Code Editor for app.py

import streamlit as st
import pandas as pd
import plotly.express as px
import plotly.graph_objects as go
from datetime import datetime

# Configuration de la page
st.set_page_config(page_title="Chemical Dashboard", layout="wide")

# Titre de l'application
st.title("Chemical Treatment Dashboard")

# Création des données exemple (à remplacer par votre connexion à la base de données réelle)
@st.cache_data
def load_data():
    # Simuler les données de CHEM_TEST.LOGBOOK_TEST
    data = pd.DataFrame({
        'CHEM_DATE': pd.date_range(start='2023-01-01', periods=100),
        'INJECTION_POINT': ['Point-' + str(i%5) for i in range(100)],
        'CHEM_PRODUCT_NAME': ['Product-' + str(i%3) for i in range(100)],
        'TREATED_FLUID': ['Fluid-' + str(i%4) for i in range(100)],
        'TREATED_FLOWRATE': [float(i) for i in range(100)],
        'TARGET_INJECTION_RATE': [float(i/2) for i in range(100)],
        'TARGET_PPM': [float(i*10) for i in range(100)],
        'IS_FIELD_TEST': [i%2 == 0 for i in range(100)],
        'IS_BATCH': [i%3 == 0 for i in range(100)],
        'BSW_TREATED_FLUID': [float(i%100) for i in range(100)],
        'TEMP_TREATED_FLUID': [float(i+20) for i in range(100)],
        'PRESSURE_TREATED_FLUID': [float(i*2) for i in range(100)],
        'TDS_TREATED_FLUID': [float(i*100) for i in range(100)],
        'PH_TREATED_FLUID': [float(i%14) for i in range(100)],
        'SUBSTITUTE_PRODUCT': ['Sub-' + str(i%3) for i in range(100)]
    })
    return data

# Chargement des données
df = load_data()

# Création des filtres dans la sidebar
st.sidebar.header('Filtres')

# Filtre de date
date_range = st.sidebar.date_input(
    "Sélectionner la période",
    [df['CHEM_DATE'].min(), df['CHEM_DATE'].max()]
)

# Filtres pour les colonnes catégorielles
injection_point = st.sidebar.multiselect(
    'Point d\'injection',
    options=df['INJECTION_POINT'].unique(),
    default=df['INJECTION_POINT'].unique()
)

product_name = st.sidebar.multiselect(
    'Nom du produit chimique',
    options=df['CHEM_PRODUCT_NAME'].unique(),
    default=df['CHEM_PRODUCT_NAME'].unique()
)

# Application des filtres
mask = (
    (df['CHEM_DATE'].dt.date >= date_range[0]) &
    (df['CHEM_DATE'].dt.date <= date_range[1]) &
    (df['INJECTION_POINT'].isin(injection_point)) &
    (df['CHEM_PRODUCT_NAME'].isin(product_name))
)
filtered_df = df.loc[mask]

# Layout du dashboard en colonnes
col1, col2 = st.columns(2)

with col1:
    # Graphique d'évolution temporelle des taux d'injection
    fig_injection = px.line(
        filtered_df,
        x='CHEM_DATE',
        y='TARGET_INJECTION_RATE',
        color='INJECTION_POINT',
        title='Évolution du taux d\'injection par point'
    )
    st.plotly_chart(fig_injection, use_container_width=True)

    # Graphique de distribution des PPM
    fig_ppm = px.box(
        filtered_df,
        x='CHEM_PRODUCT_NAME',
        y='TARGET_PPM',
        title='Distribution des PPM par produit'
    )
    st.plotly_chart(fig_ppm, use_container_width=True)

with col2:
    # Graphique de corrélation température/pression
    fig_corr = px.scatter(
        filtered_df,
        x='TEMP_TREATED_FLUID',
        y='PRESSURE_TREATED_FLUID',
        color='TREATED_FLUID',
        title='Corrélation Température/Pression par fluide traité'
    )
    st.plotly_chart(fig_corr, use_container_width=True)

    # Graphique pH
    fig_ph = px.histogram(
        filtered_df,
        x='PH_TREATED_FLUID',
        title='Distribution du pH des fluides traités'
    )
    st.plotly_chart(fig_ph, use_container_width=True)

# Métriques clés
col1, col2, col3, col4 = st.columns(4)

with col1:
    st.metric(
        "Nombre total d'injections",
        len(filtered_df)
    )

with col2:
    st.metric(
        "Moyenne PPM",
        f"{filtered_df['TARGET_PPM'].mean():.2f}"
    )

with col3:
    st.metric(
        "Température moyenne",
        f"{filtered_df['TEMP_TREATED_FLUID'].mean():.1f}°C"
    )

with col4:
    st.metric(
        "pH moyen",
        f"{filtered_df['PH_TREATED_FLUID'].mean():.1f}"
    )

# Table des données filtrées
st.subheader("Données détaillées")
st.dataframe(filtered_df)