Enable flake8-logging-format (G) linting rules via Ruff

bugbug/code_search/parser.py

@@ -42,7 +42,9 @@ def search(repo_dir, commit_hash, symbol_name):
             return []
 
         logger.error(
-            f"Error running 'hg grep' command.\nstdout:\n{e.stdout.decode()}\n\nstderr:\n{e.stderr.decode()}"
+            "Error running 'hg grep' command.\nstdout:\n %s \n\nstderr:\n %s",
+            e.stdout.decode(),
+            e.stderr.decode(),
         )
         raise
 
@@ -60,7 +62,9 @@ def search(repo_dir, commit_hash, symbol_name):
             )
         except subprocess.CalledProcessError as e:
             logger.error(
-                f"Error running 'hg cat' command.\nstdout:\n{e.stdout.decode()}\n\nstderr:\n{e.stderr.decode()}"
+                "Error running 'hg cat' command.\nstdout:\n%s\n\nstderr:\n%s",
+                e.stdout.decode(),
+                e.stderr.decode(),
             )
             raise
 

bugbug/model.py

@@ -317,7 +317,7 @@ def print_feature_importances(self, important_features, class_probabilities=None
 
         # allow maximum of 3 columns in a row to fit the page better
         COLUMNS = 3
-        logger.info("Top {} features:".format(len(top_feature_names)))
+        logger.info("Top %s features:", len(top_feature_names))
         for i in range(0, len(top_feature_names), COLUMNS):
             table = []
             for item in shap_val:
@@ -380,7 +380,7 @@ def train(self, importance_cutoff=0.15, limit=None):
             X = X[:limit]
             y = y[:limit]
 
-        logger.info(f"X: {X.shape}, y: {y.shape}")
+        logger.info("X: %s , y: %s", X.shape, y.shape)
 
         is_multilabel = isinstance(y[0], np.ndarray)
         is_binary = len(self.class_names) == 2
@@ -408,11 +408,14 @@ def train(self, importance_cutoff=0.15, limit=None):
                     "std": score.std() * 2,
                 }
                 logger.info(
-                    f"{scoring.capitalize()}: f{score.mean()} (+/- {score.std() * 2})"
+                    "%s: f%.4f (+/- %.4f)",
+                    scoring.capitalize(),
+                    score.mean(),
+                    (score.std() * 2),
                 )
 
-        logger.info(f"X_train: {X_train.shape}, y_train: {y_train.shape}")
-        logger.info(f"X_test: {X_test.shape}, y_test: {y_test.shape}")
+        logger.info("X_train: %s, y_train: %s", X_train.shape, y_train.shape)
+        logger.info("X_test: %s, y_test: %s", X_test.shape, y_test.shape)
 
         self.clf.fit(X_train, self.le.transform(y_train))
         logger.info("Number of features: %d", self.clf.steps[-1][1].n_features_in_)
@@ -480,7 +483,7 @@ def train(self, importance_cutoff=0.15, limit=None):
                 "The predictions should be multilabel"
             )
 
-        logger.info(f"No confidence threshold - {len(y_test)} classified")
+        logger.info("No confidence threshold - %d classified", len(y_test))
         if is_multilabel:
             confusion_matrix = metrics.multilabel_confusion_matrix(y_test, y_pred)
         else:
@@ -548,7 +551,9 @@ def train(self, importance_cutoff=0.15, limit=None):
                 )
 
             logger.info(
-                f"\nConfidence threshold > {confidence_threshold} - {classified_num} classified"
+                "\nConfidence threshold > %s - %d classified",
+                confidence_threshold,
+                classified_num,
             )
             if is_multilabel:
                 confusion_matrix = metrics.multilabel_confusion_matrix(
@@ -579,7 +584,7 @@ def train(self, importance_cutoff=0.15, limit=None):
             X_train = X
             y_train = y
 
-            logger.info(f"X_train: {X_train.shape}, y_train: {y_train.shape}")
+            logger.info("X_train: %s, y_train: %s", X_train.shape, y_train.shape)
 
             self.clf.fit(X_train, self.le.transform(y_train))
 

bugbug/models/rcatype.py

@@ -157,7 +157,7 @@ def get_rca_from_whiteboard(self, whiteboard_data):
             )
 
             if rca_whiteboard_split[1] not in self.RCA_LIST:
-                logger.warning(rca_whiteboard_split[1] + " not in RCA_LIST")
+                logger.warning("%s not in RCA_LIST", rca_whiteboard_split[1])
             else:
                 rca.append(rca_whiteboard_split[1])
         return rca

bugbug/models/testselect.py

@@ -85,7 +85,7 @@ def _get_cost(config: str) -> int:
         if all(s in config for s in substrings):
             return cost
 
-    logger.warning(f"Couldn't find cost for {config}")
+    logger.warning("Couldn't find cost for %s", config)
     return max(cost for _, cost in costs)
 
