Advent of Code Day 4

Advent of Code Day 4

[John Perry]

I found this one tougher, in part because I've never done much text file processing, but also because the directions were very scant on the format of the entries, and I wasn't about to comb carefully through hundreds of entries to see what all the variations might be.

I'll be very interested to see how different people read the file.

Re: Advent of Code Day 4

[Jeffrey R. Carter]

On 12/4/20 5:16 PM, John Perry wrote:
> I found this one tougher, in part because I've never done much text file processing, but also because the directions were very scant on the format of the entries, and I wasn't about to comb carefully through hundreds of entries to see what all the variations might be.
> 
> I'll be very interested to see how different people read the file.

Enumeration types are your friend.

-- 
Jeff Carter
"[O]ne can look forward to a rapid and widespread
improvement in programming practice, both from
those who use the language [Ada] and from those
who study its concepts and structures."
C. A. R. Hoare
181

Re: Advent of Code Day 4

[gautier...@hotmail.com]

> I'll be very interested to see how different people read the file.

https://github.com/zertovitch/hac/tree/master/exm/aoc/2020

Part Two has two solutions: one, aoc_2020_04_b.adb, using HAC (with its current limitations) and one which is "full Ada", aoc_2020_04_b_full_ada.adb.

Re: Advent of Code Day 4

[Maxim Reznik]

Day 4 task is dull :)

https://github.com/reznikmm/ada-howto/blob/advent-2020/md/04/04.md

Re: Advent of Code Day 4

[gautier...@hotmail.com]

> https://github.com/reznikmm/ada-howto/blob/advent-2020/md/04/04.md

Nice one!
   Index ("amb blu brn gry grn hzl oth", Text) > 0;

Oh, was 172 also considered correct for part two?!

Re: Advent of Code Day 4

[Stephen Leake]

John Perry <john.perry@usm.edu> writes:

> I found this one tougher, in part because I've never done much text
> file processing, but also because the directions were very scant on
> the format of the entries, and I wasn't about to comb carefully
> through hundreds of entries to see what all the variations might be.
>
> I'll be very interested to see how different people read the file.

I used gnat.Regpat:

   Field : constant Pattern_Matcher := Compile ("([a-z][a-z][a-z]):([^ ]+)");

   Height : constant Pattern_Matcher := Compile ("([0-9]+)(cm|in)");

   Hair_Color : constant Pattern_Matcher := Compile ("#[0-9a-f][0-9a-f][0-9a-f][0-9a-f][0-9a-f][0-9a-f]");

   Eye_Color : constant Pattern_Matcher := Compile ("amb|blu|brn|gry|grn|hzl|oth");

   ID : constant Pattern_Matcher := Compile ("[0-9][0-9][0-9][0-9][0-9][0-9][0-9][0-9][0-9]");


But apparently I've done something wrong; I counted 225 valid passports
for part 2, which the website claims is too high. My code parsed the
example valid/invalid passports correctly. I started going thru the full
input manually to check, but that's much too tedious. If we were doing
this for real, we'd have a test file with each entry labeled
valid/invalid.

I'm assuming we each get different data files? maybe not? mine has 1169
lines, first passport entry is:

iyr:2015 cid:189 ecl:oth byr:1947 hcl:#6c4ab1 eyr:2026
hgt:174cm
pid:526744288

--
-- Stephe

Re: Advent of Code Day 4

[Jeffrey R. Carter]

On 12/4/20 11:55 PM, Stephen Leake wrote:
> 
> I'm assuming we each get different data files? maybe not? mine has 1169
> lines, first passport entry is:
> 
> iyr:2015 cid:189 ecl:oth byr:1947 hcl:#6c4ab1 eyr:2026
> hgt:174cm
> pid:526744288

Mine starts

ecl:amb
pid:690616023
byr:1994 iyr:2014 hgt:172cm hcl:#c0946f eyr:2022

and has 953 lines.

