Merge pull request 'Implement Plotly Dash app with efficient dynamic resampler' (#6) from dev-dash into main

Reviewed-on: erickahmed/CANveyor#6

main.py

@@ -4,18 +4,19 @@
 
 import os
 from pathlib import Path
+from concurrent.futures import ThreadPoolExecutor
 
 import polars as pl
 import dash
 from dash import dcc, html, Input, Output
 import dash_bootstrap_components as dbc
+import numpy as np
 
 from parser import parse_log, parse_csv
 from decoder import decode_j1939_frames
 from stats.utils.extractor import load_data
-
+from stats.id_viewer import _format_can_id_vec, plot_bits
 from stats.frequency import calculate_frequency, plot_frequency
-from stats.id_viewer import prepare_data, plot_bits
 from stats.correlation import calculate_correlation, plot_correlation_heatmap
 from stats.entropy import calculate_byte_entropy, plot_entropy_heatmap
 
@@ -29,15 +30,15 @@ BUS2_DECODED = "data/parquet/bus2_decoded.parquet"
 
 def run_pipeline():
     os.makedirs("data/logs", exist_ok=True)
+    os.makedirs("data/csv", exist_ok=True)
+    os.makedirs("data/parquet", exist_ok=True)
     if not Path(BUS1_DECODED).exists() or not Path(BUS2_DECODED).exists():
         print("Parsing raw log...")
         parse_log(RAW_LOG, BUS1_CSV, BUS2_CSV)
 
         print("Converting to parquet...")
-        lf1 = parse_csv(BUS1_CSV)
+        parse_csv(BUS1_CSV).sink_parquet(BUS1_PARQUET)
-        lf1.sink_parquet(BUS1_PARQUET)
+        parse_csv(BUS2_CSV).sink_parquet(BUS2_PARQUET)
-        lf2 = parse_csv(BUS2_CSV)
-        lf2.sink_parquet(BUS2_PARQUET)
 
         print("Decoding J1939...")
         df1 = pl.read_parquet(BUS1_PARQUET)
@@ -55,10 +56,52 @@ DATA = {
     "Bus 2": load_data(BUS2_DECODED)
 }
 
+PRECOMPUTED_FIGURES = {}
+DATA_BY_ID = {}
+CORR_CACHE = {}
+
+def process_bus_data(bus, df):
+    precomp = {}
+    precomp[f"{bus}_freq"] = plot_frequency(calculate_frequency(df), title=f"{bus} Frequency")
+    precomp[f"{bus}_entropy"] = plot_entropy_heatmap(calculate_byte_entropy(df), title=f"{bus} Byte-Level Entropy")
+
+    can_id_col = 'ID' if 'ID' in df.columns else 'Identifier'
+    formatted = _format_can_id_vec(df[can_id_col])
+    df = df.assign(Formatted_ID=formatted)
+
+    df = df.sort_values(['Formatted_ID', 'Timestamp'], kind='stable')
+
+    grouped = {}
+    for can_id, group in df.groupby(by='Formatted_ID'):
+        byte_cols = [f"b{i}" for i in range(8) if f"b{i}" in group.columns]
+        if not group.empty and len(byte_cols) > 0:
+            arr = group[byte_cols].to_numpy(dtype=np.float32, copy=False)
+            if len(arr) > 1:
+                changed = np.any(arr[1:] != arr[:-1], axis=1)
+                keep = np.concatenate(([True], changed))
+                group = group.iloc[keep]
+        grouped[can_id] = (group, byte_cols)
+
+    return precomp, grouped
+
+with ThreadPoolExecutor() as executor:
+    futures = {executor.submit(process_bus_data, bus, df): bus for bus, df in DATA.items()}
+    for future in futures:
+        bus = futures[future]
+        precomp, grouped = future.result()
+        PRECOMPUTED_FIGURES.update(precomp)
+        DATA_BY_ID[bus] = grouped
+
 app = dash.Dash(__name__, external_stylesheets=[dbc.themes.BOOTSTRAP])
+app.config.suppress_callback_exceptions = True
 
 app.layout = dbc.Container([
-    html.H1("CAN Bus Analyzer", className="my-4"),
+    html.H1("CANveyor", className="my-4"),
+    dbc.Tabs([
+        dbc.Tab(label="Overview", tab_id="overview", children=[
+            html.Div(id="overview-content")
+        ]),
+        dbc.Tab(label="Statistics", tab_id="statistics", children=[
             dbc.Row([
                 dbc.Col(html.Label("Select Bus:"), width=1, className="mt-2"),
                 dbc.Col(dcc.Dropdown(
@@ -67,7 +110,7 @@ app.layout = dbc.Container([
                     value='Bus 1',
                     clearable=False
                 ), width=2),
-    ], className="mb-3"),
+            ], className="mb-3 mt-3"),
             dbc.Tabs([
                 dbc.Tab(label="Frequency", tab_id="freq"),
                 dbc.Tab(label="ID Viewer", tab_id="id_viewer"),
@@ -75,6 +118,8 @@ app.layout = dbc.Container([
                 dbc.Tab(label="Entropy", tab_id="entropy"),
             ], id="tabs", active_tab="freq"),
             html.Div(id="tab-content", className="mt-3")
+        ])
+    ], id="main-tabs", active_tab="statistics")
 ], fluid=True)
 
