« first day (2214 days earlier)   

3:48 AM
@giuliolunati Do you have a link of just what the scan-barcode.reb file looks like?
@GeekyI The only x86 androids I've seen were some TV boxes. What's yours?
 
4:31 AM
	case A_REMOVE:	// #"a" "abc"  remove/part bs "abcd"  yuk: /part ?
		if (!D_REF(2)) Trap0(RE_MISSING_ARG); // /part required
		if (Set_Bits(VAL_SERIES(value), D_ARG(3), FALSE)) break;
		Trap_Arg(D_ARG(3));
@MarkI or others more invested in the cryptozoology of Rebol than I am... what does that comment mean?
 
 
3 hours later…
7:08 AM
@HostileFork import 'android android scanBarcode _ -- maybe you're more interested in android module
 
@giuliolunati Oh, I wondered how to call it and scan the barcode in a practical program, but thanks for that too. Incidentally, what text editor are you using on the phone?
 
 
2 hours later…
9:32 AM
@giuliolunati Can you also do me a favor and delete all the remote branches on your github that have been merged or taken care of, or no longer of interest to you?
 
10:28 AM
posted on November 22, 2016 by hostilefork

This commit began as an attempt to simplify the %tmp-funcargs.h code and eliminate Find_Refines(). It evolved into a general solution for auto-generating the parameter and refinement lists for natives and actions. So instead of writing: // // foo: native [arg /ref ref-arg] // REBNATIVE(foo) { PARAM(1, arg); REFINE(2, ref); PARAM(3, ref_arg); // blah blah ARG(arg), REF(ref

 
@HostileFork I'm using Vim -- I firmly love it!
 
@giuliolunati would you mind using conceal then, and aliasing your tabs down, but using 4 spaces?
4
Q: Vim conceal indentation. Replace indentation spaces with utf-8 chars

Sri Hari Raju PenmatsaI use four spaces for indent. I would like to replace the indentation with some UTF-8 chars. What I have: 1. def foo(bar): for jaffa in bar: print jaffa 2. var foo = function(bar){ bar.forEach(function(jaffa){ console.log(jaffa); }); } What I want: 1. def foo...

I am trying to eliminate CR and TAB characters from shared files. It would take a special command line switch to accept them: Death to Carriage Return
 
@HostileFork done :-)
 
@giuliolunati If you want your tabs can be just 1 char :-)
 
@HostileFork "done" refers to delete-branches thing...
@HostileFork If I right understand, the file contains 4-space indentation, but Vim show them in a customized fashion? :-)
 
10:45 AM
@giuliolunati Yup. Haven't tried it myself but I know it is out there. onetom was advocating for this as a solution, and that the Internet age is a bad time to be having different kinds of invisible characters in your file.
I think that if we are to "keep our eyes on the prize" in a sense--by doing what other people aren't really doing--it really has to do with remembering what Rebol is supposed to be a Rebellion Against. It's a rebellion against going into your browser, clicking view source, and seeing a gigabyte of JavaScript. Or worse, a half-gigabyte of minified JavaScript.
It is not about complexity management, it is about complexity elimination. So if there is unjustifiable complexity it needs to be excised from the concerns. We know we need spaces in the medium of text, and we know how wide a space is...it is a manageable known currency of exchange.
We know that tab is sneakier, it is a navigation instruction, it is another kind of invisible, it can be used to encode entire programs in other programming languages invisibly: en.wikipedia.org/wiki/Whitespace_(programming_language)
 
 
1 hour later…
12:20 PM
@Brett your native spec in source now enhanced by the above commit, nearly excising integer parameter indexing from the C coder's concern. No need for explicit PARAM() and REFINE() definitions. It's not perfect, but it's hard to imagine getting any traction on improving matters without moving actions away from integer indexing. Closer to Modify With Confidence, eh @ShixinZeng?
 
12:47 PM
posted on November 22, 2016 by hostilefork

Noticing a pattern in problems, Ren-C added the feature that it would be possible for a function FOO with an evaluated argument to tell the difference between being called as FOO [y] vs. FOO SECOND [[x] [y]] or BLK: [y] | FOO BLK. The reason this distinction was important was to give errors like: >> if [x] [print "oops"] ** Script error: Literal block used as conditional [x] ** Wher

 
 
2 hours later…
2:45 PM
REMOVE expects its argument to be "at" the element to remove, but there's no index in a bitset.
So you have to provide the element(s) to remove, and /PART is hijacked to do that, which is yuk for sure.
It is also yuk to force a function to have to have a refinement before it can work.
All part of the natural beauty of the action model ... :/
 

« first day (2214 days earlier)