 
@@ -309,7 +309,7 @@ def select_configs(
 
     for group in groups:
         if group not in equivalence_sets:
-            logger.warning(f"No equivalence sets for group {group}")
+            logger.warning("No equivalence sets for group %s", group)
             continue
         # Create constraints to ensure at least one task from each set of equivalent
         # groups is selected.

bugbug/phabricator.py

@@ -187,7 +187,7 @@ def get_testing_project(rev: RevisionDict) -> str | None:
     ]
 
     if len(testing_projects) > 1:
-        logger.warning("Revision D{} has more than one testing tag.".format(rev["id"]))
+        logger.warning("Revision D%s has more than one testing tag.", rev["id"])
 
     if len(testing_projects) == 0:
         return None
@@ -237,7 +237,7 @@ def get_first_review_time(rev: RevisionDict) -> timedelta | None:
     ) = get_review_dates(rev)
 
     if creation_date is None:
-        logger.warning("Revision D{} has no creation date.".format(rev["id"]))
+        logger.warning("Revision D%s has no creation date.", rev["id"])
         return None
 
     if len(review_dates) == 0:
@@ -253,7 +253,7 @@ def get_first_review_time(rev: RevisionDict) -> timedelta | None:
         and first_exclusion_end_date is not None
         and first_exclusion_start_date > first_exclusion_end_date
     ):
-        logger.warning("Revision D{} was in an inconsistent state.".format(rev["id"]))
+        logger.warning("Revision D%s was in an inconsistent state.", rev["id"])
 
     if (
         first_exclusion_start_date is None
@@ -264,9 +264,8 @@ def get_first_review_time(rev: RevisionDict) -> timedelta | None:
         first_exclusion_end_date is None or first_exclusion_end_date > first_review_date
     ):
         logger.warning(
-            "Revision D{} was accepted while in 'planned changes' or 'closed' state.".format(
-                rev["id"]
-            )
+            "Revision D%s was accepted while in 'planned changes' or 'closed' state.",
+            rev["id"],
         )
         return first_review_date - creation_date
     else:
@@ -284,7 +283,7 @@ def get_pending_review_time(rev: RevisionDict) -> timedelta | None:
     creation_date, _, exclusion_start_dates, exclusion_end_dates = get_review_dates(rev)
 
     if creation_date is None:
-        logger.warning("Revision D{} has no creation date.".format(rev["id"]))
+        logger.warning("Revision D%s has no creation date.", rev["id"])
         return None
 
     last_exclusion_start_date = max(exclusion_start_dates, default=None)
@@ -295,9 +294,8 @@ def get_pending_review_time(rev: RevisionDict) -> timedelta | None:
         or last_exclusion_start_date > last_exclusion_end_date
     ):
         logger.warning(
-            "Revision D{} was in an inconsistent state (needs review, but is in an exception timespan).".format(
-                rev["id"]
-            )
+            "Revision D%s was in an inconsistent state (needs review, but is in an exception timespan).",
+            rev["id"],
         )
 
     if last_exclusion_end_date is not None:

bugbug/repository.py

@@ -670,7 +670,9 @@ def set_commit_metrics(
     try:
         get_space_metrics(commit.metrics, after_metrics["spaces"])
     except AnalysisException:
-        logger.debug(f"rust-code-analysis error on commit {commit.node}, path {path}")
+        logger.debug(
+            "rust-code-analysis error on commit %s, path %s", commit.node, path
+        )
 
     before_metrics_dict = get_total_metrics_dict()
     try:
@@ -679,7 +681,9 @@ def set_commit_metrics(
                 before_metrics_dict, before_metrics["spaces"], calc_summaries=False
             )
     except AnalysisException:
-        logger.debug(f"rust-code-analysis error on commit {commit.node}, path {path}")
+        logger.debug(
+            "rust-code-analysis error on commit %s, path %s", commit.node, path
+        )
 
     commit.metrics_diff = {
         f"{metric}_total": commit.metrics[f"{metric}_total"]
@@ -704,7 +708,10 @@ def set_commit_metrics(
             get_space_metrics(metrics_dict, func, calc_summaries=False)
         except AnalysisException:
             logger.debug(
-                f"rust-code-analysis error on commit {commit.node}, path {path}, function {func['name']}"
+                "rust-code-analysis error on commit %s, path %s, function %s}",
+                commit.node,
+                path,
+                func["name"],
             )
 
         commit.functions[path].append(
@@ -734,7 +741,7 @@ def transform(hg: hglib.client, repo_dir: str, commit: Commit) -> Commit:
     try:
         patch_data = rs_parsepatch.get_lines(patch)
     except Exception:
-        logger.error(f"Exception while analyzing {commit.node}")
+        logger.error("Exception while analyzing %s", commit.node)
         raise
 
     for stats in patch_data:
@@ -1185,7 +1192,10 @@ def update_complex_experiences(
                     )
                 else:
                     logger.warning(
-                        f"Experience missing for file {orig}, type '{commit_type}', on commit {commit.node}"
+                        "Experience missing for file %s, type '%s', on commit %s",
+                        orig,
+                        commit_type,
+                        commit.node,
                     )
 
         if (