Nightmare

Just had a nightmare. I don't remember much, but I remember that there was a woman, and she was some sort of time paradox. If I were to encounter her, it would somehow erase my own history--I'd never have existed.

I woke up, and she was standing in my room, talking to me.

That's when I actually woke up, because I think I was moaning and crying out.

What do you do when even your nightmares are stupid?

Defeating locks by 'bumping'

This is probably one of the most interesting things I've read in a while. It's about defeating consumer-grade locks with a method called 'bumping', which I guess is somewhat similar to picking, but supposedly faster and more reliable. Assuming the accuracy of the article, it's scary how easily these locks can be nondestructively defeated.

http://www.engadget.com/2006/08/24/the-lockdown-locked-but-not-secure-part-i/

At any rate, I learned a lot about how locks work.

OSCON day 5

Short day today (just two morning sessions after the keynotes). Damian Conway gave a very funny keynote about patents. He's a really good speaker.

My first session was "Extreme Perl Makeover" by Peter Scott. This was another 'best practices'-type talk. The speaker suggested a couple of useful-sounding Perl modules: Text::Outdent sounds good for doing HERE documents without having to move them over into column 1, and Inline::Files lets you have several separate data sections after __END__.

Next was "Perl Hacks You Never Knew Existed" by chromatic. Just about everything he said went right over my head. But his talk got me interested in the Attribute::Handlers and Attribute::Method modules, which implement the subroutine attributes that I've seen in Catalyst and Class::Std. I managed to overcome the urge to get chromatic to sign my new copy of Perl Testing: A Developer's Notebook.

OSCON day 4

One of this morning's keynote speaker was a fellow named Jason Scott. He made a documentary about the online bulletin board systems of the 1980s. His Web site is www.textfiles.com.

Some pretty cool sessions today. The first was "Subversion Best Practices" by Ben Collins-Sussman and Brian Fitzpatrick (both from google). A lot of what they said I'd gotten from the documentation. But they talked about a couple of features I hadn't heard of: autoversioning and autoprops. And they described a neat trick involving making the document root of a Web site be a working copy of a project, with a hook such that the working copy is updated whenever there's a change in the repository.

Next was "Low-Maintenance Perl" by Perrin Harkins. Most of his talk was along the lines of "don't do this in your code". Most of the things he discouraged were things I hadn't heard of or knew to be generally 'bad'. So that was somewhat encouraging. Damian Conway and Larry Wall were both in the audience--if this made the speaker nervous, he didn't show it.

After lunch (which was provided by OSCON [Aramark]), I went to "SQL Outer Joins for Fun and Profit" by Bill Karwin. He solved several interesting problems with outer joins. He used a syntax that I hadn't really seen before: he put row-elimination statements in the ON clause, stuff I'd only ever seen in WHERE clauses. He pointed out that the WHERE clause isn't evaluated until after the join, so it's often quite beneficial to eliminate rows in the ON clause (fewer Cartesian products that way).

Next was "Writing Maintainable Code with PHP" by Laura Thomson from OmniTI. I actually found her point of view a little puzzling. She said that, in general, she doesn't like frameworks, database abstraction layers, or templating engines. As I've come to rely pretty heavily on the latter two and am interested in starting to use the former (I have yet to find a PHP framework that I don't hate), I have trouble understanding how code which doesn't use any of those components is more maintainable than code which does delegate those tasks. But overall I thought it was a good talk, and she made lots of good arguments for creating a set of coding guidelines for your organization (how variables are named, how code is indented, documentation templates, lots of other stuff).

Then there was "Understanding ZFramework" by John Coggeshall. I confess that I sort of zoned out after I learned that the Zend Framework requires PHP 5 (I'm kind of stuck with PHP 4). However, it looks like ZF has a nice input validation component written by Chris Schiflett.

Finally, I went to "The Conway Channel 2006" with Damian Conway. He talked about a couple of modules he's been working on: List::Maker and Contextual::Return. C::R looks particularly cool. It does what wantarray() does, but also distinguishes between the different possibilities in scalar context (a string, a number, a hashref, an undef, etc). The module has lots of neat features and is very flexible.

Powell's bookstore was one of the exhibitors/vendors, and they were offering a 35% discount. So I bought a copy of Perl Testing: A Developer's Notebook. It's ordinarily around $30, and I got it for around $20.

And I talked to my high school buddy for around an hour. Sure enough, he's getting married. It was really good talking to him. I hope I'll be able to go to the wedding.

