Apparently, the dead are owed another war. But they are not. What they are owed is a good, stand-up, bruising row over the fate of America; just who determines it and for what end?
I take back what I said yesterday about this country having the world's best music: That piece "The Last Full Measure of Devotion," sung by some airbag soprano with the U.S. Army band at the Pentagon, was what P.G. Wodehouse characters call the frozen limit. All this stuff, moving as some of it is—the reading of names, for instance—brings together some noxious tendencies. One is atmospheric overkill: As I type this, for instance, a string quartet is playing "Amazing Grace" while a man and woman take turns reading. The names alone, among distant city sounds, would have done the trick. I don't mean to sound like a fucking esthete, but whoever planned this was working in accordance with an esthetic too. It reminds me of the original coverage, when TV news—see, I'm not entirely abstemious—would put dramatic music under footage of the towers collapsing. Got to keep the customers entertained.
It's a distinction -- nationalism vs. patriotism -- worth remembering. The country, it seems, is preparing for war at a significant moment in the formation of cultural memory about Sept. 11: Private grief is beginning to ease, and politicians are increasingly comfortable with drawing larger, public lessons from 9/11. The temptation to demand nationalist sentiments from people comfortable only with patriotic ones is a recurring theme in American history, and it's a temptation greatly increased in times of war.
Date: Wed, 11 Sep 2002 09:11:25 -0400 (EDT)
From: Aaron Straup Cope
To: Benjamin Trott
Subject: Re: MT - two unrelated thingies
> > I mention all of this because it seems like it would be a nifty bit of
> > featuritis for Movable Type. Actually, I mention it because it would be
> > nice to have some kind of universal glossary system and MT is rapidly
> > approaching universal status (congrats!)
>
> I agree that a glossary system would be pretty cool. Though, it's quite
> doable currently just using the MT-Macro plugin. Have you taken a look
> at that?
Ah, I see. No I hadn't. That's cool - I gather I can do something like :
<MTApplyGlossary>
<$MTEntryBody$>
</MTApplyGlossary>
MT::Template::Context->add_container_tag(MTApplyGlossary => sub {
my $ctx = shift;
my $args = shift;
my $path = $args->{path'};
if (! $path) {
my $cfg = ... " # Remember how to get MT config manager here
$path = $cfg->{GlossaryPath} ||
return $ctx->error($ctx->errstr);
}
#
my $builder = $ctx->stash('builder');
my $tokens = $ctx->stash('tokens');
my $output = "";
my $writer = XML::SAX::Writer->new(Output=>$output);
my $glossary = XML::Filter::Glossary->new(Handler=>$writer);
my $parser = XML::SAX::ParserFactory->parser(Handler=>$glossary);
# Would also need to $glossary->no_do_startend_document_methods();
$glossary->set_glossary($path);
eval { $parser->parse_string($builder->build($ctx, $tokens)); };
if ($@) {
return $ctx->error($ctx->errstr.": $@");
}
return $output;
}
Which will almost certainly break because it is very possible that the
data passed to the plugin will not be well-formed. I suppose I could just
export the code for parsing double-quoted keywords as a package function.
I'm about to add support for <my_ns:glossary id = "some keyword with
spaces" /> per a request which I'm not keen to parse with regex(p)s since
you can also do <my_ns:keyword />.
I suppose the whole thing could be hacked together using HTML::Parser to
do the parsing since it is very forgiving. Well, it's something to work
with anyway.
Question : When a plugin is called, has the FH associated with the
document already been opened or are you just building a string? If there
is a FH, is there any way that it could be passed to the plugin?
I suppose not since that would make managing nested tags/plugins
impossible. Alas.
In the unsolicited advice department, I would only mention the docs for
plugins are less than inviting. I've noticed that other people who've
written plugins have posted code which is helpful, since you can sort of
infer what 'foo' does and how to get 'bar' from them.
But, if plugins are really more exciting that just returning the value of
system calls it isn't readily apparent how.
107 ->perldoc MT::Template::Context
No documentation found for "MT::Template::Context".