starven.coffee

# Description
#   The latest details about angel and venture investment, plus information related to starting a company.
#
# Configuration:
#   
# Dependencies:
#   "chart.js": "^1.0.1-beta.2"
#   "phantomjs": "^1.9.12"
#   
# Commands:
#   hubot startup valuations - Pulls the latest startup valuations
#
# Notes:
#  I'd like to thank Nick Moran of The Full Ratchet: Startup Investing & Fundraising from the Angel and Venture Capital Experts podcast. Much of the
#  information collected was from these podcasts. You can find the podcast at https://itunes.apple.com/us/podcast/full-ratchet-startup-investing/id888084300?mt=2
#
# Author:
#   robksawyer[@<org>]

quiche = require('quiche') # https://www.npmjs.org/package/quiche
gShort = require('short-url') # https://www.npmjs.org/package/short-url

process.env.HUBOT_DATASETS_URL ||= 'https://www.quandl.com/api/v1/datasets/'
process.env.HUBOT_GOOGLE_CHART_URL ||= 'http://chart.googleapis.com/chart?'

module.exports = (robot) ->

  robot.respond /startups/i, (msg) ->
    msg.reply "I can tell you a little about startups!"

  #
  # Venture-Backed IPOs (Median Offer Amount and Media Post Value)
  # This data is from the NVCA yearbook, which includes a comprehensive analysis of U.S. venture capital industry statistics. Age in years and values in USD million.
  # 
  # Command:
  #   user > hubot vc ipos ov
  #
  robot.respond /vc ipos ov/i, (msg) ->

    curDate = new Date()
    curDate.setYear(curDate.getFullYear()-3) # Get the date three years ago
    curDate = curDate.getFullYear() + "-" + curDate.getMonth() + "-" + "01"

    robot.http(process.env.HUBOT_DATASETS_URL + "NVCA/VENTURE_10_0.json?trim_start=" + curDate)
       .header('accept', 'application/json')
       .get() (err, res, body) -> 

          if err
            msg.send "Encountered an error :( #{err}"

          if res.statusCode isnt 200
            msg.send "I wasn't able to figure out what the numbers are."
          
          rdata = JSON.parse(body) if body

          if rdata

            # Use the data that was compiled
            formattedData = {}
            formattedData.dates = ( dates[0] for dates in rdata.data)
            formattedData.a = (a[3] for a in rdata.data)
            formattedData.b = (b[6] for b in rdata.data)
            
            theDates = formattedData.dates
            xVals = [theDates[0], theDates[Math.ceil(theDates.length/2)], theDates[theDates.length-1]]

            rdata.column_names.shift() # Remove Date
            
            chart = quiche('line')
            chart.setTitle(rdata.name);
            chart.setWidth(750);
            chart.setHeight(400);
            chart.addData(formattedData.a.reverse(), rdata.column_names[2], '6899C9');
            chart.addData(formattedData.b.reverse(), rdata.column_names[5], '9E5A8D');
            chart.addAxisLabels('x', xVals.reverse());
            chart.setAutoScaling();
            imageUrl = chart.getUrl(false)

            finalUrl = "#{imageUrl}#.png"
            gShort.shorten finalUrl, (err, url) ->
              msg.send(url); # http://goo.gl/fbsS
              msg.send rdata.description
              msg.send "See the report at " + rdata.display_url

          else 

            msg.send "The dataset was too confusing, so I gave up."


