Listing posts

Displaying posts 271 - 275 of 328 in total
Fast random number generation
mouse 980 · person cloud · link
Linux
Tools
Last update
2017-08-17
2017
08-17
« — »

Here is some empirical data from my notebook, the last command is the fastest:

1
2
3
4
5
6
7
8
9
10
# ~40MB/s | OpenSSL rand
openssl rand 1000000000 | pv > /dev/null

# ~175MB/s | /dev/urandom
pv < /dev/urandom > /dev/null

# ~1200MB/s | OpenSSL enc /dev/zero
openssl enc -aes-256-ctr -nosalt \
  -pass pass:"$(dd if=/dev/urandom bs=128 count=1 2>/dev/null | base64)" \
   < /dev/zero | pv > /dev/null

Source: OpenSSL rand, OpenSSL enc /dev/zero


~~~ * ~~~

Ruby backtrace for SystemStackError
mouse 1867 · person cloud · link
Debug
Dev
HowTo
Ruby
Last update
2017-07-31
2017
07-31
«how to debug SystemStackError: stack level too deep»

In case of a SystemStackError: stack level too deep exception you do not have a backtrace to inspect, but set_trace_func comes in help!

  • put this in an initializer:
1
2
3
4
5
6
7
8
9
10
11
12
13
# /config/initializers/ruby_tracing.rb
if ENV['RUBY_TRACING']
  $rb_tracing_enable    = false # set to true just before the desired method

  $rb_tracing_file      = File.open(Rails.root.join('rb_tracing.log'), 'w')
  $rb_tracing_file.sync = true

  set_trace_func(proc{|event, file, line, id, binding, classname|
    if $rb_tracing_enable && event == 'call'
      $rb_tracing_file.puts "#{file}:#{line} #{classname}##{id}"
    end
  })
end
  • start the webserver with the RUBY_TRACING env var initialized:
1
RUBY_TRACING=true rails s
  • set the $rb_tracing_enable flag var near the code that explodes:
1
2
$rb_tracing_enable = true
a_method_that_causes_infinite_recursion_in_a_not_obvious_way()
  • inspect the /rb_tracing.log file
1
grep project_path rb_tracing.log

Source: stackoverflow, jbgo's gist


~~~ * ~~~

Ruby zip file in memory
mouse 2608 · person cloud · link
Dev
Ruby
Last update
2017-07-31
2017
07-31
«zip & gzip compression»

GZIP/GUNZIP a string:

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
class String
  # http://ruby-doc.org/stdlib-2.2.0/libdoc/zlib/rdoc/Zlib/GzipWriter.html
  # https://gist.github.com/sinisterchipmunk/1335041
  def gzipped(fname = 'file.txt')
    gz_file = StringIO.new

    z = Zlib::GzipWriter.new(gz_file)
    z.mtime     = Time.now # for gzip compat
    z.orig_name = fname    # for gzip compat
    z.write self
    z.close

    gz_file.string
  end # gzipped

  def gunzipped
    gz_file = StringIO.new self
    z = Zlib::GzipReader.new(gz_file)
    z.read
  end # gunzipped
end

Create a ZIP archive on the fly:

1
2
3
4
5
6
7
require 'zip' # https://github.com/rubyzip/rubyzip

# Zip.default_compression = Zlib::BEST_COMPRESSION

Zip::File.open("archive.zip", Zip::File::CREATE) do |zipfile|
  zipfile.get_output_stream("hello.txt"){|f| f.write "hello world" }
end

~~~ * ~~~

Rails associations: where clause with hash parameters
mouse 1906 · person cloud · link
ActiveRecord
DB
Dev
HowTo
Rails
Last update
2017-07-29
2017
07-29
«Rails AREL "where" magic! Stop using hardcoded SQL!»

Given this sample class with two relations on the same table:

1
2
3
4
def Edge < ActiveRecord::Base
    belongs_to :first , class_name: 'Node'
    belongs_to :second, class_name: 'Node'
end

If you want to query edges by the first node, you could do this:

1
Edge.joins(:first).where(nodes: {value: 1})

1
2
3
SELECT "edges".*
FROM "edges" INNER JOIN "nodes" ON "nodes"."id" = "edges"."first_id"
WHERE "nodes"."value" = 1

But if you have to query using both nodes, you can still use joins like this:

1
Edge.joins(:first, :second).where(nodes: {value: 1}, seconds_edges: {value: 2})

1
2
3
4
5
6
SELECT "edges".*
FROM "edges"
  INNER JOIN "nodes" ON "nodes"."id" = "edges"."first_id"
  INNER JOIN "nodes" "seconds_edges" ON "seconds_edges"."id" = "edges"."second_id"
WHERE "nodes"."value" = 1
  AND "seconds_edges"."value" = 2

Source: Stackoverflow


~~~ * ~~~

Rails association data: preload, eager_load, includes, join
mouse 2035 · person cloud · link
ActiveRecord
DB
Rails
Reference
Last update
2017-07-27
2017
07-27
«recap on the four methods available on rails to query/load association data»

preload

Loads data in a separate query (we cannot place conditions on the association):

1
User.preload(:posts)

1
2
SELECT users.* FROM users
SELECT posts.* FROM posts WHERE posts.user_id IN (1)


eager_load

Loads data in a single LEFT OUTER JOIN query (we can place conditions on the association):

1
User.eager_load(:posts)

1
2
3
SELECT users.id AS t0_r0, users.name AS t0_r1   , users....,
       posts.id AS t1_r0, posts.user_id AS t1_r1, posts....
FROM users LEFT OUTER JOIN posts ON posts.user_id = users.id


includes

Loads data in a separate or single query based on conditions:

1
2
User.includes(:posts)
User.includes(:posts).where(posts: {title: 'hello'})

1
2
3
4
5
6
7
8
9
-- 1st case - without conditions
SELECT users.* FROM users
SELECT posts.* FROM posts WHERE posts.user_id IN (1)

-- 2nd case - with conditions
SELECT users.id AS t0_r0, users.name AS t0_r1   , users....,
       posts.id AS t1_r0, posts.user_id AS t1_r1, posts....
FROM users LEFT OUTER JOIN posts ON posts.user_id = users.id
WHERE posts.title = 'hello'


join

Does not load any data, just uses an INNER JOIN query:

1
User.joins(:posts)

1
SELECT users.* FROM users INNER JOIN posts ON posts.user_id = users.id

Source: BigBinary