Models:
Philip Bordner
philipB/
ehrql
0
Downloads: 1
[e988c2]: / tests / autocomplete / autocomplete_definition.py

Download this file

309 lines (276 with data), 13.7 kB 

  1
  2
  3
  4
  5
  6
  7
  8
  9
 10
 11
 12
 13
 14
 15
 16
 17
 18
 19
 20
 21
 22
 23
 24
 25
 26
 27
 28
 29
 30
 31
 32
 33
 34
 35
 36
 37
 38
 39
 40
 41
 42
 43
 44
 45
 46
 47
 48
 49
 50
 51
 52
 53
 54
 55
 56
 57
 58
 59
 60
 61
 62
 63
 64
 65
 66
 67
 68
 69
 70
 71
 72
 73
 74
 75
 76
 77
 78
 79
 80
 81
 82
 83
 84
 85
 86
 87
 88
 89
 90
 91
 92
 93
 94
 95
 96
 97
 98
 99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
from ehrql import days, maximum_of, minimum_of, weeks

from ehrql.tables import core, emis, tpp





# A file to keep track of things where we get good autocomplete behaviour.

#

# Anything in this file is automatically checked by the test file test_autocomplete.py,

# though it must be formatted correctly or the tests will fail. Currently you can:

#

# - write a single command and confirm the type e.g.:

#     core.patients.date_of_birth  ## type:DatePatientSeries

#

# - assign a single command to a variable and confirm the type e.g.:

#     bool_invert = ~core.patients.exists_for_patient()  ## type:BoolPatientSeries

#

# - either of the above but spanning multiple lines e.g.

#     bool_and = (

#       core.patients.exists_for_patient() & core.patients.exists_for_patient()

#     )  ## type:BoolPatientSeries



# Currently all columns on all tables have type of "Series | Any",

# so we don't get autocomplete. Providing type hints on the Series

# class, and on the @table decorator means the column types are

# correct e.g. IntPatientSeries, DateEventSeries etc. and therefore

# we get autocomplete for all the properties and methods on each series.

core.patients.date_of_birth  ## type:DatePatientSeries

core.patients.sex  ## type:StrPatientSeries

core.ons_deaths.underlying_cause_of_death  ## type:CodePatientSeries



core.clinical_events.date  ## type:DateEventSeries

core.clinical_events.numeric_value  ## type:FloatEventSeries

core.clinical_events.snomedct_code  ## type:CodeEventSeries

tpp.apcs.all_diagnoses  ## type:MultiCodeStringEventSeries



tpp.addresses.has_postcode  ## type:BoolEventSeries

tpp.addresses.address_id  ## type:IntEventSeries





# There are some methods that always return the same type

core.clinical_events.snomedct_code.count_distinct_for_patient()  ## type:IntPatientSeries

core.clinical_events.numeric_value.mean_for_patient()  ## type:FloatPatientSeries

core.clinical_events.date.count_episodes_for_patient(weeks(1))  ## type:IntPatientSeries

core.patients.exists_for_patient()  ## type:BoolPatientSeries

core.patients.count_for_patient()  ## type:IntPatientSeries

bool_eq = days(100) == days(100)  ## type:bool

bool_neq = days(100) != days(100)  ## type:bool



# There are some things that return the same type as the calling

# object or one of the arguments



bool_and = (

    core.patients.exists_for_patient() & core.patients.exists_for_patient()

)  ## type:BoolPatientSeries

bool_or = (

    core.patients.exists_for_patient() | core.patients.exists_for_patient()

