Class: Esteem::Reindex

Inherits:
Object
  • Object
show all
Defined in:
lib/esteem/reindex.rb

Overview

Cribbed from geronime’s “es-reindex” script: github.com/geronime/es-reindex

Instance Attribute Summary collapse

Instance Method Summary collapse

Constructor Details

#initialize(source:, destination:, frame:, update:) ⇒ Reindex

Perform a new reindex job. Success or failure will be reported by the read-only “success” attribute.

Parameters:

  • [String] (Hash)

    a customizable set of options

  • [Integer] (Hash)

    a customizable set of options

  • [Boolean] (Hash)

    a customizable set of options



21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
# File 'lib/esteem/reindex.rb', line 21

def initialize source:, destination:, frame:, update:
  src, dst = source, destination
  @success = false

  type, src = src.reverse.split('/', 2).map(&:reverse)

  surl, durl, sidx, didx = '', '', '', ''
  [[src, surl, sidx], [dst, durl, didx]].each{|param, url, idx|
    if param =~ %r{^(.*)/(.*?)$}
      url.replace $1
      idx.replace $2
    else
      url.replace 'http://127.0.0.1:9200'
      idx.replace param
    end
  }

  puts "Copying '%s/%s/%s' to '%s/%s'" % [ surl, sidx, type, durl, didx ]

  scount = retried_request :get, "#{surl}/#{sidx}/#{type}/_count?q=*"
  scount = Oj.load(scount)['count'].to_i

  puts "%d documents" % scount

  if scount == 0
    puts "Hey, this index is empty! Skipping..."
    @success = true
    return
  end

  t, done = Time.now, 0
  shards = retried_request :get, "#{surl}/#{sidx}/_count?q=*"
  shards = Oj.load(shards)['_shards']['total'].to_i
  scan = retried_request(:get, "#{surl}/#{sidx}/#{type}/_search?search_type=scan&scroll=10m&size=#{frame / shards}")
  scan = Oj.load scan
  scroll_id = scan['_scroll_id']
  total = scan['hits']['total']
  printf "    %u/%u (%.1f%%) done.\r", done, total, 0

  bulk_op = update ? "index" : "create"

  loop do
    data = retried_request(:get, "#{surl}/_search/scroll?scroll=10m&scroll_id=#{scroll_id}")
    data = Oj.load data
    break if data['hits']['hits'].empty?

    scroll_id = data['_scroll_id']
    bulk = ''

    data['hits']['hits'].each do |doc|
      bulk << %Q({"#{bulk_op}": {"_index" : "#{didx}", "_id" : "#{
          doc['_id']}", "_type" : "#{doc['_type']}"}}\n)
      bulk << Oj.dump(doc['_source']) + "\n"
      done += 1
    end

    unless bulk.empty?
      bulk << "\n" # empty line in the end required
      retried_request :post, "#{durl}/_bulk", bulk
    end

    eta = total * (Time.now - t) / done
    printf "    %u/%u (%.1f%%) done in %s, E.T.A.: %s.\r",
      done, total, 100.0 * done / total, tm_len(Time.now - t), t + eta
  end

  printf "#{' ' * 80}\r    %u/%u done in %s.\n",
    done, total, tm_len(Time.now - t)



  tries = 0
  dcount = -1

  puts "Waiting for documents to appear in index..."
  until total <= dcount
    dcount = retried_request :get, "#{durl}/#{didx}/#{type}/_count"
    dcount = Oj.load(dcount)['count'].to_i
    tries += 1

    if tries > 30
      puts "Too many tries!"
      @success = false
      return
    end

    sleep 15
  end

  if done != dcount
    puts "Incomplete done! %d != %d" % [ done, dcount ]
    @success = false
    return
  end

  @success = true
end

Instance Attribute Details

#successObject (readonly)

Returns the value of attribute success.



11
12
13
# File 'lib/esteem/reindex.rb', line 11

def success
  @success
end