OSCON day 3

Started the day by learning that my workstation at work probably has a bad hard drive. When my officemate rebooted it, he saw those two magic words...

kernel panic

Oh, well. I've (probably) got good backups.

I attended several sessions today. The first was about compiling a kernel to improve speed (only the drivers you need) and security (so a cracker can't load kernel modules--the speaker advocated a monolithic kernel, if possible). The speaker (Steve Suehring) mentioned a security-related patch called grsecurity, which sounds interesting. I wonder how it compares to the openwall kernel patch (hmmm, guess that's just for 2.4).

Next was "Maximum Netfilter" by Michael Rash of Solirix. He talked about several netfilter-related programs. fwknop does something called single-packet authentication, which is a more secure (albeit less convenient) version of port-knocking.

Then I went to "The Madness of AJAX" by Andrew van der Stock (it was about AJAX security). That was actually a little scary. Not because of anything that I've coded or something a co-worker has coded (I don't feel the need to run home and rewrite a bunch of AJAX), but the speaker did several demonstrations which were just spooky. Looks like several of the PHP AJAX toolkits have significant input validation problems, which are a little too reminiscent of register_globals. I'd like to buy a book on the subject, but there don't really seem to be any books on AJAX security (too new, I guess).

I finished out the day by attending the Perl lightening talks (a bunch of 5-minute presentations). It was sort of a strange potluck, punctuated by a rather bizarre performace called "A Perl module installation in 5 unnatural acts". But it gave me a few things I'll want to read about later: App::Ack (source code searches), Perl::Critic ('use strict' on methamphentimines), and stubmail.com (a re-implementation of SMTP by the SPF guy).

And I got voicemail from an old high school buddy of mine. Haven't heard from him in nearly 5 years. I suspect that he's renewing contact to send me a wedding invitation (good for him, if that's the case [good for him, in any case]). Maybe we'll be able to stay in touch this time.

OSCON Day 2

Another couple of tutorials today. The first was "Advanced Perl DBI" by Tim Bunce. He showed up 15 minutes late, which was fairly annoying. But it was otherwise a good session. I learned lots of interesting and useful things. For example, fetchrow_arrayref() is faster than fetchrow_hashref(). And not a little bit faster, but several times faster. I like the convenience of fetchrow_hashref(), so that I can reference column values by name, rather than array index. But I learned that I can get much the same effect (and much more efficiently) using fetchrow_arrayref() and bind_columns().

I also learned that the DBI has built-in profiling capabilities, and prepare_cached() can be used in place of global statement handles.

The afternoon session was about testing Web applications. Part of the tutorial was about Grinder, a free load-testing tool (apparently, most such tools cost obscene sums of money). It looks pretty hard to use, but might be helpful. But most of the session was about a unit-testing tool called Selenium, which is also free. By contrast, it looks pretty easy to use. I downloaded it and ran the default test suite in Firefox. Two of the tests failed: they dealt with popup windows, which Firefox blocks by default. I then disabled popup blocking and re-ran the suite, and all the tests passed. Pretty cool.

Looks like my workstation at work has gone crazy. The kernel thinks that all the filesystems are read-only. I can't even reboot the thing remotely--I'll have to get my officemate to do it manually tomorrow. I never had this problem with Slackware, but I've now seen it three times with RedHats (twice on Fedora and once on CentOS). I'm wondering if it's LVM. I've never actually really used LVM features (like resizing a partition), so maybe I should go back to normal partitions.

The restaurant here in the hotel has an amber ale on draft called Drop Top. It's yummy.

OSCON day 1

I'm at OSCON this week. I flew here yesterday (without incident). There were at least three other OSCON'ers on the DFW->PDX flight. The goodies aren't as cool this year: the bag is a canvas tote (we got nifty backpacks last year), and there were no really fun toys inside. Oh, well. I registered early, so I got the _AJAX Hacks_ book.

The wireless access is pretty spotty. I remember it being more reliable last year. Maybe it'll improve as the week goes on. The internet connection in my hotel room didn't work yesterday afternoon, but it's good now (it's how I'm doing this).

My room is a lot smaller this year (which is fine). But it's a nice view--I can see Mt. Hood. We actually flew right past Mt. Hood on the way in yesterday (I was surprised by how close the plane flew to the mountain).

