# -*- coding: utf-8 -*
# Demonstrate the use of non-monotonic algebraic operators
# RA query:
# "Find those LEGO sets that do not contain any stickers."
from DB1 import *
# relevant LEGO database tables (NB: small variants)
sets = Table('sets-small.csv')
contains = Table('contains-small.csv')
bricks = Table('bricks-small.csv')
# runs in about 70s
# formulation based on set difference
# (1) Identify the LEGO bricks with stickers
q1 = select(lambda t:t['name'].find('Sticker') >= 0,
bricks)
# (2) Find the sets that contain these LEGO bricks
q2 = project(['set'],
natjoin(
contains,
q1))
# (3) These are exactly those sets that do *not* interest us
q3 = difference(project(['set'], sets), q2)
# (4) Attach set name and other set information of interest
q4 = project(['set','name'], natjoin(sets, q3))
# Algebraic plan:
#
# π[set,name]
# |
# ⋈
# / \
# sets -
# / \
# π[set] π[set]
# | |
# sets ⋈
# / \
# contains σ[name ~ 'Sticker']
# |
# bricks
# q = project(['set','name'],
# natjoin(
# sets,
# difference(
# project(['set'], sets),
# project(['set'],
# natjoin(
# contains,
# select(lambda t:find(t['name'], 'Sticker') >= 0,
# bricks))))))
# print output
for row in q4:
print(row)