core.pandas: add check for 'error' column + add empty one by default

my/core/__main__.py

@@ -168,7 +168,7 @@ def config_ok(args) -> bool:
         sys.exit(1)
 
     cfg_path = cfg.__file__# todo might be better to use __path__?
-    info(f"config file: {cfg_path}")
+    info(f"config file : {cfg_path}")
 
     import my.core as core
     try:
@@ -195,7 +195,7 @@ See https://github.com/karlicoss/HPI/blob/master/doc/SETUP.org#setting-up-module
     if mres is not None: # has mypy
         rc = mres.returncode
         if rc == 0:
-            info('mypy check: success')
+            info('mypy check  : success')
         else:
             error('mypy check: failed')
             errors.append(RuntimeError('mypy failed'))

my/core/common.py

@@ -292,16 +292,6 @@ else:
     from .py37 import fromisoformat
 
 
-if sys.version_info[:2] >= (3, 8):
-    from typing import Literal
-else:
-    if TYPE_CHECKING:
-        from typing_extensions import Literal
-    else:
-        # erm.. I guess as long as it's not crashing, whatever...
-        Literal = Union
-
-
 # TODO doctests?
 def isoparse(s: str) -> tzdatetime:
     """
@@ -313,6 +303,8 @@ def isoparse(s: str) -> tzdatetime:
     s = s[:-1] + '+00:00'
     return fromisoformat(s)
 
+from .compat import Literal
+
 
 import re
 # https://stackoverflow.com/a/295466/706389

my/core/compat.py

@@ -47,3 +47,16 @@ def _get_dal(cfg, module_name: str):
         from importlib import import_module
         return import_module(f'my.config.repos.{module_name}.dal')
 
+
+import sys
+from typing import TYPE_CHECKING
+
+if sys.version_info[:2] >= (3, 8):
+    from typing import Literal
+else:
+    if TYPE_CHECKING:
+        from typing_extensions import Literal
+    else:
+        from typing import Union
+        # erm.. I guess as long as it's not crashing, whatever...
+        Literal = Union

my/core/pandas.py

@@ -7,6 +7,9 @@ from datetime import datetime
 from pprint import pformat
 from typing import Optional, TYPE_CHECKING, Any, Iterable
 from . import warnings
+from .common import LazyLogger
+
+logger = LazyLogger(__name__)
 
 
 if TYPE_CHECKING:
@@ -14,6 +17,8 @@ if TYPE_CHECKING:
     # later will be unignored when they implement type annotations
     import pandas as pd # type: ignore
     # DataFrameT = pd.DataFrame
+    # TODO ugh. pretty annoying, having any is not very useful since it would allow arbitrary coercions..
+    # ideally want to use a type that's like Any but doesn't allow arbitrary coercions??
     DataFrameT = Any
 else:
     # in runtime, make it defensive so it works without pandas
@@ -40,21 +45,54 @@ def check_dateish(s) -> Iterable[str]:
     '''.strip()
 
 
+from .compat import Literal
+
+ErrorColPolicy = Literal[
+    'add_if_missing', # add error column if it's missing
+    'warn'          , # warn, but do not modify
+    'ignore'        , # no warnings
+]
+
+def check_error_column(df: DataFrameT, *, policy: ErrorColPolicy) -> Iterable[str]:
+    if 'error' in df:
+        return
+    if policy == 'ignore':
+        return
+
+    wmsg = '''
+No 'error' column detected. You probably forgot to handle errors defensively, which means a single bad entry might bring the whole dataframe down.
+'''.strip()
+    if policy == 'add_if_missing':
+        # todo maybe just add the warnings text as well?
+        df['error'] = None
+        wmsg += "\nAdding empty 'error' column (see 'error_col_policy' if you want to change this behaviour)"
+        pass
+
+    yield wmsg
+
+
 from typing import Any, Callable, TypeVar
 FuncT = TypeVar('FuncT', bound=Callable[..., DataFrameT])
 
-def check_dataframe(f: FuncT) -> FuncT:
-    from functools import wraps
-    @wraps(f)
-    def wrapper(*args, **kwargs) -> DataFrameT:
-        df = f(*args, **kwargs)
-        # todo make super defensive?
+# TODO ugh. typing this is a mess... shoul I use mypy_extensions.VarArg/KwArgs?? or what??
+from decorator import decorator
+@decorator
+def check_dataframe(f: FuncT, error_col_policy: ErrorColPolicy='add_if_missing', *args, **kwargs) -> DataFrameT:
+    df = f(*args, **kwargs)
+    tag = '{f.__module__}:{f.__name__}'
+    # makes sense to keep super defensive
+    try:
         for col, data in df.reset_index().iteritems():
             for w in check_dateish(data):
-                warnings.low(f"{f.__module__}:{f.__name__}, column '{col}': {w}")
-        return df
-    # https://github.com/python/mypy/issues/1927
-    return wrapper # type: ignore[return-value]
+                warnings.low(f"{tag}, column '{col}': {w}")
+    except Exception as e:
+        logger.exception(e)
+    try:
+        for w in check_error_column(df, policy=error_col_policy):
+            warnings.low(f"{tag}, {w}")
+    except Exception as e:
+        logger.exception(e)
+    return df
 
 # todo doctor: could have a suggesion to wrap dataframes with it?? discover by return type?
 

my/core/warnings.py

@@ -36,7 +36,7 @@ def _warn(message: str, *args, color=None, **kwargs) -> None:
 
 
 def low(message: str, *args, **kwargs) -> None:
-    kwargs['color'] = 'grey'
+    # kwargs['color'] = 'grey' # eh, grey is way too pale
     _warn(message, *args, **kwargs)