module Sequel::MySQL::DatasetMethods

Dataset methods shared by datasets that use MySQL databases.

Constants

APOS
APOS_RE
BACKSLASH_RE
BACKTICK
BACKTICK_RE
BINARY
BOOL_FALSE
BOOL_TRUE
CAST_BITCOMP_CLOSE
CAST_BITCOMP_OPEN
COMMA
COMMA_SEPARATOR
CONCAT
DELETE_CLAUSE_METHODS
DOUBLE_APOS
DOUBLE_BACKTICK
EMPTY_COLUMNS
EMPTY_VALUES
EQ
EQ_VALUES
EXPLAIN
EXPLAIN_EXTENDED
FOR_SHARE
FROM
GROUP_BY
IGNORE
INSERT
INSERT_CLAUSE_METHODS
LIKE
LIMIT
MATCH_AGAINST
MATCH_AGAINST_BOOLEAN
NATURAL_LEFT_JOIN
NOT_SPACE
ON_DUPLICATE_KEY_UPDATE
PAREN_CLOSE
PAREN_OPEN
QUAD_BACKSLASH
REGEXP
REPLACE
SELECT_CLAUSE_METHODS
SPACE
SQL_CALC_FOUND_ROWS
STRAIGHT_JOIN
UPDATE_CLAUSE_METHODS
WITH_ROLLUP

Public Instance Methods

calc_found_rows() click to toggle source

Sets up the select methods to use SQL_CALC_FOUND_ROWS option.

dataset.calc_found_rows.limit(10)
# SELECT SQL_CALC_FOUND_ROWS * FROM table LIMIT 10
# File lib/sequel/adapters/shared/mysql.rb, line 580
def calc_found_rows
  clone(:calc_found_rows => true)
end
complex_expression_sql_append(sql, op, args) click to toggle source

MySQL specific syntax for LIKE/REGEXP searches, as well as string concatenation.

# File lib/sequel/adapters/shared/mysql.rb, line 536
def complex_expression_sql_append(sql, op, args)
  case op
  when :IN, :"NOT IN"
    ds = args.at(1)
    if ds.is_a?(Sequel::Dataset) && ds.opts[:limit]
      super(sql, op, [args.at(0), ds.from_self])
    else
      super
    end
  when :~, :'!~', :'~*', :'!~*', :LIKE, :'NOT LIKE', :ILIKE, :'NOT ILIKE'
    sql << PAREN_OPEN
    literal_append(sql, args.at(0))
    sql << SPACE
    sql << 'NOT ' if [:'NOT LIKE', :'NOT ILIKE', :'!~', :'!~*'].include?(op)
    sql << ([:~, :'!~', :'~*', :'!~*'].include?(op) ? REGEXP : LIKE)
    sql << SPACE
    sql << BINARY if [:~, :'!~', :LIKE, :'NOT LIKE'].include?(op)
    literal_append(sql, args.at(1))
    sql << PAREN_CLOSE
  when :'||'
    if args.length > 1
      sql << CONCAT
      array_sql_append(sql, args)
    else
      literal_append(sql, args.at(0))
    end
  when :'B~'
    sql << CAST_BITCOMP_OPEN
    literal_append(sql, args.at(0))
    sql << CAST_BITCOMP_CLOSE
  else
    super
  end
end
distinct(*args) click to toggle source

Use GROUP BY instead of DISTINCT ON if arguments are provided.

# File lib/sequel/adapters/shared/mysql.rb, line 572
def distinct(*args)
  args.empty? ? super : group(*args)
end
explain(opts={}) click to toggle source

Return the results of an EXPLAIN query as a string. Options:

:extended

Use EXPLAIN EXPTENDED instead of EXPLAIN if true.

# File lib/sequel/adapters/shared/mysql.rb, line 586
def explain(opts={})
  # Load the PrettyTable class, needed for explain output
  Sequel.extension(:_pretty_table) unless defined?(Sequel::PrettyTable)

  ds = db.send(:metadata_dataset).with_sql((opts[:extended] ? EXPLAIN_EXTENDED : EXPLAIN) + select_sql).naked
  rows = ds.all
  Sequel::PrettyTable.string(rows, ds.columns)
end
for_share() click to toggle source

Return a cloned dataset which will use LOCK IN SHARE MODE to lock returned rows.

# File lib/sequel/adapters/shared/mysql.rb, line 596
def for_share
  lock_style(:share)
end
full_text_sql(cols, terms, opts = {}) click to toggle source

MySQL specific full text search syntax.

# File lib/sequel/adapters/shared/mysql.rb, line 606
def full_text_sql(cols, terms, opts = {})
  terms = terms.join(' ') if terms.is_a?(Array)
  SQL::PlaceholderLiteralString.new((opts[:boolean] ? MATCH_AGAINST_BOOLEAN : MATCH_AGAINST), [Array(cols), terms])
end
having(*cond, &block) click to toggle source

MySQL allows HAVING clause on ungrouped datasets.

# File lib/sequel/adapters/shared/mysql.rb, line 612
def having(*cond, &block)
  _filter(:having, *cond, &block)
end
insert_ignore() click to toggle source

Sets up the insert methods to use INSERT IGNORE. Useful if you have a unique key and want to just skip inserting rows that violate the unique key restriction.

dataset.insert_ignore.multi_insert(
 [{:name => 'a', :value => 1}, {:name => 'b', :value => 2}]
)
# INSERT IGNORE INTO tablename (name, value) VALUES (a, 1), (b, 2)
# File lib/sequel/adapters/shared/mysql.rb, line 645
def insert_ignore
  clone(:insert_ignore=>true)
end
join_table(type, table, expr=nil, table_alias={}, &block) click to toggle source

Transforms an CROSS JOIN to an INNER JOIN if the expr is not nil. Raises an error on use of :full_outer type, since MySQL doesn't support it.

