Prints a histogram of a binary file to file. Never stores the entire file in memory.

histogram.py

# Written with the aid of OpenAI Codex

# Implements a histogram and draws each bar with proportional number of '*' 
#
# Example Usage:
#
# data = []
# for i in range(0,10):
#    data.append(random.uniform(-3.4, 6.7))
# minVal = min(data)
# maxVal = max(data)
# numBins = 10
# maxGlyphs = 20
# histogram = Histogram( minVal, maxVal, numBins, maxGlyphs  )
# 
# for i in range(0,10):
#     histogram.add(data[i])
#
# print(histogram)
#

import struct, sys

class Histogram:
    def __init__(self, minVal, maxVal, numBins, maxGlyphs):
        self.minVal = minVal
        self.maxVal = maxVal
        self.numBins = numBins
        self.maxGlyphs = maxGlyphs
        self.bins = [0] * numBins
        self.binWidth = (maxVal - minVal) / numBins
        self.binMin = minVal
        self.binMax = minVal + self.binWidth
        self.largestBinValue = 0

    def add(self, value):
        if value < self.minVal:
            return
        if value > self.maxVal:
            return
        bin = int((value - self.minVal) / self.binWidth)
        if bin >= self.numBins:
            return
        self.bins[bin] += 1
        if self.bins[bin] > self.largestBinValue:
            self.largestBinValue = self.bins[bin]

    def __str__(self):
        result = ""
        for i in range(0, self.numBins):
            result += f'[{self.binMin:8.4f},{self.binMax:8.4f}): '
            result += "*" * int(self.bins[i] * self.maxGlyphs / self.largestBinValue)
            result += "\n"
            self.binMin += self.binWidth
            self.binMax += self.binWidth
        return result

def read_binary_floats_into_histogram( filename, numBins, numGlyphs ):
    # get minVal and maxVal in first pass
    minVal = float("inf")
    maxVal = float("-inf")
    with open(filename, "rb") as f:
        while True:
            try:
                value = struct.unpack('f', f.read(4))[0]
                if value < minVal:
                    minVal = value
                if value > maxVal:
                    maxVal = value
            except:
                break
    histogram = Histogram(minVal, maxVal, numBins, numGlyphs)
    with open(filename, "rb") as f:
        while True:
            try:
                value = struct.unpack('f', f.read(4))[0]
                histogram.add(value)
            except:
                break
    return histogram

if __name__ == '__main__':
    if len(sys.argv) < 2:
        print("Usage: python3 histogram.py <filename> [numBins] [numGlyphs]")
        sys.exit(1)
    filename = sys.argv[1]
    numBins = 10
    numGlyphs = 20
    if len(sys.argv) > 2:
        numBins = int(sys.argv[2])
    if len(sys.argv) > 3:
        numGlyphs = int(sys.argv[3])
    histogram = read_binary_floats_into_histogram(filename, numBins, numGlyphs)
    print(histogram)