-- 
Jeff Carter
"[O]ne can look forward to a rapid and widespread
improvement in programming practice, both from
those who use the language [Ada] and from those
who study its concepts and structures."
C. A. R. Hoare
181

recantation

[Bill Findlay]

On 4 Dec 2020, Jeffrey R. Carter wrote (in 
article<rqehuk$9pr$1@dont-email.me>):

> "[O]ne can look forward to a rapid and widespread improvement in programming
> practice, both from those who use the language [Ada] and from those who study
> its concepts and structures."
> C. A. R. Hoare

I have great *personal* appreciation of that declaration!

-- 
Bill Findlay

Re: Advent of Code Day 4

[Stephen Leake]

Stephen Leake <stephen_leake@stephe-leake.org> writes:

> I used gnat.Regpat:
>
>    Field : constant Pattern_Matcher := Compile ("([a-z][a-z][a-z]):([^ ]+)");
>
>    Height : constant Pattern_Matcher := Compile ("([0-9]+)(cm|in)");
>
>    Hair_Color : constant Pattern_Matcher := Compile
> ("#[0-9a-f][0-9a-f][0-9a-f][0-9a-f][0-9a-f][0-9a-f]");
>
>    Eye_Color : constant Pattern_Matcher := Compile ("amb|blu|brn|gry|grn|hzl|oth");
>
>    ID : constant Pattern_Matcher := Compile
> ("[0-9][0-9][0-9][0-9][0-9][0-9][0-9][0-9][0-9]");
>
>
> But apparently I've done something wrong;

I downloaded the solution posted by Gautier de Montmollin, and used that
to debug my code. The problem is the pattern for ID above matches 9
digits followed by anything, and one of the bad passports has a 10 digit
ID. I added a check for length = 9 (gnat regpat doesn't have a pattern
representing "end of string") and got the right answer.

-- 
-- Stephe

Re: Advent of Code Day 4

[Simon Beàn]

On Friday, December 4, 2020 at 3:20:06 PM UTC-5, Maxim Reznik wrote:
> Day 4 task is dull :) 
> 
> https://github.com/reznikmm/ada-howto/blob/advent-2020/md/04/04.md

TIL so much! Thank you!

Re: Advent of Code Day 4

[Jeffrey R. Carter]

On 12/5/20 3:06 AM, Stephen Leake wrote:
> 
> I downloaded the solution posted by Gautier de Montmollin, and used that
> to debug my code. The problem is the pattern for ID above matches 9
> digits followed by anything, and one of the bad passports has a 10 digit
> ID. I added a check for length = 9 (gnat regpat doesn't have a pattern
> representing "end of string") and got the right answer.

Right. Technically every field needs a check that it's followed by a space or 
line end, and in production code there would be other checks I would make, but 
as this is Q&D and the input is otherwise well formed, I left them out.

-- 
Jeff Carter
"Every sperm is sacred."
Monty Python's the Meaning of Life
55

Re: Advent of Code Day 4

[Stephen Leake]

"Jeffrey R. Carter" <spam.jrcarter.not@spam.not.acm.org> writes:

> On 12/5/20 3:06 AM, Stephen Leake wrote:
>> I downloaded the solution posted by Gautier de Montmollin, and used
>> that
>> to debug my code. The problem is the pattern for ID above matches 9
>> digits followed by anything, and one of the bad passports has a 10 digit
>> ID. I added a check for length = 9 (gnat regpat doesn't have a pattern
>> representing "end of string") and got the right answer.
>
> Right. Technically every field needs a check that it's followed by a
> space or line end, and in production code there would be other checks
> I would make, but as this is Q&D and the input is otherwise well
> formed, I left them out.

One lesson here is we should not rely on the examples provided in the
puzzle to be a complete test; we should review them for completeness.

-- 
-- Stephe

Re: Advent of Code Day 4

[John Perry]

On Friday, December 4, 2020 at 2:20:06 PM UTC-6, Maxim Reznik wrote:
> Day 4 task is dull :) 
> 
> https://github.com/reznikmm/ada-howto/blob/advent-2020/md/04/04.md

Flourishes like this:

   return Pasport (byr .. pid) = (byr .. pid => True);

illustrate idioms that I really want to learn, thanks for sharing.