energy-reports-pkg 0.6.14__py3-none-any.whl

mindsett_energy_reports_pkg/mdl/prepare_and_email_reports/prepare_report/generate_components/generate_energy_meter_with_benchmarking/energy_meter_with_benchmarking/energy_meter_design/__init__.py

@@ -0,0 +1,233 @@
+
+from matplotlib.offsetbox import OffsetImage, AnnotationBbox
+
+import matplotlib.image as mpimg
+import matplotlib.patches as mpatches
+import matplotlib.text as mpltext
+
+from os import path
+
+cd = {"tomato": '#FF836A',"aquablue": '#6DC2B3',"peach": '#FED6D2',"darkgrey": '#9F9D9C',"potato": '#FEF8C8',
+      "cyan": '#B6E4E1',"dimgray":'#696969',"Seafoam":'#87e0cf',"gainsboro":'#DCDCDC'}
+
+def energy_meter_design(ax, consumption_mwh_cur, consumption_mwh_pre, image_folder,
+                        conv_mwh_co2 = 0.233,
+                        conv_mwh_price = 190,
+                        fx = 0.2, # fancy box x-axis value
+                        fy = 0.3, # fancy box y-axis value
+                        fs  = 0.6, # fancy box scale factor
+                        currency = 'GBP',
+                        fontsize = 13
+                        ):
+
+    ax.set_xlim(0.2, 0.9)
+    ax.set_ylim(0.2, 0.9)
+    iz  = 0.06 * fs # icon zooming factor
+
+    # add a fancy box as the energy meter outline
+
+    fancybox_width  = fs
+    fancybox_height = fs
+    fancybox = mpatches.FancyBboxPatch(
+                [fx, fy], fancybox_width, fancybox_height, fc="w", ec='k', lw=1.5, ls= 'dotted',
+                boxstyle=mpatches.BoxStyle("Round", rounding_size=0.07, pad=0.02))
+    ax.add_artist(fancybox)
+
+    notes_fontsize = fontsize - 2
+    # text sub-title 
+    text_title = mpltext.Text(x=fx+0.45*fs, y=fy+0.7*fs, text=f'Equivalent$^*$', 
+                              va=u'baseline', color='grey', ha=u'right', fontweight='light',fontstyle='italic', fontsize=notes_fontsize)
+    ax.add_artist(text_title)
+
+    # text footnote
+
+    ## currency_symbol selection
+    currency_symbol_dict = {
+        'GBP': '£',
+        'EUR': '€',
+        'USD': '\$',
+        'SGD': 'S\$'
+    }
+    try: 
+        currency_symbol = currency_symbol_dict[currency]
+    except:
+        raise Exception('There is no symbol provided for this currency!')
+
+    ## symbol selection
+    text_footnote_with_co2 = f'* {conv_mwh_co2:1.3f}$t$ CO2e $\Leftrightarrow$ 1MWh $\Leftrightarrow$ {currency_symbol}{conv_mwh_price:1.1f}' # not used for now - 2024 Sep 23th
+    text_footnote_without_co2 = f'* 1MWh $\Leftrightarrow$ {currency_symbol}{conv_mwh_price:1.1f}'
+    text_footnote = mpltext.Text(x=fx+0.5*fs, y=fy+0.04*fs, text=text_footnote_without_co2, 
+                                 va=u'baseline', color='k', ha=u'center', fontsize=notes_fontsize)
+    ax.add_artist(text_footnote)
+
+    ## text readings settings
+    re_w = 0.80   # reading width
+    re_x = 0.127  # reading x-axis value
+    re_y = 0.80   # reading y-axis value
+    re_c = "k"    # reading colour
+    r_fs = fontsize + 3   # reading fontsize # please note that the other sizes except default 13 haven't been tested yet
+    mr_fs = fontsize + 5    # main reading fontsize # please note that the other sizes except default 13 haven't been tested yet
+
+    ## reading - energy 
+    reading = consumption_mwh_cur
+
+    reading_abs = abs(reading) # the default unit is MWh
