statistics visualize : --task_completion_cireteria引数でタスクの完了条件を指定で…

…きるようにしました (#1296)

* 最初に受入フェーズに到達した日時を記載する

* update

* 教師付開始日の基準を完了日にする

* 完了条件を記載

* merge

* fix\

* update doc

* document 修正

* task_compeltion_criteria を追加

* グラフを生成

* modify test data

* テストデータの修正

annofabcli/stat_visualization/mask_visualization_dir.py

@@ -33,7 +33,7 @@
 from annofabcli.statistics.visualization.dataframe.task_worktime_by_phase_user import TaskWorktimeByPhaseUser
 from annofabcli.statistics.visualization.dataframe.user_performance import UserPerformance
 from annofabcli.statistics.visualization.dataframe.worktime_per_date import WorktimePerDate
-from annofabcli.statistics.visualization.model import ProductionVolumeColumn
+from annofabcli.statistics.visualization.model import ProductionVolumeColumn, TaskCompletionCriteria
 from annofabcli.statistics.visualization.project_dir import ProjectDir
 
 logger = logging.getLogger(__name__)
@@ -223,12 +223,15 @@ def main(args: argparse.Namespace) -> None:
         create_custom_production_volume_list(args.custom_production_volume) if args.custom_production_volume is not None else None
     )
 
+    task_completion_criteria = TaskCompletionCriteria(args.task_completion_criteria)
     input_project_dir = ProjectDir(
         args.dir,
+        task_completion_criteria,
         custom_production_volume_list=custom_production_volume_list,
     )
     output_project_dir = ProjectDir(
         args.output_dir,
+        task_completion_criteria,
         metadata=input_project_dir.read_metadata(),
     )
     mask_visualization_dir(
@@ -271,6 +274,16 @@ def parse_args(parser: argparse.ArgumentParser) -> None:
         "column_list": [{"value": "video_duration_minute", "name": "動画長さ"}],
     }
 
+    parser.add_argument(
+        "--task_completion_criteria",
+        type=str,
+        choices=[e.value for e in TaskCompletionCriteria],
+        default=TaskCompletionCriteria.ACCEPTANCE_COMPLETED.value,
+        help="タスクの完了条件を指定します。\n"
+        "* ``acceptance_completed``: タスクが受入フェーズの完了状態であれば「タスクの完了」とみなす\n"
+        "* ``acceptance_reached``: タスクが受入フェーズに到達したら「タスクの完了」とみなす\n",
+    )
+
     parser.add_argument(
         "--custom_production_volume",
         type=str,

annofabcli/stat_visualization/merge_visualization_dir.py

@@ -35,7 +35,7 @@
     WholeProductivityPerFirstAnnotationStartedDate,
 )
 from annofabcli.statistics.visualization.dataframe.worktime_per_date import WorktimePerDate
-from annofabcli.statistics.visualization.model import ProductionVolumeColumn
+from annofabcli.statistics.visualization.model import ProductionVolumeColumn, TaskCompletionCriteria
 from annofabcli.statistics.visualization.project_dir import MergingInfo, ProjectDir
 
 logger = logging.getLogger(__name__)
@@ -45,11 +45,13 @@ class WritingVisualizationFile:
     def __init__(
         self,
         output_project_dir: ProjectDir,
+        task_completion_criteria: TaskCompletionCriteria,
         *,
         user_id_list: Optional[List[str]] = None,
         minimal_output: bool = False,
     ) -> None:
         self.output_project_dir = output_project_dir
+        self.task_completion_criteria = task_completion_criteria
         self.user_id_list = user_id_list
         self.minimal_output = minimal_output
 
@@ -126,13 +128,13 @@ def write_task_worktime_by_phase_user(self, task_worktime_by_phase_user: TaskWor
 
     @_catch_exception
     def write_performance_per_first_annotation_started_date(self, task: Task) -> None:
-        obj = WholeProductivityPerFirstAnnotationStartedDate.from_task(task)
+        obj = WholeProductivityPerFirstAnnotationStartedDate.from_task(task, self.task_completion_criteria)
         self.output_project_dir.write_whole_productivity_per_first_annotation_started_date(obj)
         self.output_project_dir.write_whole_productivity_line_graph_per_annotation_started_date(obj)
 
     @_catch_exception
     def write_merge_performance_per_date(self, task: Task, worktime_per_date: WorktimePerDate) -> None:
-        obj = WholeProductivityPerCompletedDate.from_df_wrapper(task, worktime_per_date)
+        obj = WholeProductivityPerCompletedDate.from_df_wrapper(task, worktime_per_date, task_completion_criteria=self.task_completion_criteria)
         self.output_project_dir.write_whole_productivity_per_date(obj)
         self.output_project_dir.write_whole_productivity_line_graph_per_date(obj)
 
@@ -176,7 +178,7 @@ def merge_task_worktime_by_phase_user(self) -> TaskWorktimeByPhaseUser:
         merged_obj = TaskWorktimeByPhaseUser.merge(*tmp_list, custom_production_volume_list=self.custom_production_volume_list)
         return merged_obj
 
