move transpose words to AutomationEngine

create tests for transpose words
This commit is contained in:
smilerz 2023-08-30 20:44:40 -05:00
parent 39253cfd02
commit b8317c2c29
No known key found for this signature in database
GPG Key ID: 39444C7606D47126
5 changed files with 124 additions and 92 deletions

View File

@ -1,5 +1,6 @@
import re
from django.core.cache import caches
from django.db.models import Q
from django.db.models.functions import Lower
from cookbook.models import Automation
@ -136,6 +137,38 @@ class AutomationEngine:
return tokens
def apply_transpose_automation(self, string):
"""
If two words (param_1 & param_2) are detected in sequence, swap their position in the ingredient string
:param 1: first word to detect
:param 2: second word to detect
return: new ingredient string
"""
if self.use_cache and self.transpose_words is None:
self.transpose_words = {}
TRANSPOSE_WORDS_CACHE_KEY = f'automation_transpose_words_{self.request.space.pk}'
if c := caches['default'].get(TRANSPOSE_WORDS_CACHE_KEY, None):
self.transpose_words = c
caches['default'].touch(TRANSPOSE_WORDS_CACHE_KEY, 30)
else:
i = 0
for a in Automation.objects.filter(space=self.request.space, disabled=False, type=Automation.TRANSPOSE_WORDS).only('param_1', 'param_2').order_by('order').all():
self.transpose_words[i] = [a.param_1.lower(), a.param_2.lower()]
i += 1
caches['default'].set(TRANSPOSE_WORDS_CACHE_KEY, self.transpose_words, 30)
else:
self.transpose_words = {}
tokens = [x.lower() for x in string.replace(',', ' ').split()]
if self.transpose_words:
for key, value in self.transpose_words.items():
if value[0] in tokens and value[1] in tokens:
string = re.sub(rf"\b({value[0]})\W*({value[1]})\b", r"\2 \1", string, flags=re.IGNORECASE)
else:
for rule in Automation.objects.filter(space=self.request.space, type=Automation.TRANSPOSE_WORDS, disabled=False) \
.annotate(param_1_lower=Lower('param_1'), param_2_lower=Lower('param_2')) \
.filter(param_1_lower__in=tokens, param_2_lower__in=tokens).order_by('order'):
if rule.param_1 in tokens and rule.param_2 in tokens:
string = re.sub(rf"\b({rule.param_1})\W*({rule.param_2})\b", r"\2 \1", string, flags=re.IGNORECASE)
return string
def apply_regex_replace_automation(self, string):

