[statistics visualize] 折れ線-横軸_日-全体.htmlなどの折れ線グラフに、フェーズごとの作業時間を表示するう…

…ようにしました。 (#810)

* プロットする折れ線がない場合に、ファイルを出力しないようにする

* format && lint

* フェーズごとの作業時間をプロットする

* 折れ線グラフの日付が連続になるように配置

* version up

* versionup

annofabcli/__version__.py

@@ -1 +1 @@
-__version__ = "1.66.2"
+__version__ = "1.66.3"

annofabcli/statistics/linegraph.py

@@ -28,6 +28,8 @@
 class LineGraph:
     """
     折れ線グラフに対応するクラス
+
+    TODO: 引数にx_column, y_columnを受け取るのはよくないかもしれない
     """
 
     def __init__(
@@ -119,6 +121,9 @@ def config_legend(self) -> None:
         折れ線を追加した後に、凡例の位置などを設定します。
         """
         fig = self.figure
+        if len(fig.legend) == 0:
+            return
+
         fig.legend.location = "top_left"
         fig.legend.click_policy = "hide"
 
@@ -128,9 +133,9 @@ def config_legend(self) -> None:
             fig.legend.label_height = 10
             fig.legend.glyph_height = 10
 
-        if len(fig.legend) > 0:
-            legend = fig.legend[0]
-            fig.add_layout(legend, "left")
+        # グラフの左側に凡例を表示
+        legend = fig.legend[0]
+        fig.add_layout(legend, "left")
 
     def create_button_hiding_all_lines(self) -> Button:
         """

annofabcli/statistics/visualization/dataframe/cumulative_productivity.py

@@ -85,6 +85,7 @@ def _plot(self, line_graph_list: list[LineGraph], user_id_list: list[str], outpu
         # (例) 20000件をプロットする際、この対応がないと、ファイルサイズが3倍以上になる
         required_columns = list(set(itertools.chain.from_iterable([e.required_columns for e in line_graph_list])))
 
+        line_count = 0
         for user_index, user_id in enumerate(user_id_list):
             df_subset = df[df[f"first_{self.phase.value}_user_id"] == user_id]
             if df_subset.empty:
@@ -95,9 +96,14 @@ def _plot(self, line_graph_list: list[LineGraph], user_id_list: list[str], outpu
             color = get_color_from_palette(user_index)
             username = df_subset.iloc[0][f"first_{self.phase.value}_username"]
 
+            line_count += 1
             for line_graph in line_graph_list:
                 line_graph.add_line(source, legend_label=username, color=color)
 
+        if line_count == 0:
+            logger.warning(f"プロットするデータがなかっため、'{output_file}'は出力しません。")
+            return
+
         graph_group_list = []
         for line_graph in line_graph_list:
             line_graph.config_legend()

annofabcli/statistics/visualization/dataframe/productivity_per_date.py

@@ -286,6 +286,7 @@ def plot_annotation_metrics(
 
         logger.debug(f"{output_file} を出力します。")
 
+        line_count = 0
         for user_index, user_id in enumerate(user_id_list):
             df_subset = df[df["first_annotation_user_id"] == user_id]
             if df_subset.empty:
@@ -309,6 +310,7 @@ def plot_annotation_metrics(
             color = get_color_from_palette(user_index)
             username = df_subset.iloc[0]["first_annotation_username"]
 
+            line_count += 1
             for line_graph in line_graph_list:
                 if line_graph.y_column.endswith(WEEKLY_MOVING_AVERAGE_COLUMN_SUFFIX):
                     line_graph.add_moving_average_line(
@@ -324,6 +326,10 @@ def plot_annotation_metrics(
                         color=color,
                     )
 
+        if line_count == 0:
+            logger.warning(f"プロットするデータがなかっため、'{output_file}'は出力しません。")
+            return
+
         self._plot(line_graph_list, output_file)
 
     def plot_input_data_metrics(
@@ -414,6 +420,8 @@ def plot_input_data_metrics(
 
         logger.debug(f"{output_file} を出力します。")
 
+        line_count = 0
+
         for user_index, user_id in enumerate(user_id_list):
             df_subset = df[df["first_annotation_user_id"] == user_id]
             if df_subset.empty:
@@ -438,6 +446,7 @@ def plot_input_data_metrics(
             color = get_color_from_palette(user_index)
             username = df_subset.iloc[0]["first_annotation_username"]
 
+            line_count += 1
             for line_graph in line_graph_list:
                 if line_graph.y_column.endswith(WEEKLY_MOVING_AVERAGE_COLUMN_SUFFIX):
                     line_graph.add_moving_average_line(
@@ -453,6 +462,10 @@ def plot_input_data_metrics(
                         color=color,
                     )
 
