Author: david
Date: Thu Dec 8 16:02:49 2011
New Revision: 10385
Log:
Handle Archivematica DIP filenames (UUID + filename + '.jpg')
Modified:
trunk/lib/task/archivematicaDipImportTask.class.php
Modified: trunk/lib/task/archivematicaDipImportTask.class.php
==============================================================================
--- trunk/lib/task/archivematicaDipImportTask.class.php Thu Dec 8 15:18:56
2011 (r10384)
+++ trunk/lib/task/archivematicaDipImportTask.class.php Thu Dec 8 16:02:49
2011 (r10385)
@@ -83,12 +83,39 @@
$this->logSection("Link digital objects from {$arguments['filename']}...");
- // Build hash on information_object.id, with array value if information
+ // If DIP folder is specified then digital objects will be in the
+ // "objects" subfolder
+ if (file_exists(rtrim($arguments['path'], '/').'/objects'))
+ {
+ $arguments['path'] = rtrim($arguments['path'], '/').'/objects';
+ }
+
+ // Create a lookup table to match original file name (foo.TIF) with DIP
+ // filename (979c4458-21f3-11e1-a4bd-001d09282b9d-foo.jpg)
+ $this->createFilenameLookup($arguments['path']);
+
+ // Build hash on information_object key, with array value if information
// object has multiple digital objects attached
while ($row = fgetcsv($fh, 1000))
{
$filepath = rtrim($arguments['path'], '/').'/'.$row[1];
+ // Check if filename has been changed by Archivematica
+ if (!file_exists($filepath))
+ {
+ // Substitute original image (TIFF) name with new DIP (JPEG) name
+ $filename = basename($filepath);
+
+ if (isset($this->filenames[strtolower($filename)]))
+ {
+ $filepath = str_replace($filename,
$this->filenames[strtolower($filename)], $filepath);
+ }
+ else
+ {
+ continue;
+ }
+ }
+
if (!isset($digitalObjects[$row[0]]))
{
$digitalObjects[$row[0]] = $filepath;
@@ -178,4 +205,30 @@
return $uploadDir;
}
+
+ /**
+ * Create a lookup table to match original file name (foo.TIF) with DIP
+ * filename (979c4458-21f3-11e1-a4bd-001d09282b9d-foo.jpg)
+ */
+ protected function createFilenameLookup($path)
+ {
+ if (!($files = scandir($path)))
+ {
+ throw new sfException("Couldn't read path '$path'");
+ }
+
+ foreach ($files as $file)
+ {
+ // Format should be UUID (37 hex chars or hyphen) + filename + '.jpg'
+ $pattern = '/[0-9a-f-]{37}(.+)\.jpg/';
+
+ if (is_dir($file) || !preg_match($pattern, strtolower($file), $matches))
+ {
+ continue;
+ }
+
+ // Key is original TIFF name
+ $this->filenames[$matches[1].'.tif'] = $file;
+ }
+ }
}
--
You received this message because you are subscribed to the Google Groups
"Qubit Toolkit Commits" group.
To post to this group, send email to [email protected].
To unsubscribe from this group, send email to
[email protected].
For more options, visit this group at
http://groups.google.com/group/qubit-commits?hl=en.
