refacto: rename from utils

smartpm/visuals.py

@@ -0,0 +1,359 @@
+import matplotlib.pyplot as plt
+import datetime
+import pandas as pd
+
+from collections import defaultdict
+
+def plot_percent_complete_curve(json_data):
+    """
+    Plot the progress curve from scenario data.
+    Reproduces this figure from SmartPM: https://help.smartpmtech.com/the-progress-curve
+
+    Parameters
+    ----------
+    json_data : dict
+        Dictionary containing percent complete types and scenario data with progress information.
+    """
+    percent_complete_types = json_data['percentCompleteTypes']
+    scenario_data = json_data['data']
+
+    # Extract dates and values for each type of progress
+    dates = [datetime.datetime.strptime(point['DATE'], '%Y-%m-%d') for point in scenario_data]
+
+    # Extract data points for each type based on percentCompleteTypes
+    data_series = {key: [point.get(key, None) for point in scenario_data] for key in percent_complete_types.keys()}
+
+    # Define colors and linestyles for each type
+    colors = {
+        "ACTUAL": "steelblue",
+        "SCHEDULED": "seagreen",
+        "PLANNED": "seagreen",
+        "PREDICTIVE": "steelblue",
+        "LATE_DATE_PLANNED": "firebrick",
+    }
+    markers = {
+        "ACTUAL": "s",
+        "SCHEDULED": "^",
+        "PLANNED": "o",
+        "PREDICTIVE": "v",
+        "LATE_DATE_PLANNED": "d",
+    }
+
+    linestyles = {
+        "ACTUAL": "-",
+        "SCHEDULED": "--",
+        "PLANNED": "-",
+        "PREDICTIVE": "--",
+        "LATE_DATE_PLANNED": "-",
+    }
+
+    plt.figure(figsize=(14, 6))
+    ms = 4  # marker size
+    lw = 2  # linewidth
+
+    # Plot each data series
+    for key, label in percent_complete_types.items():
+        if any(data_series[key]):
+            plt.plot(dates, data_series[key], label=label.split(" (")[0], marker=markers.get(key, 'o'), markersize=ms, linestyle=linestyles.get(key, '-'), linewidth=lw, color=colors.get(key, 'dodgerblue'))
+
+    # Shade the region between early date planned and late date planned lines if available
+    if 'PLANNED' in data_series and 'LATE_DATE_PLANNED' in data_series:
+        plt.fill_between(dates, data_series['PLANNED'], data_series.get('LATE_DATE_PLANNED', [None]*len(dates)), color='gray', alpha=0.3)
+
+    plt.title('Planned VS Actual Percent Complete')
+
+    # Customize x-axis to show the first of every month with the format "mm/dd/yy"
+    plt.gca().xaxis.set_major_formatter(plt.matplotlib.dates.DateFormatter('%m/%d/%y'))
+    plt.gca().xaxis.set_major_locator(plt.matplotlib.dates.MonthLocator(bymonthday=1, interval=2))
+
+    # Rotate the x-axis labels by -30 degrees and align them to the left
+    plt.xticks(rotation=-30, ha='left')
+
+    # Customize y-axis to show 0, 25, 50, 75, 100
+    plt.yticks([0, 25, 50, 75, 100])
+
+    # Remove top, right, and left spines
+    ax = plt.gca()
+    ax.spines['top'].set_visible(False)
+    ax.spines['right'].set_visible(False)
+    ax.spines['left'].set_visible(False)
+
+    # Draw a vertical line for the current date if it is before the last date in predictive completion
+    current_date = datetime.datetime.now()
+    if current_date < max(dates):
+        plt.axvline(x=current_date, color='black', linestyle='-', linewidth=lw + 1, label='Current Date')
+
+    # Place the legend below the x-axis with no box around it on one line
+    plt.legend(loc='upper center', bbox_to_anchor=(0.5, -0.15), ncol=7, frameon=False)
+    plt.grid(True)
+
+    plt.tight_layout()
+    plt.show()
+
+def plot_earned_schedule_curve(json_data):
+    """
+    Plot the earned days curve from the provided JSON data.