 @app.callback(
@@ -86,12 +131,10 @@ def render_content(tab, bus):
     df = DATA[bus]
 
     if tab == 'freq':
-        stats = calculate_frequency(df)
+        return dcc.Graph(figure=PRECOMPUTED_FIGURES[f"{bus}_freq"], style={'height': '80vh'})
-        fig = plot_frequency(stats, title=f"{bus} Frequency")
-        return dcc.Graph(figure=fig, style={'height': '80vh'})
 
     elif tab == 'id_viewer':
-        ids = sorted(df['ID'].unique().tolist())
+        ids = sorted(DATA_BY_ID[bus].keys())
         return html.Div([
             html.Label("Select CAN ID:"),
             dcc.Dropdown(
@@ -105,7 +148,7 @@ def render_content(tab, bus):
         ])
 
     elif tab == 'corr':
-        ids = sorted(df['ID'].unique().tolist())
+        ids = sorted(DATA_BY_ID[bus].keys())
         return html.Div([
             dbc.Row([
                 dbc.Col(html.Label("Method:"), width=1, className="mt-2"),
@@ -127,9 +170,7 @@ def render_content(tab, bus):
         ])
 
     elif tab == 'entropy':
-        entropy_df = calculate_byte_entropy(df)
+        return dcc.Graph(figure=PRECOMPUTED_FIGURES[f"{bus}_entropy"], style={'height': '80vh'})
-        fig = plot_entropy_heatmap(entropy_df, title=f"{bus} Byte-Level Entropy")
-        return dcc.Graph(figure=fig, style={'height': '80vh'})
 
     return html.Div("Tab not found")
 
@@ -142,8 +183,12 @@ def render_content(tab, bus):
 def update_id_viewer(selected_id, bus, tab):
     if tab != 'id_viewer' or not selected_id:
         return dash.no_update
-    df = DATA[bus]
+
-    filtered_df, byte_cols = prepare_data(df, selected_id)
+    grouped_data = DATA_BY_ID.get(bus, {})
+    if selected_id not in grouped_data:
+        return dash.no_update
+
+    filtered_df, byte_cols = grouped_data[selected_id]
     return plot_bits(filtered_df, byte_cols, selected_id, title=f"{bus} Byte Visualization")
 
 @app.callback(
@@ -156,12 +201,21 @@ def update_id_viewer(selected_id, bus, tab):
 def update_corr(method, target, bus, tab):
     if tab != 'corr':
         return dash.no_update
-    df = DATA[bus]
+
     target_id = None if target == 'all' or not target else target
+    cache_key = (bus, method, target_id)
+
+    if cache_key not in CORR_CACHE:
+        df = DATA[bus]
         corr_df = calculate_correlation(df, method=method, target_id=target_id)
+        CORR_CACHE[cache_key] = corr_df
+    else:
+        corr_df = CORR_CACHE[cache_key]
+
     title = f"{bus} Correlation"
     if target_id:
         title += f" ({target_id})"
+
     return plot_correlation_heatmap(corr_df, target_id=target_id, title=title)
 
 if __name__ == '__main__':

pyproject.toml

@@ -1,7 +1,15 @@
 [project]
 name = "CANveyor"
-version = "0.0.4"
+version = "0.1.0"
 description = "J1939 CAN bus parser that works in pair with CANdigger"
 readme = "README.md"
-requires-python = ">=3.14"
+requires-python = ">=3.10"
-dependencies = ["polars", "pathlib", "typing"]
+dependencies = [
+    "polars",
+    "dash",
+    "dash-bootstrap-components",
+    "numpy",
+    "pandas",
+    "plotly",
+    "plotly-resampler"
+]

stats/correlation.py

@@ -4,6 +4,7 @@
 
 import argparse
 from pathlib import Path
+from concurrent.futures import ThreadPoolExecutor
 
 import numpy as np
 import pandas as pd
@@ -27,8 +28,7 @@ def _ensure_int_bytes(df: pd.DataFrame, cols: list) -> pd.DataFrame:
     return df
 
 def calculate_correlation(df: pd.DataFrame, method: str, target_id: str | None = None) -> pd.DataFrame:
-    byte_cols = [f"b{i}" for i in range(8) if f"b{i}" in df.columns]
+    available_cols = [f"b{i}" for i in range(8) if f"b{i}" in df.columns]
-    available_cols = [col for col in byte_cols if col in df.columns]
 
     if not available_cols:
         raise ValueError("No byte columns (b0-b7) found in the DataFrame")
@@ -70,8 +70,7 @@ def calculate_correlation(df: pd.DataFrame, method: str, target_id: str | None =
     else:
         groups = []
 
