backup script

backup_films.py

@@ -0,0 +1,66 @@
+#!/usr/bin/python3
+
+import os
+import sys
+import unicodedata
+
+import ox
+import ox.api
+
+
+def get_extension(api, oshash):
+    r = api.findMedia({'query': {
+        'conditions': [{'key': 'oshash', 'value': oshash}]
+    }, 'keys': ['extension']})['data']
+    return r['items'][0]['extension']
+
+
+if __name__ == '__main__':
+    target = sys.argv[1]
+
+    site = 'pandora.cinemusespace.com'
+    api = ox.api.signin('https://%s/api/' % site)
+
+    keep = []
+    r = api.find({'range': [0, 1000], 'keys': ['id']})
+    for i in r['data']['items']:
+        item = api.get(id=i['id'], keys=['id', 'streams', 'title', 'director', 'year', 'instances'])['data']
+        director = item.get('director', ['Unknown Director'])
+        path = os.path.join('; '.join(director), '%s' % (item['title']))
+        if 'year' in item:
+            path += ' (%s)' % item['year']
+        path = ox.decode_html(path)
+        prefix = ox.decode_html(item['title'])
+        parts = []
+        if len(item['streams']) == 1:
+            id = item['streams'][0]
+            ext = get_extension(api, id)
+            name = '%s.%s' % (item['title'], ext)
+            name = ox.decode_html(name)
+            part = 1
+            parts.append([os.path.join(path, name), item['id'], id, part])
+        else:
+            part = 1
+            for id in item['streams']:
+                ext = get_extension(api, id)
+                name = '%s.Part %d.%s' % (prefix, part, ext)
+                name = ox.decode_html(name)
+                parts.append([os.path.join(path, name), item['id'], id, part])
+                part += 1
+
+        for path, id, oshash, part in parts:
+            abspath = os.path.join(target, path)
+            if os.path.exists(abspath) and ox.oshash(abspath) != oshash:
+                os.unlink(abspath)
+            if not os.path.exists(abspath):
+                url = 'https://%s/%s/download/source/%s' % (site, id, part)
+                print('downloading', abspath[len(target) + 1:])
+                api.save_url(url, abspath)
+            keep.append(unicodedata.normalize('NFD', abspath))
+
+    for root, folders, files in os.walk(target):
+        for f in files:
+            path = os.path.join(root, f)
+            if unicodedata.normalize('NFD', path) not in keep:
+                print('deleting', path)
+                os.unlink(path)