Configure file logging

.gitignore

@@ -13,6 +13,7 @@ data/nltk/
 data/training/
 data/validation/
 data/evaluation/
+data/logs/
 
 ##################
 # Python ignores #

unimore-bda-6.iml

@@ -10,6 +10,8 @@
       <excludeFolder url="file://$MODULE_DIR$/.venv" />
       <excludeFolder url="file://$MODULE_DIR$/data/evaluation" />
       <excludeFolder url="file://$MODULE_DIR$/data/training" />
+      <excludeFolder url="file://$MODULE_DIR$/data/validation" />
+      <excludeFolder url="file://$MODULE_DIR$/data/logs" />
     </content>
     <orderEntry type="jdk" jdkName="Poetry (unimore-bda-6)" jdkType="Python SDK" />
     <orderEntry type="sourceFolder" forTests="false" />

unimore_bda_6/__main__.py

@@ -1,8 +1,8 @@
 import logging
 import pymongo.errors
-from .log import install_log_handler
+from .log import install_general_log_handlers
 
-install_log_handler()
+install_general_log_handlers()
 
 from .config import config
 from .database import mongo_client_from_config, reviews_collection, sample_reviews_polar, sample_reviews_varied
@@ -36,7 +36,7 @@ def main():
         for sample_func in [sample_reviews_varied, sample_reviews_polar]:
 
             slog = logging.getLogger(f"{__name__}.{sample_func.__name__}")
-            slog.debug("Selected sample_func: %s", sample_func)
+            slog.debug("Selected sample_func: %s", sample_func.__name__)
 
             for SentimentAnalyzer in [
                 TensorflowSentimentAnalyzer,
@@ -44,7 +44,7 @@ def main():
             ]:
 
                 slog = logging.getLogger(f"{__name__}.{sample_func.__name__}.{SentimentAnalyzer.__name__}")
-                slog.debug("Selected SentimentAnalyzer: %s", SentimentAnalyzer)
+                slog.debug("Selected SentimentAnalyzer: %s", SentimentAnalyzer.__name__)
 
                 for Tokenizer in [
                     PlainTokenizer,
@@ -55,7 +55,7 @@ def main():
                 ]:
 
                     slog = logging.getLogger(f"{__name__}.{sample_func.__name__}.{SentimentAnalyzer.__name__}.{Tokenizer.__name__}")
-                    slog.debug("Selected Tokenizer: %s", Tokenizer)
+                    slog.debug("Selected Tokenizer: %s", Tokenizer.__name__)
 
                     run_counter = 0
 
@@ -66,10 +66,10 @@ def main():
                         slog.debug("Run #%d", run_counter)
 
                         try:
-                            slog.debug("Instantiating %s with %s...", SentimentAnalyzer, Tokenizer)
+                            slog.debug("Instantiating %s with %s...", SentimentAnalyzer.__name__, Tokenizer.__name__)
                             sa = SentimentAnalyzer(tokenizer=Tokenizer())
                         except TypeError:
-                            slog.warning("%s is not supported by %s, skipping run...", SentimentAnalyzer, Tokenizer)
+                            slog.warning("%s is not supported by %s, skipping run...", SentimentAnalyzer.__name__, Tokenizer.__name__)
                             break
 
                         with Caches.from_database_samples(collection=reviews, sample_func=sample_func) as datasets:

unimore_bda_6/log.py

@@ -1,22 +1,35 @@
 import collections
 import logging
 import coloredlogs
+import pathlib
 
 this_log = logging.getLogger(__name__)
 
 
-def install_log_handler(loggers: list[logging.Logger] = None):
-    if loggers is None:
-        loggers = [
-            logging.getLogger("__main__"),
-            logging.getLogger("unimore_bda_6"),
-        ]
+def install_general_log_handlers():
+    """
+    Setup the `unimore_bda_6` and ``__main__`` loggers.
 
-    for logger in loggers:
+    With colors!
+    """
+    main_logger: logging.Logger = logging.getLogger("__main__")
+
+    interesting_loggers: list[logging.Logger] = [
+        main_logger,
+        logging.getLogger("unimore_bda_6"),
+    ]
+
+    chatty_loggers: list[logging.Logger] = [
+        logging.getLogger("unimore_bda_6.database.cache"),
+        logging.getLogger("unimore_bda_6.database.datatypes"),
+    ]
+
+    this_log.debug("Installing console handlers...")
+    for logger in interesting_loggers:
         coloredlogs.install(
             logger=logger,
             level="DEBUG" if __debug__ else "INFO",
-            fmt="{asctime} | {name:<80} | {levelname:>8} | {message}",
+            fmt="{asctime} | {name} | {levelname} | {message}",
             style="{",
             level_styles=dict(
                 debug=dict(color="white"),
@@ -32,10 +45,24 @@ def install_log_handler(loggers: list[logging.Logger] = None):
             ),
             isatty=True,
         )
-        this_log.debug("Installed custom log handler on: %s", logger)
+        this_log.debug("Installed console log handler on: %s", logger)
 
-    logging.getLogger("unimore_bda_6.database.cache").setLevel("INFO")
-    logging.getLogger("unimore_bda_6.database.datatypes").setLevel("INFO")
+    this_log.debug("Silencing chatty loggers...")
+    for logger in chatty_loggers:
+        logger.setLevel("INFO")
+        this_log.debug("Silenced: %s", logger)
+
+    log_file_path = pathlib.Path("./data/logs/last_run.tsv")
+    log_directory_path = log_file_path.parent
+    this_log.debug("Ensuring %s exists...", log_directory_path)
+    log_directory_path.mkdir(parents=True, exist_ok=True)
+    this_log.debug("Ensuring %s exists...", log_file_path)
+    open(log_file_path, "w").close()
+    this_log.debug("Installing FileHandler for the __main__ logger at %s...", log_file_path)
+    file_handler = logging.FileHandler(log_file_path, mode="w")
+    file_handler.formatter = logging.Formatter("{asctime}\t{name}\t{levelname}\t{message}", style="{")
+    main_logger.addHandler(file_handler)
+    this_log.debug("Installed FileHandler for the __main__ logger at %s!", log_file_path)
 
 
 _passage_counts = collections.defaultdict(lambda: 0)
@@ -48,6 +75,6 @@ def count_passage(log: logging.Logger, key: str, mod: int):
 
 
 __all__ = (
-    "install_log_handler",
+    "install_general_log_handlers",
     "count_passage",
 )