Python visuals doesn't work on online reading view but on desktop does

Flan — Mon, 02 Dec 2024 17:33:09 GMT

Hi everyone,

Since a couple of day maybe week, i didn't noticed that one of my own python visual didn't work on the online version. On my dekstop app, nothing, it work well but only when i share the link and open it via my workspace the visual doesn't work anymore.

there erreur that i have is this one :

Script Runtime Error [S-12c5e384-a869-4a16-9c4a-12dd6770c94c][S-12c5e384-a869-4a16-9c4a-12dd6770c94c]AttributeError: 'NoneType' object has no attribute 'points_to_pixels' Please try again later or contact support. If you contact support, please provide these details. Activity ID: fb2167c7-a9c2-4229-ba95-dd2df542479d Request ID: d0f51534-072d-4d2e-a767-651859b64a82 Correlation ID: 101ac569-2fa3-434e-5997-79102e36e5a0 Time: Mon Dec 02 2024 18:25:33 GMT+0100 (Central European Standard Time) Service version: 13.0.24766.27 Client version: 2411.3.21845-train Cluster URI: https://wabi-west-europe-b-primary-redirect.analysis.windows.net/

My code is but i am not sure that it come from him because nothing change since a couple of months...

import matplotlib.pyplot as plt import pandas as pd import matplotlib.dates as mdates from matplotlib.dates import DateFormatter, MonthLocator # Supposons que dataset soit un DataFrame pandas déjà défini dataset['Date update'] = pd.to_datetime(dataset['Date update']) dataset['Forecast'] = pd.to_datetime(dataset['Forecast']) # Pivot table pivot = pd.pivot_table(dataset, aggfunc='first', index='Date update', values='Forecast', columns='Milestones name') fig, ax = plt.subplots(figsize=(12, 8)) # Plot each column for column in pivot.columns: clean = pivot[[column]].dropna() ax.plot_date(clean.index, clean[column], xdate=True, ydate=True, ls='-', lw=1, label=column) # Ajuster les limites des axes pour découpler Date update (axe x) et Forecast (axe y) min_date_update = dataset['Date update'].min() max_date_update = dataset['Date update'].max() min_forecast = dataset['Forecast'].min() max_forecast = dataset['Forecast'].max() # Dates limites pour les axes max_date_forecast = max(dataset['Forecast']) next_year_forecast = max_date_forecast.year + 1 q1_next_year_forecast = pd.Timestamp(year=next_year_forecast, month=1, day=1) min_date_forecast = min(dataset['Forecast']) min_year_forecast = min_date_forecast.year q1_actual_year_forecast = pd.Timestamp(year=min_year_forecast, month=1, day=1) min_date_update = min(dataset['Date update']) min_year_dateupdate = min_date_update.year q1_actual_year_dateupdate = pd.Timestamp(year=min_year_dateupdate, month=1, day=1) # Ajouter la ligne 45 degrés (y=x) # Créer des listes de dates pour la ligne 45 degrés line_dates = pd.date_range(start=q1_actual_year_dateupdate, end=q1_next_year_forecast, freq='D') ax.plot(line_dates, line_dates, color='black', linestyle='--', label='') # Coloration après les courbes adjusted_min_date_update = q1_actual_year_dateupdate adjusted_max_forecast = q1_next_year_forecast if len(line_dates) > 0: ax.fill_between(line_dates, adjusted_min_date_update, line_dates, color='white', zorder=2) # ----------------------------------------------------------------------------------------------- # Définir les localisateurs pour les mois (trimestres) month_locator = MonthLocator(bymonth=[1, 4, 7, 10]) plt.gca().xaxis.set_major_locator(month_locator) plt.gca().yaxis.set_major_locator(month_locator) # Création du formateur personnalisé pour afficher les trimestres class QuarterFormatter(DateFormatter): def __init__(self, fmt='%Y-%m', **kwargs): super().__init__(fmt, **kwargs) def __call__(self, x, pos=0): date = mdates.num2date(x) # Si c'est janvier, afficher l'année en dessous de Q1 if date.month == 1: return f'{date.year} - Q1' elif date.month == 4: return 'Q2' elif date.month == 7: return 'Q3' elif date.month == 10: return 'Q4' return '' # Appliquer le formateur de trimestre personnalisé quarter_formatter = QuarterFormatter() plt.gca().xaxis.set_major_formatter(quarter_formatter) plt.gca().yaxis.set_major_formatter(quarter_formatter) # ----------------------------------------------------------------------------------------------- # Rotation des étiquettes pour une meilleure lisibilité plt.xticks(rotation=90, ha='center') # Ajouter les lignes de grille pour les années uniquement start_date_update = pd.Timestamp(year=q1_actual_year_dateupdate.year, month=1, day=1) end_date_forecast = pd.Timestamp(year=q1_next_year_forecast.year, month=1, day=1) for year in range(start_date_update.year, end_date_forecast.year): january_start = pd.Timestamp(year=year, month=1, day=1) ax.axvline(x=january_start, color='gray', linestyle='--', linewidth=0.5) ax.axhline(y=january_start, color='gray', linestyle='--', linewidth=0.5) # ----------------------------------------------------------------------------------------------- # Ajuster les limites des axes plt.xlim(q1_actual_year_dateupdate, q1_next_year_forecast) # Limites pour l'axe des x (Date update) plt.ylim(q1_actual_year_dateupdate, q1_next_year_forecast) # Limites pour l'axe des y (Forecast) # ----------------------------------------------------------------------------------------------- legend = ax.legend(loc='upper left', bbox_to_anchor=(1.01, 1), frameon=False) # Ajuster la légende def adjust_legend_fontsize(ax): legend = ax.get_legend() # Récupère la légende existante if legend is None: # Si aucune légende n'est définie, arrêter return 10 fig = ax.get_window_extent() fig_height = fig.height fontsize = 10 adjusted = False while not adjusted: legend = ax.legend(loc='upper left', bbox_to_anchor=(1, 1), fontsize=fontsize) legend_bbox = legend.get_window_extent() legend_height = legend_bbox.height if legend_height <= fig_height: adjusted = True else: fontsize -= 1 if fontsize < 8: break return fontsize ax = plt.gca() fontsize = adjust_legend_fontsize(ax) myLegend = ax.legend(loc='upper left', bbox_to_anchor=(1.01, 1), frameon=False, fontsize=fontsize) # Désactiver les ajustements automatiques des marges plt.subplots_adjust(left=0.08, right=0.75, top=0.95, bottom=0.1) # Afficher le graphique plt.show()

