Skip to content

HTTPS clone URL

Subversion checkout URL

You can clone with HTTPS or Subversion.

Download ZIP
tree: f2d62c2bf9
Fetching contributors…

Cannot retrieve contributors at this time

87 lines (58 sloc) 2.867 kb
Step 1: Prepare raw data
go over the code for post_parser.rb (don't write)
write mapper_tag_counter, but start with the STDIN loop already defined
gem install active_support
write the reducer_tag_aggregator. talk about whether we want to return raw data or rollups
tail -n 3000000 posts.xml | ./mapper_tag_counter.rb | sort | ./reducer_tags_by_week.rb > tags_by_week.txt
tail -n 3000000 posts.xml | ./mapper_tag_counter.rb | sort | ./reducer_tags_last_30_days.rb | sort > tags_last_30_days.json
******************************************************************************************************************
Step 2: D3 Basics
bring up D3 website
* open source javascript visualization toolkit created by Mike Bostock
* evolution of years of experience writing Protovis
* more than a charting library
* you can build charting libraries with it
(first install JSONView on the presenter machine)
show the server (don't write)
ruby server.rb -p 3001
http://localhost:3001/tags/javascript
show section2.html and bring it up along with jsconsole
// show some data
d3.json("tags/javascript", function(data) {console.log(data)});
// put somewhere to access
d3.json("tags/javascript", function(data) {window.tagHistory = data});
// do drawing of chart to show scale, domain, range.
// scale converts the domain (the input) to a range (the output/pixels)
// range determined by the size of the chart
// domain determined by the min and max of the data. or fixed bounds
tagCountExtent = d3.extent(tagHistory.data, function(d) {return d[1];})
// d3 scales make it easy to map values to their corresponding pixel location in a visualization
tagCountScale = d3.scale.linear()
tagCountScale.domain(tagCountExtent)
tagCountScale.range([0, 300])
// or use the more idiomatic d3 way: chaining
tagCountScale.domain(tagCountExtent).range([0, 300])
tagCountExtent
tagCountScale(754)
tagCountScale(2067)
tagCountScale(1000)
tagCountScale(3000)
// bound the output
tagCountScale.clamp(true)
tagCountScale(3000)
// selecting elements
d3.select("body").append("h1").text("d3 ftw")
d3.select("body").append("h1").text("moar h1")
d3.select("h1")
d3.selectAll("h1")
d3.select("body").append("h1").text("h1 with class").attr("class", "foo")
d3.selectAll("h1.foo")
d3.json("tag_counts_last_30_days.json", function(d) {window.tagCounts = d})
d3.select("body").append("ol").selectAll("li").data(tagCounts.slice(0, 20)).enter().append("li").text(function(d) {return d[1]})
******************************************************************************************************************
Step 3: Create a bar chart
cover section3.html
# you have more code, but you have full control.
******************************************************************************************************************
Step 4: Create a time series
cover section4.html (left off on line 28)
Jump to Line
Something went wrong with that request. Please try again.