)  ## type:BoolPatientSeries

bool_invert = ~core.patients.exists_for_patient()  ## type:BoolPatientSeries

numeric_neg = -core.clinical_events.numeric_value  ## type:FloatEventSeries

core.clinical_events.date.to_first_of_year()  ## type:DateEventSeries

core.patients.date_of_birth.to_first_of_month()  ## type:DatePatientSeries

duration_negation = -days(100)  ## type:days

core.patients.sex.when_null_then(core.patients.sex)  ## type:StrPatientSeries

duration_add = days(100) + core.patients.date_of_birth  ## type:DatePatientSeries

duration_radd = core.clinical_events.date + days(100)  ## type:DateEventSeries

duration_add_duration = days(100) + days(100)  ## type:days

# duration_rsub = core.clinical_events.date - days(100)  ## type: DateEventSeries

# !!! the above doesn't work and thinks its a DateDifference. I assume

# !!! because the NotImplemented on the DateFunctions __sub__ method

# !!! is only known at runtime

# !!! commenting out until we fix it

duration_sub_duration = days(100) - days(100)  ## type: days





# There are loads of things that return a series where the typ

# (int, str, float etc.) is fixed, but it can be a PatientSeries

# or an EventSeries depending on whether the calling object is a

# PatientSeries or an EventSeries. This can be achieved with two

# overloaded methods and type hints



#

# BaseSeries

#

base_eq = core.patients.sex == core.patients.sex  ## type:BoolPatientSeries

base_ne = (

    core.clinical_events.date != core.clinical_events.date

)  ## type:BoolEventSeries

core.patients.sex.is_null()  ## type:BoolPatientSeries

core.clinical_events.date.is_not_null()  ## type:BoolEventSeries

core.patients.sex.is_in([])  ## type:BoolPatientSeries

core.clinical_events.snomedct_code.is_not_in([])  ## type:BoolEventSeries



#

# ComparableFunctions

#

comparable_lt = (

    core.clinical_events.numeric_value < core.clinical_events.numeric_value

)  ## type:BoolEventSeries

comparable_le = (

    core.clinical_events.numeric_value <= core.clinical_events.numeric_value

)  ## type:BoolEventSeries

comparable_gt = (

    core.clinical_events.numeric_value > core.clinical_events.numeric_value

)  ## type:BoolEventSeries

comparable_ge = (

    core.clinical_events.numeric_value >= core.clinical_events.numeric_value

)  ## type:BoolEventSeries



#

# StrFunctions

#

core.patients.sex.contains("m")  ## type:BoolPatientSeries



#

# NumericFunctions

#

numeric_truediv = core.clinical_events.numeric_value / 10  ## type:FloatEventSeries

numeric_rtruediv = 10 / core.clinical_events.numeric_value  ## type:FloatEventSeries

numeric_floordiv = core.clinical_events.numeric_value // 10  ## type:IntEventSeries

numeric_rfloordiv = 10 // core.clinical_events.numeric_value  ## type:IntEventSeries

core.clinical_events.numeric_value.as_int()  ## type:IntEventSeries

core.clinical_events.numeric_value.as_float()  ## type:FloatEventSeries



#

# DateFunctions

#

date_str = "2024-01-01"

core.patients.date_of_birth.is_before(date_str)  ## type:BoolPatientSeries

core.patients.date_of_birth.is_on_or_before(date_str)  ## type:BoolPatientSeries

core.patients.date_of_birth.is_after(date_str)  ## type:BoolPatientSeries

core.patients.date_of_birth.is_on_or_after(date_str)  ## type:BoolPatientSeries

core.clinical_events.date.is_between_but_not_on(

    date_str, date_str

)  ## type:BoolEventSeries

core.clinical_events.date.is_on_or_between(date_str, date_str)  ## type:BoolEventSeries

core.clinical_events.date.is_during((date_str, date_str))  ## type:BoolEventSeries





#

# MultiCodeStringFunctions

#

tpp.apcs.all_diagnoses.contains("N13")  ## type:BoolEventSeries

tpp.apcs.all_diagnoses.contains_any_of(["N13"])  ## type:BoolEventSeries



#

# Couple of random list[tuple] types

starting_on = weeks(3).starting_on("2000-01-01")[0][0]  ## type:date

ending_on = weeks(3).ending_on("2000-01-01")[0][0]  ## type:date



#

# Things that aggregate from EventSeries to PatientSeries

# but that need to maintain the type (int, float, bool etc)

core.clinical_events.numeric_value.sum_for_patient()  ## type:FloatPatientSeries

core.clinical_events.numeric_value.as_int().sum_for_patient()  ## type:IntPatientSeries