+
+    # adjust the decimal point and the unit based on the actual value
+    if reading_abs > 10:
+        reading_value = f"{reading:.0f} MWh"
+    elif reading_abs > 1:
+        reading_value = f"{reading:.1f} MWh"
+    elif reading_abs*1000 > 10:
+        reading_value = f"{reading*1000:.0f} KWh"
+    elif reading_abs*1000 > 1:
+        reading_value = f"{reading*1000:.1f} KWh"
+    elif reading_abs*1000_000 > 10:
+        reading_value = f"{reading*1000_000:.0f} Wh"
+    else:
+        reading_value = f"{reading*1000_000:.1f} Wh"
+    # reading - energy - text - monthly total
+    rt_x = re_x + 0.15 + 0.03
+    rt_y = re_y
+    readingtext = mpltext.Text(x=fx+rt_x*fs, y=fy+rt_y*fs, text=f'{reading_value}', 
+                               va=u'bottom', color=re_c, ha=u'center', fontweight="bold", fontsize=mr_fs)
+
+    # reading - energy - text - monthly change
+    rt_x = re_x + 0.65
+    rt_y = re_y + 0.02
+
+
+    change_value = (consumption_mwh_cur - consumption_mwh_pre) / abs(consumption_mwh_pre)
+    change_value_int = int(change_value*100)
+
+    # print(f'{consumption_mwh_cur = }')
+    # print(f'{consumption_mwh_pre = }')
+    # print(f'{change_value = }')
+
+
+    if change_value_int > 0: 
+        change_arrow_str = r'${\blacktriangle}$'
+    elif change_value_int < 0:
+        change_arrow_str = r'$\:\!\triangledown\:\!$'
+    else :
+        change_arrow_str = r'--'
+
+
+    change_in_percentage_value = int(abs(change_value)*100)
+
+    if change_in_percentage_value > 999:
+        change_in_percentage_value_str = r"$\,$" + ">999" + "%"
+    elif change_in_percentage_value > 99: 
+        change_in_percentage_value_str = r"$\,$" + str(change_in_percentage_value) + "%"
+    elif change_in_percentage_value > 9: 
+        change_in_percentage_value_str = " " + str(change_in_percentage_value) + r"$\,$" + "%"
+    else:
+        change_in_percentage_value_str = " " + str(change_in_percentage_value) + r"$\;$" + "%"
+
+
+    change_in_percentage = change_arrow_str + change_in_percentage_value_str
+    text_percentage = mpltext.Text(x=fx+rt_x*fs, y=fy+rt_y*fs, text=f'{change_in_percentage}', 
+                                   va=u'baseline', color='k', ha=u'center', fontweight="bold", fontsize=mr_fs)
+    ax.add_artist(text_percentage)
+
+    # horizontal line to separate the different readings
+    rl_x = re_x * 0.2       # the x-axis value of the line at the bottom of the reading 
+    rl_y = re_y - 0.15        # the y-axis value of the line at the bottom of the reading
+    rl_w = re_w * 1.2    # the width value of the line at the bottom of the reading
+    horizontal_line = mpatches.Rectangle([fx+rl_x*fs, fy+rl_y*fs], 
+                                         rl_w*fs, 0, facecolor="k", edgecolor='k', lw=1, ls= 'dotted')
+    ax.add_artist(readingtext)
+    ax.add_artist(horizontal_line)
+
+
+    ## reading - co2
+    re_y = 0.48
+    reading = consumption_mwh_cur * conv_mwh_co2
+
+    # adjust the unit based on value
+    reading_value = f"{reading:1.1f} tons"
+    if reading_value == '0.0 tons':
+        reading_value = f"{round(reading*1000)} kg"
+        
+    image_name = 'co2.png'
+
+    # reading - co2 - icon
+    ri_x = re_x + 0.09  # reading icon x-axis value
+    ri_y = re_y + 0.042
+    icon = mpimg.imread(image_folder + image_name)
