Class: MysqlTruck::Loader

Inherits:
Object
  • Object
show all
Includes:
FileUtils, Helper
Defined in:
lib/mysql_truck/loader.rb

Instance Method Summary collapse

Methods included from Helper

#base_path, #benchmark, #bucket_dir, #config, #csv_options, #db_connection_options, #formatted_time, #initialize_directories, #initialize_s3, #local_host?, #remote_host?, #remove_directories, #s3_path, #tmp_path

Constructor Details

#initialize(config) ⇒ Loader

Returns a new instance of Loader.



6
7
8
9
 
# File 'lib/mysql_truck/loader.rb', line 6

def initialize(config)
  @config = config
  initialize_s3
end

Instance Method Details

#backups ⇒ Object

Get a list of backups stored on S3.

Returns an array of s3 paths that look like:

mysql/YYYY-MM-DD

Array elements are sorted with the latest date first.



232
233
234
235
236
237
238
239
240
241
242
243
244
245
 
# File 'lib/mysql_truck/loader.rb', line 232

def backups
  unless @backups
    @backups = []
    # Backups are stored in the mysql/ directory
    @bucket.s3.interface.incrementally_list_bucket(@bucket.name, {
      :prefix => "#{bucket_dir}", :delimiter => "/"
    }) do |item|
      @backups += item[:common_prefixes]
    end
    @backups = @backups.sort { |a,b| b <=> a }
  end

  @backups
end

#download_file(key) ⇒ Object 



165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
 
# File 'lib/mysql_truck/loader.rb', line 165

def download_file(key)
  filename = File.basename(key.name)

  unless should_download_file?(filename)
    return
  end

  puts "\n#{key}"

  file = tmp_path.join(filename)
  unzipped_file = tmp_path.join(file.basename(".gz"))
  if !smartly? || (smartly? && !unzipped_file.exist?)
    print " - Downloading... "

    benchmark do
      file.open("wb") do |f|
        @bucket.s3.interface.get(@bucket.name, key.name) do |chunk|
          f.write chunk
        end
      end
    end
  else
    puts " already downloaded."
  end

  filename
end

#execute_sql_file(table, backup_date_str, file_path) ⇒ Object 



124
125
126
127
128
129
130
131
132
133
 
# File 'lib/mysql_truck/loader.rb', line 124

def execute_sql_file(table, backup_date_str, file_path)
  cat_cmd = "cat #{file_path}"
  sed_cmd = config[:date_suffix] ? "sed 's/`#{table}`/`#{table}_#{backup_date_str.gsub(/-/, "")}`/g'" : nil
  sed2_cmd = config[:date_suffix] ? "sed 's/TABLE #{table}/TABLE #{table}_#{backup_date_str.gsub(/-/, "")}/g'" : nil
  import_cmd =  "mysql #{db_connection_options}"

  benchmark do
    `#{[ cat_cmd, sed_cmd, sed2_cmd, import_cmd].compact.join(' | ')}`
  end
end

#import_csv_file(table, backup_date_str, file_path) ⇒ Object 



135
136
137
138
139
140
141
142
143
144
145
146
147
148
 
# File 'lib/mysql_truck/loader.rb', line 135

def import_csv_file(table, backup_date_str, file_path)
  if config[:date_suffix]
    old_file_path = file_path
    file_path = file_path.to_s.gsub("/#{table}.", "/#{table}_#{backup_date_str.gsub(/-/, "")}.")
    # move file
    `mv #{old_file_path} #{file_path}`
  end

  benchmark do
    `mysqlimport --local --compress #{csv_options} #{db_connection_options} #{file_path}`
  end

  file_path
end

#import_tsv_file(table, backup_date_str, file_path) ⇒ Object 



150
151
152
153
154
155
156
157
158
159
160
161
162
163
 
# File 'lib/mysql_truck/loader.rb', line 150

def import_tsv_file(table, backup_date_str, file_path)
  if config[:date_suffix]
    old_file_path = file_path
    file_path = file_path.to_s.gsub("/#{table}.data.", "/#{table}_#{backup_date_str.gsub(/-/, "")}.")
    # move file
    `mv #{old_file_path} #{file_path}`
  end

  benchmark do
    `mysqlimport --local --compress #{db_connection_options} #{file_path}`
  end

  file_path
end

#load_backup ⇒ Object 



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
118
119
120
121
122
 
# File 'lib/mysql_truck/loader.rb', line 21

def load_backup
  # Default to latest backup
  backup_date_str = config[:backup_date] || backups.first.split("/").last