core.clinical_events.numeric_value.minimum_for_patient()  ## type:FloatPatientSeries

core.clinical_events.numeric_value.as_int().minimum_for_patient()  ## type:IntPatientSeries

tpp.addresses.msoa_code.minimum_for_patient()  ## type:StrPatientSeries

core.clinical_events.date.minimum_for_patient()  ## type:DatePatientSeries



core.clinical_events.numeric_value.maximum_for_patient()  ## type:FloatPatientSeries

core.clinical_events.numeric_value.as_int().maximum_for_patient()  ## type:IntPatientSeries

tpp.addresses.msoa_code.maximum_for_patient()  ## type:StrPatientSeries

core.clinical_events.date.maximum_for_patient()  ## type:DatePatientSeries



#

# NumericFunctions which maintain the series (Event or Patient)

# and the type (int or float)



numeric_add = core.clinical_events.numeric_value + 10  ## type:FloatEventSeries

numeric_radd = 10 + core.clinical_events.numeric_value.as_int()  ## type:IntEventSeries

numeric_add_patient = (

    core.clinical_events.numeric_value.maximum_for_patient() + 10

)  ## type:FloatPatientSeries

numeric_radd_patient = (

    10 + core.clinical_events.numeric_value.as_int().maximum_for_patient()

)  ## type:IntPatientSeries

numeric_add_series = (

    core.clinical_events.numeric_value + core.clinical_events.numeric_value

)  ## type:FloatEventSeries



numeric_sub = core.clinical_events.numeric_value - 10  ## type:FloatEventSeries

numeric_rsub = 10 - core.clinical_events.numeric_value.as_int()  ## type:IntEventSeries

numeric_sub_patient = (

    core.clinical_events.numeric_value.maximum_for_patient() - 10

)  ## type:FloatPatientSeries

numeric_rsub_patient = (

    10 - core.clinical_events.numeric_value.as_int().maximum_for_patient()

)  ## type:IntPatientSeries

numeric_sub_series = (

    core.clinical_events.numeric_value - core.clinical_events.numeric_value

)  ## type:FloatEventSeries



numeric_mul = core.clinical_events.numeric_value * 10  ## type:FloatEventSeries

numeric_rmul = 10 * core.clinical_events.numeric_value.as_int()  ## type:IntEventSeries

numeric_mul_patient = (

    core.clinical_events.numeric_value.maximum_for_patient() * 10

)  ## type:FloatPatientSeries

numeric_rmul_patient = (

    10 * core.clinical_events.numeric_value.as_int().maximum_for_patient()

)  ## type:IntPatientSeries

numeric_mul_series = (

    core.clinical_events.numeric_value * core.clinical_events.numeric_value

)  ## type:FloatEventSeries



#

# Horizontal aggregations

# The type checker casts eveything to the first series. But the only

# type we can easily get is the first arg. So if the first thing is

# a series then that's fine. Otherwise we ignore

#

max_of_float = maximum_of(

    core.clinical_events.numeric_value, 10

)  ## type:FloatEventSeries

max_of_int = maximum_of(

    core.clinical_events.numeric_value.as_int(), 10

)  ## type:IntEventSeries

max_of_date = maximum_of(

    core.clinical_events.date, "2024-01-01"

)  ## type:DateEventSeries

max_of_float_patient = maximum_of(

    core.clinical_events.numeric_value.maximum_for_patient(), 10

)  ## type:FloatPatientSeries

max_of_int_patient = maximum_of(

    core.clinical_events.numeric_value.maximum_for_patient().as_int(), 10

)  ## type:IntPatientSeries

max_of_date_patient = maximum_of(

    core.patients.date_of_birth, "2024-01-01"

)  ## type:DatePatientSeries

min_of_float = minimum_of(

    core.clinical_events.numeric_value, 10

)  ## type:FloatEventSeries

min_of_int = minimum_of(

    core.clinical_events.numeric_value.as_int(), 10

)  ## type:IntEventSeries

min_of_date = minimum_of(

    core.clinical_events.date, "2024-01-01"

