#!/usr/bin/env ruby

#
# This program produces depth range plots from tables exported from the database gene_sample
#
# 2009-06-01                   Michael Matschiner <michaelmatschiner@mac.com>
#
# run this script with a table produced by the below SQL query. Copy the part between
# =begin and =end into the Command window of Sequel Pro and press 'Run All'. Then use
# the Sequel Pro menu to export the results table in CSV format (File > Export > Custom
# Query Result > CSV file... ).
=begin
SELECT organism.latin_name, ROUND(MIN(station.min_water_depth + (station.max_water_depth-station.min_water_depth)/2.0),1), ROUND(MAX(station.min_water_depth + (station.max_water_depth-station.min_water_depth)/2.0),1) , ROUND(MIN(station.mean_water_depth),1), ROUND(MAX(station.mean_water_depth),1)
FROM organism INNER JOIN specimen ON specimen.organism_id = organism.id
              INNER JOIN station ON specimen.station_id = station.id
GROUP BY organism.latin_name
=end
# Run this script by typing
# 'ruby depth_range_plotter.rb input_filename'
# in the Terminal


# Class "Array2" is used to build a two-dimensional matrix which stores
# organism names and water depths as given in the CSV file.
#
class Array2
  def initialize
    @store = [[]]
  end
  def [](a,b)
    if @store[a]==nil ||
      @store[a][b]==nil
      return nil
    else
      return @store[a][b]
    end
  end
  def []=(a,b,x)
    @store[a] = [] if @store[a]==nil
    @store[a][b] = x
  end
end


# The method to_s() is added to class Float for economical output. This way,
# the number of decimals included in floats can be chosen.
class Float
  alias_method :orig_to_s, :to_s
  def to_s(arg = nil)
    if arg.nil?
      orig_to_s
    else
      sprintf("%.#{arg}f", self)
    end
  end
end



# The main program starts here. A number of default values is set. See below for dimX.
progName = "depth_range_plotter.rb"
vertTicks = true
vertTickLength = 3
horiTickLength = 3
strokeWidth = 1.0
barwidth = 10  # change this value to stretch the whole plot in x direction.
fontSize = 10  # change this value to increase or decrease the font size.
rightMargin = 100
topMargin = 120
bottomMargin = fontSize/2
dimY = 400


# The input file is the first command line argument.
fileIn = ARGV[0]


# The input file is read as a single string. Line endings are detected,
# and used to split the string into an array of strings.
#
str = IO.read( fileIn )
ary = str.split(/\n/) if str.include?("\n")

# The header line is deleted
#
ary.shift

# The dimensions of the table is detected.
#
rowNum = ary.size
colNum = 5
inquotes = false

mat = Array2.new
rowNum.times do |r|
  tmpstr = ary[r]
  commapositions = []
  tmpstr.length.times do |l|
    if tmpstr[l..l] == '"'
      if inquotes == false
        inquotes = true
      else
        inquotes = false
      end
    elsif tmpstr[l..l] == ","
      commapositions << l if inquotes == false
    end
  end
  mat[r,0] = tmpstr[0..commapositions[0]-1].delete '"'
  mat[r,1] = tmpstr[commapositions[0]+1..commapositions[1]-1]
  mat[r,2] = tmpstr[commapositions[1]+1..commapositions[2]-1]
  mat[r,3] = tmpstr[commapositions[2]+1..commapositions[3]-1]
  mat[r,4] = tmpstr[commapositions[3]+1..tmpstr.length].chomp
end

organism_names = []
depth_range_min = []
depth_range_max = []
rowNum.times do |r|
  organism_names << mat[r,0]
  potential_mins = []
  potential_mins << mat[r,1].to_f unless mat[r,1] == "NULL"
  potential_mins << mat[r,2].to_f unless mat[r,2] == "NULL"
  potential_mins << mat[r,3].to_f unless mat[r,3] == "NULL"
  potential_mins << mat[r,4].to_f unless mat[r,4] == "NULL"
  depth_range_min << potential_mins.min
  potential_maxs = []
  potential_maxs << mat[r,1].to_f unless mat[r,1] == "NULL"
  potential_maxs << mat[r,2].to_f unless mat[r,2] == "NULL"
  potential_maxs << mat[r,3].to_f unless mat[r,3] == "NULL"
  potential_maxs << mat[r,4].to_f unless mat[r,4] == "NULL"
  depth_range_max << potential_maxs.max
end

