« first day (452 days earlier)      last day (2042 days later) » 

1:07 AM
@Dennis Can you pull ecpp when you have a minute (and update the C/C++ versions if necessary)? Thanks!
 
@MDXF The old Hello World uses vanilla C and works just fine. I meant the warnings I pasted. None of these gcda files exist on the server.
 
@Dennis Oh, hmm... What is TIO's gcc version?
 
@MDXF gcc version 7.2.1 20170915 (Red Hat 7.2.1-2) (GCC)
 
1:44 AM
@MDXF The problem seems to be that the gcda files are generated when running ecpp. The sandbox isn't allowed to do that.
What are those, anyway?
 
They're coverage files for codecov.io. I'm trying to find a way to keep them out of TIO's Makefile without too much hassle.
And I'm stupid. I can make codecov do make codecov.
Just a minute, fixing that...
Or better yet, CFLAGS=-coverage. Pushed, @Dennis can you pull please?
 
@MDXF @HyperNeutrino @cairdcoinheringaahing @ATaco Done.
 
@Dennis Thanks! I'm guessing it works now, then?
 
@Pavel Workstation or Server?
 
yay thanks :D
 
2:01 AM
 
2:17 AM
@Dennis Sweet! Thanks :)
Oh, and I just remembered, there's an easier way to use print "Hello, World!", without all of that #rule and #def print +++ nonsense.
#replace print puts. ಠ_ಠ
 
 
1 hour later…
3:35 AM
@MDXF How does #replace differ from #define
 
@Pavel #def matches multiple tokens regardless of names (with the exception of operators). So #def a*b mul(a,b) will match x*y.
#replace also matches tokens (well, only one) but only cares about the name. #replace print puts will replace print with puts, but not foo.
#def print puts will attempt to replace every instance of a variable with itself.
And seg fault.
Which is on purpose. :P
 
@MDXF Ah, so you hijacked the default #define behaviour.
ಠ_ಠ
 
Oh oops, you asked how #replace differs from #define, not #def.
#replace can replace everything. #define can replace variable/function names.
 
By that do you mean it works in strings?
That doesn't seem all that useful.
 
No, it doesn't work in strings. It replaces entire tokens.
 
3:44 AM
Oh, so like #replace 0 1 is a thing that works?
 
You can #replace << < to mess with someone, for example.
@Pavel Yes.
And #replace "badword" "goodword" is a thing. You can also use it to quickly replace tokens in a search file without preprocessing everything.
 
neat
 
For example, if you want to change every instance of MACRO_FOO to MACRO_BAR, you could put #replace MACRO_FOO MACRO_BAR at the top of a file, and it will automatically replace it for you.
(It comes with a built-in prettyprinter too)
 
 
14 hours later…
5:28 PM
@Dennis when I press the Escape key when on TIO, it comes up with a message saying "You've pressed the Escape key and entered command mode. Press Escape again to exit". What is command mode?
(I'm on Firefox on a Mac if that changes anything)
 
@cairdcoinheringaahing It enables hotkeys for all the buttons so you don't have to click on them
 
@cairdcoinheringaahing For instance Esc followed by R runs the program, Esc followed by S takes you to the permalinks, Esc followed by H takes you to the Home page.
 
6:10 PM
Does anyone know if there's a way to change the input encoding format on TIO? In case it matters, I'm specifically concerned about Befunge, which defaults to using the UTF-8 encoding.
Given the 2D nature of the language, a variable length encoding makes no sense, but I couldn't see any way to change it.
 
6:45 PM
@Dennis could you please pull Ohm v2? also, while you're at it, could you change the repository URL from github.com/MiningPotatoes/Ohm to github.com/nickbclifford/Ohm ? thanks
 
@JamesHolderness You ask Dennis to change the encoding to one that makes more sense, however the Befunge interpreter probably reads the code as UTF-8 anyway.
 
@Pavel Most Befunge interpreters (including all of the ones on TIO) will interpret the source as one byte per cell, so a multibyte character will end up using multiple cells, thus messing up the alignment of any other code that follows it.
 
7:00 PM
@JamesHolderness The interpreter reads UTF-8 and behaves in an (admitedly akward) way. It doesn't read it as some other encoding.
 
7:44 PM
@Pavel That doesn't make any sense. The source doesn't have an inherent encoding - it's just a sequence of bytes on disk. The encoding is determined entirely by the editor that's viewing it, or the interpreter that is executing it.
I know of only one Befunge implementation that will interpret a source file as UTF-8, and it's not one of the ones on TIO.
 
¯\_(ツ)_/¯
 

« first day (452 days earlier)      last day (2042 days later) »