### Exercise 6.7
### Make a nested dictionary from a file
### OCL

# First read the file into a list, with each list
# element being a whole line in the file
infile = open('human_evolution.txt', 'r')
lines = infile.readlines()
infile.close()

# To be able to split each line into individual columns,
# we first determine from the file header where (i.e. in
# what position) each column starts and stops. The start
# points are given by (a) 0 (the first column), (b) the 
# position where "(mill. yrs)" starts (the second column), 
# etc. The stop positions can easily be found from the 
# start positions; for example, the first column stops 
# right before the start of the second column. We assume
# that the last column stops at position 80. 

s = lines[1]
start = [0, s.index('(mill. yrs)'), s.index('height (m)'),
         s.index('mass (kg)'), s.index('(cm**3)')]
stop = start[1:len(start)] + [80]

# We also need to determine what is the first and last line
# of data, since the file also contains some files with other
# content. We notice that the data start with a line "---..." 
# and stop with a similar line. 

k = 0
while lines[k][0] != '-':  
    k = k + 1             
first = k + 1            
k = first                 
while lines[k][0] != '-': 
    k = k + 1             
last = k - 1              

# We create an empty dictionary 'humans' and then go 
# sequentially through each line of data, extract each 
# of the columns and put into variables species, period, ...
# We then create a "mini-dictionary" representing all 
# the data in a single line, and insert it into the larger
# dictionary 'humans'. 

humans = {}
for i in range(first, last+1):
    species = lines[i][start[0]:stop[0]].strip()
    period = lines[i][start[1]:stop[1]].strip()
    height = lines[i][start[2]:stop[2]].strip()
    weight = lines[i][start[3]:stop[3]].strip()
    volume = lines[i][start[4]:stop[4]].strip()
    humans[species] = {'period': period, 'height': height,
         'weight': weight, 'volume': volume}

# Finally, we print the contents of the dictionary 'humans' 
# as a nicely formatted table on the screen. The exact width
# of each column was determined by a little bit of trial and
# error. 

# First the header
s = '%-23s %-13s %-13s %-13s %-25s' % \
        ('species', 'period', 'height', 'weight', 'volume')
print(s)

# Then the table contents
for sp in humans:
    d = humans[sp]
    period = d['period']
    height = d['height']
    weight = d['weight']
    volume = d['volume']
    s = '%-23s %-13s %-13s %-13s %-25s' % \
        (sp, period, height, weight, volume)
    print(s)