Let's analyze the str_replace parameters one by one.
range('-',T)
The range() function returns an array that has the elements spanning from the first parameter to the second parameter. Characters are considered by their ASCII values, so the result is
Array
(
[0] => -
[1] => .
[2] => /
[3] => 0
[4] => 1
[5] => 2
[6] => 3
[7] => 4
[8] => 5
[9] => 6
[10] => 7
[11] => 8
[12] => 9
[13] => :
[14] => ;
[15] => <
[16] => =
[17] => >
[18] => ?
[19] => @
[20] => A
[21] => B
[22] => C
[23] => D
[24] => E
[25] => F
[26] => G
[27] => H
[28] => I
[29] => J
[30] => K
[31] => L
[32] => M
[33] => N
[34] => O
[35] => P
[36] => Q
[37] => R
[38] => S
[39] => T
)
Why T instead of "T"? PHP has a misfeature that makes it evaluate undefined constants as strings with the same content as the constant's name. The constant T isn't defined so it's the same as "T" which saves two characters for code golf purposes. The same goes for q later. If the server has error reporting on, it will show a warning about an undefined constant.
split(q,"I justCannaLE?2Gotta >u=Msta=.q...");
This splits the string into an array at the q characters. Again, this makes for shorter code than using an array literal. The result:
Array
(
[0] =>
I justCannaLE?2Gotta >u=Msta=.
[1] =>
Ng1Nlet? downNrun<rH=5desMt?N>cryNsayRoodbyeNtE< lie5hurt?
[2] =>
We'T3n each@Jor s8lSg6r hear9<ch: but6;Lo7hyL7BInsideCe both3Cha9Ro: S
We3KeRa45we;QplB
[3] => 1)O)NgiT, nPgiT
(G
[4] => iT? up
[5] => howFJeel:
[6] => know
[7] => me
[8] => <=
[9] =>
YH
[10] => 8s
[11] => o
[12] => t's been
[13] => ing
[14] => 're
[15] => a
[16] => nd
[17] => make?
[18] => yH
[19] => othM
[20] => A
[21] => ay it
[22] => w
[23] => D
[24] => ell
[25] => I'm
[26] => G
[27] => ou
[28] => I
[29] => f
[30] => Lh
[31] => t
[32] => er
[33] =>
NP
[34] =>
(Ooh
[35] => eTrQ
[36] => RSna
[37] => g
[38] => on
[39] => ve
)
The final parameter is the target string.
"We; n7trangMsL8loT63Ke rules5s8d8I
AJull commit4nt'sChatFKink: of6CHldn'tRetKisJrom<ny@Ruy-/A= if?<sk 42DS'tLE 4?;Lo8bli=L7ee..
O,R1)O,R001)/-.."
If you pass arrays to str_replace() as needle and haystack, the replacement is done one at a time. For simplicity's sake, let's take just "We; n7trangMs" as the taget string and start replacing from ;. The first step after replacing "7" with "8s" (the corresponding replacement in the second array):
"We; n8strangMs"
Then replace "8" with "o "
"We; no strangMs"
";" with "'re"
"We're no strangMs"
"M" with "er"
"We're no strangers"
In short, it's a basic compression algorithm where you find character sequences that repeat inside the original text and replace them with a single character. When decompressing that character is replaced with the original sequence. By running the progress iteratively you can compress the once-compressed text again ("o s" => "8s" => "7").
