thoth@reef.cis.ufl.edu (Gilligan) (10/05/90)
Is there any way to specify the endiannes of the data you are
unpacking? I am writing a perl script to decode Moria save files (for
unspecified purposes :) and unfortunately for me the data is written
in Intel order (LSB first). I can not use unpack (as far as I know)
on the data from our Suns (MSB first). If I could get unpack to
perform the byte shuffling for me it would squish my code by a factor
of two.
Does anyone know how to do this?
Could it be as simple as
read(STDIN, $_, 2*2);
($exp, $maxexp) = unpack("s-2", $_);
Also, what is the filehandle for <>, I hate having to type
"moriainterp.perl < moria.dc".
--
--
"Until it's on my desk, it's vaporware" (`it' is the NeXT)lwall@jpl-devvax.JPL.NASA.GOV (Larry Wall) (10/05/90)
In article <THOTH.90Oct4162237@reef.cis.ufl.edu> thoth@reef.cis.ufl.edu (Gilligan) writes: : : Is there any way to specify the endiannes of the data you are : unpacking? I am writing a perl script to decode Moria save files (for : unspecified purposes :) and unfortunately for me the data is written : in Intel order (LSB first). I can not use unpack (as far as I know) : on the data from our Suns (MSB first). If I could get unpack to : perform the byte shuffling for me it would squish my code by a factor : of two. : Does anyone know how to do this? : : Could it be as simple as : : read(STDIN, $_, 2*2); : ($exp, $maxexp) = unpack("s-2", $_); There's no way to do this with unpack on a big-endian machine at the moment. The best you can do, as far as I can see, is to say sub swab { $_[0] =~ s/([\0-\377])([\0-\377])/$2$1/g; } read(STDIN, $_, 2*2); &swab($_); ($exp, $maxexp) = unpack("s2", $_); : Also, what is the filehandle for <>, I hate having to type : "moriainterp.perl < moria.dc". It's ARGV, but you can't do a read() on it and keep the other semantics of <>. You want something more like open(STDIN,shift). Larry
cameron@usage.csd.oz (Cameron Simpson,foo) (10/06/90)
From article <9821@jpl-devvax.JPL.NASA.GOV>, by lwall@jpl-devvax.JPL.NASA.GOV (Larry Wall): | In article <THOTH.90Oct4162237@reef.cis.ufl.edu> thoth@reef.cis.ufl.edu (Gilligan) writes: | :Is there any way to specify the endiannes of the data you are unpacking? | | There's no way to do this with unpack on a big-endian machine at the moment. | The best you can do, as far as I can see, is to say | sub swab { | $_[0] =~ s/([\0-\377])([\0-\377])/$2$1/g; | } [...] Um, why can't I write sub swab { $_[0] =~ s/(.)(.)/$2$1/g; } - Cameron Simpson, cameron@spectrum.cs.unsw.oz.au
lwall@jpl-devvax.JPL.NASA.GOV (Larry Wall) (10/07/90)
In article <885@usage.csd.unsw.oz.au> cameron@spectrum.cs.unsw.oz.au (Cameron Simpson) writes:
: From article <9821@jpl-devvax.JPL.NASA.GOV>, by lwall@jpl-devvax.JPL.NASA.GOV (Larry Wall):
: | sub swab {
: | $_[0] =~ s/([\0-\377])([\0-\377])/$2$1/g;
: | }
: [...]
:
: Um, why can't I write
: sub swab {
: $_[0] =~ s/(.)(.)/$2$1/g;
: }
Because . doesn't match \n. [\0-\377] is the most efficient way to match
everything currently. Maybe \e should match everything.
And \E would of course match nothing. :-)
Larrycameron@usage.csd.oz (Cameron Simpson,foo) (10/07/90)
From article <9847@jpl-devvax.JPL.NASA.GOV>, by lwall@jpl-devvax.JPL.NASA.GOV (Larry Wall): | In article <885@usage.csd.unsw.oz.au> cameron@spectrum.cs.unsw.oz.au (Cameron Simpson) writes: | : From article <9821@jpl-devvax.JPL.NASA.GOV>, by lwall@jpl-devvax.JPL.NASA.GOV (Larry Wall): | : | $_[0] =~ s/([\0-\377])([\0-\377])/$2$1/g; | : | : Um, why can't I write | : $_[0] =~ s/(.)(.)/$2$1/g; | | Because . doesn't match \n. Goes back and Rs the FM more closely. Oh the embarrassment. | [\0-\377] is the most efficient way to match | everything currently. Maybe \e should match everything. Nah. What's wrong with [^]? | And \E would of course match nothing. :-) Correspondingly, []. Haven't tried it yet (does so). Ouch. The ] at the start of the pattern is taken as part of the range. A feature. Well, it was a nice idea... - Cameron Simpson cameron@spectrum.cs.unsw.oz.au "To every problem there is a simple, obvious, wrong solution."
brnstnd@kramden.acf.nyu.edu (Dan Bernstein) (10/08/90)
In article <887@usage.csd.unsw.oz.au> cameron@spectrum.cs.unsw.oz.au (Cameron Simpson) writes: > From article <9847@jpl-devvax.JPL.NASA.GOV>, by lwall@jpl-devvax.JPL.NASA.GOV (Larry Wall): > | [\0-\377] is the most efficient way to match > | everything currently. Maybe \e should match everything. > Nah. What's wrong with [^]? How do you specify ] in a [^...] expression, then? You mention the same problem with []... ---Dan
worley@compass.com (Dale Worley) (10/08/90)
X-Name: Cameron Simpson,foo
Um, why can't I write
sub swab {
$_[0] =~ s/(.)(.)/$2$1/g;
}
Basically, because it doesn't work. This swaps every pair of bytes,
but doesn't swap the byte pairs, etc. The goal is to turn ABCD into
DCBA.
