Day two introduced some staples of staples of not only Advent of Code but also of programming problems in general. The first is input parsing. For this problem you get lines of input like this:

1-3 a: abcde
1-3 b: cdefg
2-9 c: cccccccc

or in general

number_1-number_2 Letter: String

There are a few ways to handle this. One is to brute force it. In Python maybe something like:

sample_line="4-15 f: abcdefg"
sample_list = sample_line.split()
numbers=sample_list[0]
letter=sample_list[1]
string=sample_list[2]

number_list = numbers.split('-')
num1 = int(number_list[0])
num2 = int(number_list[1])

letter = letter[0]

print(num1, num2, letter, string)

or more concisely:

sample_line="4-15 f: abcdefg"

(numbers,letter,string) = sample_line.split()

(num1,num2) = [int(x) for x in numbers.split('-')]

letter = letter[0]

print(num1, num2, letter, string)

If you're working in a language like C, tools like scanf can make things easier. I generally turn to Regular Expressions and if you're doing some AOC with your classes this is a great time to introduce regex. Basically, you set up a pattern with special symbols to represent things like sequences of digits and then the regex matcher does the hard work.

Here's the code to

1
2
3
4
5
6
7
8

parser_expression= "(\d+)-(\d+) ([a-z]): ([a-z]*)"
data = []
for line in open("../data/day02.dat").readlines():
    x = re.search(parser_expression,line)
    (mini,maxi,letter,password) = x.groups()
    mini = int(mini)
    maxi = int(maxi)
    data.append([mini,maxi,letter,password])

The parser_expression in line 1 is the regular expression. Each section in parentheses is a "group" or a pattern that will be extracted. The \d for instance means a digit and the + adger it means 1 or more digits so basically that will match any positive integer. The [a-z] matches a single character and the *after the final character match means zero or more of them. Once you know regular expressions, parsing lines like this becomes very easy. The only thing that made the above code messy at all was that I wanted to convert the two numbers into integers rather than leave them as strings.

There are a couple of things worth thinking about here. First is know your libraries. If you know regex the parsing is basically all done. If not, you have a bit of work. On the other hand, you're probably not going to ever have to parse input exactly like this again so it doesn't make sense writing a super robust set of parsing functions tied in with this data specification. That's the second thing. Learning what to leave as building blocks and what to write into libraries of your own. It's most definitely worth writing a set of routines that can be reused. Maybe something to parse dates in a standard format would be an example. On the other hand, it's also worth knowing when you probably won't be able to reuse things or when generalizing to a library or set of functions you have to spend too much time on a hundred options to make in general purpose.

Now, part 1 of the problem itself is nice because there are a lot of ways to do it. You could loop through the password and count the number of times a letter occurs:

count = 0
subpassword = password[mini-1,maxi]
for letter in subpassword:
    if letter == what_im_looking_for:
        count = count + 1

or with a list comprehension count = [x for x in password if x == waht_im_looking_for ].

You could be more general and build a hash table of counts and then pull what you want:

def frequencies(word):
    d={}
    for letter in word:
        d.setdefault(letter,0)
        d[letter]=d[letter]+1
    return d

This is admittedly overkill for the problem but one might need this for part 2. I coded my original solution in Clojure and Clojure has a function that does this in one shot: frequencies so I just used that and pulled the letter I needed.

Once you have the count, it's simple enough to see if it's in the range.

Part 2 should have been easy but for me it was a lesson in reading. You had to pull characters from the password at two specific indices and see if one or the other, but not both were the letter in question. Basically an xor. The first mistake one could make has to do with the indexing and the second in using a regular or which students will be more familiar with.

My mistake is that I misread the question as requiring that you have one instance of the letter within the range from the low index to the high index not one instance at either of those two points Needless to say, lots of wasted time and feeling silly.

So that's day 2. The question is very approachable for APCS-A and I'd say even CS0 students. The interesting part to teach would be dealing with the input, regular expressions, and how much to generalize into utility functions when writing code over time.

Hope to have time to solve and write up more of these but getting to crunch time in the semester.