Last active
May 16, 2016 01:36
-
-
Save bohwaz/cbfba841e990ca85d414a89199a2f928 to your computer and use it in GitHub Desktop.
Download large images of art from National Gallery of Victoria (NGV) website
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
#!/usr/bin/php | |
<?php | |
/** | |
* Download a large size art image from National Gallery of Victoria (NGV) website | |
* Copyleft (C) 2016 BohwaZ http://bohwaz.net/ (Public domain) | |
*/ | |
if (empty($argv[1])) | |
{ | |
echo "Usage: " . $argv[0] . " NGV_URL" . PHP_EOL; | |
exit; | |
} | |
$url = $argv[1]; | |
if (!preg_match('!^https?://www.ngv.vic.gov.au/explore/collection/work/(\d+)/!i', $url, $match)) | |
{ | |
echo "Invalid URL supplied." . PHP_EOL; | |
exit; | |
} | |
$body = file_get_contents($url); | |
if (!preg_match('/imgWidth\s*=\s*(\d+)/i', $body, $width) | |
|| !preg_match('/imgHeight\s*=\s*(\d+)/i', $body, $height) | |
|| !preg_match('/var\s+url\s*=\s*\'(.*?)\'/i',$body, $tiles_url) | |
|| !preg_match('!<title>(.*?)</title>!i',$body, $title)) | |
{ | |
echo "Can not find image size, title or tile URL" . PHP_EOL; | |
exit; | |
} | |
$width = (int)$width[1]; | |
$height = (int)$height[1]; | |
$tiles_url = $tiles_url[1] . 'TileGroup0/'; | |
$comment = preg_replace('/\|\s*NGV.*$/', '', $title[1]) . ' -- Archived from ' . $url; | |
$filename = preg_replace('/\|\s*NGV.*$/', '', $title[1]); | |
$filename = str_replace('|', '-', $filename); | |
$filename = trim(preg_replace('/[^\w\d -]+/', '', $filename)) . '.jpg'; | |
$tiles_x = ceil($width / 256); | |
$tiles_y = ceil($height / 256); | |
$tileSize = 256; | |
$levels = 0; | |
// Getting tile size | |
while ($width > $tileSize || $height > $tileSize) | |
{ | |
$levels++; | |
$tileSize += $tileSize; | |
} | |
$z = $levels; | |
echo "Size: $width x $height ($tiles_x X $tiles_y Tiles, max zoom = $z)" . PHP_EOL; | |
$tmpdir = sys_get_temp_dir() . '/ngv-dl.' . uniqid(); | |
mkdir($tmpdir); | |
echo "Downloading tiles"; | |
$i = 0; | |
for ($y = 0; $y < $tiles_y; $y++) | |
{ | |
for ($x = 0; $x < $tiles_x; $x++) | |
{ | |
echo "."; | |
copy($tiles_url . $z . '-' . $x . '-' . $y . '.jpg', $tmpdir . '/tile-' . sprintf('%05d', $i++) . '.jpg'); | |
} | |
} | |
echo PHP_EOL; | |
echo "Assembling tiles..."; | |
$cmd = 'montage -quality 95 -mode concatenate -tile ' . ($tiles_x) . 'x' . ($tiles_y) . ' ' . $tmpdir . '/tile-*.jpg ' . escapeshellarg($filename); | |
#echo $cmd . PHP_EOL; | |
system($cmd); | |
// Adding comment | |
system('exiv2 -c ' . escapeshellarg($comment) . ' ' . escapeshellarg($filename)); | |
system('rm -rf ' . escapeshellarg($tmpdir)); | |
echo "OK" . PHP_EOL; |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment