240 lines
10 KiB
Python
240 lines
10 KiB
Python
import corrlib.find as find
|
|
import sqlite3
|
|
from pathlib import Path
|
|
import corrlib.initialization as cinit
|
|
import pytest
|
|
import pandas as pd
|
|
import datalad.api as dl
|
|
|
|
|
|
def make_sql(path: Path) -> Path:
|
|
db = path / "test.db"
|
|
cinit._create_db(db)
|
|
return db
|
|
|
|
|
|
def test_find_lookup_by_one_alias(tmp_path: Path) -> None:
|
|
db = make_sql(tmp_path)
|
|
conn = sqlite3.connect(db)
|
|
c = conn.cursor()
|
|
uuid = "test_uuid"
|
|
alias_str = "fun_project"
|
|
tag_str = "tt"
|
|
owner = "tester"
|
|
code = "test_code"
|
|
c.execute("INSERT INTO projects (id, aliases, customTags, owner, code, created_at, updated_at) VALUES (?, ?, ?, ?, ?, datetime('now'), datetime('now'))",
|
|
(uuid, alias_str, tag_str, owner, code))
|
|
conn.commit()
|
|
assert uuid == find._project_lookup_by_alias(db, "fun_project")
|
|
uuid = "test_uuid2"
|
|
alias_str = "fun_project"
|
|
c.execute("INSERT INTO projects (id, aliases, customTags, owner, code, created_at, updated_at) VALUES (?, ?, ?, ?, ?, datetime('now'), datetime('now'))",
|
|
(uuid, alias_str, tag_str, owner, code))
|
|
conn.commit()
|
|
with pytest.raises(Exception):
|
|
assert uuid == find._project_lookup_by_alias(db, "fun_project")
|
|
conn.close()
|
|
|
|
|
|
def test_find_project(tmp_path: Path) -> None:
|
|
cinit.create(tmp_path)
|
|
db = tmp_path / "backlogger.db"
|
|
dl.unlock(str(db), dataset=str(tmp_path))
|
|
conn = sqlite3.connect(db)
|
|
c = conn.cursor()
|
|
uuid = "test_uuid"
|
|
alias_str = "fun_project"
|
|
tag_str = "tt"
|
|
owner = "tester"
|
|
code = "test_code"
|
|
c.execute("INSERT INTO projects (id, aliases, customTags, owner, code, created_at, updated_at) VALUES (?, ?, ?, ?, ?, datetime('now'), datetime('now'))",
|
|
(uuid, alias_str, tag_str, owner, code))
|
|
conn.commit()
|
|
|
|
assert uuid == find.find_project(tmp_path, "fun_project")
|
|
|
|
uuid = "test_uuid2"
|
|
alias_str = "fun_project"
|
|
c.execute("INSERT INTO projects (id, aliases, customTags, owner, code, created_at, updated_at) VALUES (?, ?, ?, ?, ?, datetime('now'), datetime('now'))",
|
|
(uuid, alias_str, tag_str, owner, code))
|
|
conn.commit()
|
|
|
|
with pytest.raises(Exception):
|
|
assert uuid == find._project_lookup_by_alias(tmp_path, "fun_project")
|
|
conn.close()
|
|
|
|
|
|
def test_find_lookup_by_id(tmp_path: Path) -> None:
|
|
db = make_sql(tmp_path)
|
|
conn = sqlite3.connect(db)
|
|
c = conn.cursor()
|
|
uuid = "test_uuid"
|
|
alias_str = "fun_project"
|
|
tag_str = "tt"
|
|
owner = "tester"
|
|
code = "test_code"
|
|
c.execute("INSERT INTO projects (id, aliases, customTags, owner, code, created_at, updated_at) VALUES (?, ?, ?, ?, ?, datetime('now'), datetime('now'))",
|
|
(uuid, alias_str, tag_str, owner, code))
|
|
conn.commit()
|
|
conn.close()
|
|
result = find._project_lookup_by_id(db, uuid)[0]
|
|
assert uuid == result[0]
|
|
assert alias_str == result[1]
|
|
assert tag_str == result[2]
|
|
assert owner == result[3]
|
|
assert code == result[4]
|
|
|
|
|
|
def test_db_lookup(tmp_path: Path) -> None:
|
|
db = make_sql(tmp_path)
|
|
conn = sqlite3.connect(db)
|
|
c = conn.cursor()
|
|
|
|
corr = "f_A"
|
|
ensemble = "SF_A"
|
|
code = "openQCD"
|
|
meas_path = "archive/SF_A/f_A/Project_A.json.gz::asdfasdfasdf"
|
|
uuid = "Project_A"
|
|
pars = "{par_A: 3.0, par_B: 5.0}"
|
|
parameter_file = "projects/Project_A/myinput.in"
|
|
c.execute("INSERT INTO backlogs (name, ensemble, code, path, project, parameters, parameter_file, created_at, updated_at) VALUES (?, ?, ?, ?, ?, ?, ?, datetime('now'), datetime('now'))",
|
|
(corr, ensemble, code, meas_path, uuid, pars, parameter_file))
|
|
conn.commit()
|
|
|
|
results = find._db_lookup(db, ensemble, corr, code)
