load annotations in chunks

ontology/update_keywords.py

@@ -9,24 +9,35 @@ site = 'pandora.cinemusespace.com'
 api = ox.api.signin('https://%s/api/' % site)
 
 keywords = collections.Counter()
-for annotation in api.findAnnotations({
+query = {
-    'query': {
         'conditions': [{
             'key': 'layer',
             'value': 'keywords',
             'operator': '=='
         }],
         'operator': '&'
-    },
+}
-    'keys': ['id', 'in', 'out', 'value', 'user', 'created'],
+count = api.findAnnotations({'query': query})['data']['items']
-    'range': [0, 500000]
+position = 0
-})['data']['items']:
+chunk = 1000
-    if annotation['id'].startswith('BA/'):
+
-        continue
+while position < count:
-    keyword = annotation['value']
+    r = api.findAnnotations({
-    if ': ' not in keyword:
+        'query': query,
-        keyword = 'other: ' + keyword
+        'keys': ['id', 'in', 'out', 'value', 'user', 'created'],
-    keywords[keyword] += 1
+        'sort': [{'key': 'public_id', 'operator': '+'}],
+        'range': [position, position+chunk]
+    })
+    if 'data' not in r:
+        print('failed', r)
+    for annotation in r['data']['items']:
+        if annotation['id'].startswith('BA/'):
+            continue
+        keyword = annotation['value']
+        if ': ' not in keyword:
+            keyword = 'other: ' + keyword
+        keywords[keyword] += 1
+    position += chunk
 
 with open('keywords.json', 'w') as fd:
     json.dump(keywords, fd, indent=4, ensure_ascii=False, sort_keys=True)