If you have an idea, that coudl help me !!
thanks you very much in advance, and have a nice day !

Flan

Re: Python visuals doesn't work on online reading view but on desktop does

lbendlin — Mon, 02 Dec 2024 22:17:59 GMT

Maybe read through the release notes of these two libraries - could be that a recent version update has caused that.

Re: Python visuals doesn't work on online reading view but on desktop does

Flan — Tue, 03 Dec 2024 10:17:42 GMT

Hi !
Unfortunately, even with librabries updated, it still doesn't work.
It's weird because others visual work so i guess there is an issue with this one in particular... 😞

Re: Python visuals doesn't work on online reading view but on desktop does

lbendlin — Tue, 03 Dec 2024 13:45:58 GMT

If you have a Pro license you can open a Pro ticket at https://admin.powerplatform.microsoft.com/newsupportticket/powerbi
Otherwise you can raise an issue at https://community.fabric.microsoft.com/t5/Issues/idb-p/Issues .

topic Python visuals doesn't work on online reading view but on desktop does in Custom Visuals Development Discussion

Python visuals doesn't work on online reading view but on desktop does

Re: Python visuals doesn't work on online reading view but on desktop does

Re: Python visuals doesn't work on online reading view but on desktop does

Re: Python visuals doesn't work on online reading view but on desktop does