|
|
assert len(results) == 1
|
|
results = find._db_lookup(db, "SF_B", corr, code)
|
|
assert results.empty
|
|
results = find._db_lookup(db, ensemble, "g_A", code)
|
|
assert results.empty
|
|
results = find._db_lookup(db, ensemble, corr, "sfcf")
|
|
assert results.empty
|
|
results = find._db_lookup(db, ensemble, corr, code, project = "Project_A")
|
|
assert len(results) == 1
|
|
results = find._db_lookup(db, ensemble, corr, code, project = "Project_B")
|
|
assert results.empty
|
|
results = find._db_lookup(db, ensemble, corr, code, parameters = pars)
|
|
assert len(results) == 1
|
|
results = find._db_lookup(db, ensemble, corr, code, parameters = "{par_A: 3.0, par_B: 4.0}")
|
|
assert results.empty
|
|
|
|
corr = "g_A"
|
|
ensemble = "SF_A"
|
|
code = "openQCD"
|
|
meas_path = "archive/SF_A/f_A/Project_A.json.gz::asdfasdfasdf"
|
|
uuid = "Project_A"
|
|
pars = "{par_A: 3.0, par_B: 4.0}"
|
|
parameter_file = "projects/Project_A/myinput.in"
|
|
c.execute("INSERT INTO backlogs (name, ensemble, code, path, project, parameters, parameter_file, created_at, updated_at) VALUES (?, ?, ?, ?, ?, ?, ?, datetime('now'), datetime('now'))",
|
|
(corr, ensemble, code, meas_path, uuid, pars, parameter_file))
|
|
conn.commit()
|
|
|
|
corr = "f_A"
|
|
results = find._db_lookup(db, ensemble, corr, code)
|
|
assert len(results) == 1
|
|
results = find._db_lookup(db, "SF_B", corr, code)
|
|
assert results.empty
|
|
results = find._db_lookup(db, ensemble, "g_A", code)
|
|
assert len(results) == 1
|
|
results = find._db_lookup(db, ensemble, corr, "sfcf")
|
|
assert results.empty
|
|
results = find._db_lookup(db, ensemble, corr, code, project = "Project_A")
|
|
assert len(results) == 1
|
|
results = find._db_lookup(db, ensemble, "g_A", code, project = "Project_A")
|
|
assert len(results) == 1
|
|
results = find._db_lookup(db, ensemble, corr, code, project = "Project_B")
|
|
assert results.empty
|
|
results = find._db_lookup(db, ensemble, "g_A", code, project = "Project_B")
|
|
assert results.empty
|
|
results = find._db_lookup(db, ensemble, corr, code, parameters = pars)
|
|
assert results.empty
|
|
results = find._db_lookup(db, ensemble, "g_A", code, parameters = "{par_A: 3.0, par_B: 4.0}")
|
|
assert len(results) == 1
|
|
|
|
conn.close()
|
|
|
|
|
|
def test_time_filter() -> None:
|
|
record_A = ["f_A", "ensA", "sfcf", "archive/SF_A/f_A/Project_A.json.gz::asdfasdfasdf", "SF_A", '{par_A: 5.0, par_B: 5.0}', "projects/SF_A/input.in",
|
|
'2025-03-26 12:55:18.229966', '2025-03-26 12:55:18.229966'] # only created
|
|
record_B = ["f_A", "ensA", "sfcf", "archive/SF_A/f_A/Project_A.json.gz::asdfasdfasdf", "SF_A", '{par_A: 5.0, par_B: 5.0}', "projects/SF_A/input.in",
|
|
'2025-03-26 12:55:18.229966', '2025-04-26 12:55:18.229966'] # created and updated
|
|
record_C = ["f_A", "ensA", "sfcf", "archive/SF_A/f_A/Project_A.json.gz::asdfasdfasdf", "SF_A", '{par_A: 5.0, par_B: 5.0}', "projects/SF_A/input.in",
|
|
'2026-03-26 12:55:18.229966', '2026-05-26 12:55:18.229966'] # created and updated later
|
|
record_D = ["f_A", "ensA", "sfcf", "archive/SF_A/f_A/Project_A.json.gz::asdfasdfasdf", "SF_A", '{par_A: 5.0, par_B: 5.0}', "projects/SF_A/input.in",
|
|
'2026-03-26 12:55:18.229966', '2026-03-27 12:55:18.229966']
|
|
record_E = ["f_A", "ensA", "sfcf", "archive/SF_A/f_A/Project_A.json.gz::asdfasdfasdf", "SF_A", '{par_A: 5.0, par_B: 5.0}', "projects/SF_A/input.in",
|
|
'2024-03-26 12:55:18.229966', '2024-03-26 12:55:18.229966'] # only created, earlier
|
|
record_F = ["f_A", "ensA", "sfcf", "archive/SF_A/f_A/Project_A.json.gz::asdfasdfasdf", "SF_A", '{par_A: 5.0, par_B: 5.0}', "projects/SF_A/input.in",
|
|
'2026-03-26 12:55:18.229966', '2024-03-26 12:55:18.229966'] # this is invalid...