+    ibox = OffsetImage(icon, zoom=2*iz)
+    readingicon = AnnotationBbox(ibox, (fx+ri_x*fs, fy+ri_y*fs), frameon = False)
+
+    # reading - co2 - text
+    rt_x = re_x + 0.342
+    rt_y = re_y
+    readingtext = mpltext.Text(x=fx+rt_x*fs, y=fy+rt_y*fs, text=f'{reading_value}', 
+                               va=u'bottom', color=re_c, ha=u'left', fontweight="bold", fontsize=r_fs)
+
+    # horizontal line to separate the different readings
+    rl_x = re_x         # the x-axis value of the line at the bottom of the reading 
+    rl_y = re_y - 0.07      # the y-axis value of the line at the bottom of the reading
+    rl_w = re_w * 0.9   # the width value of the line at the bottom of the reading
+    horizontal_line = mpatches.Rectangle([fx+rl_x*fs, fy+rl_y*fs], 
+                                         rl_w*fs, 0, facecolor="k", edgecolor='k', lw=1, ls= 'dotted')
+
+    ax.add_artist(readingicon)
+    ax.add_artist(readingtext)
+    ax.add_artist(horizontal_line)
+
+
+    ## reading - billing
+    re_y = 0.23
+    reading = consumption_mwh_cur * conv_mwh_price
+    reading_value = f"{reading:,.2f}".replace(".", ". ") 
+
+    # adjust the size for visual effect
+    if len(reading_value) > 8:
+        reading_value = reading_value.replace(". ", ".") 
+        if len(reading_value) > 8:
+            reading_value = reading_value[:-2]
+
+    # reading - billing - icon
+
+    ## icon selection
+    currency_icon_dict = { # please note that the format has to be 512px png
+        'GBP': 'pound-sterling.png',
+        'EUR': 'euros.png',
+        'USD': 'dollar-symbol.png', # https://www.flaticon.com/free-icon/dollar-symbol_1140418?term=dollar+symbol&page=1&position=22&origin=search&related_id=1140418
+        'SGD': 'singapore-dollar.png' # https://www.flaticon.com/free-icon/singapore-dollar_9921422?term=singapore+dollar&page=3&position=73&origin=search&related_id=9921422
+    }
+    try: 
+        image_name = currency_icon_dict[currency]
+    except:
+        raise Exception('There is no icon provided for this currency!')
+
+    ## icon plotting
+    ri_x = re_x + 0.09  # reading icon x-axis value
+    ri_y = re_y + 0.06
+    icon = mpimg.imread(image_folder + image_name)
+    ibox = OffsetImage(icon, zoom=1.6*iz)
+    readingicon = AnnotationBbox(ibox, (fx+ri_x*fs, fy+ri_y*fs), frameon = False)
+
+    # reading - billing - text
+    rt_x = re_x + 0.342
+    rt_y = re_y
+    readingtext = mpltext.Text(x=fx+rt_x*fs, y=fy+rt_y*fs, text=f'{reading_value}', 
+                               va=u'bottom', color=re_c, ha=u'left', fontweight="bold", fontsize=r_fs)
+
+    # horizontal line to separate the different readings
+    rl_x = re_x          # the x-axis value of the line at the bottom of the reading 
+    rl_y = re_y - 0.07        # the y-axis value of the line at the bottom of the reading
+    rl_w = re_w * 0.9    # the width value of the line at the bottom of the reading
+    horizontal_line = mpatches.Rectangle([fx+rl_x*fs, fy+rl_y*fs], 
+                                         rl_w*fs, 0, facecolor="k", edgecolor='k', lw=1, ls= 'dotted')
+
+    ax.add_artist(readingicon)
+    ax.add_artist(readingtext)
+    ax.add_artist(horizontal_line)
+
+
+    ## benchmarking design
+    re_y = 0.65
+    ri_y = re_y + 0.07*fs
+
+    ax.axis('equal')