  #
  # Venture-Backed IPOs (Mean and Median Age at IPO)
  # This data is from the NVCA yearbook, which includes a comprehensive analysis of U.S. venture capital industry statistics. Age in years and values in USD million.
  # 
  # Command:
  #   user > hubot vc ipos age
  #
  robot.respond /vc ipos age/i, (msg) ->

    curDate = new Date()
    curDate.setYear(curDate.getFullYear()-3) # Get the date three years ago
    curDate = curDate.getFullYear() + "-" + curDate.getMonth() + "-" + "01"

    robot.http(process.env.HUBOT_DATASETS_URL + "NVCA/VENTURE_10_0.json?trim_start=" + curDate)
       .header('accept', 'application/json')
       .get() (err, res, body) -> 

          if err
            msg.send "Encountered an error :( #{err}"

          if res.statusCode isnt 200
            msg.send "I wasn't able to figure out what the numbers are."
          
          rdata = JSON.parse(body) if body

          if rdata

            # Use the data that was compiled
            formattedData = {}
            formattedData.dates = ( dates[0] for dates in rdata.data)
            formattedData.a = (a[8] for a in rdata.data)
            formattedData.b = (b[9] for b in rdata.data)
            
            theDates = formattedData.dates
            xVals = [theDates[0], theDates[Math.ceil(theDates.length/2)], theDates[theDates.length-1]]

            rdata.column_names.shift() # Remove Date
            
            chart = quiche('line')
            chart.setTitle(rdata.name);
            chart.setWidth(750);
            chart.setHeight(400);
            chart.addData(formattedData.a.reverse(), rdata.column_names[7], '6899C9');
            chart.addData(formattedData.b.reverse(), rdata.column_names[8], '9E5A8D');
            chart.addAxisLabels('x', xVals.reverse());
            chart.setAutoScaling();
            imageUrl = chart.getUrl(false)

            finalUrl = "#{imageUrl}#.png"
            gShort.shorten finalUrl, (err, url) ->
              msg.send(url); # http://goo.gl/fbsS
              msg.send rdata.description
              msg.send "See the report at " + rdata.display_url

          else 

            msg.send "The dataset was too confusing, so I gave up."

  #
  # Venture-Backed IPOs (Number of IPOs and Offer Amount)
  # This data is from the NVCA yearbook, which includes a comprehensive analysis of U.S. venture capital industry statistics. Age in years and values in USD million.
  # 
  # Command:
  #   user > hubot vc ipos
  #
  robot.respond /vc ipos offer/i, (msg) ->

    curDate = new Date()
    curDate.setYear(curDate.getFullYear()-3) # Get the date three years ago
    curDate = curDate.getFullYear() + "-" + curDate.getMonth() + "-" + "01"

    robot.http(process.env.HUBOT_DATASETS_URL + "NVCA/VENTURE_10_0.json?trim_start=" + curDate)
       .header('accept', 'application/json')
       .get() (err, res, body) -> 

          if err
            msg.send "Encountered an error :( #{err}"

          if res.statusCode isnt 200
            msg.send "I wasn't able to figure out what the numbers are."
          
          rdata = JSON.parse(body) if body

          if rdata

            # Use the data that was compiled
            formattedData = {}
            formattedData.dates = ( dates[0] for dates in rdata.data)
            formattedData.a = (a[1] for a in rdata.data)
            formattedData.b = (b[2] for b in rdata.data)
            
            theDates = formattedData.dates
            xVals = [theDates[0], theDates[Math.ceil(theDates.length/2)], theDates[theDates.length-1]]

            rdata.column_names.shift() # Remove Date
            
            chart = quiche('line')
            chart.setTitle(rdata.name);
            chart.setWidth(750);
            chart.setHeight(400);
            chart.addData(formattedData.a.reverse(), rdata.column_names[0], '6899C9');
            chart.addData(formattedData.b.reverse(), rdata.column_names[1], '9E5A8D');
            chart.addAxisLabels('x', xVals.reverse());
            chart.setAutoScaling();
            imageUrl = chart.getUrl(false)

            finalUrl = "#{imageUrl}#.png"
            gShort.shorten finalUrl, (err, url) ->
              msg.send(url); # http://goo.gl/fbsS
              msg.send rdata.description
              msg.send "See the report at " + rdata.display_url

          else 

            msg.send "The dataset was too confusing, so I gave up."
  