View File

@ -2,12 +2,8 @@ import re
import string
import unicodedata
from django.core.cache import caches
from django.db.models import Q
from django.db.models.functions import Lower
from cookbook.helper.automation_helper import AutomationEngine
from cookbook.models import Automation, Food, Ingredient, Unit
from cookbook.models import Food, Ingredient, Unit
class IngredientParser:
@ -16,7 +12,7 @@ class IngredientParser:
# food_aliases = {}
# unit_aliases = {}
# never_unit = {}
transpose_words = {}
# transpose_words = {}
automation = None
def __init__(self, request, cache_mode=True, ignore_automations=False):
@ -29,7 +25,7 @@ class IngredientParser:
self.request = request
self.ignore_rules = ignore_automations
self.automation = AutomationEngine(self.request, use_cache=cache_mode)
if cache_mode:
# if cache_mode:
# FOOD_CACHE_KEY = f'automation_food_alias_{self.request.space.pk}'
# if c := caches['default'].get(FOOD_CACHE_KEY, None):
# self.food_aliases = c
@ -59,21 +55,21 @@ class IngredientParser:
# caches['default'].set(NEVER_UNIT_CACHE_KEY, self.never_unit, 30)
# TODO migrated to automation engine
TRANSPOSE_WORDS_CACHE_KEY = f'automation_transpose_words_{self.request.space.pk}'
if c := caches['default'].get(TRANSPOSE_WORDS_CACHE_KEY, None):
self.transpose_words = c
caches['default'].touch(TRANSPOSE_WORDS_CACHE_KEY, 30)
else:
i = 0
for a in Automation.objects.filter(space=self.request.space, disabled=False, type=Automation.TRANSPOSE_WORDS).only('param_1', 'param_2').order_by('order').all():
self.transpose_words[i] = [a.param_1.lower(), a.param_2.lower()]
i += 1
caches['default'].set(TRANSPOSE_WORDS_CACHE_KEY, self.transpose_words, 30)
else:
# TRANSPOSE_WORDS_CACHE_KEY = f'automation_transpose_words_{self.request.space.pk}'
# if c := caches['default'].get(TRANSPOSE_WORDS_CACHE_KEY, None):
# self.transpose_words = c
# caches['default'].touch(TRANSPOSE_WORDS_CACHE_KEY, 30)
# else:
# i = 0
# for a in Automation.objects.filter(space=self.request.space, disabled=False, type=Automation.TRANSPOSE_WORDS).only('param_1', 'param_2').order_by('order').all():
# self.transpose_words[i] = [a.param_1.lower(), a.param_2.lower()]
# i += 1
# caches['default'].set(TRANSPOSE_WORDS_CACHE_KEY, self.transpose_words, 30)
# else:
# self.food_aliases = {}
# self.unit_aliases = {}
# self.never_unit = {}
self.transpose_words = {}
# self.transpose_words = {}
# def apply_food_automation(self, food):
# """
@ -277,33 +273,32 @@ class IngredientParser:
# return tokens
# TODO migrated to automation engine
def apply_transpose_words_automations(self, ingredient):
"""
If two words (param_1 & param_2) are detected in sequence, swap their position in the ingredient string
:param 1: first word to detect
:param 2: second word to detect
return: new ingredient string
"""
# def apply_transpose_words_automations(self, ingredient):
# """
# If two words (param_1 & param_2) are detected in sequence, swap their position in the ingredient string
# :param 1: first word to detect
# :param 2: second word to detect
# return: new ingredient string
# """
if self.ignore_rules:
return ingredient
# if self.ignore_rules:
# return ingredient
else:
tokens = [x.lower() for x in ingredient.replace(',', ' ').split()]
if self.transpose_words:
filtered_rules = {}
for key, value in self.transpose_words.items():
if value[0] in tokens and value[1] in tokens:
filtered_rules[key] = value
for k, v in filtered_rules.items():
ingredient = re.sub(rf"\b({v[0]})\W*({v[1]})\b", r"\2 \1", ingredient, flags=re.IGNORECASE)
else:
for rule in Automation.objects.filter(space=self.request.space, type=Automation.TRANSPOSE_WORDS, disabled=False) \
.annotate(param_1_lower=Lower('param_1'), param_2_lower=Lower('param_2')) \
.filter(Q(Q(param_1_lower__in=tokens) | Q(param_2_lower__in=tokens))).order_by('order'):
ingredient = re.sub(rf"\b({rule.param_1})\W*({rule.param_1})\b", r"\2 \1", ingredient, flags=re.IGNORECASE)
return ingredient
# else:
# tokens = [x.lower() for x in ingredient.replace(',', ' ').split()]
# if self.transpose_words:
# filtered_rules = {}
# for key, value in self.transpose_words.items():
# if value[0] in tokens and value[1] in tokens:
# filtered_rules[key] = value
# for k, v in filtered_rules.items():
# ingredient = re.sub(rf"\b({v[0]})\W*({v[1]})\b", r"\2 \1", ingredient, flags=re.IGNORECASE)
# else:
# for rule in Automation.objects.filter(space=self.request.space, type=Automation.TRANSPOSE_WORDS, disabled=False) \
# .annotate(param_1_lower=Lower('param_1'), param_2_lower=Lower('param_2')) \
# .filter(Q(Q(param_1_lower__in=tokens) | Q(param_2_lower__in=tokens))).order_by('order'):
# ingredient = re.sub(rf"\b({rule.param_1})\W*({rule.param_1})\b", r"\2 \1", ingredient, flags=re.IGNORECASE)
# return ingredient
def parse(self, ingredient):
"""
@ -345,8 +340,8 @@ class IngredientParser:
if re.match('([0-9])+([A-z])+\\s', ingredient):
ingredient = re.sub(r'(?<=([a-z])|\d)(?=(?(1)\d|[a-z]))', ' ', ingredient)
# TODO migrated to automation engine
ingredient = self.apply_transpose_words_automations(ingredient)
if not self.ignore_rules:
ingredient = self.automation.apply_transpose_words_automations(ingredient)
tokens = ingredient.split() # split at each space into tokens
if len(tokens) == 1:
@ -360,7 +355,7 @@ class IngredientParser:
# three arguments if it already has a unit there can't be
# a fraction for the amount
if len(tokens) > 2:
# TODO migrated to automation engine
if not self.ignore_rules:
tokens = self.automation.apply_never_unit_automation(tokens)
try:
if unit is not None:

View File

@ -2,7 +2,6 @@ import re
import traceback
from html import unescape
from django.core.cache import caches
from django.utils.dateparse import parse_duration
from django.utils.translation import gettext as _
from isodate import parse_duration as iso_parse_duration

View File

@ -6,16 +6,6 @@ from django_scopes import scope
from cookbook.helper.automation_helper import AutomationEngine
from cookbook.models import Automation
# TODO test case sensitive match, assert update value
# TODO test case insensitive match, assert update value
# TODO test no match, assert not update value
# TODO test accent insensitive match, assert not update value
# @pytest.fixture()
# def automation_food(space_1):
# return Keyword.objects.get_or_create(name='test_1', space=space_1)[0]
@pytest.mark.parametrize("arg", [
['Match', True],
@ -103,8 +93,24 @@ def test_never_unit_automation(u1_s1, arg):
assert automation.apply_never_unit_automation(arg[0]) == arg[2]
# def test_transpose_automation():
# assert True == True
@pytest.mark.parametrize("arg", [
['second first', 'first second'],
['longer string second first longer string', 'longer string first second longer string'],
['second fails first', 'second fails first'],
])
def test_transpose_automation(u1_s1, arg):
user = auth.get_user(u1_s1)
space = user.userspace_set.first().space
request = RequestFactory()
request.user = user
request.space = space
automation = AutomationEngine(request, False)
with scope(space=space):
Automation.objects.get_or_create(name='transpose words test', type=Automation.TRANSPOSE_WORDS, param_1='second', param_2='first', created_by=user, space=space)
assert automation.apply_transpose_automation(arg[0]) == arg[1]
assert True == True
# # for some reason this tests cant run due to some kind of encoding issue, needs to be fixed
# # def test_description_replace_automation(u1_s1, space_1):
# # if 'cookbook' in os.getcwd():

View File

@ -1,4 +1,3 @@
import pytest
from django.contrib import auth
from django.test import RequestFactory
from django_scopes import scope