# Organism names and depth ranges are sorted.
#
i = 0
while i < organism_names.size-1 do
  if (depth_range_min[i] + depth_range_max[i]) > (depth_range_min[i+1] + depth_range_max[i+1])
    organism_names[i],organism_names[i+1] = organism_names[i+1],organism_names[i]
    depth_range_min[i],depth_range_min[i+1] = depth_range_min[i+1],depth_range_min[i]
    depth_range_max[i],depth_range_max[i+1] = depth_range_max[i+1],depth_range_max[i]
    i -= 1 unless i == 0
  else
    i += 1
  end
end


depth_range_max_max = depth_range_max.max
if depth_range_max_max > 1000
  y_axis_max = ((depth_range_max_max/1000).ceil)*1000
elsif depth_range_max_max > 100
  y_axis_max = ((depth_range_max_max/100).ceil)*100
elsif depth_range_max_max > 10
  y_axis_max = ((depth_range_max_max/10).ceil)*10
else
  y_axis_max = depth_range_max_max.ceil
end


dimX = (rowNum)*2*barwidth + rightMargin



# The SVG output is prepared.
#
svgOutput = String.new
svgOutput << "<?xml version=\"1.0\" standalone=\"no\"?>\n"
svgOutput << "<!DOCTYPE svg PUBLIC \"-//W3C//DTD SVG 1.0//EN\" \"http://www.w3.org/TR/2001/REC-SVG-20010904/DTD/svg10.dtd\">\n"
svgOutput << "<svg xmlns=\"http://www.w3.org/2000/svg\" width=\"#{dimX}\" height=\"#{dimY}\" viewBox=\"0 0 #{dimX} #{dimY}\"\n"
svgOutput << "  xmlns:xlink=\"htttp://www.w3.org/1999/xlink\">\n\n"
svgOutput << "<!--created with #{progName}-->\n"
svgOutput << "\n\n"


#svgOutput << "<!--Bars-->\n"
#rowNum.times do |r|
#  clustNum.times do |c|
#    svgOutput << "<rect style=\"fill:black\" x=\"#{((dimX-2*strokeWidth+vertLinesWidth)*((r).to_f/rowNum.to_f)+strokeWidth-vertLinesWidth/2.0).to_s(2)}\" y=\"#{(strokeWidth+sum*(dimY-2*strokeWidth)*(1.0/all)).to_s(2)}\" width=\"#{((dimX-2*strokeWidth)/rowNum).to_s(2)}\" height=\"#{(((dimY-2*strokeWidth)*mat[r,c+5].to_f)*(1.0/all)).to_s(2)}\" />\n"
#    sum += mat[r,c+5].to_f
#  end
#end
#svgOutput << "\n\n"
#

svgOutput << "<!--Depth ranges-->\n"
rowNum.times do |r|
  if depth_range_max[r]-depth_range_min[r] > y_axis_max/100
    svgOutput << "<line id=\"#{r+1}\" style=\"stroke:black; stroke-width:#{barwidth}px\" x1=\"#{barwidth+(2*r*barwidth)}\" y1=\"#{topMargin+(dimY-topMargin-bottomMargin)*(depth_range_min[r]/y_axis_max)}\" x2=\"#{barwidth+2*r*barwidth}\" y2=\"#{topMargin+(dimY-topMargin-bottomMargin)*(depth_range_max[r]/y_axis_max)}\"/>\n"
  else
    svgOutput << "<circle id=\"#{r+1}\" cx=\"#{barwidth+2*r*barwidth}\" cy=\"#{topMargin+(dimY-topMargin-bottomMargin)*(depth_range_min[r]/y_axis_max)}\" r=\"#{barwidth/2.0}\" fill=\"black\"/>\n"
  end
end  
svgOutput << "\n\n"