  #
  # Angel Investments
  # Data compiled from the Center for Venture Research annual reports. Total Investmments in USD billion.
  # 
  # Command:
  #   user > hubot active angels
  #
  robot.respond /active angels/i, (msg) ->

    curDate = new Date()
    curDate.setYear(curDate.getFullYear()-3) # Get the date three years ago
    curDate = curDate.getFullYear() + "-" + curDate.getMonth() + "-" + "01"

    robot.http(process.env.HUBOT_DATASETS_URL + "CVR/ANGEL_TOTALS.json?trim_start=" + curDate)
       .header('accept', 'application/json')
       .get() (err, res, body) -> 

          if err
            msg.send "Encountered an error :( #{err}"

          if res.statusCode isnt 200
            msg.send "I wasn't able to figure out what the numbers are."
          
          rdata = JSON.parse(body) if body

          if rdata

            # Use the data that was compiled
            #msg.send "Please wait a few seconds. Now creating..."

            formattedData = {}
            formattedData.dates = ( dates[0] for dates in rdata.data)
            # formattedData.a = (a[1] for a in rdata.data) #Total Investments ($B)
            # formattedData.b = (b[2] for b in rdata.data) #Total Ventures Receiving Funds
            formattedData.c = (c[3] for c in rdata.data) #Active Investors
            
            theDates = formattedData.dates
            xVals = [theDates[0], theDates[Math.ceil(theDates.length/2)], theDates[theDates.length-1]]

            rdata.column_names.shift() # Remove Date
            
            chart = quiche('line')
            chart.setTitle(rdata.name);
            chart.setWidth(750);
            chart.setHeight(400);
            # chart.addData(formattedData.a.reverse(), rdata.column_names[0], '6899C9');
            # chart.addData(formattedData.b.reverse(), rdata.column_names[1], '9E5A8D'); #Number known
            chart.addData(formattedData.c.reverse(), rdata.column_names[2], '50A450');
            chart.addAxisLabels('x', xVals.reverse());
            chart.setAutoScaling();
            #chart.setTransparentBackground();
            imageUrl = chart.getUrl(false)

            finalUrl = "#{imageUrl}#.png"
            gShort.shorten finalUrl, (err, url) ->
              msg.send(url); # http://goo.gl/fbsS
              msg.send rdata.description
              msg.send "See the report at " + rdata.display_url

          else 

            msg.send "The dataset was too confusing, so I gave up."

  #
  # Angel Investments
  # Data compiled from the Center for Venture Research annual reports. Total Investmments in USD billion.
  # 
  # Command:
  #   user > hubot angel totals
  #
  robot.respond /angel totals/i, (msg) ->

    curDate = new Date()
    curDate.setYear(curDate.getFullYear()-3) # Get the date three years ago
    curDate = curDate.getFullYear() + "-" + curDate.getMonth() + "-" + "01"

    robot.http(process.env.HUBOT_DATASETS_URL + "CVR/ANGEL_TOTALS.json?trim_start=" + curDate)
       .header('accept', 'application/json')
       .get() (err, res, body) -> 

          if err
            msg.send "Encountered an error :( #{err}"

          if res.statusCode isnt 200
            msg.send "I wasn't able to figure out what the numbers are."
          
          rdata = JSON.parse(body) if body

          if rdata

            # Use the data that was compiled
            #msg.send "Please wait a few seconds. Now creating..."

            formattedData = {}
            formattedData.dates = ( dates[0] for dates in rdata.data)
            formattedData.a = (a[1] for a in rdata.data) #Total Investments ($B)
            # formattedData.b = (b[2] for b in rdata.data) #Total Ventures Receiving Funds
            #formattedData.c = (c[3] for c in rdata.data) #Active Investors
            
            theDates = formattedData.dates
            xVals = [theDates[0], theDates[Math.ceil(theDates.length/2)], theDates[theDates.length-1]]