-    def create_merging_info(self) -> MergingInfo:
+    def create_merging_info(self, task_completion_criteria: TaskCompletionCriteria) -> MergingInfo:
         """
         `project_info.json`の内容から、どのようにマージしたかを示す情報を作成する。
         """
@@ -186,12 +188,15 @@ def create_merging_info(self) -> MergingInfo:
             project_info = project_dir.read_project_info()
             project_info_list.append(project_info)
 
-        merge_info = MergingInfo(target_dir_list=target_dir_list, project_info_list=project_info_list)
+        merge_info = MergingInfo(
+            target_dir_list=target_dir_list, project_info_list=project_info_list, task_completion_criteria=task_completion_criteria
+        )
         return merge_info
 
 
 def merge_visualization_dir(  # pylint: disable=too-many-statements
     project_dir_list: List[ProjectDir],
+    task_completion_criteria: TaskCompletionCriteria,
     output_project_dir: ProjectDir,
     *,
     custom_production_volume_list: Optional[list[ProductionVolumeColumn]] = None,
@@ -200,7 +205,7 @@ def merge_visualization_dir(  # pylint: disable=too-many-statements
 ) -> None:
     merging_obj = MergingVisualizationFile(project_dir_list, custom_production_volume_list=custom_production_volume_list)
 
-    merging_info = merging_obj.create_merging_info()
+    merging_info = merging_obj.create_merging_info(task_completion_criteria)
     output_project_dir.metadata = merging_info.to_dict(encode_json=True)
 
     # 基本となるCSVファイルを読み込みマージする
@@ -210,7 +215,9 @@ def merge_visualization_dir(  # pylint: disable=too-many-statements
 
     user_performance = UserPerformance.from_df_wrapper(task_worktime_by_phase_user=task_worktime_by_phase_user, worktime_per_date=worktime_per_date)
     whole_performance = WholePerformance.from_df_wrapper(task_worktime_by_phase_user=task_worktime_by_phase_user, worktime_per_date=worktime_per_date)
-    writing_obj = WritingVisualizationFile(output_project_dir, user_id_list=user_id_list, minimal_output=minimal_output)
+    writing_obj = WritingVisualizationFile(
+        output_project_dir, user_id_list=user_id_list, minimal_output=minimal_output, task_completion_criteria=task_completion_criteria
+    )
 
     writing_obj.write_task_list_and_histogram(task)
     writing_obj.write_worktime_per_date(worktime_per_date)
@@ -258,20 +265,31 @@ def main(args: argparse.Namespace) -> None:
     custom_production_volume_list = (
         create_custom_production_volume_list(args.custom_production_volume) if args.custom_production_volume is not None else None
     )
-
+    task_completion_criteria = TaskCompletionCriteria(args.task_completion_criteria)
     merge_visualization_dir(
-        project_dir_list=[ProjectDir(e) for e in args.dir],
+        project_dir_list=[ProjectDir(e, task_completion_criteria) for e in args.dir],
+        task_completion_criteria=task_completion_criteria,
         user_id_list=user_id_list,
         custom_production_volume_list=custom_production_volume_list,
         minimal_output=args.minimal,
-        output_project_dir=ProjectDir(args.output_dir),
+        output_project_dir=ProjectDir(args.output_dir, task_completion_criteria),
     )
 
 
 def parse_args(parser: argparse.ArgumentParser) -> None:
     parser.add_argument("--dir", type=Path, nargs="+", required=True, help="マージ対象ディレクトリ。2つ以上指定してください。")
     parser.add_argument("-o", "--output_dir", type=Path, required=True, help="出力先ディレクトリ。配下にプロジェクト名のディレクトリが出力される。")
 
+    parser.add_argument(
+        "--task_completion_criteria",
+        type=str,
+        choices=[e.value for e in TaskCompletionCriteria],
+        default=TaskCompletionCriteria.ACCEPTANCE_COMPLETED.value,
+        help="タスクの完了条件を指定します。\n"
+        "* ``acceptance_completed``: タスクが受入フェーズの完了状態であれば「タスクの完了」とみなす\n"
+        "* ``acceptance_reached``: タスクが受入フェーズに到達したら「タスクの完了」とみなす\n",
+    )
+
     parser.add_argument(
         "-u",
         "--user_id",

annofabcli/stat_visualization/summarize_whole_performance_csv.py

@@ -11,7 +11,7 @@
     get_json_from_args,
 )
 from annofabcli.statistics.visualization.dataframe.project_performance import ProjectPerformance
-from annofabcli.statistics.visualization.model import ProductionVolumeColumn
+from annofabcli.statistics.visualization.model import ProductionVolumeColumn, TaskCompletionCriteria
 from annofabcli.statistics.visualization.project_dir import ProjectDir
 
 logger = logging.getLogger(__name__)
@@ -31,7 +31,10 @@ def create_custom_production_volume_list(cli_value: str) -> list[ProductionVolum
 
