fran barton @ludictechnologies.eurosky.social · 25/08/2026They bag these loaves up the wrong way round don't they? I've only just realised this has been bothering me for a long while. When you open the bag via the twisty end you want to get to the big slices first, not the crust. Something something #bigendian #amirite #uknews 210 2
fran barton @ludictechnologies.eurosky.social · 10/07/2026Spain's defence also asking "Where's de kettle here?!" 130 3
fran barton @ludictechnologies.eurosky.social · 26/02/2026Everyone in Britain knows Gorton and Denton 000 0
fran barton @ludictechnologies.eurosky.social · 08/02/2026It made me think of the meme (I'm sure I'm not the first Sacred Harp singer to do this) 110 1
fran barton @ludictechnologies.eurosky.social · 01/02/2026A new regular #SacredHarp singing is born! We've got a small group together in #Stroud, Gloucestershire, meeting to sing on the first Friday of the month, starting this Friday at the Baptist Hall on Union St., GL5 2HA. I can't wait. All welcome. Email sing@stroudwater.info to find out more. 140 4
fran barton @ludictechnologies.eurosky.social · 08/12/2025Day 7 done. I made very heavy weather of part 2 as I was skimping on fully understanding the logic required to get the counts right. Happy to have solved another day by reducing over (imho) fairly neat functions that encapsulate and clarify the logic required. 030 3
fran barton @ludictechnologies.eurosky.social · 06/12/2025Another enjoyable solve. I was happy I could use a couple of steps from part 1 again in part 2. They just had slightly different wrangles in the middle part. I was happy to get to use `pop!` and a splat - the onomatopoeic side of coding... (As a kid I loved that scene in Star Wars, scary and gross!) 110 1
fran barton @ludictechnologies.eurosky.social · 05/12/2025#AdventOfCode Day 5: Cafeteria I ended up with nice, functional solutions that I felt I couldn't have written much more elegantly. Happy. But it did take me a while to get the logic for part 2 right. Today was the first time I have experimented with named tuples in Julia. Code on my Codeberg repo. 110 1
fran barton @ludictechnologies.eurosky.social · 05/12/2025Day 4. I had used #JuliaLang 's CartesianIndices function in previous years but this time I wanted to base my whole solution around it. I felt pleased with my solution but also feel like I just go at problems in a very direct, obvious way sometimes, lacking the incisive imagination I see from others 120 2
fran barton @ludictechnologies.eurosky.social · 03/12/2025But verily my code, though logically correct, was not actually viable for part 2 within a reasonable span of time. So I wrote a fresh function with a different logic. And this was in turn backwards-applicable to part 1, so I'm just sharing that function in the code snapshot here. 100 1
fran barton @ludictechnologies.eurosky.social · 01/12/2025Decided to do a code snapshot with codesnapshot.com as that always looks nice - and my Codeberg link above didn't get a visual. Code in the alt text too. 000 0
fran barton @ludictechnologies.eurosky.social · 24/11/2025Ooh didn't know about this. Looks lovely. Despite the info on their site... 010 1
fran barton @ludictechnologies.eurosky.social · 20/09/2025After the celiidh, setting the chairs back out ready for the morning. We've had a wonderful day singing from the green book! 131 3
fran barton @ludictechnologies.eurosky.social · 19/06/2025I find myself very unsatisfied with the "either India... or Australia" construction here. It makes it sound like England turn up at the start of each match not entirely sure who they are going to find they're playing against. I think the standard was set by the maths in the first sentence. #EngVInd 010 1
fran barton @ludictechnologies.eurosky.social · 09/12/2024Latest adventures in "how long will it take me to get to Eastville in #Bristol on a Saturday morning by public transport"? It's 36km away, and a 45 minute drive. Or, on great British public transport, 4.5 hours there and back including at least 8 trains and buses. 000 0
fran barton @ludictechnologies.eurosky.social · 18/09/2024Bluesky now has over 10 million users, and I was #9,283,937! 040 4
![### Day 7: Laboratories
array = [ .==(x, "^") for x in split.(filter(contains("^"), input), "") ];
# part 1 --------------------------
process_arr = function(x::@NamedTuple{last::BitVector, t::Int64}, y::BitVector)
# take all the last beams, count the splits (clashes with y)
# remove the splits and also keep any beams that didn't hit a split (misses)
splits = x[:last] .& y # beams split
misses = x[:last] .& .~y # beams continue down
process_splits(x::BitVector) = (x >> 1 .| x << 1)
new_beams = process_splits(splits) .| misses
(last = new_beams, t = x[:t] + sum(splits))
end;
# We can pass in array[1] as init array, since we know from inspection that
# there is just 1 beam (from S), which is above the only splitter in array[1]
reduce(process_arr, array, init = (last = array[1], t = 0))[:t]
# part 2 --------------------------
process_arr = function(last::Vector{Int64}, y::BitVector)
x = .>(last, 0) # inherited beams
splits = [ x[i] & y[i] ? last[i] : 0 for i in eachindex(last) ]
misses = [ x[i] & ~y[i] ? last[i] : 0 for i in eachindex(last) ]
# circshift would be risky if we had data at the ends of our input grid, but
# we can see from inspection that isn't going to happen till the last step.
new_routes = [ circshift(splits, 1), circshift(splits, -1), misses ]
+(new_routes...)