            rdata.column_names.shift() # Remove Date
            
            chart = quiche('line')
            chart.setTitle(rdata.name);
            chart.setWidth(750);
            chart.setHeight(400);
            chart.addData(formattedData.a.reverse(), rdata.column_names[0], '6899C9');
            # chart.addData(formattedData.b.reverse(), rdata.column_names[1], '9E5A8D'); #Number known
            # chart.addData(formattedData.c.reverse(), rdata.column_names[2], '50A450');
            chart.addAxisLabels('x', xVals.reverse());
            chart.setAutoScaling();
            #chart.setTransparentBackground();
            imageUrl = chart.getUrl(false)

            finalUrl = "#{imageUrl}#.png"
            gShort.shorten finalUrl, (err, url) ->
              msg.send(url); # http://goo.gl/fbsS
              msg.send rdata.description
              msg.send "See the report at " + rdata.display_url

          else 

            msg.send "The dataset was too confusing, so I gave up."

  #
  # Venture-Backed M & A Exits
  # This data is from the NVCA yearbook, which includes a comprehensive analysis of U.S. venture capital industry statistics. Price and Average are in USD millions.
  # 
  # Command:
  #   user > hubot vc exits
  #
  robot.respond /vc exits/i, (msg) ->

    curDate = new Date()
    curDate.setYear(curDate.getFullYear()-3) # Get the date three years ago
    curDate = curDate.getFullYear() + "-" + curDate.getMonth() + "-" + "01"

    robot.http(process.env.HUBOT_DATASETS_URL + "NVCA/VENTURE_10_0.json?trim_start=" + curDate)
       .header('accept', 'application/json')
       .get() (err, res, body) -> 

          if err
            msg.send "Encountered an error :( #{err}"

          if res.statusCode isnt 200
            msg.send "I wasn't able to figure out what the numbers are."
          
          rdata = JSON.parse(body) if body

          if rdata

            # Use the data that was compiled
            #msg.send "Please wait a few seconds. Now creating..."

            formattedData = {}
            formattedData.dates = ( dates[0] for dates in rdata.data)
            # formattedData.a = (a[1] for a in rdata.data)
            formattedData.b = (b[2] for b in rdata.data)
            # formattedData.c = (c[3] for c in rdata.data)
            # formattedData.d = (d[4] for d in rdata.data)
            
            theDates = formattedData.dates
            xVals = [theDates[0], theDates[Math.ceil(theDates.length/2)], theDates[theDates.length-1]]

            rdata.column_names.shift() # Remove Date
            
            chart = quiche('line')
            chart.setTitle("Venture-Backed Merger and Acquisition Exits");
            chart.setWidth(750);
            chart.setHeight(400);
            # chart.addData(formattedData.a.reverse(), rdata.column_names[0], '6899C9');
            chart.addData(formattedData.b.reverse(), rdata.column_names[1], '9E5A8D'); #Number known
            # chart.addData(formattedData.c.reverse(), rdata.column_names[2], '50A450');
            # chart.addData(formattedData.d.reverse(), rdata.column_names[3], 'E57F3C');
            chart.addAxisLabels('x', xVals.reverse());
            chart.setAutoScaling();
            #chart.setTransparentBackground();
            imageUrl = chart.getUrl(false)

            finalUrl = "#{imageUrl}#.png"
            gShort.shorten finalUrl, (err, url) ->
              msg.send(url); # http://goo.gl/fbsS
              msg.send rdata.description
              msg.send "See the report at " + rdata.display_url

          else 

            msg.send "The dataset was too confusing, so I gave up."