mindsett_energy_reports_pkg/mdl/prepare_and_email_reports/prepare_report/generate_components/generate_energy_meter_with_benchmarking/preprocessing_for_energy_meter_with_benchmarking/__init__.py

@@ -0,0 +1,23 @@
+
+
+def preprocessing_for_energy_meter_with_benchmarking(df_meta_with_value_building,
+                                                     period_column="period",  
+                                                     kwh_column="kwh"):
+
+    df_period_total = df_meta_with_value_building.groupby([period_column])[kwh_column].sum().div(1000).tail(2)
+
+    # for avoiding the case that we don't have data for the previous week
+    if df_period_total.shape[0] < 2:
+        consumption_mwh_cur = df_period_total.to_list()[0]
+        consumption_mwh_pre = consumption_mwh_cur 
+    else:
+        consumption_mwh_pre, consumption_mwh_cur = df_period_total.to_list()
+
+    period_freqstr = df_period_total.index[-1].freqstr
+
+    if period_freqstr == "M":
+        days_in_period = df_period_total.index[-1].daysinmonth
+    else:
+        days_in_period = 7
+
+    return consumption_mwh_cur, consumption_mwh_pre, days_in_period

mindsett_energy_reports_pkg/mdl/prepare_and_email_reports/prepare_report/generate_components/generate_insight_statements/__init__.py

@@ -0,0 +1,42 @@
+from pathlib import Path
+import pickle
+
+from .preprocessing_for_statement import preprocessing_for_statement
+from .insight_statements import insight_statements
+
+
+def generate_insight_statements(db, df_meta_with_value, 
+                                asset_group = "asset_class",
+                                # fixed_group_to_filter = [],
+                                directory_to_savefile=None): # todo: update the default value for directory
+
+    df_for_statements = preprocessing_for_statement(df_meta_with_value, asset_group=asset_group)
+
+    # print('df_for_statements: ', df_for_statements) # debug
+
+    if df_for_statements is not None:
+        # get the readable name of period frequency
+        current_period_obj = df_meta_with_value.period.max()
+        period_type = current_period_obj.freqstr[0]
+        period_freq_str_dict = {'W': 'week', 'M': 'month'}
+        period_freq_str = period_freq_str_dict[period_type]
+
+        statements_list = insight_statements(db, df_for_statements, df_meta_with_value, period_freq_str=period_freq_str)
+    else:
+        statements_list = ['Auto observations for 24/7 sites are not yet available!']
+
+    # Specify the directory to save figures, if it does not exist, create it
+    # Path(directory_to_savefile).mkdir(parents=True, exist_ok=True)
+
+    # with open(directory_to_savefile+'statements.pkl', 'wb') as f:
+    #     pickle.dump(statements_list, f)
+
+    pkl_name = 'statements.pkl'
+    if directory_to_savefile == None:
+        # obj_object = io.BytesIO()
+        return {pkl_name: statements_list}
+    else:
+        pkl_path = directory_to_savefile + pkl_name
+        with open(pkl_path, 'wb') as f:
+            pickle.dump(statements_list, f)
+        return {pkl_name: pkl_path}

mindsett_energy_reports_pkg/mdl/prepare_and_email_reports/prepare_report/generate_components/generate_insight_statements/insight_statements/__init__.py

@@ -0,0 +1,41 @@
+
+from .statement_for_biggest_ooh import statement_for_biggest_ooh
+from .statement_for_total_ooh import statement_for_total_ooh
+from .statement_for_avg_action_time import statement_for_avg_action_time
+
+
+def insight_statements(db, df_for_statements, df_meta_with_value, period_freq_str='week'):   #df_meta_with_value is only used to get metadata information
+    
+    statements_list = []
+
+    statement_str_total_ooh = statement_for_total_ooh(df_for_statements, 
+                                                      period_freq_str=period_freq_str)
+    statements_list.append(statement_str_total_ooh)
+
+    # preparation for the third statement
+
+    thing_name = 'Pizza Oven'
+    
+    if thing_name in df_meta_with_value['thing_name'].str.lstrip().str.rstrip("0123456789 ").unique():
+
+        building_name = df_meta_with_value['building_name'].unique()[0]
+        max_period = df_meta_with_value["period"].max()
+        start_time_str = max_period.start_time
+        end_time_str = max_period.end_time
+
+        statement_str_avg_action_time = statement_for_avg_action_time(db, building_name, thing_name, start_time_str, end_time_str,
+                                                                      action=1, 
+                                                                      period_freq_str=period_freq_str) # None will be returned if no on/off data is found
+        if statement_str_avg_action_time  is not None: 
+
+            statements_list.append(statement_str_avg_action_time)
+
+    # Statement for biggest OOH
+
+    # print('df_for_statements: ', df_for_statements)
+
+    statement_str_biggest_ooh = statement_for_biggest_ooh(df_for_statements, 
+                                                          period_freq_str=period_freq_str)
+    statements_list.append(statement_str_biggest_ooh)
+        
+    return statements_list