-    out = np.zeros((len(unique_ids), n_cols), dtype=np.float64)
+    def _process_group(sub):
-    for gi, sub in enumerate(groups):
         mask = ~np.isnan(sub).any(axis=1)
         sub = sub[mask]
         if len(sub) > 1:
@@ -81,7 +80,11 @@ def calculate_correlation(df: pd.DataFrame, method: str, target_id: str | None =
                 c = np.abs(np.corrcoef(sub, rowvar=False))
             np.nan_to_num(c, copy=False, nan=0.0)
             np.fill_diagonal(c, 0.0)
-            out[gi] = c.max(axis=0)
+            return c.max(axis=0)
+        return np.zeros(n_cols, dtype=np.float64)
+
+    with ThreadPoolExecutor() as executor:
+        out = np.array(list(executor.map(_process_group, groups)))
 
     result = pd.DataFrame(out, index=unique_ids, columns=available_cols)
     result.index.name = 'Identifier'

stats/entropy.py

@@ -4,6 +4,7 @@
 
 import argparse
 from pathlib import Path
+from concurrent.futures import ThreadPoolExecutor
 
 import numpy as np
 import pandas as pd
@@ -36,8 +37,7 @@ def _entropy_col(a: np.ndarray) -> float:
     return float(-np.sum(p * np.log2(p)))
 
 def calculate_byte_entropy(df: pd.DataFrame) -> pd.DataFrame:
-    byte_cols = [f"b{i}" for i in range(8) if f"b{i}" in df.columns]
+    available_cols = [f"b{i}" for i in range(8) if f"b{i}" in df.columns]
-    available_cols = byte_cols
     if not available_cols:
         raise ValueError("No byte columns (b0-b7) found in the DataFrame")
 
@@ -64,10 +64,14 @@ def calculate_byte_entropy(df: pd.DataFrame) -> pd.DataFrame:
     else:
         groups = []
 
-    out = np.zeros((len(unique_ids), n_cols), dtype=np.float64)
+    def _process_group(sub):
-    for gi, sub in enumerate(groups):
+        res = np.zeros(n_cols, dtype=np.float64)
         for ci in range(n_cols):
-            out[gi, ci] = _entropy_col(sub[:, ci])
+            res[ci] = _entropy_col(sub[:, ci])
+        return res
+
+    with ThreadPoolExecutor() as executor:
+        out = np.array(list(executor.map(_process_group, groups)))
 
     result = pd.DataFrame(out, index=unique_ids, columns=available_cols)
     result.index.name = 'Identifier'

stats/frequency.py

@@ -18,7 +18,6 @@ def _format_can_id_vec(s: pd.Series) -> pd.Series:
 def calculate_frequency(df: pd.DataFrame) -> pd.DataFrame:
     can_id_col = 'ID' if 'ID' in df.columns else 'Identifier'
     formatted = _format_can_id_vec(df[can_id_col])
-    df['Formatted_ID'] = formatted
 
     counts = formatted.value_counts()
     freq_df = pd.DataFrame({

stats/id_viewer.py

@@ -7,6 +7,7 @@ from pathlib import Path
 import numpy as np
 import pandas as pd
 import plotly.graph_objects as go
+from plotly_resampler import FigureResampler
 from stats.utils.extractor import load_data
 
 def _format_can_id_vec(s: pd.Series) -> pd.Series:
@@ -40,22 +41,24 @@ def prepare_data(df, target_id):
     return filtered, byte_cols
 
 def plot_bits(df, byte_cols, can_id, title):
-    fig = go.Figure()
+    fig = FigureResampler(
+        resampled_trace_prefix_suffix=("", ""),
+        show_mean_aggregation_size=False
+    )
     colors = ['#e41a1c', '#377eb8', '#4daf4a', '#984ea3', '#ff7f00', '#ffff33', '#a65628', '#f781bf']
     n = len(byte_cols)
 
     x = df['Timestamp'].to_numpy() if not df.empty else np.array([])
     for i, col in enumerate(byte_cols):
         y = df[col].to_numpy(dtype=np.float32, copy=False) if not df.empty else np.array([])
-        fig.add_trace(go.Scattergl(
+
-            x=x,
+        fig.add_trace(go.Scatter(
-            y=y,
             mode='lines',
             line=dict(shape='hv', width=2, color=colors[i % len(colors)]),
             name=col.upper(),
             legendgroup=col.upper(),
             hovertemplate=f"<b>{col.upper()}</b><br>Time: %{{x}}<br>Value: %{{y}}<extra></extra>",
-        ))
+        ), hf_x=x, hf_y=y)
 
     all_button = dict(label='ALL', method='restyle', args=[{'visible': [True] * n}])
     none_button = dict(label='NONE', method='restyle', args=[{'visible': ['legendonly'] * n}])