Help With $Read And ""s In Files

I am trying to read a text file that contains font entries from the registry from mIRC. The registry entries use quotation marks which mIRC seems to ignore. Any file I attempt to read with ""s in the various lines, it seems to just ignore those lines.

My font scanner uses a run command to export the current Windows installed fonts. Then I wanted to read that file to build a list. However, the read will not read much of the file, I am assuming due to the quote marks.

alias testfonts {
  echo -at Loading Fonts from the Windows registry..
  GetFonts
  .timer 1 2 LoadFonts
}

alias -l mirc.fontfile { return $shortfn($mircdir) $+ fonts.dat }

alias GetFonts {
  run -n reg export "HKEY_LOCAL_MACHINE\SOFTWARE\Microsoft\Windows NT\CurrentVersion\Fonts" $mirc.fontfile
}

alias LoadFonts {
  var %i $lines($mirc.fontfile), %r 0 
  while (%r <= %i) {
    echo -at %r - $read($mirc.fontfile, %r)
    inc %r
  }
}

Using /testfonts gets me nothing but a bunch of blank lines. I have used /fopen and $fread to no avail. Is there a way to read this file without using binary?

PS - I am assuming it is the quote marks because when i read mirc.ini, it skips lines with them.

Have you used something like example: $noqt(whatever) ?

 

I tried just now.. no bueno.. it must be a mirc file processing thing.

when you view the output in notepad it seems to output in a line by line way but if you use a better editor you will see it outputs the registry file in a different format. That could be why you are unable to get the fonts in a readable manner.

seems most are just using HKEY_CURRENT_USER\Software\Microsoft\Windows\CurrentVersion\Explorer\Shell Folders\\Fonts to get the systems font folder path and doing a findfile on .TFF in that folder in which they save and read from.

This is a very basic way to do it. Note you will need to improve it to meet your needs and this assumes your font folder is c windows font

alias testfonts {
  write -c fonts.txt
  var %getfonts $file($findfile(C:\WINDOWS\Fonts,*.ttf,0,write fonts.txt $+($replace($truetype($1-).fullname,$chr(32),\<img src='http://www.tg007.net/forum/public/style_emoticons/<#EMO_DIR#>/cool.gif' class='bbc_emoticon' alt='B)' />,%getfonts)))
}
; Credits to Kamek for these 2 aliases - http://www.mircscripts.org/
alias truetype {
  if (!$isfile($1)) { return }
  var %fn = $iif(("*" iswm $1), $1, $+(", $1, ")), %ntables, %i = 1, %p, %namepos, %namelen, %nid = 1
  if ($findtok(copyright family subfamily id fullname version postscript trademark manufacturer designer - urlvendor urldesigner, $prop, 32)) { %nid = $calc($ifmatch - 1) }
  bread %fn 0 8192 &font
  if ($bvar(&font, 1, 4) != 0 1 0 0) { return }
  %ntables = $bvar(&font, 5).nword
  while (%i <= %ntables) {
    %p = $calc(13 + (%i - 1) * 16)
    if (%p > 8192) { return }
    if ($bvar(&font, %p, 4).text === name) { %namepos = $bvar(&font, $calc(%p + 8)).nlong | %namelen = $bvar(&font, $calc(%p + 12)).nlong | break }
    inc %i
  }
  if (!%namepos) { return }
  if (%namelen > 8192) { %namelen = 8192 }
  bread %fn %namepos %namelen &font
  var %nrecs = $bvar(&font, 3).nword, %storepos = $calc(%namepos + $bvar(&font, 5).nword), %i = 1
  while (%i <= %nrecs) {
    %p = $calc(7 + (%i - 1) * 12)
    if ($bvar(&font, %p).nword = 3) && ($bvar(&font, $calc(%p + 6)).nword = %nid) {
      var %len = $bvar(&font, $calc(%p + 8)).nword, %peid = $bvar(&font, $calc(%p + 2)).nword
      bread %fn $calc(%storepos + $bvar(&font, $calc(%p + 10)).nword) %len &font
      return $uni2ansi($bvar(&font, 1, %len))
    }
    inc %i
  }
}

alias uni2ansi {
  var %unicode = $1, %i = 1, %t = $numtok(%unicode, 32), %s = i, %c
  while (%i <= %t) {
    %c = $gettok(%unicode, $+(%i, -, $calc(%i + 2)), 32)
    if ($gettok(%c, 1, 32) = 0) { %c = $chr($gettok(%c, 2, 32)) }
    else { %c = ? }
    %s = $left(%s, -1) $+ %c $+ i
    inc %i 2
  }
  return $left(%s, -1)
}

DCX has a very easy font dialog. Only take's a line of code and the dcx.dll.

I'm stuck between a rock and a hard place; somewhere between using findfile for a slow laggy method and having a customizable dialog. My objective is to be able to find all system fonts as fast as REG EXPORT works while making a custom dialog.

So are you saying that the registry output file is stored in unicode or with a binary header? I will look into that.

The DCX method has some of the features I want. But if I make something, it should have what I want. Good option though.

i went back and included a snippet. it may not be perfect but seems to do ok.

I'm stuck between a rock and a hard place; somewhere between using findfile for a slow laggy method and having a customizable dialog. My objective is to be able to find all system fonts as fast as REG EXPORT works while making a custom dialog.

 

So are you saying that the registry output file is stored in unicode or with a binary header? I will look into that.

 

The DCX method has some of the features I want. But if I make something, it should have what I want. Good option though.

 

You should check out the /filter command then. It's amazingly fast and stable.

Or write to a hidden window or hash table.

Fixed ini file handling bug when reading section/item/data that are enclosed in quotes. New mirc 7 , maybe that's what you're looking for .

Well it was the header of the file and the method in which it was stored. I could not get anything to read the file as is.

So since I was already running a command, I made a batch file to TYPE registry_export.file > clean.file after the registry export. Not so pretty, but it works.