+        if line_count == 0:
+            logger.warning(f"プロットするデータがなかっため、'{output_file}'は出力しません。")
+            return
+
         self._plot(line_graph_list, output_file)
 
     def to_csv(self, output_file: Path) -> None:
@@ -645,6 +658,7 @@ def plot_annotation_metrics(
 
         logger.debug(f"{output_file} を出力します。")
 
+        line_count = 0
         for user_index, user_id in enumerate(user_id_list):
             df_subset = df[df["first_inspection_user_id"] == user_id]
             if df_subset.empty:
@@ -680,6 +694,10 @@ def plot_annotation_metrics(
                         color=color,
                     )
 
+        if line_count == 0:
+            logger.warning(f"プロットするデータがなかっため、'{output_file}'は出力しません。")
+            return
+
         self._plot(line_graph_list, output_file)
 
     def plot_input_data_metrics(
@@ -752,6 +770,7 @@ def plot_input_data_metrics(
 
         logger.debug(f"{output_file} を出力します。")
 
+        line_count = 0
         for user_index, user_id in enumerate(user_id_list):
             df_subset = df[df["first_inspection_user_id"] == user_id]
             if df_subset.empty:
@@ -772,6 +791,7 @@ def plot_input_data_metrics(
             color = get_color_from_palette(user_index)
             username = df_subset.iloc[0]["first_inspection_username"]
 
+            line_count += 1
             for line_graph in line_graph_list:
                 if line_graph.y_column.endswith(WEEKLY_MOVING_AVERAGE_COLUMN_SUFFIX):
                     line_graph.add_moving_average_line(
@@ -787,6 +807,10 @@ def plot_input_data_metrics(
                         color=color,
                     )
 
+        if line_count == 0:
+            logger.warning(f"プロットするデータがなかっため、'{output_file}'は出力しません。")
+            return
+
         self._plot(line_graph_list, output_file)
 
     def to_csv(self, output_file: Path) -> None:
@@ -972,6 +996,7 @@ def plot_annotation_metrics(
 
         logger.debug(f"{output_file} を出力します。")
 
+        line_count = 0
         for user_index, user_id in enumerate(user_id_list):
             df_subset = df[df["first_acceptance_user_id"] == user_id]
             if df_subset.empty:
@@ -993,6 +1018,7 @@ def plot_annotation_metrics(
             color = get_color_from_palette(user_index)
             username = df_subset.iloc[0]["first_acceptance_username"]
 
+            line_count += 1
             for line_graph in line_graph_list:
                 if line_graph.y_column.endswith(WEEKLY_MOVING_AVERAGE_COLUMN_SUFFIX):
                     line_graph.add_moving_average_line(
@@ -1008,6 +1034,10 @@ def plot_annotation_metrics(
                         color=color,
                     )
 
+        if line_count == 0:
+            logger.warning(f"プロットするデータがなかっため、'{output_file}'は出力しません。")
+            return
+
         self._plot(line_graph_list, output_file)
 
     def plot_input_data_metrics(
@@ -1079,6 +1109,7 @@ def plot_input_data_metrics(
 
         logger.debug(f"{output_file} を出力します。")
 
+        line_count = 0
         for user_index, user_id in enumerate(user_id_list):
             df_subset = df[df["first_acceptance_user_id"] == user_id]
             if df_subset.empty:
@@ -1099,6 +1130,7 @@ def plot_input_data_metrics(
             color = get_color_from_palette(user_index)
             username = df_subset.iloc[0]["first_acceptance_username"]
 
+            line_count += 1
             for line_graph in line_graph_list:
                 if line_graph.y_column.endswith(WEEKLY_MOVING_AVERAGE_COLUMN_SUFFIX):
                     line_graph.add_moving_average_line(
@@ -1114,6 +1146,10 @@ def plot_input_data_metrics(
                         color=color,
                     )
 
+        if line_count == 0:
+            logger.warning(f"プロットするデータがなかっため、'{output_file}'は出力しません。")
+            return
+
         self._plot(line_graph_list, output_file)
 
     def to_csv(self, output_file: Path) -> None:

annofabcli/statistics/visualization/dataframe/whole_productivity_per_date.py

@@ -308,7 +308,15 @@ def add_velocity_columns(df: pandas.DataFrame):
                 df["monitored_worktime_hour"]
             ) / get_weekly_sum(df["task_count"])
 
