yaronn/wopr

WOPR

WOPR is a simple markup language for creating rich terminal reports, presentations and infographics.

Put a report on the web (e.g. gist) and view it via curl:

$> curl -N tty.zone/\[0-2\]\?auto\&cols=$((COLUMNS))

(If you experience firewall issues replace tty.zone with ec2-23-21-64-152.compute-1.amazonaws.com or use a local viewer)

Created by Yaron Naveh (@YaronNaveh)

##Writing your first terminal report##

Here is a simple report with a bar chart:

<document>
  <page>
    <item col="0" row="0" colSpan="5" rowSpan="4">
      <bar maxHeight="5" data-titles="A,B,C" data-data="2,5,3" />
    </item>
  </page>
</document>

You have 3 options to view this report:

Option 1: POST it to the wopr online viewer

$> curl --data '<document><page><item col="0" row="0" colSpan="5" rowSpan="4"><bar maxHeight="5" data-titles="A,B,C" data-data="2,5,3" /></item></page></document>' tty.zone\?cols=$((COLUMNS))

If you experience firewall issues replace tty.zone with ec2-23-21-64-152.compute-1.amazonaws.com.

Note: The online viewer is a reference implementation. Do not send it secret data but rather create your own.

Option 2: POST it from external url

Save the report content in some url (e.g. gist) and then:

$> a=$(curl -s https://gist.githubusercontent.com/yaronn/e6eec6d0e7adac63c83f/raw/50aca544d26a32aa189e790635c8679067017948/gistfile1.xml); curl --data "$a" tty.zone\?cols=$((COLUMNS))

(note you need the gist raw url)

If you experience firewall issues replace tty.zone with ec2-23-21-64-152.compute-1.amazonaws.com.

Note: The online viewer is a reference implementation. Do not send it secret data but rather create your own.

Option 3: Via the local viewer

Save the report xml to report.xml and then:

$> npm install -g wopr
$> wopr report.xml

Note the local viewer does not send anything online and does not require network.

##Markup Basics#

Pages

A document is a set of pages:

<document>
  <page>
    ...
  </page>
  <page>
    ...
  </page>
</document>

Layout

A page is a 12x12 grid in which you can position different widgets:

<document>
  <page>
    <item col="0" row="0" colSpan="3" rowSpan="3">
      <bar maxHeight="5" data-titles="A,B,C" data-data="2,5,3" />
    </item>
    <item col="5" row="9" colSpan="1" rowSpan="1">
      <box content="some text" />
    </item>
  </page>
</document>

Here, the bar widget is in the first column and row (0-based indexing) and spans three columns and rows. The box element is in the same page but in a different position.

Widgets

The available widgets are the ones that exist in the blessed and blessed-contrib projects. You can infer the xml representation of a javascript widget using a simple convention. Assume that you would instantiate some blessed widget with this javascript:

blessed.widget({ string: "5"
               , int: 1
               , intArray: [1,2,3]
               , stringArray: ["a", "b", "c"]
               , multiArray: [ [1,2,3], [4,5,6] ]
               , complexArray: [ {a: 1, b: [1,2] }, {a: 3, b: [3,4]} ]
               , object: { innerProp: 1, multiArray: [ [1,2], [3,4] ] }
})

Then here is how you would represent it in xml:

<widget string="5" int="1" intArray="1,2,3" stringArray="a,b,c" object-innerProp="1">
  <multiArray>
    1,2,3
    4,5,6
  </multiArray>
  <object-multiArray>
    1,2
    3,4
  </object-multiArray>
  <complexArray>
    <m a="1" b="1,2" />
    <m a="3" b="3,4" />
  </complexArray>
</widget>

You can also look at the demo xml to get more samples.

##Viewing Reports##

Depending on how you use a report, you have a few ways to view it. On Windows you will probably only be able to use the third option and need to install the fonts for best view.

Option 1: POST it to the wopr online viewer

$> curl --data '<document><page><item col="0" row="0" colSpan="5" rowSpan="4"><bar maxHeight="5" data-titles="A,B,C" data-data="2,5,3" /></item></page></document>' tty.zone\?cols=$((COLUMNS))

If you experience firewall issues replace tty.zone with ec2-23-21-64-152.compute-1.amazonaws.com.

Option 2: POST it from external url

Save the report content in some url (e.g. gist) and then:

$> a=$(curl -s https://gist.githubusercontent.com/yaronn/e6eec6d0e7adac63c83f/raw/50aca544d26a32aa189e790635c8679067017948/gistfile1.xml); curl --data "$a" tty.zone\?cols=$((COLUMNS))

(note you need the gist raw url)

If you experience firewall issues replace tty.zone with ec2-23-21-64-152.compute-1.amazonaws.com.

Tip: If you use a url shortener (e.g. bit.ly) add the -L flag to curl to follow redirects.

Option 3: via the local viewer

Save the report xml to report.xml and then:

$> npm install -g wopr
$> wopr report.xml

Note the local viewer does not send anything online and does not require network.

Tip: Maximize the terminal before viewing the report for best viewing experience
Tip: If you CTRL+C in the middle or rendering your cursoe might disappear. Restore it by running again and letting the render complete or with $> echo '\033[?25h'

View customization When using the online reports, you might need to adjust the slides size based on your font / resolution or use non-xterm terminal. tty.zone supports the following query params:

curl -N tty.zone\?\&cols=200\&rows=50\&terminal=xterm

You can infer them automatically from your environment:

curl -N tty.zone\?\&cols=$((COLUMNS))\&rows=$((LINES-5))\&terminal=${TERM}

It is best to escape all special characters (e.g. ? &) as seen in the above samples, since some shells will require this (zsh).

Pages

When viewing a report with the local viewer you can advance slides with the Return or Space keys. When using the online viewer you have 2 options:

Option 1: Manually advance slides with Return or Space:

p=0; while true; do curl tty.zone/$((p++))\?cols=$((COLUMNS)); read; done

Option 2: Slides advance automatically every 5 seconds:

curl -N tty.zone/\[0-2\]\?auto\&cols=$((COLUMNS))

Where 0 is the index of the first slide and 2 of the last slide. Keep the brackets in the url (they are not to express optional argument) and escape them as in the above sample.

Tip: disable curl buffering with the -N flag

You can also view a specific slide (#4 in this case):

curl --data '<document>...</document>' tty.zone/4\?cols=$((COLUMNS))

##License## MIT

More Information

Created by Yaron Naveh (twitter, blog)