svgOutput << "<!--Organism names-->\n"
rowNum.times do |r|
  svgOutput << "<text x=\"#{barwidth+2*r*barwidth}\" y=\"#{topMargin}\" font-family=\"Helvetica,sans-serif\" font-size=\"#{fontSize}px\" transform=\"rotate(-45,#{barwidth+2*r*barwidth},#{topMargin})\">#{organism_names[r]}</text>"
end  
svgOutput << "\n\n"



svgOutput << "<!--Horizontal ticks-->\n"
svgOutput << "<text x=\"#{dimX-rightMargin+2.5}\" y=\"#{topMargin+0.3*fontSize}\" font-family=\"Helvetica,sans-serif\" font-size=\"#{fontSize}px\">0</text>"
if y_axis_max > 3000
  (y_axis_max/1000).times do |j|
    svgOutput << "<line style=\"stroke:black; stroke-width:#{strokeWidth}px\" x1=\"#{dimX-rightMargin-horiTickLength}\" y1=\"#{topMargin+(dimY-topMargin-bottomMargin)*(((j+1)*1000)/y_axis_max.to_f)}\" x2=\"#{dimX-rightMargin}\" y2=\"#{topMargin+(dimY-topMargin-bottomMargin)*(((j+1)*1000)/y_axis_max.to_f)}\"/>\n" unless j+1 == (y_axis_max/1000)
    svgOutput << "<text x=\"#{dimX-rightMargin+2.5}\" y=\"#{topMargin+(dimY-topMargin-bottomMargin)*(((j+1)*1000)/y_axis_max.to_f)+0.3*fontSize}\" font-family=\"Helvetica,sans-serif\" font-size=\"#{fontSize}px\">#{(j+1)*1000}</text>"
  end
elsif y_axis_max > 300
  (y_axis_max/100).times do |j|
    svgOutput << "<line style=\"stroke:black; stroke-width:#{strokeWidth}px\" x1=\"#{dimX-rightMargin-horiTickLength}\" y1=\"#{topMargin+(dimY-topMargin-bottomMargin)*(((j+1)*100)/y_axis_max.to_f)}\" x2=\"#{dimX-rightMargin}\" y2=\"#{topMargin+(dimY-topMargin-bottomMargin)*(((j+1)*100)/y_axis_max.to_f)}\"/>\n" unless j+1 == (y_axis_max/100)
    svgOutput << "<text x=\"#{dimX-rightMargin+2.5}\" y=\"#{topMargin+(dimY-topMargin-bottomMargin)*(((j+1)*100)/y_axis_max.to_f)+0.3*fontSize}\" font-family=\"Helvetica,sans-serif\" font-size=\"#{fontSize}px\">#{(j+1)*100}</text>"
  end
elsif y_axis_max > 300
  (y_axis_max/10).times do |j|
    svgOutput << "<line style=\"stroke:black; stroke-width:#{strokeWidth}px\" x1=\"#{dimX-rightMargin-horiTickLength}\" y1=\"#{topMargin+(dimY-topMargin-bottomMargin)*(((j+1)*10)/y_axis_max.to_f)}\" x2=\"#{dimX-rightMargin}\" y2=\"#{topMargin+(dimY-topMargin-bottomMargin)*(((j+1)*10)/y_axis_max.to_f)}\"/>\n" unless j+1 == (y_axis_max/10)
    svgOutput << "<text x=\"#{dimX-rightMargin+2.5}\" y=\"#{topMargin+(dimY-topMargin-bottomMargin)*(((j+1)*10)/y_axis_max.to_f)+0.3*fontSize}\" font-family=\"Helvetica,sans-serif\" font-size=\"#{fontSize}px\">#{(j+1)*10}</text>"
  end
else
  y_axis_max.times do |j|
    svgOutput << "<line style=\"stroke:black; stroke-width:#{strokeWidth}px\" x1=\"#{dimX-rightMargin-horiTickLength}\" y1=\"#{topMargin+(dimY-topMargin-bottomMargin)*((j+1)/y_axis_max.to_f)}\" x2=\"#{dimX-rightMargin}\" y2=\"#{topMargin+(dimY-topMargin-bottomMargin)*((j+1)/y_axis_max.to_f)}\"/>\n" unless j+1 == y_axis_max
    svgOutput << "<text x=\"#{dimX-rightMargin+2.5}\" y=\"#{topMargin+(dimY-topMargin-bottomMargin)*((j+1)/y_axis_max.to_f)+0.3*fontSize}\" font-family=\"Helvetica,sans-serif\" font-size=\"#{fontSize}px\">#{j+1}</text>"
  end
end
svgOutput << "\n\n"

