use uniqueness for complete db check

corrlib/find.py

@@ -6,8 +6,12 @@ import numpy as np
 from .input.implementations import codes
 from .tools import k2m, get_db_file
 from .tracker import get
-from typing import Any, Optional
+from .integrity import check_time_validity
+from typing import Any, Optional, Union
 from pathlib import Path
+import datetime as dt
+from collections.abc import Callable
+import warnings
 
 
 def _project_lookup_by_alias(db: Path, alias: str) -> str:
@@ -38,7 +42,7 @@ def _project_lookup_by_alias(db: Path, alias: str) -> str:
     return str(results[0][0])
 
 
-def _project_lookup_by_id(db: Path, uuid: str) -> list[tuple[str, str]]:
+def _project_lookup_by_id(db: Path, uuid: str) -> list[tuple[str, ...]]:
     """
     Return the project information available in the database by UUID.
 
@@ -62,8 +66,56 @@ def _project_lookup_by_id(db: Path, uuid: str) -> list[tuple[str, str]]:
     return results
 
 
-def _db_lookup(db: Path, ensemble: str, correlator_name: str, code: str, project: Optional[str]=None,  parameters: Optional[str]=None,
-               created_before: Optional[str]=None, created_after: Optional[Any]=None, updated_before: Optional[Any]=None, updated_after: Optional[Any]=None) -> pd.DataFrame:
+def _time_filter(results: pd.DataFrame,  created_before: Optional[str]=None, created_after: Optional[Any]=None, updated_before: Optional[Any]=None, updated_after: Optional[Any]=None) -> pd.DataFrame:
+    """
+    Filter the results from the database in terms of the creation and update times.
+
+    Parameters
+    ----------
+    results: pd.DataFrame
+        The dataframe holding the unfilteres results from the database.
+    created_before: str
+        Contraint on the creation date in datetime.datetime.isoformat. Note that this is exclusive. The creation date has to be truly before the date and time given.
+    created_after: str
+        Contraint on the creation date in datetime.datetime.isoformat. Note that this is exclusive. The creation date has to be truly after the date and time given.
+    updated_before: str
+        Contraint on the creation date in datetime.datetime.isoformat. Note that this is exclusive. The date of the last update has to be truly before the date and time given.
+    updated_after: str
+        Contraint on the creation date in datetime.datetime.isoformat. Note that this is exclusive. The date of the last update has to be truly after the date and time given.
+    """
+    drops = []
+    for ind in range(len(results)):
+        result = results.iloc[ind]
+        created_at = dt.datetime.fromisoformat(result['created_at'])
+        updated_at = dt.datetime.fromisoformat(result['updated_at'])
+        db_times_valid = check_time_validity(created_at=created_at, updated_at=updated_at)
+        if not db_times_valid:
+            raise ValueError('Time stamps not valid for result with path', result["path"])
+
+        if created_before is not None:
+            date_created_before = dt.datetime.fromisoformat(created_before)
+            if date_created_before < created_at:
+                drops.append(ind)
+                continue
+        if created_after is not None:
+            date_created_after = dt.datetime.fromisoformat(created_after)
+            if date_created_after > created_at:
+                drops.append(ind)
+                continue
+        if updated_before is not None:
+            date_updated_before = dt.datetime.fromisoformat(updated_before)
+            if date_updated_before < updated_at:
+                drops.append(ind)
+                continue
+        if updated_after is not None:
+            date_updated_after = dt.datetime.fromisoformat(updated_after)
+            if date_updated_after > updated_at:
+                drops.append(ind)
+                continue
+    return results.drop(drops)
+
+
+def _db_lookup(db: Path, ensemble: str, correlator_name: str, code: str, project: Optional[str]=None,  parameters: Optional[str]=None) -> pd.DataFrame:
     """
     Look up a correlator record in the database by the data given to the method.
 