  #
  # Venture Capital Investments By Stage - Number of Deals (Quarterly)
  # This data is from the NVCA yearbook, which includes a comprehensive analysis of U.S. venture capital industry statistics.
  # 
  # Command:
  #   user > hubot vc deal #
  #
  robot.respond /vc deal (#|numbers)/i, (msg) ->

    curDate = new Date()
    curDate.setYear(curDate.getFullYear()-3) # Get the date three years ago
    curDate = curDate.getFullYear() + "-" + curDate.getMonth() + "-" + "01"

    robot.http(process.env.HUBOT_DATASETS_URL + "NVCA/VENTURE_3_09D.json?trim_start=" + curDate)
       .header('accept', 'application/json')
       .get() (err, res, body) -> 

          if err
            msg.send "Encountered an error :( #{err}"

          if res.statusCode isnt 200
            msg.send "I wasn't able to figure out what the numbers are."
          
          rdata = JSON.parse(body) if body

          if rdata

            # Use the data that was compiled
            #msg.send "Please wait a few seconds. Now creating..."

            formattedData = {}
            formattedData.dates = ( dates[0] for dates in rdata.data)
            formattedData.a = (a[1] for a in rdata.data)
            formattedData.b = (b[2] for b in rdata.data)
            formattedData.c = (c[3] for c in rdata.data)
            formattedData.d = (d[4] for d in rdata.data)
            
            theDates = formattedData.dates
            xVals = [theDates[0], theDates[Math.ceil(theDates.length/2)], theDates[theDates.length-1]]

            rdata.column_names.shift() # Remove Date
            
            chart = quiche('line')
            chart.setTitle(rdata.name);
            chart.setWidth(750);
            chart.setHeight(400);
            chart.addData(formattedData.a.reverse(), rdata.column_names[0], '6899C9');
            chart.addData(formattedData.b.reverse(), rdata.column_names[1], '9E5A8D');
            chart.addData(formattedData.c.reverse(), rdata.column_names[2], '50A450');
            chart.addData(formattedData.d.reverse(), rdata.column_names[3], 'E57F3C');
            chart.addAxisLabels('x', xVals.reverse());
            chart.setAutoScaling();
            #chart.setTransparentBackground();
            imageUrl = chart.getUrl(false)

            finalUrl = "#{imageUrl}#.png"
            gShort.shorten finalUrl, (err, url) ->
              msg.send(url); # http://goo.gl/fbsS
              msg.send rdata.description
              msg.send "See the report at " + rdata.display_url

          else 

            msg.send "The dataset was too confusing, so I gave up."


  #
  # Venture Capital - Deals by Series (Quarterly)
  # Startup financing is typically done in several rounds, as the startup company grows and its capital needs evolve. The first institutional round is 
  # usually called the "Series A round". Later rounds are called Series B, C, D, etc.
  # 
  # Command:
  #   user > hubot vc deals
  #
  robot.respond /vc (deals|deals by series)/i, (msg) ->

    curDate = new Date()
    curDate.setYear(curDate.getFullYear()-3) # Get the date three years ago
    curDate = curDate.getFullYear() + "-" + curDate.getMonth() + "-" + "01"

    robot.http(process.env.HUBOT_DATASETS_URL + "COOLEY/VC_DEALS_BY_SERIES.json?trim_start=" + curDate + "&collapse=quarterly")
       .header('accept', 'application/json')
       .get() (err, res, body) -> 

          if err
            msg.send "Encountered an error :( #{err}"

          if res.statusCode isnt 200
            msg.send "I wasn't able to figure out what the numbers are."
          
          rdata = JSON.parse(body) if body

          if rdata

            # Use the data that was compiled
            #msg.send "Please wait a few seconds. Now creating..."

            formattedData = {}
            formattedData.dates = ( dates[0] for dates in rdata.data)
            formattedData.series_a = (a[1] for a in rdata.data)
            formattedData.series_b = (b[2] for b in rdata.data)
            formattedData.series_c = (c[3] for c in rdata.data)
            formattedData.series_d = (d[4] for d in rdata.data)
            
            theDates = formattedData.dates
            xVals = [theDates[0], theDates[Math.ceil(theDates.length/2)], theDates[theDates.length-1]]
            #theDates = theDates.join(',').replace(/\-/g,'').split(',') # Remove the - so that charts will read as a number

            series_a = formattedData.series_a 
            series_b = formattedData.series_b 
            series_c = formattedData.series_c
            series_d = formattedData.series_d

