''' Purpose: for user-specified dataset, and two user-specified column labels
        column1 and column2:
            determines the minimum value min1 in column1 in the dataset
            determines col2 -- the column2 value whose column1 value equals min1.

            prints labeled explanatory information about output
'''

# get ahold of helpful web functions
import url

# specify base web folder for datasets
CSV_REPOSITORY = 'http://www.cs.virginia.edu/~cs1112/datasets/csv/'

# get dataset of interest
reply = input( 'Enter the name of a data set: ' )

name = reply.strip()

link = CSV_REPOSITORY + name

dataset = url.get_dataset( link )

# get the labels for the columns of interest
label1 = input( 'Enter a column label for the data set: ' )
label2 = input( 'Enter a column label for the data set: ' )

print()

# clean up the labels
label1 = label1.strip()
label2 = label2.strip()

# identify header and data from the dataset

header = dataset[ 0 ]               # header is the first row of the dataset
table  = dataset[ 1 : ]             # other rows are our table of data values

# determine the indices index1 and index2 of label1 and label2 in the header
index1 = header.index( label1 )
index2 = header.index( label2 )

# need to find the min value min1 in the column of the table whose index is index1

# to find it, let's first build a list values1 of the values in that column

values1 = []                        # because we are going to accumulate them
                                    # we need to start off with an empty list

# need to add the index1 value of each row of the table
for row in table :                  # loop to consider the rows one by one

    # get the index1 value v from the current row
    v = row[ index1 ]

    # add v to the list of values1
    values1.append( v )

# print what we know so far
print( 'Analyzing', name )

print()

print( label1, 'values :', values1 )

print()

# now let's find the minimum value min1 in values1
min1 = min( values1 )               # built-in min() can find value min1 for us

# print what we now know
print( 'min', label1, 'value is', min1 )

print()

# now supposed to find the index2 column value for the row whose index1
# column value is min1

# key insight: 
#   the first value in values1 equals the value of the index1 column of the
#       first row in table

#   the second value in values1 equals the value of the index1 column of the
#       second row in table

#   the third ...

# implication:
#   if we can find the index i into values1 that equals min1, then the row
#       with index i in table has the index2 value we are looking for


i = values1.index( min1 )           # find where min1 occurs in values1

row_of_interest = table[ i ]        # find the corresponding row of table

# print what we now know

print( 'dataset row of interest:', row_of_interest )

# ready to determine the value col2 -- that is, the value in column index2 from
# the row of interst

col2 = row_of_interest[ index2 ]    # grab the value from the column of interest

# print what we now know

print( label2, ':', col2 )