mindsett_energy_reports_pkg/mdl/prepare_and_email_reports/prepare_report/generate_components/generate_insight_statements/insight_statements/statement_for_avg_action_time/__init__.py

@@ -0,0 +1,61 @@
+from datetime import timedelta
+import pandas as pd
+# from sqlalchemy import create_engine
+
+from mdl.utils_ import enriching_time_features
+
+def statement_for_avg_action_time(db, building_name, thing_name, start_time, end_time,
+                                  action=1,
+                                  period_freq_str='week'):
+
+    
+
+    time_restriction = f"""(time >= '{start_time}') and (time < '{end_time}')"""
+
+    statement_list = [f""""root_name"='{building_name}'"""]
+    statement_full = " and ".join(statement_list)
+
+    query = f"""select time,action,root_name,thing_name from {db.table_on_off.name} where {statement_full} and {time_restriction};"""
+
+    engine = db.engine
+
+    with engine.connect() as conn:
+        df_on_off = pd.read_sql_query(query, con=conn)
+
+    df_on_off['thing_name'] = df_on_off['thing_name'].str.lstrip().str.rstrip("0123456789 ")
+
+    df_on_off_selected = df_on_off.loc[df_on_off['thing_name']==thing_name]
+    
+
+    if df_on_off_selected.shape[0] > 0: # handle the  case that no on/off data is returned
+
+        df_on_off_selected = enriching_time_features(df_on_off_selected)
+
+        df_on_off_avg = df_on_off_selected.groupby(['action']).time_of_day_in_float.mean()
+
+        avg_start_time = str(timedelta(hours=df_on_off_avg[action])).split('.')[0][:-3]
+
+        # AM/PM to the time
+
+        time_list = avg_start_time.split(":")
+
+        hour_digit = int(time_list[0])
+        if hour_digit < 12:
+            letter = " AM" 
+        else: 
+            letter = " PM"
+            if hour_digit >= 13:
+                time_list[0] = str(hour_digit-12)
+
+        time_list.append(letter)
+        time_list[1] = ":"+time_list[1]
+        avg_start_time_with_letter = "".join(time_list)   
+        
+        start_finish_dict = {1: 'start', -1: 'finish'}
+
+
+        statement = f"{avg_start_time_with_letter} was the average {start_finish_dict[action]} time for the {thing_name} over this {period_freq_str}." # temp fix
+    else: 
+        statement = None
+    
+    return statement

mindsett_energy_reports_pkg/mdl/prepare_and_email_reports/prepare_report/generate_components/generate_insight_statements/insight_statements/statement_for_biggest_ooh/__init__.py

@@ -0,0 +1,31 @@
+
+
+def statement_for_biggest_ooh(df_asset_group_period_sum_others, number_for_pick_out=3,  period_freq_str='week'):
+
+    # print('df_asset_group_period_sum_others: ', df_asset_group_period_sum_others) # debug
+
+    if (number_for_pick_out+1) > df_asset_group_period_sum_others.shape[0]: # there are two rows for "Total" and "Others", this is to avoid them to be shown.
+
+        number_for_pick_out = df_asset_group_period_sum_others.shape[0] - 2
+
+        print(f'[WARN]: in func [statement_for_biggest_ooh], the no. of entries in df_asset_group_period_sum_others [{df_asset_group_period_sum_others.shape[0]}] is smaller then the expected number_for_pick_out [ {number_for_pick_out}]', number_for_pick_out)
+
+    df_ooh_biggest = df_asset_group_period_sum_others.head(number_for_pick_out+1).tail(number_for_pick_out).drop(columns=['gt_pct_others','sum_for_sort'])
+    
+    # pdb.set_trace()
+
+    statement = f"""The largest out-of-hours consumers of energy over the previous {period_freq_str} were: """ #temp fix
+
+    for index, item in enumerate(df_ooh_biggest['sum'].round().astype('int').items()):
+
+        if item[0] not in ['Total', 'Others']: # only add valid entries, not total and others.
+
+            statement_item = "  \t \t \t \t"+str(index+1)+'. '+item[0]+' '+str(item[1])+' kwh,'
+            
+            statement += statement_item
+        else:
+            print(f'[WARN]: statement_for_biggest_ooh - Please note that ooh category [{item[0]}] has been skipped')
+
+    statement = statement[:-1]+'.'
+    
+    return statement

