
|---|
| 427 | # last;
|
|---|
| 428 | # }
|
|---|
| 429 | # }
|
|---|
| 430 | # }
|
|---|
| 431 | # }
|
|---|
| 432 | # }
|
|---|
| 433 |
|
|---|
| 434 | # grab the last part of the link (ignoring the start and the &e option)
|
|---|
| 435 | # for my $search(0 .. (length($found) - 3))
|
|---|
| 436 | # {
|
|---|
| 437 | # if((substr($found, $search, 3) eq '?a=')||(substr($found, $search, 3) eq '&a='))
|
|---|
| 438 | # {
|
|---|
| 439 | # $short_link = substr($found, $search);
|
|---|
| 440 | # last;
|
|---|
| 441 | # }
|
|---|
| 442 | # }
|
|---|
| 443 |
|
|---|
| 444 | ($short_link) = $found =~ /\?(.*)$/;
|
|---|
| 445 | $short_link =~ s/(^|&)e=[^&]*/$1/;
|
|---|
| 446 |
|
|---|
| 447 |
|
|---|
| 448 | # this filters out multiple copies of for example the help page, which has #something at the end of its links
|
|---|
| 449 | # now do this first with regular expression above -- Stefan
|
|---|
| 450 |
|
|---|
| 451 | # for my $search(0 .. length($found))
|
|---|
| 452 | # {
|
|---|
| 453 | # if ((substr($found, $search, 1)) eq '#')
|
|---|
| 454 | # {
|
|---|
| 455 | # $ok++;
|
|---|
| 456 | # last;
|
|---|
| 457 | # }
|
|---|
| 458 | # }
|
|---|
| 459 |
|
|---|
| 460 |
|
|---|
| 461 |
|
|---|
| 462 | # compare the found link to the links we've stored in the arrays (compares both full link and partial link)
|
|---|
| 463 | for my $search(0 .. $#linkz_list)
|
|---|
| 464 | {
|
|---|
| 465 | return if ($found eq $linkz_list[$search]);
|
|---|
| 466 | return if ($short_link eq $short_linkz_list[$search]);
|
|---|
| 467 | }
|
|---|
| 468 |
|
|---|
| 469 | # if found link is not in links array, add it
|
|---|
| 470 | open (DUMP, ">>links.txt") or die " ** Can't open links.txt!: $!\n";
|
|---|
| 471 | print DUMP $found;
|
|---|
| 472 | close(DUMP);
|
|---|
| 473 |
|
|---|
| 474 | $linkz_list[$linkz_pointer] = $found;
|
|---|
| 475 | $short_linkz_list[$linkz_pointer] = $short_link;
|
|---|
| 476 | $linkz_pointer++;
|
|---|
| 477 | }
|
|---|
| 478 |
|
|---|
| 479 | sub do_image_dirs
|
|---|
| 480 | {
|
|---|
| 481 | local($found) = @_;
|
|---|
| 482 | my $count = 0;
|
|---|
| 483 | my @br_index;
|
|---|
| 484 | my $image_dir = "";
|
|---|
| 485 | my $new_dir = 0;
|
|---|
| 486 |
|
|---|
| 487 | for my $search(1 .. (length($found) - 1 ))
|
|---|
| 488 | {
|
|---|
| 489 | $bracket = substr($found, ($search * - 1), 1);
|
|---|
| 490 | if ($bracket eq '/')
|
|---|
| 491 | {
|
|---|
| 492 | $count++;
|
|---|
| 493 | $br_index[$count] = $search;
|
|---|
| 494 | }
|
|---|
| 495 | if($count == 2)
|
|---|
| 496 | {
|
|---|
| 497 | $image_dir = substr($found, ($br_index[2] * -1) , ($br_index[2] - $br_index[1]));
|
|---|
| 498 | }
|
|---|
| 499 | }
|
|---|
| 500 |
|
|---|