+    Reproduces this figure: https://help.smartpmtech.com/earned-baseline-days
+
+    Parameters
+    ----------
+    json_data : dict
+        Dictionary containing the date and days data.
+    """
+    data = json_data['data']
+
+    # Extract dates and values for each type of days
+    dates = [datetime.datetime.strptime(point['date'], '%Y-%m-%d') for point in data]
+    earned_days = [point['earnedDays'] for point in data]
+    planned_days = [point['plannedDays'] for point in data]
+    predictive_days = [point['predictiveDays'] if point['predictiveDays'] is not None else None for point in data]
+
+    plt.figure(figsize=(14, 6))
+    ms = 4  # marker size
+    lw = 2  # linewidth
+
+    plt.plot(dates, earned_days, label='Earned Days', marker='o', markersize=ms, linestyle='-', linewidth=lw, color="seagreen")
+    plt.plot(dates, planned_days, label='Planned Days', marker='d', markersize=ms, linestyle='-', linewidth=lw, color="firebrick")
+    if any(predictive_days):
+        plt.plot(dates, predictive_days, label='Predictive Days', marker='s', markersize=ms, linestyle='-', linewidth=lw, color="goldenrod")
+
+    plt.title('Earned Baseline Days')
+
+    # Customize x-axis to show the first of every month with the format "mm/dd/yy"
+    plt.gca().xaxis.set_major_formatter(plt.matplotlib.dates.DateFormatter('%m/%d/%y'))
+    plt.gca().xaxis.set_major_locator(plt.matplotlib.dates.MonthLocator(bymonthday=1, interval=2))
+
+    # Rotate the x-axis labels by -30 degrees and align them to the left
+    plt.xticks(rotation=-30, ha='left')
+
+    # Remove top, right, and left spines
+    ax = plt.gca()
+    ax.spines['top'].set_visible(False)
+    ax.spines['right'].set_visible(False)
+    ax.spines['left'].set_visible(False)
+
+    # Draw a vertical line for the current date if it is before the last date in data
+    current_date = datetime.datetime.now()
+    if current_date < max(dates):
+        plt.axvline(x=current_date, color='black', linestyle='-', linewidth=lw + 1, label='Current Date')
+
+    # Place the legend below the x-axis with no box around it on one line
+    plt.legend(loc='upper center', bbox_to_anchor=(0.5, -0.15), ncol=7, frameon=False)
+    plt.grid(True)
+
+    plt.tight_layout()
+    plt.show()
+
+def plot_schedule_delay(json_data):
+    """
+    Plot the schedule delay curve from the provided JSON data.
+
+    Parameters
+    ----------
+    json_data : dict
+        Dictionary containing the schedule variance data.
+    """
+    data = json_data
+
+    # Extract dates and values for each type of variance
+    dates = [datetime.datetime.strptime(point['dataDate'], '%Y-%m-%dT%H:%M:%S') for point in data]
+    end_date_variance = [point['endDateVariance']['cumulative'] for point in data]
+    critical_path_delay = [point['criticalPathDelay']['cumulative'] for point in data]
+    acceleration = [point['delayRecovery']['cumulative']*-1 for point in data] # values are inverse in response
+
+    plt.figure(figsize=(14, 6))
+    ms = 4  # marker size
+    lw = 2  # linewidth
+
+    plt.plot(dates, end_date_variance, label='End Date Variance', marker='o', markersize=ms, linestyle='-', linewidth=lw, color="goldenrod")
+    plt.plot(dates, critical_path_delay, label='Critical Path Delay', marker='d', markersize=ms, linestyle='-', linewidth=lw, color="firebrick")
+    plt.plot(dates, acceleration, label='Acceleration', marker='s', markersize=ms, linestyle='-', linewidth=lw, color="seagreen")
+
+    # Shade the area between each curve and the x-axis
+    plt.fill_between(dates, end_date_variance, color="goldenrod", alpha=0.7)
+    plt.fill_between(dates, critical_path_delay, color="firebrick", alpha=0.7)
+    plt.fill_between(dates, acceleration, color="seagreen", alpha=0.7)
+
+    plt.title('Schedule Delay Over Time')
+
+    # Customize x-axis to show the first of every month with the format "mm/dd/yy"
