Merge pull request 'feat/minteg' (#39) from feat/minteg into develop

Reviewed-on: #39

corrlib/cli.py

@@ -108,7 +108,7 @@ def find(
     ),
     ) -> None:
     """
-    Find a record in the backlog at hand. Through specifying it's ensemble and the measured correlator.
+    Find a record in the given backlog.
     """
     results = find_record(path, ensemble, corr, code)
     if results.empty:
@@ -147,6 +147,9 @@ def check(path: Path = typer.Option(
         "-d",
         ),
     ) -> None:
+    """
+    Check the integrity of the repository.
+    """
     full_integrity_check(path)
 
 

corrlib/integrity.py

@@ -1,15 +1,34 @@
 import datetime as dt
 from pathlib import Path
-from .tools import get_db_file
+from .tools import get_db_file, CONFIG_FILENAME
 import pandas as pd
 import sqlite3
 from .tracker import get
 import pyerrors.input.json as pj
+import os
+from configparser import ConfigParser
 
 from typing import Any
 
 
+path_opts = ['db', 'projects_path', 'archive_path', 'toml_imports_path', 'import_scripts_path']
+
+
 def has_valid_times(result: pd.Series) -> bool:
+    """
+    Check, whether the result at hand has time-stamps that are sensible:
+    A recored is created first, then updated, with both times laying in the past.
+
+    Parameters
+    ----------
+    result: pd.Series
+        The result to check
+
+    Returns
+    -------
+    b: bool
+        True, if the timestamps make sense.
+    """
     # we expect created_at <= updated_at <= now
     created_at = dt.datetime.fromisoformat(result['created_at'])
     updated_at = dt.datetime.fromisoformat(result['updated_at'])
@@ -20,15 +39,41 @@ def has_valid_times(result: pd.Series) -> bool:
     return True
 
 def are_keys_unique(db: Path, table: str, col: str) -> bool:
+    """
+    Check whether the strings listed in a column of a given table are unique.
+
+    Parameters
+    ----------
+    db: Path
+        The database to check.
+    table: str
+        The table to check.
+    col: str
+        The column to be checked for uniqueness.
+
+    Returns
+    -------
+    b: bool
+        True, if the strings are unique.
+    """
     conn = sqlite3.connect(db)
     c = conn.cursor()
-    c.execute(f"SELECT COUNT( DISTINCT CAST(path AS nvarchar(4000))), COUNT({col}) FROM {table};")
+    c.execute(f"SELECT COUNT( DISTINCT CAST({col} 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:
+    """
+    Check intergrity of the database by checking the uniqueness of the record keys used to load the records
+    and ensuring that the timestamps of each record is sensible. Throws an error, if issues are detected.
+
+    Parameters
+    ----------
+    path: Path
+        Path to the backlog-library to check.
+    """
     db = get_db_file(path)
 
     if not are_keys_unique(path / db, 'backlogs', 'path'):
@@ -41,11 +86,21 @@ def check_db_integrity(path: Path) -> None:
     for _, result in results.iterrows():
         if not has_valid_times(result):
             raise ValueError(f"Result with id {result[id]} has wrong time signatures.")
-    print("DB:\t✅")
     return
 
 
 def _check_db2paths(path: Path, meas_paths: list[str]) -> None:
+    """
+    Check whether for each record in the given by meas_paths, we can find the data in the file as we expect.
+    Also check, whether there are unreachable records in the files. If either of the issues arise, throws an error.
+
+    Parameters
+    ----------
+    path: Path
+        Path to the backlog-library to check.
+    meas_paths: list[str]
+        List of measurement paths to check.
+    """
     needed_data: dict[str, list[str]] = {}
     for mpath in meas_paths:
         file = mpath.split("::")[0]
@@ -67,11 +122,19 @@ def _check_db2paths(path: Path, meas_paths: list[str]) -> None:
             for key in needed_data[file]:
                 if key not in filedict.keys():
                     raise ValueError(f"Did not find data for key {key} that should be in file {file}.")
-    print("Links:\t✅")
     return
 
 
 def check_db_file_links(path: Path) -> None:
+    """
+    Check whether for each record in the given correlator library, we can find the data in the file as we expect.
+    Also check, whether there are unreachable records in the files. If either of the issues arise, throws an error.
+
+    Parameters
+    ----------
+    path: Path
+        Path to the backlog-library to check.
+    """
     db = get_db_file(path)
     search_expr = "SELECT path FROM 'backlogs'"
     conn = sqlite3.connect(path / db)
@@ -79,9 +142,90 @@ def check_db_file_links(path: Path) -> None:
     _check_db2paths(path, list(results))
 
 
+def check_path_and_config(path: Path) -> None:
+    """
+    Check whether the given path exists and the cinfigureation file can be found.
+
+    Parameters
+    ----------
+    path: Path
+        Path to the backlog-library to check.
+    """
+    if not os.path.exists(path):
+        raise FileNotFoundError(f"Corrlib path {path} does not exist.")
+    config_path = path / CONFIG_FILENAME
+    if not os.path.exists(config_path):
+        raise FileNotFoundError(f"Configuration file {config_path} not found.")
+
+
+def check_config_validity(path: Path) -> None:
+    """
+    Check whether the configuration file of the given corrlib-dataset path is valid.
+
+    Parameters
+    ----------
+    path: Path
+        Path to the backlog-library to check.
+    """
+    config = ConfigParser()
+    config_path = path / CONFIG_FILENAME
+    if os.path.exists(config_path):
+        config.read(config_path)
+    else:
+        raise FileNotFoundError("Configuration file not found.")
+
+    if config.has_section('core'):
+        core_opts = ['version', 'tracker', 'cached']
+        has_core_opts = [config.has_option('core', opt) for opt in core_opts]
+        if not all(has_core_opts):
+            raise ValueError("One of the options in the 'core' section  ('version', 'tracker', 'cached') is missing.")
+
+    if config.has_section('paths'):
+        has_path_opts = [config.has_option('paths', opt) for opt in path_opts]
+        if not all(has_path_opts):
+            raise ValueError("One of the options in the 'path' section  ('db', 'projects_path', 'archive_path', 'toml_imports_path', 'import_scripts_path') is missing.")
+
+
+def check_paths(path: Path) -> None:
+    """
+    Check whether all paths demanded by the 'paths' section of the configuration-file exist.
+
+    Parameters
+    ----------
+    path: Path
+        Path to the backlog-library to check.
+    """
+    config = ConfigParser()
+    config_path = path / CONFIG_FILENAME
+    if os.path.exists(config_path):
+        config.read(config_path)
+    else:
+        raise FileNotFoundError("Configuration file not found.")
+    has_paths = [os.path.exists(path / config.get('paths', opt)) for opt in path_opts]
+    if not all(has_paths):
+        raise FileNotFoundError("One of the paths specified in the configuration file is not present.")
+
+
 def full_integrity_check(path: Path) -> None:
+    """
+    Aggregate all checks for easy validation of the backlog-library.
+
+    Parameters
+    ----------
+    path: Path
+        Path to the backlog-library to check.
+    """
+    print("Run full integrity check...")
+    check_path_and_config(path)
+    print("(1/5) Path and config-file exist: ✅")
+    check_config_validity(path)
+    print("(2/5) Configuration is valid: ✅")
+    check_paths(path)
+    print("(3/5) Needed paths exist: ✅")
     check_db_integrity(path)
+    print("(4/5) Database is sane: ✅")
     check_db_file_links(path)
-    print("Full:\t✅")
+    print("(5/5) DB2File and File2DB-links are sound: ✅")
+    print("Full integrity check: ✅")