go-sparkline

Install

# install via go tools
go get github.com/muhqu/go-sparkline
go install github.com/muhqu/go-sparkline

# verify
go-sparkline --help

Help

$ go-sparkline --help
usage: go-sparkline [<flags>] [<values>]

Flags:
  --help             Show help.
  -s, --stream       stream
  --animate          start animation
  --lazy             ignore parse errors
  --char-size=7:17   Pixel size of a single character. Can also be set via env
                     ITERM_CHARACTER_SIZE. The default 7:17 corresponds to 12p Monaco.
  --rows=3           height in number of rows
  --renderer=sparks  available renderers: line, sparks, vlines

Args:
  [<values>]  Numeric values to render. Can also be read from stdin.

Examples

simple numbers

go-sparkline 16 19 18 12 7 4 7 15 25 33 35 32 26 21
# or
echo 16 19 18 12 7 4 7 15 25 33 35 32 26 21 | go-sparkline

simple json array

echo '[16,19,18,12,7,4,7,15,25,33,35,32,26,21]' | go-sparkline

streamed json arrays

(
echo '[16,19,18]'; sleep 1;
echo '[12,7,4]'; sleep 1;
echo '[7,15,25]'; sleep 1;
echo '[33,35,32]'; sleep 1;
echo '[26,21]';
) | go-sparkline --stream

streamed ping graph

ping -n -i 0.3 localhost \
 | awk 'BEGIN{FS="time=|ms"}/time=/{printf "%d\n",$2*1000;fflush()}' \
 | go-sparkline --stream

aws cloudwatch metric data

$ aws cloudwatch get-metric-statistics \
    --namespace AWS/ELB \
    --metric-name RequestCount \
    --end-time "2015-04-15T21:15:00Z" \
    --start-time "2015-04-15T17:45:00Z" \
    --period 900 \
    --statistics Sum \
    | tee cloudwatch.json \
    | go-sparkline
$ head cloudwatch.json; echo '...'; tail cloudwatch.json;
{
    "Datapoints": [
        {
            "Timestamp": "2015-04-15T17:45:00Z",
            "Sum": 37823.0,
            "Unit": "Count"
        },
        {
            "Timestamp": "2015-04-15T11:30:00Z",
            "Sum": 12413.0,
...
            "Unit": "Count"
        },
        {
            "Timestamp": "2015-04-15T21:15:00Z",
            "Sum": 29428.0,
            "Unit": "Count"
        }
    ],
    "Label": "RequestCount"
}

FAQ

Q: What?! How does it display inline images directly in the terminal?!

A: iTerm2 supports a bunch of propritary ESC seq. The nightly build even includes one to render images directly into the terminal.

ESC ] 1337 ; File = [optional arguments] : base-64 encoded file contents ^G

Autor

	© 2015 by Mathias Leppich github.com/muhqu, @muhqu