@@ -105,20 +157,84 @@ def _db_lookup(db: Path, ensemble: str, correlator_name: str, code: str, project
         search_expr += f" AND code = '{code}'"
     if parameters:
         search_expr += f" AND parameters = '{parameters}'"
-    if created_before:
-        search_expr += f" AND created_at < '{created_before}'"
-    if created_after:
-        search_expr += f" AND created_at > '{created_after}'"
-    if updated_before:
-        search_expr += f" AND updated_at < '{updated_before}'"
-    if updated_after:
-        search_expr += f" AND updated_at > '{updated_after}'"
     conn = sqlite3.connect(db)
     results = pd.read_sql(search_expr, conn)
     conn.close()
     return results
 
 
+def _sfcf_drop(param: dict[str, Any], **kwargs: Any) -> bool:
+        if 'offset' in kwargs:
+            if kwargs.get('offset') != param['offset']:
+                return True
+        if 'quark_kappas' in kwargs:
+            kappas = kwargs['quark_kappas']
+            if (not np.isclose(kappas[0], param['quarks'][0]['mass']) or not np.isclose(kappas[1], param['quarks'][1]['mass'])):
+                return True
+        if 'quark_masses' in kwargs:
+            masses = kwargs['quark_masses']
+            if (not np.isclose(masses[0], k2m(param['quarks'][0]['mass'])) or not np.isclose(masses[1], k2m(param['quarks'][1]['mass']))):
+                return True
+        if 'qk1' in kwargs:
+            quark_kappa1 = kwargs['qk1']
+            if not isinstance(quark_kappa1, list):
+                if (not np.isclose(quark_kappa1, param['quarks'][0]['mass'])):
+                    return True
+            else:
+                if len(quark_kappa1) == 2:
+                    if (quark_kappa1[0] > param['quarks'][0]['mass']) or (quark_kappa1[1] < param['quarks'][0]['mass']):
+                        return True
+                else:
+                    raise ValueError("quark_kappa1 has to have length 2")
+        if 'qk2' in kwargs:
+            quark_kappa2 = kwargs['qk2']
+            if not isinstance(quark_kappa2, list):
+                if (not np.isclose(quark_kappa2, param['quarks'][1]['mass'])):
+                    return True
+            else:
+                if len(quark_kappa2) == 2:
+                    if (quark_kappa2[0] > param['quarks'][1]['mass']) or (quark_kappa2[1] < param['quarks'][1]['mass']):
+                        return True
+                else:
+                    raise ValueError("quark_kappa2 has to have length 2")
+        if 'qm1' in kwargs:
+            quark_mass1 = kwargs['qm1']
+            if not isinstance(quark_mass1, list):
+                if (not np.isclose(quark_mass1, k2m(param['quarks'][0]['mass']))):
+                    return True
+            else:
+                if len(quark_mass1) == 2:
+                    if (quark_mass1[0] > k2m(param['quarks'][0]['mass'])) or (quark_mass1[1] < k2m(param['quarks'][0]['mass'])):
+                        return True
+                else:
+                    raise ValueError("quark_mass1 has to have length 2")
+        if 'qm2' in kwargs:
+            quark_mass2 = kwargs['qm2']
+            if not isinstance(quark_mass2, list):
+                if (not np.isclose(quark_mass2, k2m(param['quarks'][1]['mass']))):
+                    return True
+            else:
+                if len(quark_mass2) == 2:
+                    if (quark_mass2[0] > k2m(param['quarks'][1]['mass'])) or (quark_mass2[1] < k2m(param['quarks'][1]['mass'])):
+                        return True
+                else:
+                    raise ValueError("quark_mass2 has to have length 2")
+        if 'quark_thetas' in kwargs:
+            quark_thetas = kwargs['quark_thetas']
+            if (quark_thetas[0] != param['quarks'][0]['thetas'] and quark_thetas[1] != param['quarks'][1]['thetas']) or (quark_thetas[0] != param['quarks'][1]['thetas'] and quark_thetas[1] != param['quarks'][0]['thetas']):
+                return True
+        # careful, this is not save, when multiple contributions are present!
+        if 'wf1' in kwargs:
+            wf1 = kwargs['wf1']
+            if not (np.isclose(wf1[0][0], param['wf1'][0][0], 1e-8) and np.isclose(wf1[0][1][0], param['wf1'][0][1][0], 1e-8) and np.isclose(wf1[0][1][1], param['wf1'][0][1][1], 1e-8)):
+                return True
+        if 'wf2' in kwargs:
+            wf2 = kwargs['wf2']
+            if not (np.isclose(wf2[0][0], param['wf2'][0][0], 1e-8) and np.isclose(wf2[0][1][0], param['wf2'][0][1][0], 1e-8) and np.isclose(wf2[0][1][1], param['wf2'][0][1][1], 1e-8)):
+                return True
+        return False
+
+
 def sfcf_filter(results: pd.DataFrame, **kwargs: Any) -> pd.DataFrame:
     r"""
     Filter method for the Database entries holding SFCF calculations.
@@ -136,9 +252,9 @@ def sfcf_filter(results: pd.DataFrame, **kwargs: Any) -> pd.DataFrame:
     qk2: float, optional
         Mass parameter $\kappa_2$ of the first quark.
     qm1: float, optional
-        Bare quak mass $m_1$ of the first quark.
+        Bare quark mass $m_1$ of the first quark.
     qm2: float, optional
-        Bare quak mass $m_1$ of the first quark.
+        Bare quark mass $m_2$ of the first quark.
     quarks_thetas: list[list[float]], optional
     wf1: optional
     wf2: optional
@@ -148,101 +264,80 @@ def sfcf_filter(results: pd.DataFrame, **kwargs: Any) -> pd.DataFrame:
     results: pd.DataFrame
         The filtered DataFrame, only holding the records that fit to the parameters given.
     """
+
     drops = []
     for ind in range(len(results)):
         result = results.iloc[ind]
         param = json.loads(result['parameters'])
-        if 'offset' in kwargs:
-            if kwargs.get('offset') != param['offset']:
+        if _sfcf_drop(param, **kwargs):
             drops.append(ind)
-                continue
-        if 'quark_kappas' in kwargs:
-            kappas = kwargs['quark_kappas']
-            if (not np.isclose(kappas[0], param['quarks'][0]['mass']) or not np.isclose(kappas[1], param['quarks'][1]['mass'])):
-                drops.append(ind)
-                continue
-        if 'quark_masses' in kwargs:
-            masses = kwargs['quark_masses']
-            if (not np.isclose(masses[0], k2m(param['quarks'][0]['mass'])) or not np.isclose(masses[1], k2m(param['quarks'][1]['mass']))):
-                drops.append(ind)
-                continue
-        if 'qk1' in kwargs:
-            quark_kappa1 = kwargs['qk1']
-            if not isinstance(quark_kappa1, list):
-                if (not np.isclose(quark_kappa1, param['quarks'][0]['mass'])):
-                    drops.append(ind)
-                    continue
-            else:
-                if len(quark_kappa1) == 2:
-                    if (quark_kappa1[0] > param['quarks'][0]['mass']) or (quark_kappa1[1] < param['quarks'][0]['mass']):
-                        drops.append(ind)
-                        continue
-        if 'qk2' in kwargs:
-            quark_kappa2 = kwargs['qk2']
-            if not isinstance(quark_kappa2, list):
-                if (not np.isclose(quark_kappa2, param['quarks'][1]['mass'])):
-                    drops.append(ind)
-                    continue
-            else:
-                if len(quark_kappa2) == 2:
-                    if (quark_kappa2[0] > param['quarks'][1]['mass']) or (quark_kappa2[1] < param['quarks'][1]['mass']):
-                        drops.append(ind)
-                        continue
-        if 'qm1' in kwargs:
-            quark_mass1 = kwargs['qm1']
-            if not isinstance(quark_mass1, list):
-                if (not np.isclose(quark_mass1, k2m(param['quarks'][0]['mass']))):
-                    drops.append(ind)
-                    continue
-            else:
-                if len(quark_mass1) == 2:
-                    if (quark_mass1[0] > k2m(param['quarks'][0]['mass'])) or (quark_mass1[1] < k2m(param['quarks'][0]['mass'])):
-                        drops.append(ind)
-                        continue
-        if 'qm2' in kwargs:
-            quark_mass2 = kwargs['qm2']
-            if not isinstance(quark_mass2, list):
-                if (not np.isclose(quark_mass2, k2m(param['quarks'][1]['mass']))):
-                    drops.append(ind)
-                    continue
-            else:
-                if len(quark_mass2) == 2:
-                    if (quark_mass2[0] > k2m(param['quarks'][1]['mass'])) or (quark_mass2[1] < k2m(param['quarks'][1]['mass'])):
-                        drops.append(ind)
-                        continue
-        if 'quark_thetas' in kwargs:
-            quark_thetas = kwargs['quark_thetas']
-            if (quark_thetas[0] != param['quarks'][0]['thetas'] and quark_thetas[1] != param['quarks'][1]['thetas']) or (quark_thetas[0] != param['quarks'][1]['thetas'] and quark_thetas[1] != param['quarks'][0]['thetas']):
-                drops.append(ind)
-                continue
-        # careful, this is not save, when multiple contributions are present!
-        if 'wf1' in kwargs:
-            wf1 = kwargs['wf1']
-            if not (np.isclose(wf1[0][0], param['wf1'][0][0], 1e-8) and np.isclose(wf1[0][1][0], param['wf1'][0][1][0], 1e-8) and np.isclose(wf1[0][1][1], param['wf1'][0][1][1], 1e-8)):
-                drops.append(ind)
-                continue
-        if 'wf2' in kwargs:
-            wf2 = kwargs['wf2']
-            if not (np.isclose(wf2[0][0], param['wf2'][0][0], 1e-8) and np.isclose(wf2[0][1][0], param['wf2'][0][1][0], 1e-8) and np.isclose(wf2[0][1][1], param['wf2'][0][1][1], 1e-8)):
-                drops.append(ind)
-                continue
     return results.drop(drops)
 
