es-reindex.py

#!/usr/bin/env python3

# A script which creates a new index using a new mapping and create an alias
# for the old index name.

from argparse import ArgumentParser
from urllib.request import Request, build_opener, install_opener, HTTPSHandler
import json
import base64
from os import getenv
import subprocess

# Avoid checking for certificate since we execute this only inside CERN.
# Drop the following stanza if not.
import ssl

if hasattr(ssl, "_create_unverified_context"):
    ssl._create_default_https_context = ssl._create_unverified_context


def format(s, **kwds):
    return s % kwds


def writeToES(server, data):
    url = format("%(server)s/_bulk", server=server)
    handler = HTTPSHandler(debuglevel=0)
    opener = build_opener(handler)
    install_opener(opener)

    new_request = Request(url, data)
    new_request.get_method = lambda: "PUT"
    base64string = base64.encodebytes(getenv("ES_AUTH").encode()).replace(b"\n", b"").decode()
    new_request.add_header("Authorization", "Basic %s" % base64string)
    try:
        response = opener.open(new_request)
        print(response.read())
    except Exception as e:
        print(e)
        exit(1)


# - Get the index via scan and scroll.
# - Push items to the new index.
# - Delete old index.
# - Create alias from the old index to the new one.
if __name__ == "__main__":
    # First get all the available indices
    parser = ArgumentParser()
    parser.add_argument(
        "--server", "-s", dest="server", default="localhost:9200", help="the elasticsearch server"
    )
    parser.add_argument(
        "--dry-run",
        "-n",
        dest="dryrun",
        action="store_true",
        default=False,
        help="do not change the DB.",
    )
    parser.add_argument("source")
    parser.add_argument("dest")
    args = parser.parse_args()
    proxy_string = ""

    if not getenv("ES_AUTH"):
        print("ES_AUTH not specified")
        exit(1)

    user_string = "--user %s" % getenv("ES_AUTH")

    query = {"query": {"match_all": {}}, "size": 1000}
    cmd = format(
        "curl -s %(user_string)s '%(server)s/%(source)s/_search?search_type=scan&scroll=1m' -d'%(query)s'",
        source=args.source,
        query=json.dumps(query),
        server=args.server,
        user_string=user_string,
    )
    print(cmd)
    err, out = subprocess.getstatusoutput(cmd)
    if err:
        print("Error while getting indices")
        print(out)
        exit(0)
    result = json.loads(out)

    while True:
        cmd = format(
            "curl -s %(user_string)s '%(server)s/_search/scroll?scroll=1m' -d'%(scroll_id)s'",
            source=args.source,
            query=json.dumps(query),
            server=args.server,
            user_string=user_string,
            scroll_id=result["_scroll_id"],
        )
        err, out = subprocess.getstatusoutput(cmd)
        if err:
            print("Error while getting entries")
            print(out)
            exit(1)
        result = json.loads(out)
        if result.get("status", 200) != 200:
            print(out)
            exit(1)

        # Exit when there are not results
        if not len(result["hits"]["hits"]):
            break
        line = ""
        for item in result["hits"]["hits"]:
            cmd = format(
                '{ "create": { "_index": "%(index)s", "_type": "%(obj_type)s", "_id": "%(obj_id)s" }}',
                index=args.dest,
                obj_type=item["_type"],
                obj_id=item["_id"],
            )
            payload = json.dumps(item["_source"])
            line += cmd + "\n" + payload + "\n"
            if len(line) < 500000:
                continue
            #  print json.dumps(json.loads(cmd))
            #  print json.dumps(json.loads(payload))
            writeToES(server=args.server, data=line)
            line = ""
        if line:
            writeToES(server=args.server, data=line)