r3wp [groups: 83 posts: 189283]
  • Home
  • Script library
  • AltME Archive
  • Mailing list
  • Articles Index
  • Site search
 

World: r3wp

[Core] Discuss core issues

Gregg
16-Jul-2009
[14235]
Yes, that's a known issue.
Graham
20-Jul-2009
[14236x3]
What happened to the open/async ?
as per http://www.rebol.net/docs/async-examples.html
I read somewhere that if you do any other IO during async, that will 
break the async.  True??
Maarten
20-Jul-2009
[14239]
Within an async handler things might get unpredictable. If you have 
a normal event loop, including async in the wait-list, I think you 
should be good.
Graham
20-Jul-2009
[14240x3]
I was wondering how to download a 120Mb file from S3
and wanting to do it async while write/binary/append in the async 
handler
was open/async removed after 2.5.5 ??
Maarten
20-Jul-2009
[14243x2]
Yes.
It interfered with the garbage collector and was one of the reasons 
to start R3 development iirc (the whole port subsystem, actually).
Pekr
20-Jul-2009
[14245x3]
Graham - you don't need the async mode in order to behave in an async 
like manner ...
I can send you short script called "multiserver", which is able to 
accept connections from various IPs, save content to file, it simply 
multiplexes on opened ports ....
the script is very short, commented. It served as example for my 
friend learning REBOL.
Graham
20-Jul-2009
[14248x2]
That's okay Pekr ... I was just trying to see if I could use the 
style as written by Carl.
Maarten ... when did that become common knowledge??  I had endless 
issues trying to upload files async, and it only worked by turning 
off GC  .... but of course killed my program with huge memory use 
:(
Ashley
20-Jul-2009
[14250x2]
Anyone have a nifty func to "untrim" a string? Something that converts:

	"ArialBold"

to:

	"Arial Bold"

Best I can come up with is a horrible replace loop:

	foreach char "ABC ..." [
		replace/all next string char join " " char
	]

 replace/all string "  " " " ; handle case where words were already 
 space seperated
replace/all next -> replace/all/case next
Graham
20-Jul-2009
[14252]
you want to insert spaces before each caps ?
Ashley
20-Jul-2009
[14253]
Yep
Graham
20-Jul-2009
[14254]
can't you use parse
Ashley
20-Jul-2009
[14255]
One of the few things I've yet to master in REBOL! ;)
Graham
20-Jul-2009
[14256x3]
caps: charset [ #"A" - #"Z" ]
non-caps: complement caps

parse "ArialBold" [ some [ copy fontname name (print fontname)  ]]
>> parse "ArialBold" [ some [ copy fontname name (repend out [ fontname 
" " ])  ]]
== true
>> out
== "Arial Bold "
close enuf ??
Ashley
20-Jul-2009
[14259]
** Script Error: name has no value
Graham
20-Jul-2009
[14260x2]
name: [ caps some non-caps ]
out: ""
cut and paste works poorly in windows :(
Ashley
20-Jul-2009
[14262x2]
and Mac ...
Thanks, that's enough for me to start with.
Graham
20-Jul-2009
[14264]
>> parse next s: "ArialBoldItalic" [ some [ some non-caps [ end | 
mark: ( insert mark #" " ) skip ]] ]
== true
>> s
== "Arial Bold Italic"
Ashley
20-Jul-2009
[14265]
Even better!
Graham
20-Jul-2009
[14266]
I'm waiting now for Ladislav's answer !
Ashley
20-Jul-2009
[14267]
In 12 bytes or less ;)
Pekr
20-Jul-2009
[14268]
not sure if correct, but:

caps: charset [ #"A" - #"Z" ]
non-caps: complement caps
s: "aaaArialBoldItalic"

parse/all s [any [mark: caps (insert mark #" ") skip | skip] end]
Ashley
20-Jul-2009
[14269]
Slight complexity, "HelveticaCY" has to parse as "Helvetica CY" not 
"Helvetica C Y"
Graham
20-Jul-2009
[14270x4]
>> parse next s: "ArialBoldItalicYY" [ some [ some non-caps [ end 
| mark: ( insert mark #" " ) 2 skip ]] ]
== true
>> s
== "Arial Bold Italic YY"
>> parse/all s: "abcArialBoldItalicsCY" [some [mark: caps (insert 
mark #" ") 2 skip | skip] end]
== true
>> s
== "abc Arial Bold Italics CY"
I think you should have 'some and not 'any as there should always 
be at least one space to be inserted.
Pekr's rule looks shorter because you don't need the non-caps
Ashley
20-Jul-2009
[14274]
'some works. The 4 variations it has to handle are:

	"Arial"
	"ArialBold"
	"ArialBoldItalic"
	"ArialCY"

so putting a 'next prior to 's seems to handle all these.
Pekr
20-Jul-2009
[14275x2]
'any will work too ...
putting 'next there serves the purpose, but is an ugly hack :-)
Graham
20-Jul-2009
[14277]
why is it a hack??
Pekr
20-Jul-2009
[14278]
you should solve it by parse and parse only - that is the challenge 
:-)
Graham
20-Jul-2009
[14279x2]
we know we only have to parse part of the string ... not all of it
I find it easier to bypass edge conditions rather than program for 
them :)
Ashley
20-Jul-2009
[14281x2]
I'm with G on this one.
Here's the finished code (which obtains REBOL compatable font names 
under Mac):

	fonts: copy []
	caps: make bitset! [#"A" - #"Z"]

 foreach file compose [(read %/System/Library/Fonts/) (read %/Library/Fonts/)] 
 [
		if %.dfont = suffix? file [
			s: form first parse file "."

   parse next s [any [mark: caps (insert mark #" ") 2 skip | skip] end]
			insert tail fonts s
		]
	]
	remove-each font-name fonts: sort unique fonts [

  (size-text make face [text: "A" font: make face/font [name: font-name 
  size: 10]]) =

  size-text make face [text: "A" font: make face/font [name: font-name 
  size: 12]]
	]


(the windows func to do this is http://www.reboltech.com/library/scripts/get-fonts-windows.r
)
Pekr
20-Jul-2009
[14283]
there's no edge condition. What if the name would not begin with 
a capital letter? I would parse all string and instead of 'next I 
would use 'trim, which would tream initial space, in case first letter 
is capital :-) But if it is the rule, that the first letter is always 
being a capital, then your solution is absolutly correct ...
Ashley
20-Jul-2009
[14284]
First letter is always a capital ... except where it's a # but those 
fonts are not usable by REBOL anyway.