            rdata.column_names.shift() # Remove Date
            
            chart = quiche('line')
            chart.setTitle(rdata.name);
            chart.setWidth(750);
            chart.setHeight(400);
            chart.addData(series_a.reverse(), rdata.column_names[0], '6899C9');
            chart.addData(series_b.reverse(), rdata.column_names[1], '9E5A8D');
            chart.addData(series_c.reverse(), rdata.column_names[2], '50A450');
            chart.addData(series_d.reverse(), rdata.column_names[3], 'E57F3C');
            chart.addAxisLabels('x', xVals.reverse());
            chart.setAutoScaling();
            #chart.setTransparentBackground();
            imageUrl = chart.getUrl(false)

            finalUrl = "#{imageUrl}#.png"
            gShort.shorten finalUrl, (err, url) ->
              msg.send(url); # http://goo.gl/fbsS
              msg.send rdata.description
              msg.send "See the report at " + rdata.display_url

          else 

            msg.send "The dataset was too confusing, so I gave up."

  #
  # Venture Capital - Median Pre-Money Valuations by Series
  # This data provides a summary of data reflecting our experience in venture capital financing terms and trends. Information is taken from transactions 
  # in which Cooley served as counsel to either the issuing company or the investors. Values in USD million.
  # 
  # Command:
  #   user > hubot vc premoney
  #
  robot.respond /vc (premoney|pre)/i, (msg) ->

    curDate = new Date()
    curDate.setYear(curDate.getFullYear()-3) # Get the date three years ago
    curDate = curDate.getFullYear() + "-" + curDate.getMonth() + "-" + "01"
    
    robot.http(process.env.HUBOT_DATASETS_URL + "COOLEY/VC_VALUE_BY_SERIES.json?trim_start=" + curDate + "&collapse=quarterly")
       .header('accept', 'application/json')
       .get() (err, res, body) -> 

          if err
            msg.send "Encountered an error :( #{err}"

          if res.statusCode isnt 200
            msg.send "I wasn't able to figure out what the numbers are."
          
          rdata = JSON.parse(body) if body

          if rdata

            # Use the data that was compiled
            #msg.send "Please wait a few seconds. Now creating..."

            formattedData = {}
            formattedData.dates = ( dates[0] for dates in rdata.data)
            formattedData.series_a = (a[1] for a in rdata.data)
            formattedData.series_b = (b[2] for b in rdata.data)
            formattedData.series_c = (c[3] for c in rdata.data)
            formattedData.series_d = (d[4] for d in rdata.data)
            
            theDates = formattedData.dates
            xVals = [theDates[0], theDates[Math.ceil(theDates.length/2)], theDates[theDates.length-1]]
            #theDates = theDates.join(',').replace(/\-/g,'').split(',') # Remove the - so that charts will read as a number

            series_a = formattedData.series_a 
            series_b = formattedData.series_b 
            series_c = formattedData.series_c
            series_d = formattedData.series_d

            rdata.column_names.shift() # Remove Date
            
            chart = quiche('line')
            chart.setTitle(rdata.name);
            chart.setWidth(750);
            chart.setHeight(400);
            chart.addData(series_a.reverse(), rdata.column_names[0], '6899C9');
            chart.addData(series_b.reverse(), rdata.column_names[1], '9E5A8D');
            chart.addData(series_c.reverse(), rdata.column_names[2], '50A450');
            chart.addData(series_d.reverse(), rdata.column_names[3], 'E57F3C');
            chart.addAxisLabels('x', xVals.reverse());
            chart.setAutoScaling();
            #chart.setTransparentBackground();
            imageUrl = chart.getUrl(false)

            finalUrl = "#{imageUrl}#.png"
            gShort.shorten finalUrl, (err, url) ->
              msg.send(url); # http://goo.gl/fbsS
              msg.send rdata.description
              msg.send "See the report at " + rdata.display_url

          else 

            msg.send "The dataset was too confusing, so I gave up."
  