  # Set directory where backup is downloaded to
  @time = Time.new(*backup_date_str.split("-"))
  initialize_directories

  puts "Download & decompressing backups"
  puts "-------------------"

  @bucket.keys(:prefix => s3_path).each do |key|
    next if key.to_s.match(/\/$/)
    next unless (filename = download_file(key))

    if tmp_path.join(filename).exist?
      print " - Inflating #{filename} ... "

      if File.extname(filename).match(".lzo")
        decompress_cmd = "lzop -d -f -U"
      else
        decompress_cmd = "gunzip -f"
      end
      benchmark { `#{decompress_cmd} #{tmp_path.join(filename)}` }
    end
  end

  # Load data
  puts "\nLoading schema and data by table"
  puts "--------------------------------"

  files = Dir["#{tmp_path}/*"]
  tables = files.map { |f| File.basename(f).gsub(/\..*$/, '') }.uniq
  total = tables.size
  count = 0

  # Find all table names and process
  tables.sort.each do |table|
    count += 1
    puts "\nProcessing #{table} (#{count}/#{total})"

    old_schema_file = tmp_path.join("#{table}.sql")
    schema_file     = tmp_path.join("#{table}.no_index.sql")
    index_file      = tmp_path.join("#{table}.indices.sql")
    data_file       = tmp_path.join("#{table}.data.sql")
    csv_data_file   = tmp_path.join("#{table}.csv")
    tsv_data_file   = tmp_path.join("#{table}.data.tsv")

    # Create table
    if schema_file.exist?
      print " - Loading schema: #{File.basename(schema_file)} ... "
      execute_sql_file(table, backup_date_str, schema_file)
      schema_file.delete

    elsif old_schema_file.exist?
      print " - Loading schema: #{File.basename(old_schema_file)} ... "
      execute_sql_file(table, backup_date_str, old_schema_file)
      old_schema_file.delete

    else
      puts "ERROR: no schema file!"
      next
    end

    # Load data
    if !config[:skip_all_data]
      if data_file.exist?
        print " - Importing #{File.basename(data_file)} ... "
        execute_sql_file(table, backup_date_str, data_file)
        data_file.delete

      elsif csv_data_file.exist?
        print " - Importing #{File.basename(csv_data_file)} ... "
        csv_data_file = import_csv_file(table, backup_date_str, csv_data_file)
        File.delete(csv_data_file)

      elsif tsv_data_file.exist?
        print " - Importing #{File.basename(tsv_data_file)} ... "
        tsv_data_file = import_tsv_file(table, backup_date_str, tsv_data_file)
        File.delete(tsv_data_file)
      end
    end

    # Add indices
    if index_file.exist?
      print " - Adding indices: #{File.basename(index_file)} ... "
      execute_sql_file(table, backup_date_str, index_file)
      index_file.delete
    end

  end

  puts "Backup loaded."

  # This isn't in an ensure block because we want to keep around
  # downloads if there's a failure importing a table.
  # remove_directories

rescue Exception => e
  puts e.message
  puts e.backtrace.join("\n")
end

#only_tables ⇒ Object

only import these tables schema+data



17
18
19
 
# File 'lib/mysql_truck/loader.rb', line 17

def only_tables
  config[:only_tables] || []
end

#should_download_file?(filename) ⇒ Boolean

Returns:

  • (Boolean) 


193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
 
# File 'lib/mysql_truck/loader.rb', line 193

def should_download_file?(filename)
  table_name = filename.gsub(/\..*\..*$/, '')
  if only_tables.empty? and skip_data_for_tables.empty? and !config[:skip_all_data]
    return true
  end

  # If we're targetting specific tables, then we always want both
  # schema and csv files.
  if !only_tables.empty?
    return only_tables.include?(table_name)
  end

  if filename.match(/\.data\.(sql|tsv)\.(lzo|gz)$/)
    return false if config[:skip_all_data]
    is_data = true
    is_schema = false
  else
    is_data = false
    is_schema = true
  end

  return true if is_schema && config[:skip_all_data]

  if !skip_data_for_tables.empty?
    if is_schema || (is_data && !skip_data_for_tables.include?(table_name))
      return true
    end
  end

  false
end

#skip_data_for_tables ⇒ Object

only import schema for these tables



12
13
14
 
# File 'lib/mysql_truck/loader.rb', line 12

def skip_data_for_tables
  config[:skip_data_for_tables] || []
end

#smartly? ⇒ Boolean

Returns:

  • (Boolean) 


247
248
249
 
# File 'lib/mysql_truck/loader.rb', line 247

def smartly?
  config[:smartly]
end