data-viz（数据可视化）

Data Visualization

Create terminal-based charts and visualizations from CSV, JSON, or piped data.

Quick Visualizations with YouPlot

YouPlot (uplot) creates Unicode charts in the terminal.

Bar Chart

echo -e "Apple,30\
Banana,45\
Cherry,20\
Date,35" | uplot bar -d, -t "Fruit Sales"

Line Chart

seq 1 20 | awk '{print $1, sin($1/3)*10+10}' | uplot line -t "Sine Wave"

Histogram

awk 'BEGIN{for(i=0;i<1000;i++)print rand()}' | uplot hist -t "Random Distribution" -n 20

Scatter Plot

awk 'BEGIN{for(i=0;i<100;i++)print rand()*100, rand()*100}' | uplot scatter -t "Random Points"

From CSV Files

# Bar chart from CSV
cat sales.csv | uplot bar -d, -H -t "Monthly Sales"

# Line chart with headers
cat timeseries.csv | uplot line -d, -H -t "Stock Price"

From JSON (with jq)

# Extract data from JSON and plot
curl -s "https://api.example.com/data" | jq -r '.items[] | "\(.name),\(.value)"' | uplot bar -d,

Termgraph (Python Alternative)

Simple horizontal bar charts:

echo -e "2020 50\
2021 75\
2022 90\
2023 120" | termgraph

With colors:

echo -e "Sales 150\
Costs 80\
Profit 70" | termgraph --color green

Gnuplot (Advanced)

For publication-quality charts:

# Quick line plot
gnuplot -e "set terminal dumb; plot sin(x)"

# From data file
gnuplot -e "set terminal dumb; plot 'data.txt' with lines"

Sparklines

Inline mini-charts:

# Using spark (if installed)
echo "1 5 22 13 5" | spark
# Output: ▁▂█▅▂

# Pure bash sparkline
data="1 5 22 13 5"; min=$(echo $data | tr ' ' '\
' | sort -n | head -1); max=$(echo $data | tr ' ' '\
' | sort -n | tail -1); for n in $data; do printf "\u258$((7-7*($n-$min)/($max-$min)))"; done; echo

ASCII Tables

Format data as tables:

# Using column
echo -e "Name,Score,Grade\
Alice,95,A\
Bob,82,B\
Carol,78,C" | column -t -s,

# Using csvlook (csvkit)
cat data.csv | csvlook

Real-World Examples

Stock Price Chart

# Fetch and plot stock data (using Alpha Vantage free API)
curl -s "https://www.alphavantage.co/query?function=TIME_SERIES_DAILY&symbol=AAPL&apikey=demo" | \
  jq -r '.["Time Series (Daily)"] | to_entries | .[:20] | reverse | .[] | "\(.key) \(.value["4. close"])"' | \
  uplot line -t "AAPL Stock Price"

System Metrics

# CPU usage over time
for i in {1..20}; do
  top -bn1 | grep "Cpu(s)" | awk '{print 100-$8}'
  sleep 1
done | uplot line -t "CPU Usage %"

API Response Times

# Measure and plot response times
for i in {1..10}; do
  curl -s -o /dev/null -w "%{time_total}\
" https://example.com
done | uplot line -t "Response Time (s)"

Tips

• Use -d, for comma-delimited data, -d'\ ' for tabs
• Use -H when your data has headers
• Pipe through head or tail to limit data points
• Combine with jq for JSON data extraction
• Use watch for live updating charts: watch -n1 'command | uplot bar'