davetapley

#!/bin/bash

SANDBOX_DIR="cabal-dev"
SDIST_DIR="./sdist/"
SDIST_SUFFIX=".tar.gz"

PACKAGES_CONF_DIR=$SANDBOX_DIR"/packages-"`ghc --numeric-version`".conf/"

find $SDIST_DIR -name *$SDIST_SUFFIX -print0 | while read -d $'\0' sdist_path
do

davetapley

require 'rspec'
require 'savon_spec'

RSpec.configure do |config|
  config.include Savon::Spec::Macros
end

describe 'with client' do

  let(:client) do

davetapley

SELECT nspname || '.' || relname AS "relation",
    pg_size_pretty(pg_relation_size(C.oid)) AS "size"
  FROM pg_class C
  LEFT JOIN pg_namespace N ON (N.oid = C.relnamespace)
  WHERE nspname NOT IN ('pg_catalog', 'information_schema')
  ORDER BY pg_relation_size(C.oid) DESC
  LIMIT 20;

davetapley

OMITTED NORMAL MAIL HEADERS

X-Github-Reason: comment
List-Archive: https://github.com/dylankeil/chronos
List-Unsubscribe: <mailto:unsub+i-XXXXXXXXXXXXXXXXXXXXXXXXXXXXXxx@reply.github.com>
List-Id: dylankeil/chronos <chronos.dylankeil.github.com>
List-Post: <mailto:reply+i-XXXXXXXXXXXXXXXXXXXXXXXXXXXXXX@reply.github.com>
X-Github-Recipient: dukedave

davetapley

=> VACUUM VERBOSE location_events;
INFO:  vacuuming "public.location_events"
INFO:  scanned index "location_events_pkey" to remove 3882346 row versions
DETAIL:  CPU 0.91s/7.16u sec elapsed 13.93 sec.
INFO:  scanned index "index_location_events_on_user_id" to remove 3882346 row versions
DETAIL:  CPU 1.68s/9.01u sec elapsed 21.61 sec.
INFO:  scanned index "index_location_events_on_user_id_where_not_archived" to remove 3882346 row versions
DETAIL:  CPU 0.00s/1.70u sec elapsed 1.66 sec.
INFO:  "location_events": removed 3882346 row versions in 92228 pages
DETAIL:  CPU 0.21s/1.98u sec elapsed 8.32 sec.

davetapley

(13:27:40) dukedave: Lazy post: Any comments on my question here, the single answer, and my comment on that answer: http://dba.stackexchange.com/questions/25890/does-updating-a-partially-indexed-field-add-or-remove-it-from-an-index-in-postgr
(13:29:05) RhodiumToad: dukedave: your answer seems correct
(13:29:19) supplicant: I think dukedave is the commenter?
(13:29:31) dukedave: Yes, I'm the asker and commenter :)
(13:29:48) RhodiumToad: dukedave: ok, so the answer you received is correct :-)
(13:29:50) dukedave: The answer certainly makes sense, but I'm still seeing the partial index grow
(13:29:56) dukedave: Which seems in consistent 
(13:30:19) RhodiumToad: dukedave: no, not really. it'll depend on the exact pattern of updates. here's why:
(13:30:44) dukedave: The purpose of making it partial was that it only has non-archived rows in it, which should be a reasonably fixed number
(13:30:44) RhodiumToad: dukedave: I'm assuming we're inserting in strictly increasing order of ids, since that's what usually happ

davetapley

// ==UserScript==
// @name        Mittens
// @namespace   http://userscripts.org/users/489276
// @include     http*
// @version     1
// @grant       None
// ==/UserScript==


// Thanks to: http://stackoverflow.com/a/5494405/21115http://stackoverflow.com/a/5494405/21115

davetapley

; <<>> DiG 9.8.1-P1 <<>> git.tddium.com
;; global options: +cmd
;; Got answer:
;; ->>HEADER<<- opcode: QUERY, status: NOERROR, id: 51959
;; flags: qr rd ra; QUERY: 1, ANSWER: 3, AUTHORITY: 0, ADDITIONAL: 0

;; QUESTION SECTION:
;git.tddium.com.                        IN      A

;; ANSWER SECTION:

davetapley

# Normal has_many / belongs_to:
>> a = c.permitted_ips
=> []

# What we got back was just an Array:
>> a.class
=> Array

# Now build a new record:
>> c.permitted_ips.build :address => '123123'

davetapley

class Range

  def overlaps?(other)
    (self.first <= other.last) and (other.first <= self.last)
  end

end