| 501 | my $dirs_to_wipe = substr($found, $br_index[$#br_index - 2] * - 1, $br_index[$#br_index - 2] - $br_index[2] + 1)."\n";
|
|---|
| 502 |
|
|---|
| 503 | for my $counter(0 .. $#image_dirs_list)
|
|---|
| 504 | {
|
|---|
| 505 | if($dirs_to_wipe eq $image_dirs_list[$counter])
|
|---|
| 506 | {
|
|---|
| 507 | $new_dir++;
|
|---|
| 508 | }
|
|---|
| 509 | }
|
|---|
| 510 |
|
|---|
| 511 | if ($new_dir == 0)
|
|---|
| 512 | {
|
|---|
| 513 | open (IMAGE_DIRS, ">>image_dirs.txt") or die " ** Can't open image_dirs.txt!: $!\n";
|
|---|
| 514 | print IMAGE_DIRS $dirs_to_wipe;
|
|---|
| 515 | close(IMAGE_DIRS);
|
|---|
| 516 | $image_dirs_list[$image_dirs_pointer] = $dirs_to_wipe;
|
|---|
| 517 | $image_dirs_pointer++;
|
|---|
| 518 | }
|
|---|
| 519 |
|
|---|
| 520 | print " ",substr($finaldir, 0 ,length($finaldir) - 1).$image_dir.substr($found, ($br_index[1] * - 1), length($found) - (length($found) - $br_index[1])),"\n";
|
|---|
| 521 |
|
|---|
| 522 | return $image_dir;
|
|---|
| 523 | }
|
|---|
| 524 |
|
|---|
| 525 | sub its_an_image
|
|---|
| 526 | {
|
|---|
| 527 | local($found, $outpdir) = @_;
|
|---|
| 528 | local($kommand);
|
|---|
| 529 | my $new = 0;
|
|---|
| 530 |
|
|---|
| 531 | my $temp_found = $found . "\n";
|
|---|
| 532 |
|
|---|
| 533 | # check if the image is in the list
|
|---|
| 534 | for my $counter(0 .. $#image_list)
|
|---|
| 535 | {
|
|---|
| 536 | if($temp_found eq $image_list[$counter])
|
|---|
| 537 | {
|
|---|
| 538 | $new++;
|
|---|
| 539 | }
|
|---|
| 540 | }
|
|---|
| 541 |
|
|---|
| 542 | # only download the image if its not in the list..
|
|---|
| 543 | if($new == 0)
|
|---|
| 544 | {
|
|---|
| 545 | my $image_dir = &do_image_dirs;
|
|---|
| 546 | my $temp_outputdir = $outpdir;
|
|---|
| 547 | if (substr($temp_outputdir, -1, 1) eq "/")
|
|---|
| 548 | {
|
|---|
| 549 | substr($temp_outputdir, -1, 1) = "";
|
|---|
| 550 | }
|
|---|
| 551 |
|
|---|
| 552 | # wget is set to 'q - quiet' and 'nc - dont clobber existing files'
|
|---|
| 553 | $kommand = $wget.' -qnc --directory-prefix='.$temp_outputdir.$image_dir.' "'.$found.'"';
|
|---|
| 554 | system ("$kommand");
|
|---|
| 555 |
|
|---|
| 556 | open (IMAGES, ">>images.txt") or die " ** Can't open images.txt!: $!\n";
|
|---|
| 557 | print IMAGES $temp_found;
|
|---|
| 558 | close(IMAGES);
|
|---|
| 559 |
|
|---|
| 560 | $image_list[$image_pointer] = $temp_found;
|
|---|
| 561 | $image_pointer++;
|
|---|
| 562 |
|
|---|
| 563 | # grab corresponding ON pictures for navigation bar if we've just dl-ed the OFF picture
|
|---|
| 564 | if(substr($found , -6) eq "of.gif")
|
|---|
| 565 | {
|
|---|
| 566 | substr($found, -6, 6) = "on.gif";
|
|---|
| 567 | &its_an_image($found, $outpdir);
|
|---|
| 568 | }
|
|---|
| 569 | }
|
|---|
| 570 | }
|
|---|
