fix: threshold renamings

evaluation-lightweight.sh

@@ -3,6 +3,6 @@
 args=""
 projects="projects/baasbox"
 
-python3 -m renas.repository_analyzer $projects
+python3 -m renas.repository_analyzer $projects -threshold 1
 python3 -m renas.recommendation $projects
 python3 -m renas.evaluator -sim -pre $projects

renas/preliminaryResearch.sh

@@ -5,7 +5,7 @@ projects=("baasbox" "cordova-plugin-local-notifications" "morphia" "spring-integ
 
 for line in "${projects[@]}"
 do
-python3 -m renas.repository_analyzer "projects/${line}"
+python3 -m renas.repository_analyzer "projects/${line}" -threshold 1
 python3 -m renas.recommendation "projects/${line}"
 args="${args} projects/${line}"
 done

renas/relationship/analyzer.py

@@ -38,13 +38,12 @@ def set_logger(level):
     return root_logger
 
 
-def filter_data(data, threshold):
+def filter_data(data, threshold: int):
     LOGGER.info("filter data")
     commits = data.groupby("commit").size()
-    if threshold:
-        commits = commits[commits > 3]
-    else:
-        commits = commits[commits > 1]
+
+    commits = commits[commits > threshold]
+    LOGGER.info(f"threshold = more than {threshold} renames")
     LOGGER.info(f"total {commits.sum()} renames")
     LOGGER.info(f"pick {len(commits)} commits")
     return data[data["commit"].isin(commits.index)]
@@ -86,7 +85,7 @@ def git_archive_wrapper(arg):
     return git_archive(*arg)
 
 
-def main(root: pathlib.Path, rename_data: pd.DataFrame, threshold: bool):
+def main(root: pathlib.Path, rename_data: pd.DataFrame, threshold: int):
     set_logger(INFO)
     try:
         rename_data = filter_data(rename_data, threshold)

renas/repository_analyzer.py

@@ -15,9 +15,9 @@ def set_argument():
     )
     parser.add_argument(
         "-threshold",
-        help="use commit which has more than 3 renames",
-        action="store_true",
-        default=False,
+        help="use commit which has more than specifying renames",
+        action="store",
+        default=0,
     )
     parser.add_argument(
         "-f",
@@ -38,7 +38,7 @@ def main(root, args):
         dump(root, rename_data)
     else:
         rename_data = pd.read_json(rename_path, orient="records")
-    analyzer.main(root, rename_data, args.threshold)
+    analyzer.main(root, rename_data, int(args.threshold))
 
 
 def dump(root, data: pd.DataFrame):

renas/researchQuestion.sh

@@ -5,7 +5,7 @@ projects=("testng" "jackson-databind" "rest.li" "Activiti" "k-9" "genie" "eucaly
 
 for line in "${projects[@]}"
 do
-python3 -m renas.repository_analyzer "projects/${line}"
+python3 -m renas.repository_analyzer "projects/${line}" -threshold 3
 python3 -m renas.recommendation "projects/${line}"
 args="${args} projects/${line}"
 done

renas/researchQuestionManually.sh

@@ -5,7 +5,7 @@ projects=("ratpack" "argouml")
 
 for line in "${projects[@]}"
 do
-python3 -m renas.repository_analyzer "projects/${line}"
+python3 -m renas.repository_analyzer "projects/${line}" -threshold 1
 python3 -m renas.recommendation "projects/${line}"
 args="${args} projects/${line}"
 done