load annotations in chunks

This commit is contained in:
j 2018-11-15 15:30:49 +00:00
parent 0b5d56ed94
commit ebf2ba4dbd
1 changed files with 23 additions and 12 deletions

View File

@ -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'],
'range': [0, 500000]
})['data']['items']:
if annotation['id'].startswith('BA/'):
continue
keyword = annotation['value']
if ': ' not in keyword:
keyword = 'other: ' + keyword
keywords[keyword] += 1
}
count = api.findAnnotations({'query': query})['data']['items']
position = 0
chunk = 1000
while position < count:
r = api.findAnnotations({
'query': query,
'keys': ['id', 'in', 'out', 'value', 'user', 'created'],
'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)