# File lib/sequel/adapters/shared/mysql.rb, line 618
def join_table(type, table, expr=nil, table_alias={}, &block)
  type = :inner if (type == :cross) && !expr.nil?
  raise(Sequel::Error, "MySQL doesn't support FULL OUTER JOIN") if type == :full_outer
  super(type, table, expr, table_alias, &block)
end
join_type_sql(join_type) click to toggle source

Transforms :natural_inner to NATURAL LEFT JOIN and straight to STRAIGHT_JOIN.

# File lib/sequel/adapters/shared/mysql.rb, line 626
def join_type_sql(join_type)
  case join_type
  when :straight
    STRAIGHT_JOIN
  when :natural_inner
    NATURAL_LEFT_JOIN
  else
    super
  end
end
multi_insert_sql(columns, values) click to toggle source

MySQL specific syntax for inserting multiple values at once.

# File lib/sequel/adapters/shared/mysql.rb, line 673
def multi_insert_sql(columns, values)
  sql = LiteralString.new('VALUES ')
  expression_list_append(sql, values.map{|r| Array(r)})
  [insert_sql(columns, sql)]
end
multi_replace(*values) click to toggle source

Replace multiple rows in a single query.

# File lib/sequel/adapters/shared/mysql.rb, line 696
def multi_replace(*values)
  clone(:replace=>true).multi_insert(*values)
end
on_duplicate_key_update(*args) click to toggle source

Sets up the insert methods to use ON DUPLICATE KEY UPDATE If you pass no arguments, ALL fields will be updated with the new values. If you pass the fields you want then ONLY those field will be updated.

Useful if you have a unique key and want to update inserting rows that violate the unique key restriction.

dataset.on_duplicate_key_update.multi_insert(
 [{:name => 'a', :value => 1}, {:name => 'b', :value => 2}]
)
# INSERT INTO tablename (name, value) VALUES (a, 1), (b, 2)
# ON DUPLICATE KEY UPDATE name=VALUES(name), value=VALUES(value)

dataset.on_duplicate_key_update(:value).multi_insert(
  [{:name => 'a', :value => 1}, {:name => 'b', :value => 2}]
)
# INSERT INTO tablename (name, value) VALUES (a, 1), (b, 2)
# ON DUPLICATE KEY UPDATE value=VALUES(value)
# File lib/sequel/adapters/shared/mysql.rb, line 668
def on_duplicate_key_update(*args)
  clone(:on_duplicate_key_update => args)
end
quoted_identifier_append(sql, c) click to toggle source

MySQL uses the nonstandard ` (backtick) for quoting identifiers.

# File lib/sequel/adapters/shared/mysql.rb, line 680
def quoted_identifier_append(sql, c)
  sql << BACKTICK << c.to_s.gsub(BACKTICK_RE, DOUBLE_BACKTICK) << BACKTICK
end
replace(*values) click to toggle source

Execute a REPLACE statement on the database.

# File lib/sequel/adapters/shared/mysql.rb, line 685
def replace(*values)
  execute_insert(replace_sql(*values))
end
replace_sql(*values) click to toggle source

MySQL specific syntax for REPLACE (aka UPSERT, or update if exists, insert if it doesn't).

# File lib/sequel/adapters/shared/mysql.rb, line 691
def replace_sql(*values)
  clone(:replace=>true).insert_sql(*values)
end
supports_distinct_on?() click to toggle source

MySQL can emulate DISTINCT ON with its non-standard GROUP BY implementation, though the rows returned cannot be made deterministic through ordering.

# File lib/sequel/adapters/shared/mysql.rb, line 702
def supports_distinct_on?
  true
end
supports_group_rollup?() click to toggle source

MySQL supports GROUP BY WITH ROLLUP (but not CUBE)

# File lib/sequel/adapters/shared/mysql.rb, line 707
def supports_group_rollup?
  true
end
supports_intersect_except?() click to toggle source

MySQL does not support INTERSECT or EXCEPT

# File lib/sequel/adapters/shared/mysql.rb, line 712
def supports_intersect_except?
  false
end
supports_modifying_joins?() click to toggle source

MySQL supports modifying joined datasets

# File lib/sequel/adapters/shared/mysql.rb, line 717
def supports_modifying_joins?
  true
end
supports_ordered_distinct_on?() click to toggle source

MySQL's DISTINCT ON emulation using GROUP BY does not respect the queries ORDER BY clause.

# File lib/sequel/adapters/shared/mysql.rb, line 723
def supports_ordered_distinct_on?
  false
end
supports_regexp?() click to toggle source

MySQL supports pattern matching via regular expressions

# File lib/sequel/adapters/shared/mysql.rb, line 728
def supports_regexp?
  true
end
supports_timestamp_usecs?() click to toggle source

MySQL does support fractional timestamps in literal timestamps, but it ignores them. Also, using them seems to cause problems on 1.9. Since they are ignored anyway, not using them is probably best.

# File lib/sequel/adapters/shared/mysql.rb, line 735
def supports_timestamp_usecs?
  false
end
update_ignore() click to toggle source

Sets up the update methods to use UPDATE IGNORE. Useful if you have a unique key and want to just skip updating rows that violate the unique key restriction.

dataset.update_ignore.update({:name => 'a', :value => 1})
# UPDATE IGNORE tablename SET name = 'a', value = 1
# File lib/sequel/adapters/shared/mysql.rb, line 745
def update_ignore
  clone(:update_ignore=>true)
end