{"id":12,"date":"2007-02-03T20:29:36","date_gmt":"2007-02-03T19:29:36","guid":{"rendered":"http:\/\/schuerig.de\/michael\/blog\/?p=12"},"modified":"2021-11-24T09:20:10","modified_gmt":"2021-11-24T08:20:10","slug":"ar-enumerable","status":"publish","type":"post","link":"https:\/\/www.schuerig.de\/michael\/blog\/index.php\/2007\/02\/03\/ar-enumerable\/","title":{"rendered":"ActiveRecord Enumerable"},"content":{"rendered":"

Ever wanted to iterate over a huge number of objects and found
\nthat find(:all)<\/code> doesn’t cut it? Then this plugin
\nmight be for you. Install it with<\/p>\n

<\/p>\n

\r\n$ script\/plugin install svn:\/\/rubyforge.org\/var\/svn\/ar-enumerable\/ar_enumerable\/trunk\r\n<\/pre>\n
<\/code><\/p>\n
First, a note of caution. This plugin has not seen much practical
\nuse yet, in its current state it is only slightly above a
\nproof-of-concept. There are several tests included that cover most
\nof the functionality. If you intend to use this plugin, please
\nmake sure to independently check that it does what you want.<\/p>\n
Enumerating ActiveRecord objects<\/h3>\n
Let’s start with a boring example, to get the general idea<\/p>\n
<\/p>\n
\r\nPerson.where(['lastname = ?', 'Smith']).map(&:firstname)\r\n<\/pre>\n
<\/code><\/p>\n
Slightly more interesting<\/p>\n
<\/p>\n
\r\nPerson.where(['lastname = ?', 'Smith']).inject(0) do |income, person|\r\n  income += person.income\r\nend\r\n<\/pre>\n
<\/code><\/p>\n
<\/p>\n
\r\nsiblings = Person.where(\r\n    ['lastname = ?', 'Smith'], \r\n    :include => :siblings).inject([0, 0]) do |memo, person|\r\n  memo[0] += person.siblings.count\r\n  memo[1] += 1\r\n  memo\r\nend\r\nsiblings[0].to_f \/ siblings[1]\r\n<\/pre>\n
<\/code><\/p>\n
The point here is that your database may contain millions of Smiths,
\nfor what it’s worth, the entire population of the earth could have
\nturned into Smiths, still, only a limited number of them would be
\nloaded into memory at the same time.<\/p>\n
Internally, enumeration works in two completely different ways.
\nBy default, a chunk of objects is retrieved from the database using
\nOFFSET and LIMIT clauses. This way is supported by all databases,
\nhowever it can be slow.
\nTherefore, the preferred alternative is to use a standard SQL cursor
\nfor iterating over the result set of a query. It would simply look like
\nthis<\/p>\n
<\/p>\n
\r\nPerson.where(['lastname = ?', 'Smith'], :use_cursor => true).inject(0) do |income, person|\r\n  income += person.income\r\nend\r\n<\/pre>\n
<\/code><\/p>\n
A further parameter is how many records (or rows, in the case of cursors)
\nare fetched from the database in one access. The default is 1000. You can
\nset it explicitly like this<\/p>\n
<\/p>\n
\r\nPerson.where(['lastname = ?', 'Smith'], :fetch_count => 100).inject(0) do |income, person|\r\n  income += person.income\r\nend\r\n<\/pre>\n
<\/code><\/p>\n
Performance Considerations<\/h3>\n
Some informal tests suggest that for large result sets without
\nincluded associations cursor-based enumeration can be faster than
\nOFFSET\/LIMIT-based enumeration by a factor of 3 to 5.
\nWhen associations are included, things become more blurry. In this
\ncase, cursor-based enumeration imposes an ordering on the result
\nset that can slow down things considerably.
\nThe fetch count has considerable influence on performance, basically,
\nyou can buy speed by spending memory.<\/p>\n","protected":false},"excerpt":{"rendered":"
Ever wanted to iterate over a huge number of objects and found that find(:all) doesn’t cut it? Then this plugin might be for you. Install it with $ script\/plugin install svn:\/\/rubyforge.org\/var\/svn\/ar-enumerable\/ar_enumerable\/trunk First, a note of caution. This plugin has not … Continue reading →<\/span><\/a><\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"closed","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[5],"tags":[],"class_list":["post-12","post","type-post","status-publish","format-standard","hentry","category-rails"],"_links":{"self":[{"href":"https:\/\/www.schuerig.de\/michael\/blog\/index.php\/wp-json\/wp\/v2\/posts\/12","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.schuerig.de\/michael\/blog\/index.php\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.schuerig.de\/michael\/blog\/index.php\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.schuerig.de\/michael\/blog\/index.php\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/www.schuerig.de\/michael\/blog\/index.php\/wp-json\/wp\/v2\/comments?post=12"}],"version-history":[{"count":1,"href":"https:\/\/www.schuerig.de\/michael\/blog\/index.php\/wp-json\/wp\/v2\/posts\/12\/revisions"}],"predecessor-version":[{"id":149,"href":"https:\/\/www.schuerig.de\/michael\/blog\/index.php\/wp-json\/wp\/v2\/posts\/12\/revisions\/149"}],"wp:attachment":[{"href":"https:\/\/www.schuerig.de\/michael\/blog\/index.php\/wp-json\/wp\/v2\/media?parent=12"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.schuerig.de\/michael\/blog\/index.php\/wp-json\/wp\/v2\/categories?post=12"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.schuerig.de\/michael\/blog\/index.php\/wp-json\/wp\/v2\/tags?post=12"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}