His Deeds Are Dust

I have a Samsung X120 notebook. Commiserate with me, please, for ever since F13′s kernel went from 2.6.33 to 2.6.34, I’ve been without ACPI. Anything other than acpi=off in the boot settings produced a giant stack trace that scrolled off the tiny screen so fast and so early in the boot that only videoing the screen would have enabled me to transcribe it. Even boot_delay didn’t work, because that won’t delay every line of output when the kernel is curling up in a corner and dying.

It’s a bug in the Samsung BIOS no doubt, and it has an upstream kernel report, but the bisect points to a sane commit that’s good for everyone else.

Running without ACPI isn’t a realistic option on a laptop. I wouldn’t know when to plug it back into the mains, and shutting down requires rather more persuasion than should be necessary, so last night it was time to patch the kernel.

Fedora Wiki to the rescue. I followed the instructions on building a custom kernel to the letter, and eleven hours later, I had a functioning laptop. I’ll break that down, lest you think me hard of understanding: five minutes to read the page, 20 minutes to grab the kernel src.rpm and make the appropriate patch, and 10 hours and 35 minutes to compile the entire thing on the laptop. I went to bed for the last bit, after ensuring that the mains was plugged in and on so that my laptop wouldn’t do the same.

So, to everyone who has had a hand in that page on the wiki, thank you.

I’ve just been looking at packaging a font for Fedora, and have found that there is no separate licence file, though the font metadata contains the full licence text. I should ask upstream to put a copy of the licence in their archive, the next time they do a release, but what if they aren’t interested? How would I extract the licence text?

The simplest starting point seems to be TTX, which can take a TrueType font and convert some or all of it to an XML document. From then, anything that can parse XML or plain text should help. I’ll just extract the TrueType “name” table:

$ ttx -t name font.ttf

This creates a file called font.ttx. Annoying, I can’t send the output to stdout, and I can’t ask that it overwrites any old file of the same name; it’ll just insist on creating font.ttx, font#1.ttx, font#2.ttx and so on.

If we take a look at font.ttx, we can see the field I’m interested in, in a namerecord element with attribute nameID of 13:

<namerecord nameID="13" platformID="1" platEncID="0" langID="0x0">
       This font software is copyright (c) 2007, Frixxon Wanglewurx. All rights reserved.&#13;&#10;"Frixxon Muck" is a Reserved ...
</namerecord>

The entire licence is extracted as one line in the XML output, with carriage returns and line feeds encoded as 
 and 
, respectively. There may be more than one copy of this licence, with different platformID values.

To go from here to a text file with no more than 80 characters per line and LF endings, I could either use sed or xsltproc. Let’s take a look at how both would work. Here’s getlicence.sed:

/<namerecord nameID="13"/ {
  n
  s/^ \+//
  s/&#13;//g
  s/&#9;/\t/g
  s/&#10;/\n/g
  s/&amp;/\&/g
  p
  q
}

Which reads as: find the namerecord, skip to the next line, strip leading spaces, convert some XML entities to text, print it and quit.

I’d invoke this with:

$ sed -nf getlicence.sed font.ttx | fold -s > LICENSE.txt

Yes, despite using the British spelling of licence normally, I’ll use license for packaging.

The XML entity conversion is a bit of a faff which a proper XML tool could take care of for me. Here’s getlicence.xsl:

<xsl:stylesheet version="1.0" xmlns:xsl="http://www.w3.org/1999/XSL/Transform">
<xsl:output method="text"/>
<xsl:template match='/'>
  <xsl:value-of select="//namerecord[@nameID='13'][position()=1]"/>
</xsl:template>
</xsl:stylesheet>

I’d use this with:

$ xsltproc getlicence.xsl font.ttx | dos2unix | fold -s > LICENSE.txt

There are plenty more tiny hacks that TTX and XSLT can help with. I haven’t yet tried many round trip conversions with TTX, but if I lose my way in FontForge’s Font Information menus and dialogs, I could be tempted.

I still have a VT100 terminal, but it’s in storage. I figured I could pretend that it was on my desk if I made a font that looked like the old beast, including the gaps between scan lines.

Once I’d started, I needed the reverse video form of it, and the forms correctly underlined, and double width, and double height and double width. Blinking is more problematic

The challenge with the double height, double width font is that the VT100 had escape sequences that made particular character rows either display just the top half of characters, or the bottom half. To display characters double height, you had to put the same characters on two consecutive rows, and set the line attributes correctly. In fact, you could make up some funky alien characters by setting the line attributes and typing different characters on each row.

In order to reproduce this effect (yes, I’m carrying on digging), I had to make an upper half font, and a lower half font. And that gives me a problem with Fontconfig, because that takes a look through fonts as it caches them, and marks glyphs as broken if they don’t make any marks, but they are encoded at positions which aren’t space characters in Unicode. My “upper half” font doesn’t have any marks for the underscore glyph, and my “lower half” font doesn’t have any marks for the double quotes, for example.

There isn’t a way of telling Fontconfig that my font isn’t broken just because it didn’t fancy making any marks for a particular glyph, so I’ll have to do some pre-processing when generating “screen shots” of my old terminal. I can locally configure Fontconfig to not do this, but that wouldn’t help anyone else.