|
|
|
|
data = [record_A, record_B, record_C, record_D, record_E]
|
|
cols = ["name",
|
|
"ensemble",
|
|
"code",
|
|
"path",
|
|
"project",
|
|
"parameters",
|
|
"parameter_file",
|
|
"created_at",
|
|
"updated_at"]
|
|
df = pd.DataFrame(data,columns=cols)
|
|
|
|
results = find._time_filter(df, created_before='2023-03-26 12:55:18.229966')
|
|
assert results.empty
|
|
results = find._time_filter(df, created_before='2027-03-26 12:55:18.229966')
|
|
assert len(results) == 5
|
|
results = find._time_filter(df, created_before='2026-03-25 12:55:18.229966')
|
|
assert len(results) == 3
|
|
results = find._time_filter(df, created_before='2026-03-26 12:55:18.229965')
|
|
assert len(results) == 3
|
|
results = find._time_filter(df, created_before='2025-03-04 12:55:18.229965')
|
|
assert len(results) == 1
|
|
|
|
results = find._time_filter(df, created_after='2023-03-26 12:55:18.229966')
|
|
assert len(results) == 5
|
|
results = find._time_filter(df, created_after='2027-03-26 12:55:18.229966')
|
|
assert results.empty
|
|
results = find._time_filter(df, created_after='2026-03-25 12:55:18.229966')
|
|
assert len(results) == 2
|
|
results = find._time_filter(df, created_after='2026-03-26 12:55:18.229965')
|
|
assert len(results) == 2
|
|
results = find._time_filter(df, created_after='2025-03-04 12:55:18.229965')
|
|
assert len(results) == 4
|
|
|
|
results = find._time_filter(df, updated_before='2023-03-26 12:55:18.229966')
|
|
assert results.empty
|
|
results = find._time_filter(df, updated_before='2027-03-26 12:55:18.229966')
|
|
assert len(results) == 5
|
|
results = find._time_filter(df, updated_before='2026-03-25 12:55:18.229966')
|
|
assert len(results) == 3
|
|
results = find._time_filter(df, updated_before='2026-03-26 12:55:18.229965')
|
|
assert len(results) == 3
|
|
results = find._time_filter(df, updated_before='2025-03-04 12:55:18.229965')
|
|
assert len(results) == 1
|
|
|
|
results = find._time_filter(df, updated_after='2023-03-26 12:55:18.229966')
|
|
assert len(results) == 5
|
|
results = find._time_filter(df, updated_after='2027-03-26 12:55:18.229966')
|
|
assert results.empty
|
|
results = find._time_filter(df, updated_after='2026-03-25 12:55:18.229966')
|
|
assert len(results) == 2
|
|
results = find._time_filter(df, updated_after='2026-03-26 12:55:18.229965')
|
|
assert len(results) == 2
|
|
results = find._time_filter(df, updated_after='2025-03-04 12:55:18.229965')
|
|
assert len(results) == 4
|
|
|
|
data = [record_A, record_B, record_C, record_D, record_F]
|
|
cols = ["name",
|
|
"ensemble",
|
|
"code",
|
|
"path",
|
|
"project",
|
|
"parameters",
|
|
"parameter_file",
|
|
"created_at",
|
|
"updated_at"]
|
|
df = pd.DataFrame(data,columns=cols)
|
|
|
|
with pytest.raises(ValueError):
|
|
results = find._time_filter(df, created_before='2023-03-26 12:55:18.229966')
|