I guess it's only appropriate

As I tried to explain last time, I consider this habit to be a unilateral affair.  Until now, I'd no need to create a decoder.  Out of curiosity, I decided to go ahead and make one; it was easy enough.  To think all this time I hadn't known of base2dec().  I've been doing shit the hard way too many times.  Oh well.  Learning things is good.

?

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22

%% text from unicode pictogram (arbitrary-base encoding) clc; clear all; format compact   instring='▄▆▂▆ ▄█▄▄ ▄▆█▂ ▄▆█▂ ▄█▂█ ▄▆▆▂ ▄▆▆▄ ▄█▄▂'; charmap='▂▄▆█'; % length(charmap)=b space=' '; % thin space   b=length(charmap); for n=0:1:b-1;     instring(findstr(instring,charmap(n+1)))=num2str(n); end   instring(findstr(instring,space))=' '; instring=str2num(instring); % simplifies indexing outstring='';   for n=1:1:length(instring);     outstring=[outstring char(base2dec(num2str(instring(n)),b))]; end   outstring

It works well enough.  I suppose I could make it automagically find a minimal character set from the input pictogram string, but it wouldn't have any way of guessing the likely order.  Granted, even a human would have to guess if the charmap used in encoding weren't conspicuously ordered.