+    plt.gca().xaxis.set_major_formatter(plt.matplotlib.dates.DateFormatter('%m/%d/%y'))
+    plt.gca().xaxis.set_major_locator(plt.matplotlib.dates.MonthLocator(bymonthday=1, interval=2))
+
+    # Rotate the x-axis labels by -30 degrees and align them to the left
+    plt.xticks(rotation=-30, ha='left')
+
+    # Remove top, right, and left spines
+    ax = plt.gca()
+    ax.spines['top'].set_visible(False)
+    ax.spines['right'].set_visible(False)
+    ax.spines['left'].set_visible(False)
+
+    # Draw a vertical line for the current date if it is before the last date in data
+    current_date = datetime.datetime.now()
+    if current_date < max(dates):
+        plt.axvline(x=current_date, color='black', linestyle='-', linewidth=lw + 1, label='Current Date')
+
+    # Place the legend below the x-axis with no box around it on one line
+    plt.legend(loc='upper center', bbox_to_anchor=(0.5, -0.15), ncol=7, frameon=False)
+    plt.grid(True)
+
+    plt.tight_layout()
+    plt.show()
+
+def plot_schedule_changes(json_data):
+    """
+    Plot the schedule changes over time from the provided JSON data.
+
+    Parameters
+    ----------
+    json_data : list of dict
+        List of dictionaries containing the schedule change data.
+    """
+    # Extract dates and metrics
+    dates = [datetime.datetime.strptime(entry['dataDate'], '%Y-%m-%dT%H:%M:%S') for entry in json_data]
+
+    metrics = json_data[0]['metrics'].keys()
+
+    # Initialize a dictionary to hold lists of metric values
+    metric_values = {metric: [] for metric in metrics}
+
+    # Populate the metric values dictionary
+    for entry in json_data:
+        for metric in metrics:
+            metric_values[metric].append(entry['metrics'][metric])
+
+    # Define colors and linestyles for each metric
+    color_linestyle_dict = {
+        "CriticalChanges": {"color": 'red', "marker": 'o'},
+        "NearCriticalChanges": {"color": 'goldenrod', "marker": 'd'},
+        "ActivityChanges": {"color": 'seagreen', "marker": 's'},
+        "LogicChanges": {"color": 'blue', "marker": '^'},
+        "CalendarChanges": {"color": 'dodgerblue', "marker": 'v'},
+        "DurationChanges": {"color": 'steelblue', "marker": 'o'},
+        "DelayedActivityChanges": {"color": 'firebrick', "marker": 'd'},
+    }
+
+    plt.figure(figsize=(14, 10))
+    ms = 4  # marker size
+    lw = 2  # linewidth
+
+    for metric, style in color_linestyle_dict.items():
+        plt.plot(dates, metric_values[metric], label=metric, marker=style["marker"], markersize=ms, linewidth=lw, color=style["color"])
+
+    plt.title('Schedule Changes Over Time')
+
+    # Customize x-axis to show the first of every month with the format "mm/dd/yy"
+    plt.gca().xaxis.set_major_formatter(plt.matplotlib.dates.DateFormatter('%m/%d/%y'))
+    plt.gca().xaxis.set_major_locator(plt.matplotlib.dates.MonthLocator(bymonthday=1, interval=2))
+
+    # Rotate the x-axis labels by -30 degrees and align them to the left
+    plt.xticks(rotation=-30, ha='left')
+
+    # Remove top, right, and left spines
+    ax = plt.gca()
+    ax.spines['top'].set_visible(False)
+    ax.spines['right'].set_visible(False)
+    ax.spines['left'].set_visible(False)
+
+    # Draw a vertical line for the current date if it is before the last date in data
+    current_date = datetime.datetime.now()
+    if current_date < max(dates):
+        plt.axvline(x=current_date, color='black', linestyle='-', linewidth=lw + 1, label='Current Date')
+
+    # Place the legend below the x-axis with no box around it on one line
+    plt.legend(loc='upper center', bbox_to_anchor=(0.5, -0.15), ncol=4, frameon=False)
+    plt.grid(True)
+
+    plt.tight_layout()
+    plt.show()
+
+def plot_activity_distribution_by_month(json_data, scenario_details):
+    """
+    Plot the activity start and finish dates by month from the provided JSON data.