  #
  # Venture Capital - Total Deal Volumes (Quarterly)
  # This data provides a summary of data reflecting our experience in venture capital financing terms and trends. 
  # Information is taken from transactions in which Cooley served as counsel to either the issuing company or the investors. Amount raised in USD million.
  # 
  # Command:
  #   user > hubot vc deal volumes
  #
  robot.respond /vc (deal volumes|dv)/i, (msg) ->

    curDate = new Date()
    curDate.setYear(curDate.getFullYear()-3) # Get the date three years ago
    curDate = curDate.getFullYear() + "-" + curDate.getMonth() + "-" + "01"
    
    robot.http(process.env.HUBOT_DATASETS_URL + "COOLEY/VC_TOTAL_DEAL_VOLUME.json?trim_start=" + curDate + "&collapse=quarterly")
       .header('accept', 'application/json')
       .get() (err, res, body) -> 

          if err
            msg.send "Encountered an error :( #{err}"

          if res.statusCode isnt 200
            msg.send "I wasn't able to figure out what the numbers are."
          
          rdata = JSON.parse(body) if body

          if rdata

            # Use the data that was compiled

            formattedData = {}
            formattedData.dates = ( dates[0] for dates in rdata.data)
            formattedData.a = (a[1] for a in rdata.data)
            formattedData.b = (b[2] for b in rdata.data)
            
            theDates = formattedData.dates
            xVals = [theDates[0], theDates[Math.ceil(theDates.length/2)], theDates[theDates.length-1]]

            rdata.column_names.shift() # Remove Date
            
            chart = quiche('line')
            chart.setTitle(rdata.name);
            chart.setWidth(750);
            chart.setHeight(400);
            chart.addData(formattedData.a.reverse(), rdata.column_names[0], '6899C9');
            chart.addData(formattedData.b.reverse(), rdata.column_names[1], '9E5A8D');

            chart.addAxisLabels('x', xVals.reverse());
            chart.setAutoScaling();
            #chart.setTransparentBackground();
            imageUrl = chart.getUrl(false)

            finalUrl = "#{imageUrl}#.png"
            gShort.shorten finalUrl, (err, url) ->
              msg.send(url); # http://goo.gl/fbsS
              msg.send rdata.description
              msg.send "See the report at " + rdata.display_url

          else 

            msg.send "The dataset was too confusing, so I gave up."

  #
  # Venture Capital - Percentage of Up Rounds (Quarterly)
  # This data provides a summary of data reflecting our experience in venture capital financing terms and trends. Information is taken from 
  # transactions in which Cooley served as counsel to either the issuing company or the investors.
  # 
  # Command:
  #   user > hubot vc ups
  #
  robot.respond /vc ups/i, (msg) ->

    curDate = new Date()
    curDate.setYear(curDate.getFullYear()-3) # Get the date three years ago
    curDate = curDate.getFullYear() + "-" + curDate.getMonth() + "-" + "01"

    robot.http(process.env.HUBOT_DATASETS_URL + "COOLEY/VC_UPROUNDS.json?trim_start=" + curDate + "&collapse=quarterly")
       .header('accept', 'application/json')
       .get() (err, res, body) -> 

          if err
            msg.send "Encountered an error :( #{err}"

          if res.statusCode isnt 200
            msg.send "I wasn't able to figure out what the numbers are."
          
          rdata = JSON.parse(body) if body

          if rdata

            # Use the data that was compiled

            formattedData = {}
            formattedData.dates = ( dates[0] for dates in rdata.data)
            formattedData.a = (a[1] for a in rdata.data)
            
            theDates = formattedData.dates
            xVals = [theDates[0], theDates[Math.ceil(theDates.length/2)], theDates[theDates.length-1]]

            rdata.column_names.shift() # Remove Date
            
            chart = quiche('line')
            chart.setTitle(rdata.name);
            chart.setWidth(750);
            chart.setHeight(400);
            chart.addData(formattedData.a.reverse(), rdata.column_names[0], '6899C9');

            chart.addAxisLabels('x', xVals.reverse());
            chart.setAutoScaling();
            #chart.setTransparentBackground();
            imageUrl = chart.getUrl(false)

            finalUrl = "#{imageUrl}#.png"
            gShort.shorten finalUrl, (err, url) ->
              msg.send(url); # http://goo.gl/fbsS
              msg.send rdata.description
              msg.send "See the report at " + rdata.display_url

          else 

            msg.send "The dataset was too confusing, so I gave up."