mindsett_energy_reports_pkg/mdl/prepare_and_email_reports/prepare_report/generate_components/generate_insight_statements/insight_statements/statement_for_total_ooh/__init__.py

@@ -0,0 +1,18 @@
+
+
+def statement_for_total_ooh(df_asset_group_period_sum_others, row_index_for_total='Total', period_freq_str='week'):
+
+    sub_pct_value = df_asset_group_period_sum_others['sub_pct'][row_index_for_total]
+    sub_pct_abs_value = round(abs(sub_pct_value * 100))
+
+    if sub_pct_abs_value > 1:
+        if sub_pct_value > 0:
+            statement_direction = "up"
+        else:
+            statement_direction = "down"
+        statement = f"""The out-of-hours use has gone {statement_direction} by {sub_pct_abs_value}% compared to the {period_freq_str} before.""" #temp fix
+
+    else:   
+        statement = f"""The out-of-hours use has been similar to the {period_freq_str} before.""" #temp fix
+        
+    return statement

mindsett_energy_reports_pkg/mdl/prepare_and_email_reports/prepare_report/generate_components/generate_insight_statements/preprocessing_for_statement/__init__.py

@@ -0,0 +1,85 @@
+
+from datetime import date
+import pandas as pd
+
+from mdl.utils_ import get_group_with_others
+
+def preprocessing_for_statement(df_meta_with_value, 
+                                asset_group='asset_class',
+                                # fixed_group_to_filter = [], # should be all in lower case charactor # applying the variable column directly
+                                row_index_for_total = "Total", 
+                                reading_interval_in_mins=10,
+                                pct_level_tobe_others = 0):
+    
+    #Conversion into MWh
+    w_to_kw_para = 1./1000
+    min_to_hour_para = 1./60
+    
+    wm_to_kwh_parameter = w_to_kw_para * min_to_hour_para
+    reading_to_kwh_parameter = reading_interval_in_mins * wm_to_kwh_parameter
+    
+    df_meta_with_value[asset_group] = df_meta_with_value[asset_group].str.strip()
+    # print('df_meta_with_value: ', df_meta_with_value)
+
+    df_meta_with_value_variable = df_meta_with_value.loc[df_meta_with_value['variable']==True]
+
+    sr_pivot_asset_group = df_meta_with_value_variable.groupby([asset_group, 'period', 'out_of_hours'])["W"].sum() * reading_to_kwh_parameter  # Div 1000 for convertion to MWh
+
+    # print('sr_pivot_asset_groupindex: ', sr_pivot_asset_group.index)
+    sr_pivot_asset_group_unstack_ooh = sr_pivot_asset_group.unstack(["out_of_hours"])
+
+    if True not in sr_pivot_asset_group_unstack_ooh.columns:
+        return None
+    else:
+        df_pivot_asset_group_by_period_full = sr_pivot_asset_group_unstack_ooh[True].unstack(["period"])
+
+        df_pivot_asset_group_by_period_no_fixed = df_pivot_asset_group_by_period_full
+
+        # print("df_pivot_asset_group_by_period: ", df_pivot_asset_group_by_period) # debug
+        
+        # using the 'period' information from the dataframe
+        period_range = df_pivot_asset_group_by_period_no_fixed.columns
+        period_current = period_range[-1]
+
+        # for avoiding the case that we don't have data for the previous week
+        if len(period_range) < 2:
+            period_tobe_compared = period_range[-1]
+        else:
+            period_tobe_compared = period_range[-2]
+
+        # todo: check whether the result can be obtained by unstack directly
+        # remove the assets that are generating energy
+        df_pivot_asset_group_by_period = df_pivot_asset_group_by_period_no_fixed.loc[df_pivot_asset_group_by_period_no_fixed[period_current]>0].copy()