 
+def openQCD_filter(results:pd.DataFrame, **kwargs: Any) -> pd.DataFrame:
+    """
+    Filter for parameters of openQCD.
+
+    Parameters
+    ----------
+    results: pd.DataFrame
+        The unfiltered list of results from the database.
+
+    Returns
+    -------
+    results: pd.DataFrame
+        The filtered results.
+
+    """
+    warnings.warn("A filter for openQCD parameters is no implemented yet.", Warning)
+
+    return results
+
+
+def _code_filter(results: pd.DataFrame, code: str, **kwargs: Any) -> pd.DataFrame:
+    """
+    Abstraction of the filters for the different codes that are available.
+    At the moment, only openQCD and SFCF are known.
+    The possible key words for the parameters can be seen in the descriptionso f the code-specific filters.
+
+    Parameters
+    ----------
+    results: pd.DataFrame
+        The unfiltered list of results from the database.
+    code: str
+        The name of the code that produced the record at hand.
+    kwargs:
+        The keyworkd args that are handed over to the code-specific filters.
+
+    Returns
+    -------
+    results: pd.DataFrame
+        The filtered results.
+    """
+    if code == "sfcf":
+        return sfcf_filter(results, **kwargs)
+    elif code == "openQCD":
+        return openQCD_filter(results, **kwargs)
+    else:
+        raise ValueError(f"Code {code} is not known.")
+
+
 def find_record(path: Path, ensemble: str, correlator_name: str, code: str, project: Optional[str]=None, parameters: Optional[str]=None,
-                created_before: Optional[str]=None, created_after: Optional[str]=None, updated_before: Optional[str]=None, updated_after: Optional[str]=None, revision: Optional[str]=None, **kwargs: Any) -> pd.DataFrame:
+                created_before: Optional[str]=None, created_after: Optional[str]=None, updated_before: Optional[str]=None, updated_after: Optional[str]=None,
+                revision: Optional[str]=None,
+                customFilter: Optional[Callable[[pd.DataFrame], pd.DataFrame]] = None,
+                **kwargs: Any) -> pd.DataFrame:
     db_file = get_db_file(path)
     db = path / db_file
     if code not in codes:
         raise ValueError("Code " + code + "unknown, take one of the following:" + ", ".join(codes))
     get(path, db_file)
-    results = _db_lookup(db, ensemble, correlator_name,code, project, parameters=parameters, created_before=created_before, created_after=created_after, updated_before=updated_before, updated_after=updated_after)
-    if code == "sfcf":
-        results = sfcf_filter(results, **kwargs)
-    elif code == "openQCD":
-        pass
-    else:
-        raise Exception
+    results = _db_lookup(db, ensemble, correlator_name,code, project, parameters=parameters)
+    if any([arg is not None for arg in [created_before, created_after, updated_before, updated_after]]):
+        results = _time_filter(results, created_before, created_after, updated_before, updated_after)
+    results = _code_filter(results, code, **kwargs)
+    if customFilter is not None:
+        results =  customFilter(results)
     print("Found " + str(len(results)) + " result" + ("s" if len(results)>1 else ""))
     return results.reset_index()
 