  #
  # The 7 Step Startup Investment Process for Venture Capitalists
  # It's not as easy as they make it out to be on Shark Tank. In the words of Meek Mill, "there are levels to this shit."
  # 
  # Command:
  #   user > hubot stp
  #
  robot.respond /startup investment process|stp/i, (msg) ->
      process_details = "The 7 Step Startup Investment Process for Venture Capitalists\n"
      process_details += "1. Identify the startup to invest in.\n"
      process_details += "2. Initial screening - Get to know more about the business and the entrepreneur.\n"
      process_details += "3. Second screening – This could involve workshops, a deeper dive, specific Q&A and in these an investor may bring in multiple subject matter experts that are trusted or that are leading other investments from within the firm.\n"
      process_details += "4. Light due dilligence\n"
      process_details += "5. The partner meeting - This is where the investment thesis is initially presented by the investor to the partners. This could be in person, socializing the deal, or by circulating a memo.\n"
      process_details += "6. Entrepreneur invited to present to partners - This is a good sign that the partners liked the investment thesis and are ready to invest.\n"
      process_details += "7. Terms and due dilligence - Signing the paperwork and covering terms.\n"
      msg.send process_details


  #
  # Investment Resources
  # This what the investors are reading and following.
  # 
  # Command:
  #   user > hubot investment resources
  #
  robot.respond /investment resources|ir/i, (msg) ->
      books_ar = [
        "Angel Investing: The Gust Guide to Making Money and Having Fun Investing in Startups by David S. Rose (http://amzn.to/1sJTcS8)"
        "Venture Deals: Be Smarter Than Your Lawyer and Venture Capitalist by Brad Feld (http://amzn.to/1bia6Qa)"
        "The Alliance Framework by Reid Hoffman, Ben Casnocha and Chris Yeh (http://www.theallianceframework.com)"
        "Early Exits: Exit Strategies for Entrepreneurs and Angel Investors (But Maybe Not Venture Capitalists) by Basil Peters (http://amzn.to/1wYvUxQ)"
        "The Definitive Guide to Raising Money from Angels by Bill Payne (Download @ http://bit.ly/1uqWS0e)"
      ]
      books_str = "-- Books --\n"
      for i in [0..books_ar.length-1] by 1
        books_str += (i+1) + ". " + books_ar[i] + "\n"

      blogs_ar = [
        "Venture Hacks (http://venturehacks.com)"
        "Adventures in Capitalism (http://adventuresincapitalism.com)"
      ]
      blogs_str = "-- Blogs -- \n"
      for i in [0..blogs_ar.length-1] by 1
        blogs_str += (i+1) + ". " + blogs_ar[i] + "\n"

      people_ar = [
        "Bill Payne"
        "John O. Huston"
        "David S. Rose"
        "Chris Yeh"
        "Howard Tullman"
        "George Deeb"
        "Charlie O'Donnell"
        "Jeffrey Carter"
        "Michael Gruber"
        "Gabriel Weinberg"
        "Glen Gottfried"
        "Ann Winblad"
        "Erin Griffith"
        "Troy Henikoff"
        "Dave Berkus"
        "Imran Ahmad"
        "Jerry Neumann"
        "Rob Go"
      ]
      people_str = "-- People -- \n"
      for i in [0..people_ar.length-1] by 1
        people_str += (i+1) + ". " + people_ar[i] + "\n"

      msg.send books_str + "\n" +  blogs_str + "\n" + people_str