Dale Worley Compass, Inc. worley@compass.com
--
The great unsolved problem of feminsm is to decide whether we need to
make men more like women, or women more like men.dale@convex.com (Dale Lancaster) (10/10/90)
In <9847@jpl-devvax.JPL.NASA.GOV> lwall@jpl-devvax.JPL.NASA.GOV (Larry Wall) writes: >Because . doesn't match \n. [\0-\377] is the most efficient way to match >everything currently. Maybe \e should match everything. >And \E would of course match nothing. :-) >Larry I would rather see \* match everything and \e be ESC as it is in other utilities. And maybe \!* matches nothing? :-) dml
cimarron@erewhon.postgres.Berkeley.EDU (Cimarron D. Taylor </>) (10/10/90)
| From: dale@convex.com (Dale Lancaster) | Newsgroups: comp.lang.perl | Subject: Re: unpack and endianness | Date: 9 Oct 90 17:20:43 GMT | | I would rather see \* match everything and \e be ESC as it is in | other utilities. And maybe \!* matches nothing? :-) | | dml But doesn't \* already mean "turn off the special meaning of *". i.e. match only an asterix? Cimarron Taylor Electronics Research Laboratory / POSTGRES project University of California, Berkeley cimarron@postgres.berkeley.edu
mdb@ESD.3Com.COM (Mark D. Baushke) (10/10/90)
On 9 Oct 90 17:20:43 GMT, dale@convex.com (Dale Lancaster) said: Dale> In <9847@jpl-devvax.JPL.NASA.GOV> lwall@jpl-devvax.JPL.NASA.GOV Dale> (Larry Wall) writes: >Because . doesn't match \n. [\0-\377] is the most efficient way to match >everything currently. Maybe \e should match everything. >And \E would of course match nothing. :-) >Larry Dale> I would rather see \* match everything and \e be ESC as it is in Dale> other utilities. And maybe \!* matches nothing? :-) Dale> dml Please do not create any non-alphanumeric metacharacters. It would break the quoting of a pattern that might contain metacharacters using $pattern =~ s/(\W)/\\$1/g; -- Mark D. Baushke mdb@ESD.3Com.COM
lwall@jpl-devvax.JPL.NASA.GOV (Larry Wall) (10/10/90)
In article <MDB.90Oct10082856@kosciusko.ESD.3Com.COM> mdb@ESD.3Com.COM (Mark D. Baushke) writes: : On 9 Oct 90 17:20:43 GMT, dale@convex.com (Dale Lancaster) said: : : Dale> In <9847@jpl-devvax.JPL.NASA.GOV> lwall@jpl-devvax.JPL.NASA.GOV : Dale> (Larry Wall) writes: : : >Because . doesn't match \n. [\0-\377] is the most efficient way to match : >everything currently. Maybe \e should match everything. : : >And \E would of course match nothing. :-) : : >Larry : : Dale> I would rather see \* match everything and \e be ESC as it is in : Dale> other utilities. And maybe \!* matches nothing? :-) : : Dale> dml : : Please do not create any non-alphanumeric metacharacters. It would : break the quoting of a pattern that might contain metacharacters using : $pattern =~ s/(\W)/\\$1/g; 'Sides, there's millions of scripts out there that already use \* to mean a literal *. And \e doesn't mean ESC to me, it means \. What utilities does it mean ESC in? /etc/termcap uses \E, which is close. I HAVE considered making \a mean \007 (since K&R2 has it), but there's gotta be a limit somewhere. [Strange, that never stopped you before.] Oh, shaddap! Larry
epeterso@houligan.encore.com (Eric Peterson) (10/11/90)
lwall@jpl-devvax.JPL.NASA.GOV (Larry Wall) writes: | And \e doesn't mean ESC to me, it means \. What utilities does it mean | ESC in? /etc/termcap uses \E, which is close. Well, GNU Emacs does this, if you can call something its size a "utility" :-) Eric -- Eric Peterson <> epeterson@encore.com <> uunet!encore!epeterson Encore Computer Corp. * Ft. Lauderdale, Florida * (305) 587-2900 x 5208 Why did Constantinople get the works? Gung'f abobql'f ohfvarff ohg gur Ghexf.
dan@kfw.COM (Dan Mick) (10/12/90)
In article <CIMARRON.90Oct10011141@erewhon.postgres.Berkeley.EDU> cimarron@erewhon.postgres.Berkeley.EDU (Cimarron D. Taylor </>) writes: > But doesn't \* already mean "turn off the special meaning of *". > i.e. match only an asterix? What's an asterix? Why would you want perl to match a cartoon character?