Can you get a list of OSX fonts somehow?

driven

Isn't fc-list on every mac?

TT... don't you read PM's or the posts... [I am laughing]

reminds me of when you answer some of your 'clients'

The whole reason I went round in circles is fc=list has never been 'standard' on a mac.

a lot of people use to have xcode and it came with that as part of fontconfig for X11,

or if you have Wings3D, Gimp, Inkscape or Imagemagics you'd have it, so a few SU'ers will. but they all have FontBook it's a mac core app....

driven

TO RECAP
This is the fastest I can get a usable list from X11
if I remove .split[-2] from fontPath it uses /usr/local/bin and is twice as fast, same result... so the one could be used for any fc-list installed, without hardcoding a path.

 fontPath=(`find /*/*/bin/fc-list`).split[-2]
fontList=(`#{fontPath}  ; file family | grep \/Library\/Fonts`).split("\n").collect { |f| f.split(";")[1] }.collect { |f| f.split(",")[0].strip}.to_a.uniq.compact.sort[5..-5]

the .sort[5..-5] cleans out the dot files at top and some other cruft the the bottom...
john
EDIT: I missed a .strip that stop them running... leading whitespace.
on test a couple are getting through that shouldn't
*** macFonts group Error! *** i = 38 chunk = 4 font = Bitstream Charter Error #<TypeError: reference to deleted Group> Error: #<TypeError: (eval):282:inname=': reference to deleted Group>
(eval):299
(eval):282
(eval):274:in times' (eval):274

thomthom

@driven said:

TT... don't you read PM's or the posts... [I am laughing]

I've become so confused on this whole topic.

@driven said:

This is the fastest I can get a usable list from X11

fontPath=(find ///bin/fc-list).split[-2]
That command was very slow on my mac.

@driven said:

the .sort[5..-5] cleans out the dot files at top and some other cruft the the bottom...
Can you be sure this is the same on all machines? I'm hesitant to use magic numbers.

OSX is giving me headackes in regard to this plugin. I'm tempted to just drop it all together. Taking too much time.

At the moment I'm looking at first attempting fc-list, if it's installed. (Might have to do a search - so I have to cache that.) Then fall back to AppleScript and FontBook - also slow so it also needs a cache list (meaning the list won't automatically keep in sync.)

sigh

And that's not including the crashing...

driven

TT, maybe you want to move this, but
I was looking for a UFT8 font filter and came across a simple test code for checking console encoding.

I thought Ruby Console was UFT8, but it's not?

I made a little test script, and ran it both from console and from plugins.
same result from both...

>  msg = (`perl -Mcharnames=;full -CS -wle 'print "\N{EURO SIGN}"'`).to_s
 puts msg
 rply = (`locale`).to_s
 puts rply
 result = UI.messagebox msg, MB_YESNO
 if result == 6 # Yes
   UI.messagebox("Sketchup dosen't use UFT8, it uses \n" + rply)
 end
N{EURO SIGN}
LANG=
LC_COLLATE="C"
LC_CTYPE="C"
LC_MESSAGES="C"
LC_MONETARY="C"
LC_NUMERIC="C"
LC_TIME="C"
LC_ALL=
1

for comparison Terminal.app results

%(#008000)[johns_iMac at upstairs in ~
$ locale
LANG="en_US"
LC_COLLATE="en_US.UTF-8"
LC_CTYPE="en_US.UTF-8"
LC_MESSAGES="en_US.UTF-8"
LC_MONETARY="en_US.UTF-8"
LC_NUMERIC="en_US.UTF-8"
LC_TIME="en_US.UTF-8"
LC_ALL="en_US.UTF-8"
johns_iMac at upstairs in ~
$ perl -Mcharnames=:full -CS -wle 'print "\N{EURO SIGN}"'
€]
john

thomthom

It's UTF-8.

Test example:
'ø'.unpack('C*')

Returns:
[195, 184]

Which is the correct byte values in UTF-8 encoding.

195 indicate the Latin1 page 184 points to ø on that page.

Don't know what the data you got from them commands where. But just by looking at the test data byte per byte you can tell it's UTF-8.

driven

yes that does return correctly,
which is why I can't understand why some thing get lost in transition...
same font Stix(NON-Unicde)
Top happens from Ruby Console as well as WebDialogs
Bellow is the Built in Tool
john

thomthom

hm... I've not have any problems with it.

Do you see this in the 3d text plugin I sent you?

driven

@thomthom said:

Do you see this in the 3d text plugin I sent you?

Top one... I was being discrete

` > a = "2撖죺".unpack('C*')
b = a.pack('C*')
c = puts a.inspect
puts b.inspect

[50, 230, 146, 150, 236, 163, 186]
"2撖죺"
nil`
the puts should read [0, 50, 100, 150, 200, 250] "2撖죺"

TIG

Why are you using "C*" in the unpack and pack - doesn't that extract a character as an unsigned integer.
Shouldn't it be "U*" - which extracts UTF-8 characters as unsigned integers ?

thomthom

Because I actually want to see each byte. Not the Unicode ID.

thomthom

@driven said:

the puts should read [0, 50, 100, 150, 200, 250] "2撖죺"

?
Where are these numbers from?