 def main(args: argparse.Namespace) -> None:
     root_dir: Path = args.dir
-    project_dir_list = [ProjectDir(elm) for elm in root_dir.iterdir() if elm.is_dir()]
+    # task_completion_criteriaは何でもよいので、とりあえずACCEPTANCE_COMPLETEDを指定
+    project_dir_list = [
+        ProjectDir(elm, task_completion_criteria=TaskCompletionCriteria.ACCEPTANCE_COMPLETED) for elm in root_dir.iterdir() if elm.is_dir()
+    ]
 
     custom_production_volume_list = (
         create_custom_production_volume_list(args.custom_production_volume) if args.custom_production_volume is not None else None

annofabcli/stat_visualization/write_graph.py

@@ -21,7 +21,7 @@
     InspectorProductivityPerDate,
 )
 from annofabcli.statistics.visualization.dataframe.task import Task
-from annofabcli.statistics.visualization.model import ProductionVolumeColumn
+from annofabcli.statistics.visualization.model import ProductionVolumeColumn, TaskCompletionCriteria
 from annofabcli.statistics.visualization.project_dir import ProjectDir
 
 logger = logging.getLogger(__name__)
@@ -129,8 +129,9 @@ def main(args: argparse.Namespace) -> None:
         create_custom_production_volume_list(args.custom_production_volume) if args.custom_production_volume is not None else None
     )
 
-    input_project_dir = ProjectDir(args.dir, custom_production_volume_list=custom_production_volume_list)
-    output_project_dir = ProjectDir(args.output_dir, metadata=input_project_dir.read_metadata())
+    task_completion_criteria = TaskCompletionCriteria(args.task_completion_criteria)
+    input_project_dir = ProjectDir(args.dir, task_completion_criteria, custom_production_volume_list=custom_production_volume_list)
+    output_project_dir = ProjectDir(args.output_dir, task_completion_criteria, metadata=input_project_dir.read_metadata())
     main_obj = WritingGraph(
         project_dir=input_project_dir,
         output_project_dir=output_project_dir,
@@ -164,6 +165,16 @@ def parse_args(parser: argparse.ArgumentParser) -> None:
         help="必要最小限のファイルを出力します。",
     )
 
+    parser.add_argument(
+        "--task_completion_criteria",
+        type=str,
+        choices=[e.value for e in TaskCompletionCriteria],
+        default=TaskCompletionCriteria.ACCEPTANCE_COMPLETED.value,
+        help="タスクの完了条件を指定します。\n"
+        "* ``acceptance_completed``: タスクが受入フェーズの完了状態であれば「タスクの完了」とみなす\n"
+        "* ``acceptance_reached``: タスクが受入フェーズに到達したら「タスクの完了」とみなす\n",
+    )
+
     custom_production_volume_sample = {
         "column_list": [{"value": "video_duration_minute", "name": "動画長さ"}],
     }

annofabcli/stat_visualization/write_performance_rating_csv.py

@@ -21,7 +21,7 @@
     ProjectPerformance,
     ProjectWorktimePerMonth,
 )
-from annofabcli.statistics.visualization.model import ProductionVolumeColumn, WorktimeColumn
+from annofabcli.statistics.visualization.model import ProductionVolumeColumn, TaskCompletionCriteria, WorktimeColumn
 from annofabcli.statistics.visualization.project_dir import ProjectDir
 
 logger = logging.getLogger(__name__)
@@ -326,7 +326,11 @@ def create_rating_df(
                 custom_production_volume_list_by_directory.get(p_project_dir.name) if custom_production_volume_list_by_directory is not None else None
             )
             project_title = p_project_dir.name
-            project_dir = ProjectDir(p_project_dir, custom_production_volume_list=custom_production_volume_list)
+            project_dir = ProjectDir(
+                p_project_dir,
+                task_completion_criteria=TaskCompletionCriteria.ACCEPTANCE_COMPLETED,
+                custom_production_volume_list=custom_production_volume_list,
+            )
             project_dir_list.append(project_dir)
 
             try:
@@ -471,7 +475,8 @@ def create_user_df(target_dir: Path) -> pandas.DataFrame:
         if not p_project_dir.is_dir():
             continue
 
-        project_dir = ProjectDir(p_project_dir)
+        # task_completion_criteriaは何でもよいので、とりあえずACCEPTANCE_COMPLETEDを指定
+        project_dir = ProjectDir(p_project_dir, task_completion_criteria=TaskCompletionCriteria.ACCEPTANCE_COMPLETED)
 
         try:
             user_performance = project_dir.read_user_performance()

annofabcli/statistics/visualization/dataframe/task.py

@@ -110,7 +110,8 @@ def columns(self) -> list[str]:
             "first_acceptance_username",
             "first_acceptance_worktime_hour",
             "first_acceptance_started_datetime",
-            # 最後の受入
+            # 受入フェーズの日時
+            "first_acceptance_reached_datetime",
             "first_acceptance_completed_datetime",
             # 作業時間に関する内容
             "worktime_hour",
@@ -229,6 +230,7 @@ def empty(cls, *, custom_production_volume_list: Optional[list[ProductionVolumeC
             "first_acceptance_user_id",
             "first_acceptance_username",
             "first_acceptance_started_datetime",
+            "first_acceptance_reached_datetime",
             "first_acceptance_completed_datetime",
         ]