+        df_pivot_asset_group_by_period_renamed = df_pivot_asset_group_by_period.loc[:,period_current].to_frame().rename(columns={period_current: "sum"})
+        df_pivot_asset_group_by_period_renamed["sum_pre"] = df_pivot_asset_group_by_period.loc[:,period_tobe_compared]
+
+        sr_total = df_pivot_asset_group_by_period_renamed.loc[df_pivot_asset_group_by_period_renamed['sum']>0].sum() # exclude the negative values
+
+        df_total = sr_total.to_frame().transpose()
+        df_total.index = [row_index_for_total]
+
+        df_total.index.name = df_pivot_asset_group_by_period_renamed.index.name
+        df_pivot_asset_group_by_period_renamed = pd.concat([df_pivot_asset_group_by_period_renamed, df_total])
+        
+        df_pivot_asset_group_by_period_renamed['sub'] = df_pivot_asset_group_by_period_renamed.loc[:,'sum'] - df_pivot_asset_group_by_period_renamed.loc[:,'sum_pre']
+
+        df_pivot_asset_group_by_period_renamed["pct"] = df_pivot_asset_group_by_period_renamed["sum"]/df_pivot_asset_group_by_period_renamed.loc[row_index_for_total, "sum"]
+        df_pivot_asset_group_by_period_renamed["gt_pct_others"] = df_pivot_asset_group_by_period_renamed["pct"].gt(pct_level_tobe_others)
+
+        df_asset_group_period_sum = df_pivot_asset_group_by_period_renamed.reset_index()
+
+        df_asset_group_period_sum["group_with_others"] = df_asset_group_period_sum.apply(get_group_with_others,asset_group=asset_group,axis=1)
+
+        df_asset_group_period_sum_others = df_asset_group_period_sum.groupby(["group_with_others"]).sum(numeric_only=True)
+
+        df_asset_group_period_sum_others["sum_for_sort"] = df_asset_group_period_sum_others["sum"] 
+
+        df_asset_group_period_sum_others.loc["Others", "sum_for_sort"] = 0
+
+        df_asset_group_period_sum_others.sort_values(["sum_for_sort"], ascending=False, inplace=True)
+        df_asset_group_period_sum_others["sub_pct"] = df_asset_group_period_sum_others["sub"]/df_asset_group_period_sum_others["sum_pre"]
+        
+        return df_asset_group_period_sum_others
+
+    
+

mindsett_energy_reports_pkg/mdl/prepare_and_email_reports/prepare_report/generate_components/generate_period_barchart/__init__.py

@@ -0,0 +1,29 @@
+import matplotlib.pyplot as plt
+import io
+
+from .preprocessing_for_period_barchart import preprocessing_for_period_barchart
+from .plot_shifted_period_barchart import plot_shifted_period_barchart
+
+
+def generate_period_barchart(df_meta_with_value_building,
+                            fontsize = None,
+                          directory_to_savefig=None):
+                          
+    df_grouped_working_hours_period_unstacked = preprocessing_for_period_barchart(df_meta_with_value_building)
+
+    # for testing
+    # df_grouped_working_hours_period_unstacked[True] = -0.3 * df_grouped_working_hours_period_unstacked[False]
+    
+    plot_shifted_period_barchart(df_grouped_working_hours_period_unstacked, fontsize=fontsize)
+
+    png_name = 'period_barchart.png'
+    if directory_to_savefig == None:
+        png_object = io.BytesIO()
+        plt.savefig(png_object, format='png', dpi=200)
+        plt.close()
+        return {png_name: png_object}
+    else:
+        png_path = directory_to_savefig + png_name
+        plt.savefig(png_path, format='png', dpi=200)
+        plt.close()
+        return {png_name: png_path}

mindsett_energy_reports_pkg/mdl/prepare_and_email_reports/prepare_report/generate_components/generate_period_barchart/plot_shifted_period_barchart/__init__.py