+
+    Parameters
+    ----------
+    json_data : list of dict
+        List of dictionaries containing the activity data.
+    """
+    # Data date
+    data_date = datetime.datetime.strptime(scenario_details["dataDate"], "%Y-%m-%d")
+
+    # Initialize a dictionary to hold counts for each month and category
+    counts = defaultdict(lambda: defaultdict(int))
+
+    # Extract and count the relevant dates
+    for entry in json_data:
+        dates = {
+            'Baseline Starts': entry['baseline'].get('startDate'),
+            'Baseline Finishes': entry['baseline'].get('finishDate')
+        }
+        # Start Dates
+        if pd.to_datetime(entry['actualStartDate']):
+            dates['Current Starts (Actual)'] = entry['actualStartDate']
+
+        if pd.to_datetime(entry['startDate']) is not None and pd.to_datetime(entry['startDate']) >= data_date:
+            dates['Current Starts (Planned)'] = entry['startDate']
+
+        # Finish Dates
+        if pd.to_datetime(entry['actualFinishDate']):
+            dates['Current Finishes (Actual)'] = entry['actualFinishDate']
+        elif pd.to_datetime(entry['finishDate']) is not None:
+            dates['Current Finishes (Planned)'] = entry['finishDate']
+
+        for date_type, date_str in dates.items():
+            if date_str:
+                date_obj = pd.to_datetime(date_str)
+                month_year = (date_obj).strftime('%Y-%m')
+                counts[month_year][date_type] += 1
+
+    # Convert the counts dictionary to a DataFrame for easier plotting
+    df_counts = pd.DataFrame(counts).fillna(0).T
+    df_counts.index = pd.to_datetime(df_counts.index)
+
+    # Define colors for each date type
+    colors = {
+        'Current Starts (Actual)': 'lightsteelblue',
+        'Current Finishes (Actual)': 'cornflowerblue',
+        'Current Starts (Planned)': 'lightgreen',
+        'Current Finishes (Planned)': 'darkgreen',
+        'Baseline Starts': 'lightgrey',
+        'Baseline Finishes': 'darkgrey'
+    }
+
+    # Plotting
+    _, ax = plt.subplots(figsize=(14, 8))
+    width = 3  # width of the bars
+
+    # Plot bars for each date type
+    for i, date_type in enumerate(df_counts.columns):
+        offset = (i - (len(df_counts.columns) - 1) / 2) * width  # center the bars around the first of the month
+        ax.bar(df_counts.index + pd.DateOffset(months=-1, days=offset), df_counts[date_type], width=width, label=date_type, color=colors[date_type])
+
+    # Customize x-axis to show the first of every month with the format "mm/yyyy"
+    ax.xaxis.set_major_formatter(plt.matplotlib.dates.DateFormatter('%b-%y'))
+    ax.xaxis.set_major_locator(plt.matplotlib.dates.MonthLocator())
+
+    # Rotate the x-axis labels by -30 degrees and align them to the right
+    plt.xticks(rotation=-30, ha='right')
+
+    # Remove top, right, and left spines
+    ax.spines['top'].set_visible(False)
+    ax.spines['right'].set_visible(False)
+    ax.spines['left'].set_visible(False)
+
+    # Draw a vertical line for the current date and add the current date text
+    ax.axvline(x=data_date + pd.DateOffset(months=-1), color='black', linestyle='--', linewidth=1)
+    max_height = df_counts.max().max()
+    ax.text(data_date + pd.DateOffset(months=-1, days=1), max_height, f"Data Date: {data_date.strftime('%d %b-%y')}", rotation=-90, verticalalignment='top', horizontalalignment='left')
+
+    # Place the legend below the x-axis with no box around it on one line
+    plt.legend(loc='upper center', bbox_to_anchor=(0.5, -0.10), ncol=6, frameon=False)
+    plt.grid(True, axis='y')
+
+    plt.title('Monthly Activity Start & Finish Distribution')
+    plt.tight_layout()
+    plt.show()
+
+    return df_counts.sort_index()