Elasticsearch-Toronto Meetup: Zero Downtime (April 21, 2015)

elasticsearch-python reindex helper exmaple

# source: https://github.com/elastic/elasticsearch-py/blob/master/elasticsearch/helpers/__init__.py
def reindex(client, source_index, target_index, query=None, target_client=None, chunk_size=500, scroll='5m', scan_kwargs={}, bulk_kwargs={}):
    target_client = client if target_client is None else target_client
    docs = helpers.scan(client, query=query, index=source_index, scroll=scroll, **scan_kwargs)
    def _change_doc_index(hits, index):
        for h in hits:
            h['_index'] = index
            # SOME MODIFICATION TO THE DOCUMENT: 
            h['_source']['first_name'] = h['_source']['username'].split()[0]
            yield h
    kwargs = {
        'stats_only': True,
    }
    kwargs.update(bulk_kwargs)
    return helpers.bulk(target_client, _change_doc_index(docs, target_index), chunk_size=chunk_size, **kwargs)