@@ -0,0 +1,49 @@
+import matplotlib.pyplot as plt
+
+from .set_x_label_prepare_xtick import set_x_label_prepare_xtick
+from .setup_y_tick_lim_label_zero_line import setup_y_tick_lim_label_zero_line
+from .set_x_tick_and_lim import set_x_tick_and_lim
+
+from mdl.utils_ import get_pos_bot_range_unit_format
+from mdl.utils_ import plot_stacked_barchart
+
+def plot_shifted_period_barchart(df_grouped_working_hours_period_unstacked, 
+                                          fs=None,
+                                          fontsize=None):
+
+    if fs is None:
+        fs = (3.51, 3.07)
+
+    plt.style.use('seaborn-v0_8-white')
+    fig, ax = plt.subplots(1, 1, figsize=fs)        
+
+    df_grouped_working_hours_period_reset_with_sign = df_grouped_working_hours_period_unstacked.reset_index(drop=True)
+    freqstr = df_grouped_working_hours_period_unstacked.index.freqstr
+
+    df_grouped_working_hours_sorted_reset_pos, \
+        df_grouped_working_hours_sorted_reset_bot, \
+            ylim, \
+                display_unit, \
+                    formatter = get_pos_bot_range_unit_format(df_grouped_working_hours_period_reset_with_sign, 
+                                                                        pad_pct_above_max=0.2)
+
+    setup_y_tick_lim_label_zero_line(ax, ylim, display_unit, formatter, 
+                                    freqstr=freqstr,
+                                    fontsize=fontsize)
+
+    x_ticks_labels = set_x_label_prepare_xtick(ax, df_grouped_working_hours_period_unstacked,
+                                               fontsize=fontsize)
+
+    plot_stacked_barchart(ax, df_grouped_working_hours_sorted_reset_pos, 
+                        bottom=df_grouped_working_hours_sorted_reset_bot,
+                        tick_range=ylim)
+
+    set_x_tick_and_lim(ax, x_ticks_labels, df_grouped_working_hours_period_unstacked, 
+                       fontsize=fontsize)
+
+    fig.tight_layout(rect =(0.01, 0, 1, 1))
+    subplots_adjust_left = None
+    subplots_adjust_right = 0.68 # 0.92 # 
+
+    fig.subplots_adjust(left=subplots_adjust_left,
+                        right=subplots_adjust_right)

mindsett_energy_reports_pkg/mdl/prepare_and_email_reports/prepare_report/generate_components/generate_period_barchart/plot_shifted_period_barchart/set_x_label_prepare_xtick/__init__.py

@@ -0,0 +1,24 @@
+
+import matplotlib.pyplot as plt
+
+
+def set_x_label_prepare_xtick(ax, df_grouped_working_hours_period_unstacked, 
+                              fontsize=None):
+    
+    if fontsize is None:
+        fontsize = 13
+
+    if df_grouped_working_hours_period_unstacked.index.freqstr == 'M':
+        x_ticks_labels = df_grouped_working_hours_period_unstacked.index.strftime("%b %y").tolist()
+        ax.set_xlabel("Month", labelpad = 13, fontsize = fontsize)
+        plt.xticks(rotation=45) # please note: plt cannot be replace by ax, as it will throw errors
+    else:
+        ls_week_str = (df_grouped_working_hours_period_unstacked.index).strftime("%W").tolist()
+        x_ticks_labels = [f"{(int(week)+1):02d}" for week in ls_week_str]   # add 1 week to the current_period_obj so that the start week is not Week 00
+        ax.set_xlabel("Week Number", labelpad = 13, fontsize = fontsize)
+        plt.xticks(rotation=0)
+
+    x_ticks_labels.insert(0,"")
+    x_ticks_labels.append("")
+
+    return x_ticks_labels