corrlib/integrity.py

@@ -0,0 +1,45 @@
+import datetime as dt
+from pathlib import Path
+from .tools import get_db_file
+import pandas as pd
+import sqlite3
+
+
+def has_valid_times(result: pd.DataFrame) -> bool:
+    # we expect created_at <= updated_at <= now
+    created_at = dt.datetime.fromisoformat(result['created_at'])
+    updated_at = dt.datetime.fromisoformat(result['updated_at'])
+    if created_at > updated_at:
+        return False
+    if updated_at > dt.datetime.now():
+        return False
+    return True
+
+def are_keys_unique(db: Path, table: str, col: str) -> bool:
+    conn = sqlite3.connect(db)
+    c = conn.cursor()
+    c.execute(f"SELECT COUNT( DISTINCT CAST(path AS nvarchar(4000))), COUNT({col}) FROM {table};")
+    results = c.fetchall()[0]
+    conn.close()
+    return bool(results[0] == results[1])
+
+
+def check_db_integrity(path: Path) -> None:
+    db = get_db_file(path)
+
+    if not are_keys_unique(db, 'backlogs', 'path'):
+        raise Exception("The paths the backlog table of the database links are not unique.")
+
+    search_expr = "SELECT * FROM 'backlogs'"
+    conn = sqlite3.connect(db)
+    results = pd.read_sql(search_expr, conn)
+
+    for result in results:
+        if not has_valid_times(result):
+            raise ValueError(f"Result with id {result[id]} has wrong time signatures.")
+
+
+
+def full_integrity_check(path: Path) -> None:
+    check_db_integrity(path)
+