)  ## type:DateEventSeries

min_of_float_patient = minimum_of(

    core.clinical_events.numeric_value.minimum_for_patient(), 10

)  ## type:FloatPatientSeries

min_of_int_patient = minimum_of(

    core.clinical_events.numeric_value.minimum_for_patient().as_int(), 10

)  ## type:IntPatientSeries

min_of_date_patient = minimum_of(

    core.patients.date_of_birth, "2024-01-01"

)  ## type:DatePatientSeries



# properties

core.patients.date_of_birth.day  ## type:IntPatientSeries

core.patients.date_of_birth.month  ## type:IntPatientSeries

core.patients.date_of_birth.year  ## type:IntPatientSeries

core.clinical_events.date.day  ## type:IntEventSeries

core.clinical_events.date.month  ## type:IntEventSeries

core.clinical_events.date.year  ## type:IntEventSeries



# Table methods not yet tested

tpp.patients.is_alive_on(date_str)  ## type:BoolPatientSeries

tpp.patients.is_dead_on(date_str)  ## type:BoolPatientSeries

tpp.decision_support_values.electronic_frailty_index()  ## type:decision_support_values

tpp.practice_registrations.spanning_with_systmone(

    date_str, date_str

)  ## type:practice_registrations

emis.patients.has_practice_registration_spanning(

    date_str, date_str

)  ## type:BoolPatientSeries

core.patients.is_alive_on(date_str)  ## type:BoolPatientSeries

core.patients.is_dead_on(date_str)  ## type:BoolPatientSeries

core.practice_registrations.exists_for_patient_on(date_str)  ## type:BoolPatientSeries

core.practice_registrations.spanning(date_str, date_str)  ## type:practice_registrations



# query_language series methods not yet tested

core.clinical_events.date.is_after(date_str)  ## type: BoolEventSeries

core.clinical_events.date.is_before(date_str)  ## type: BoolEventSeries

core.clinical_events.date.is_on_or_after(date_str)  ## type: BoolEventSeries

core.clinical_events.date.is_on_or_before(date_str)  ## type: BoolEventSeries

core.clinical_events.date.to_first_of_month()  ## type: DateEventSeries

core.clinical_events.snomedct_code.is_in([])  ## type: BoolEventSeries

core.clinical_events.snomedct_code.is_null()  ## type: BoolEventSeries

core.clinical_events.snomedct_code.is_null().as_int()  ## type: IntEventSeries

core.clinical_events.snomedct_code.when_null_then(3)  ## type: CodeEventSeries

core.patients.date_of_birth.day.as_float()  ## type: FloatPatientSeries

core.patients.date_of_birth.day.as_int()  ## type: IntPatientSeries

core.patients.date_of_birth.is_between_but_not_on(

    date_str, date_str

)  ## type: BoolPatientSeries

core.patients.date_of_birth.is_during((date_str, date_str))  ## type: BoolPatientSeries

core.patients.date_of_birth.is_on_or_between(

    date_str, date_str

)  ## type: BoolPatientSeries

core.patients.date_of_birth.to_first_of_year()  ## type: DatePatientSeries

core.patients.sex.is_not_in(["male"])  ## type: BoolPatientSeries

core.patients.sex.is_not_null()  ## type: BoolPatientSeries

core.patients.sex.is_null().as_int()  ## type: IntPatientSeries

tpp.apcs.all_diagnoses.is_in([])  ## type:NoReturn

tpp.apcs.all_diagnoses.is_not_in([])  ## type:NoReturn

tpp.addresses.msoa_code.contains([])  ## type: BoolEventSeries



# query_language non-series methods

core.clinical_events.where(

    core.clinical_events.snomedct_code.is_in([])

)  ## type:clinical_events

core.clinical_events.except_where(

    core.clinical_events.snomedct_code.is_in([])

)  ## type:clinical_events



# Duration methods

days(100).starting_on("2045-01-01")  ## type: list[tuple[date, date]]

days(100).ending_on("2045-01-01")  ## type: list[tuple[date, date]]