-            for column in ["task_count", "input_data_count", "actual_worktime_hour", "monitored_worktime_hour"]:
+            for column in [
+                "task_count",
+                "input_data_count",
+                "actual_worktime_hour",
+                "monitored_worktime_hour",
+                "monitored_annotation_worktime_hour",
+                "monitored_inspection_worktime_hour",
+                "monitored_acceptance_worktime_hour",
+            ]:
                 df[f"{column}{WEEKLY_MOVING_AVERAGE_COLUMN_SUFFIX}"] = get_weekly_moving_average(df[column])
 
         def create_figure(title: str, y_axis_label: str) -> bokeh.plotting.Figure:
@@ -452,10 +460,7 @@ def create_input_data_figure():
         fig_info_list = [
             {
                 "figure": create_figure(title="日ごとの作業時間", y_axis_label="作業時間[hour]"),
-                "y_info_list": [
-                    {"column": "actual_worktime_hour", "legend": "実績作業時間"},
-                    {"column": "monitored_worktime_hour", "legend": "計測作業時間"},
-                ],
+                "y_info_list": [{"column": f"{e[0]}_hour", "legend": f"{e[1]}"} for e in phase_prefix],
             },
             {
                 "figure": create_figure(title="日ごとのタスクあたり作業時間", y_axis_label="タスクあたり作業時間[hour/task]"),
@@ -499,9 +504,6 @@ def create_input_data_figure():
             "input_data_count",
             "actual_worktime_hour",
             "monitored_worktime_hour",
-            "cumsum_task_count",
-            "cumsum_input_data_count",
-            "cumsum_actual_worktime_hour",
             "actual_worktime_hour/task_count",
             "actual_worktime_minute/input_data_count",
             "actual_worktime_minute/annotation_count",
@@ -643,12 +645,72 @@ def create_input_data_figure():
 
             return fig
 
+        def create_worktime_figure():
+            x_column_name = "dt_date"
+            fig = create_figure(title="日ごとの累積作業時間", y_axis_label="作業時間[hour]")
+
+            # 値をプロット
+            index = 0
+            plot_line_and_circle(
+                fig,
+                x_column_name=x_column_name,
+                y_column_name="cumsum_actual_worktime_hour",
+                source=source,
+                color=get_color_from_small_palette(index),
+                legend_label="実績作業時間",
+            )
+
+            index += 1
+            plot_line_and_circle(
+                fig,
+                x_column_name=x_column_name,
+                y_column_name="cumsum_monitored_worktime_hour",
+                source=source,
+                color=get_color_from_small_palette(index),
+                legend_label="計測作業時間",
+            )
+
+            index += 1
+            plot_line_and_circle(
+                fig,
+                x_column_name=x_column_name,
+                y_column_name="cumsum_monitored_annotation_worktime_hour",
+                source=source,
+                color=get_color_from_small_palette(index),
+                legend_label="計測作業時間(教師付)",
+            )
+
+            index += 1
+            plot_line_and_circle(
+                fig,
+                x_column_name=x_column_name,
+                y_column_name="cumsum_monitored_inspection_worktime_hour",
+                source=source,
+                color=get_color_from_small_palette(index),
+                legend_label="計測作業時間(検査)",
+            )
+
+            index += 1
+            plot_line_and_circle(
+                fig,
+                x_column_name=x_column_name,
+                y_column_name="cumsum_monitored_acceptance_worktime_hour",
+                source=source,
+                color=get_color_from_small_palette(index),
+                legend_label="計測作業時間(受入)",
+            )
+
+            return fig
+
         if not self._validate_df_for_output(output_file):
             return
 
         df = self.df.copy()
         df["dt_date"] = df["date"].map(lambda e: parse(e).date())
-        df["cumsum_monitored_worktime_hour"] = df["monitored_worktime_hour"].cumsum()
+
+        df["cumsum_monitored_annotation_worktime_hour"] = df["monitored_annotation_worktime_hour"].cumsum()
+        df["cumsum_monitored_inspection_worktime_hour"] = df["monitored_inspection_worktime_hour"].cumsum()
+        df["cumsum_monitored_acceptance_worktime_hour"] = df["monitored_acceptance_worktime_hour"].cumsum()
 
         logger.debug(f"{output_file} を出力します。")
 
@@ -664,10 +726,13 @@ def create_input_data_figure():
             "cumsum_input_data_count",
             "cumsum_actual_worktime_hour",
             "cumsum_monitored_worktime_hour",
+            "cumsum_monitored_annotation_worktime_hour",
+            "cumsum_monitored_inspection_worktime_hour",
+            "cumsum_monitored_acceptance_worktime_hour",
         ]
         hover_tool = create_hover_tool(tooltip_item)
 
-        fig_list = [create_task_figure(), create_input_data_figure()]
+        fig_list = [create_task_figure(), create_input_data_figure(), create_worktime_figure()]
 
         for fig in fig_list:
             fig.add_tools(hover_tool)