Why are you expecting a NULL byte? (That's usually a string termination in C.)

driven

I think this may be the root of my issues
if you unpack(C), then pack(U) sh*t happens

> a = "2撖죺".unpack('C*') b = a.pack('U*') c = puts a.inspect puts b.inspect [50, 230, 146, 150, 236, 163, 186] "2æì£º" nil
and visa-versa
> a = "2撖죺".unpack('U*') b = a.pack('C*') c = puts a.inspect puts b.inspect [50, 25750, 51450] "2\226\372" nil
I think the first is happening somewhere

driven

font = Arial Unicode MS
SU Top again
WD bottom
look familiar

driven

@thomthom said:

Where are these numbers from?

http://codereview.stackexchange.com/questions/3569/pack-and-unpack-bytes-to-strings

thomthom

@driven said:

@thomthom said:

Where are these numbers from?

http://codereview.stackexchange.com/questions/3569/pack-and-unpack-bytes-to-strings

That's from a questions that doesn't really make sense.

Also:

@unknownuser said:

Seeing as JavaScript has 16-bit strings, I packed two bytes per character.

Two byte per character isn't UTF-8.

driven

I was looking for test code and only grabbed the example, I got there from the Stackoverflow 'fix' that referred back.

Do these work as 3D Text on the PC...

I ran the full gamete of unpack().pack() scenario's in console and a mismatch is the only way to get the same results as the straight input.
john

Dan Rathbun

@thomthom said:

fonts.sort! %(#008000)[# (!) Not UTF-8 compatible! But better than nothing.]

(1) Since this will run only on Mac, which is Unicode aware, can't you pass the list to a command shell and use the shell's built-in sort filter ??

For instance on WIN, in DOS command shell, you can filter output by piping it through the sort filter, thus:
doc_list = %x(dir "~/documents/myproject" | sort)
or similar.

You'd need to build a plain text list from the array, each element being a line, separated by " \n"

(2) Alternative ... build an array copy using pack, sort it, then & unpack back to strings.

I think I'm late posting this... you guys are posting machines!

thomthom

If there is such a command then I'd guess that would work. But I'm not familiar with OSX terminal. Maybe John knows?

I can also use JS to sort it - since I'm displaying the list in a WebDialog.

Dan Rathbun

@thomthom said:

But I'm not familiar with OSX terminal.

Shell Scripting Primer: Command Line Primer

tcsh(1) OS X Manual Page[*]

Filename substitution

If a word contains any of the characters *`', ?', `` [' or {`' or begins with the character ~`' it is a candidate for filename substitution, also known as globbing''. This word is then regarded as a pattern (glob-pattern''), and replaced with an alphabetically sorted list of file names which match the pattern.

Also do a Find on " ls-F", it is a built-in and supposed to be faster than " ls -F"

OS X Manual Page: ls command Reference

driven

This is probably my last effort on this.... circles,
but, I finally figured out find in Ruby Console, escapes, escapes, escapes...

THE THIRD WAY... no Font Book... no fc-list... just plain old find >> mdls... not the fastest, but not too bad.
inculed is Dan's tester, these all created 3D text from console, only Font Book and usr/local/bin/fc-list do that out of the can.

a=(`find /System/Library/Fonts\ /Library/Fonts\ ~/Library/Fonts\ \\( -name "*.ttf" -o -name "*.otf" \\) -type f`).split("\n").map! { |f| f.gsub(" ", "\\ ")} #need this or something to catch spaces in filenames
b=(a.collect{|x| `mdls  -name com_apple_ats_name_family -raw #{x}`}).map! { |f| f.split(",")[0]} #the other items in each array are unicode strings for other languages, if you want those use Font Book
c=b.map { |f| f[/[("\s]+([^"\n]+)[)"\s]+/m,1] }.uniq!.sort![2..-1] #[0] is empty, [1] is a dot file, could remove them
macFonts = c
chunksize = 1
chunk = 1
limit = macFonts.length
model = Sketchup.active_model
 
fsize = 1.0
linespacing = 1.2
 
bold = false
italic = false
thick = 0.05
filled = true
quality = 0.0
 
i = 0
 
while i < limit
 
  begin
    #
    model.start_operation("3D Fontnames (#{chunk})")
      #
      chunksize.times do |n|
        #
        break if i == limit
        #
        item = macFonts[i]
        grp = model.entities.add_group()
        grp.entities.add_3d_text( "#{item}", TextAlignCenter, "#{item}",
          bold, italic, fsize, quality, 0.0, filled, thick )
        grp.name= item
        grp.move!( Geom;;Transformation.new(Geom;;Vector3d.new(0,-(i*linespacing),0)) )
        #
        i += 1
        #
      end # chunk
      #
    model.commit_operation()
    #
  rescue Exception => e
    puts("\n*** macFonts group Error! ***")
    puts("  i = #{i}")
    puts("  chunk = #{chunk}")
    puts("  font = #{macFonts[i]}\n")
    model.abort_operation()
    puts("Error #<#{e.class.name}; #{e.message}>")
    puts(e.backtrace) if $VERBOSE
    raise
  end
 
  chunk += 1
 
end # while