tests/find_test.py

@@ -3,6 +3,8 @@ 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:
@@ -10,6 +12,7 @@ def make_sql(path: Path) -> Path:
     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)
@@ -31,3 +34,381 @@ def test_find_lookup_by_one_alias(tmp_path: Path) -> None:
     with pytest.raises(Exception):
         assert uuid == find._project_lookup_by_alias(db, "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_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')
+
+
+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_sfcf_drop() -> None:
+    parameters0 = {
+                    'offset': [0,0,0],
+                    'quarks': [{'mass': 1, 'thetas': [0,0,0]}, {'mass': 2, 'thetas': [0,0,1]}], # m0s = -3.5, -3.75
+                    'wf1': [[1, [0, 0]], [0.5, [1, 0]], [.75, [.5, .5]]],
+                    'wf2': [[1, [2, 1]], [2, [0.5, -0.5]], [.5, [.75, .72]]],
+                   }
+
+    assert not find._sfcf_drop(parameters0, offset=[0,0,0])
+    assert find._sfcf_drop(parameters0, offset=[1,0,0])
+
+    assert not find._sfcf_drop(parameters0, quark_kappas = [1, 2])
+    assert find._sfcf_drop(parameters0, quark_kappas = [-3.1, -3.72])
+
+    assert not find._sfcf_drop(parameters0, quark_masses = [-3.5, -3.75])
+    assert find._sfcf_drop(parameters0, quark_masses = [-3.1, -3.72])
+
+    assert not find._sfcf_drop(parameters0, qk1 = 1)
+    assert not find._sfcf_drop(parameters0, qk2 = 2)
+    assert find._sfcf_drop(parameters0, qk1 = 2)
+    assert find._sfcf_drop(parameters0, qk2 = 1)
+
+    assert not find._sfcf_drop(parameters0, qk1 = [0.5,1.5])
+    assert not find._sfcf_drop(parameters0, qk2 = [1.5,2.5])
+    assert find._sfcf_drop(parameters0, qk1 = 2)
+    assert find._sfcf_drop(parameters0, qk2 = 1)
+    with pytest.raises(ValueError):
+        assert not find._sfcf_drop(parameters0, qk1 = [0.5,1,5])
+    with pytest.raises(ValueError):
+        assert not find._sfcf_drop(parameters0, qk2 = [1,5,2.5])
+
+    assert find._sfcf_drop(parameters0, qm1 = 1.2)
+    assert find._sfcf_drop(parameters0, qm2 = 2.2)
+    assert not find._sfcf_drop(parameters0, qm1 = -3.5)
+    assert not find._sfcf_drop(parameters0, qm2 = -3.75)
+
+    assert find._sfcf_drop(parameters0, qm2 = 1.2)
+    assert find._sfcf_drop(parameters0, qm1 = 2.2)
+    with pytest.raises(ValueError):
+        assert not find._sfcf_drop(parameters0, qm1 = [0.5,1,5])
+    with pytest.raises(ValueError):
+        assert not find._sfcf_drop(parameters0, qm2 = [1,5,2.5])
+
+
+def test_openQCD_filter() -> None:
+    record_0 = ["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']
+    record_1 = ["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']
+    record_2 = ["f_P", "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']
+    record_3 = ["f_P", "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']
+    data = [
+        record_0,
+        record_1,
+        record_2,
+        record_3,
+    ]
+    cols = ["name",
+            "ensemble",
+            "code",
+            "path",
+            "project",
+            "parameters",
+            "parameter_file",
+            "created_at",
+            "updated_at"]
+    df = pd.DataFrame(data,columns=cols)
+
+    with pytest.warns(Warning):
+        find.openQCD_filter(df, a = "asdf")
+
+
+def test_code_filter() -> None:
+    record_0 = ["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']
+    record_1 = ["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']
+    record_2 = ["f_P", "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']
+    record_3 = ["f_P", "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']
+    record_4 = ["f_A", "ensA", "openQCD", "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']
+    record_5 = ["f_A", "ensA", "openQCD", "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']
+    record_6 = ["f_P", "ensA", "openQCD", "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']
+    record_7 = ["f_P", "ensA", "openQCD", "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']
+    record_8 = ["f_P", "ensA", "openQCD", "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']
+    data = [
+        record_0,
+        record_1,
+        record_2,
+        record_3,
+    ]
+    cols = ["name",
+            "ensemble",
+            "code",
+            "path",
+            "project",
+            "parameters",
+            "parameter_file",
+            "created_at",
+            "updated_at"]
+    df = pd.DataFrame(data,columns=cols)
+
+    res = find._code_filter(df, "sfcf")
+    assert len(res) == 4
+
+    data = [
+        record_4,
+        record_5,
+        record_6,
+        record_7,
+        record_8,
+    ]
+    cols = ["name",
+            "ensemble",
+            "code",
+            "path",
+            "project",
+            "parameters",
+            "parameter_file",
+            "created_at",
+            "updated_at"]
+    df = pd.DataFrame(data,columns=cols)
+
+    res = find._code_filter(df, "openQCD")
+    assert len(res) == 5
+    with pytest.raises(ValueError):
+        res = find._code_filter(df, "asdf")
+
+
+def test_find_record() -> None:
+    assert True
+
+
+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_list_projects(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))
+    uuid = "test_uuid2"
+    alias_str = "fun_project2"
+    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))
+    uuid = "test_uuid3"
+    alias_str = "fun_project3"
+    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))
+    uuid = "test_uuid4"
+    alias_str = "fun_project4"
+    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()
+    results = find.list_projects(tmp_path)
+    assert len(results) == 4
+    for i in range(4):
+        assert len(results[i]) == 2