end;
sum(reduce(process_arr, array, init = Int.(array[1])))](https://cdn.bsky.app/img/feed_thumbnail/plain/did:plc:ggaksd42j4ome3yqtatmvdrp/bafkreiacqv2xsbo7kvlaluc5gm2mrsjzgpasfm73k3rgwl5bhkkarvffxe)
![### Day 6: Trash Compactor
# part 1 --------------------------
rows = split.(strip.(input), r"\s+");
ops = string.(filter(in(["+", "*"]), pop!(rows)));
nums = [ parse.(Int, v) for v in eachrow(reduce(hcat, nums)) ];
process_vec(op::String, vec::Vector{Int64}) = op == "+" ? +(vec...) : *(vec...);
mapreduce(process_vec, +, ops, nums)
# part 2 --------------------------
rows = reverse.(split.(input, "")); # split to each character
ops = string.(filter(in(["+", "*"]), pop!(rows))); # same as p1 but now reversed
slices = join.(eachrow(reduce(hcat, rows)));
# Blocks of numbers are not equal size (some only have 2 or 3 numbers, not the
# full 4), so we gotta split 'em up by the empty (" ") columns, then get rid
# of the empties.
strs = split.(split(join(replace(slices, " " => "-"), "|"), "|-|"), "|");
nums = [ parse.(Int, v) for v in strs ];
mapreduce(process_vec, +, ops, nums)](https://cdn.bsky.app/img/feed_thumbnail/plain/did:plc:ggaksd42j4ome3yqtatmvdrp/bafkreibutmcvnab2en5b3zzlj7oswvamtnicgqd53a72drqutynqfgkrk4)
![### Day 5: Cafeteria
input = readlines("input/day05.txt");
ranges = filter(x -> occursin("-", x), input);
ingrids = parse.(Int, filter(x -> occursin(r"^\d+$", x), input));
to_range(x::String) = range(parse.(Int, split(x, "-"))...);
ranges = sort!(to_range.(ranges), by = minimum);
# part 1 --------------------------
sum(map(y -> any(x -> in(y, x), ranges), ingrids))
# part 2 --------------------------
get_new_max = function(x::@NamedTuple{t::Int64, mx::Int64}, y::UnitRange{Int64})
mx = maximum([x[:mx], maximum(y)])
len = mx - maximum([minimum(y) - 1, x[:mx]])
(t = x[:t] + len, mx = mx)
end;
reduce(get_new_max, ranges, init = (t = 0, mx = 0))[:t]](https://cdn.bsky.app/img/feed_thumbnail/plain/did:plc:ggaksd42j4ome3yqtatmvdrp/bafkreicwdty6rmvpegl7ssmfdl4ipxyuoy2fn6psgxjedkpibd4twlxw5a)
![### Day 4: Printing Department
# part 1 --------------------------
check_accessible = function (ind::CartesianIndex{2}, mat::Matrix{Char})
surround_rect = CartesianIndices((-1:1:1, -1:1:1))
surrounding_inds = map(x -> ind .+ x, surround_rect[setdiff(1:9, 5)])
real_inds = filter(x -> checkbounds(Bool, mat, x), surrounding_inds)
rolls_n = length(filter(.==('@'), mat[real_inds]))
mat[ind] == '@' && rolls_n < 4
end;
count_rolls = function (mat::Matrix{Char})
sum(map(x -> check_accessible(x, mat), CartesianIndices(mat)))
end;
count_rolls(prepare_input(input))
# part 2 --------------------------
count_removed = function (mat::Matrix{Char}, removed::Int64 = 0)
inds = CartesianIndices(mat)
to_remove = findall(x -> check_accessible(x, mat), inds)
if length(to_remove) == 0
removed
else
removed += length(to_remove)
mat_values = map(x -> (x in to_remove ? '.' : mat[x]), inds)
mat = reshape(mat_values, size(mat))
count_removed(mat, removed)
end
end;
count_removed(prepare_input(input))](https://cdn.bsky.app/img/feed_thumbnail/plain/did:plc:ggaksd42j4ome3yqtatmvdrp/bafkreiavcjbitbx4bwyouvmyrujjvnkku7gw35par4qi7pvddvepws7cdq)
![### Day 3: Lobby
# part 2 --------------------------
using .Iterators: take, drop, peel
reduce_to_max = function (line::String, len::Int64 = 12)
set = collect(take(line, len))
rst = drop(line, len)
rng = 1:(len+1)
while length(collect(rst)) > 0
stt, rst = peel(rst)
append!(set, stt)
pms = map(x -> join(set[setdiff(rng, x)]), rng)
set = only.(split(string(first(findmax(parse.(Int, pms)))), ""))
end
parse(Int, join(set))
end;
sum(reduce_to_max.(input))
# and this is backwards-compatible with part 1 as well, of course:
sum(map(x -> reduce_to_max(x, 2), input))](https://cdn.bsky.app/img/feed_thumbnail/plain/did:plc:ggaksd42j4ome3yqtatmvdrp/bafkreigdfdyypwmk6lphwidmb43olj6w4wlll55aph2j32dkg2ljvetakm)
![### Day 1: Secret Entrance
# part 1 --------------------------
parse_input(input) = parse.(Int, replace.(input, "L" => "-", "R" => ""));
create_ends(input) = accumulate(+, pushfirst!(parse_input(input), 50));
get_password(input::Vector{String}) = sum(.==(.%(create_ends(input), 100), 0));
# part 2 --------------------------
count_zeroes = function (a::Int, b::Int)
seq = b >= a ? ((a+1):b) : (b:(a-1))
sum(.==(.%(seq, 100), 0))
end;
using .Iterators: drop;
get_password2 = function (input::Vector{String})
ends = create_ends(input)
sum([count_zeroes(ends[i-1], ends[i]) for i in drop(eachindex(ends), 1)])
end;](https://cdn.bsky.app/img/feed_thumbnail/plain/did:plc:ggaksd42j4ome3yqtatmvdrp/bafkreielcbuv6bwimwfdxp4ecue32hfz6jzsva57z2m5no3jyvunhqvpfu)