svgOutput << "<!--Y axis label-->\n"
svgOutput << "<text x=\"#{dimX-rightMargin+3.8*fontSize}\" y=\"#{topMargin+0.5*(dimY-topMargin-bottomMargin)+2*fontSize}\" font-family=\"Helvetica,sans-serif\" font-size=\"#{fontSize}px\" transform=\"rotate(-90,#{dimX-rightMargin+3.8*fontSize},#{topMargin+0.5*(dimY-topMargin-bottomMargin)+2*fontSize})\">Depth (m)</text>"


#if vertTicks
#  svgOutput << "<!--Vertical ticks between individuals-->\n"
#  (rowNum-1).times do |r|
#    unless mat[r,3].to_i < mat[r+1,3].to_i
#      svgOutput << "<line id=\"#{r+1}o\" style=\"stroke:black; stroke-width:#{vertLinesWidth}px\" x1=\"#{((dimX-2*strokeWidth+vertLinesWidth)*((r+1).to_f/rowNum.to_f)+strokeWidth-vertLinesWidth/2.0).to_s(2)}\" y1=\"#{strokeWidth}\" x2=\"#{((dimX-2*strokeWidth+vertLinesWidth)*((r+1).to_f/rowNum.to_f)+strokeWidth-vertLinesWidth/2.0).to_s(2)}\" y2=\"#{strokeWidth+vertTickLength}\"/>\n"
#      svgOutput << "<line id=\"#{r+1}u\" style=\"stroke:black; stroke-width:#{vertLinesWidth}px\" x1=\"#{((dimX-2*strokeWidth+vertLinesWidth)*((r+1).to_f/rowNum.to_f)+strokeWidth-vertLinesWidth/2.0).to_s(2)}\" y1=\"#{dimY-strokeWidth-vertTickLength}\" x2=\"#{((dimX-2*strokeWidth+vertLinesWidth)*((r+1).to_f/rowNum.to_f)+strokeWidth-vertLinesWidth/2.0).to_s(2)}\" y2=\"#{dimY-strokeWidth}\"/>\n"
#    end
#  end
#  svgOutput << "\n\n"
#end


svgOutput << "<!--Frame-->\n"
svgOutput << "<rect style=\"stroke:black; stroke-width:#{strokeWidth}px; fill:none\" x=\"#{strokeWidth/2.0}\" y=\"#{(strokeWidth/2.0)+topMargin}\" width=\"#{dimX-strokeWidth-rightMargin}\" height=\"#{dimY-strokeWidth-topMargin-bottomMargin}\" />\n"
svgOutput << "\n\n"


svgOutput << "</svg>"



# The string "output" is written to a new file with the same name as the input file,
# but with "_depth_range.svg" added to the name.
#
fileOutName = fileIn.scan(/\w+./)[0].chomp(".") + "_depth_range" + ".svg"
fileOut = File.new(fileOutName, "w")
fileOut.print svgOutput.to_s
puts "depth range plot written to \"#{fileOutName}\""