This morning's tutorial was MySQL optimization. It was very informative. The speaker took the approach of optimizing queries over server configuration tuning (although he talked about that, too). Lots of good information. Looks like there's a possible InnoDB replacement engine coming out later this year called Falcon--that's good news, in case Oracle changes the InnoDB licensing or something (that came up in a question this morning--the speaker [Jay Pipes] said that it's "business as usual" for two years).

This afternoon was 'Higher-Order Perl' by Mark-Jason Dominus. He's really funny. The first half of the talk was really good (lots of the second half went right over my head). He talked about iterators, which was very interesting (good alternative to File::Find). He started his talk (which he gave barefoot) by throwing two chairs off the podium.

Perl's Class::Accessor

If you find yourself writing a Perl module for a class with a large number of attributes and a correspondingly large number of accessors and mutators ('getters' and 'setters'), you might want to try Class::Accessor. With Class::Accessor as a base class, your module can just declare its attributes, and Class::Accessor will generate all the accessors and mutators (and the constructor) automatically.

(Your Perl distribution may not include Class::Accessor, in which case you'd need to install it from CPAN.)

Here's an example of how you might use Class::Accessor:

package MyPerson;
use base qw/ Class::Accessor /;
MyPerson->mk_accessors(
  qw/
    fname
    mname
    lname
    birthdate
    address
  /
);

1;

Then in your Perl program:

use MyPerson;

my $person = MyPerson->new(
  {
    fname     => 'Homer',
    mname     => 'Jay',
    lname     => 'Simpson',
    birthdate => '1 January 1970',
  }
);

my $birthdate = $person->birthdate(); # $birthdate is now '1 January 1970'
$person->address('742 Evergreen Terrace, Springfield'); # sets the address attribute

Note that you didn't have to explicitly write a constructor--your code just gives new() a hashref of the attributes you want to set.

You can now even use MyPerson as a base class. Another package can inherit from MyPerson and add its own attributes:

package MyCustomer;
use base qw/ MyPerson /;
MyCustomer->mk_accessors(
  qw/
    billing_address
    shipping_address
  /
);

1;

The documentation for Class::Accessor is worth reading--it offers ways to declare read-only and write-only attributes (the former would only have accessors, and the latter would only have mutators).

Lightswitches are evil

I really hate standard lightswitches.

Evidently, I have this irrational, deep-seated aversion to touching lightswitches. Dunno why. But when I want to turn one on or off, I just naturally swing at it with the intention of barely and briefly touching it with my fingertip.

Trouble is, I usually miss. Which angers me. So I swing at it again, only harder, to punish it for evading me. So, of course, I miss again, which angers me more. This continues until I typically end up hitting the lightswitch with a shoe or something.

If I had any sense, or if I were a normal person, I'd just calmly and sanely sweep my whole hand down (or up) over the lightswitch. But that apparently entails more physical contact than my neurosis can withstand.

Maybe I need one of these: http://www.gogglemarks.net/index.php?action=display&tag=fightswitch

Recovering Grammer Snob

I just read June Casagrande's _Grammer Snobs Are Great Big Meanies_ (Penguin, 2006). It's a guide to avoiding common grammatical errors and to dealing with people who think they know more about the English language than they actually do.

The book is awesome: very helpful, and very, very funny. It's only about $14 retail, and worth every bit. I borrowed it from the library, but I'm going to buy a copy. (The book has a website at http://www.grammarsnobs.com/.)

Here are a few notes (to myself) that I jotted down while reading the book.

'correct' (perhaps not unanimously):
e-mail
website
preventive (not preventative)
'entitled' for rights/privileges, 'titled' for the name of a book

The _Chicago Manual of Style_ (used for books) recommends using the Oxford comma. The _AP Stylebook_ (used for newspapers [and, arguably, blogs]) recommends omitting the Oxford comma. (The Oxford comma is the second comma in the following sentence: This blog is simultaneously pedagogical, pedantic, and pedestrian.)

Newspapers tend to use _Webster's New World College Dictionary_ (too bad for colleges of the old world), and books tend to use the _Merriam-Webster Collegiate Dictionary_.

A screed is a long discourse. And a cool word.

Use 'which' for nonrestrictve clauses, and 'that' for restrictive clauses:

1. This blog, which has a silly and misleading title, will be useful and/or interesting to a very small number of readers (on a good day).
2. The _Best of Blondie_ CD that I bought last night is really bitchin'.
   

'prurient' means interesting (even lurid). 'purient' (to my dismay) does not appear to be a word at all. (I blame dog food manufacturers.)

Web 2.0 URLs w/ Apache

(Yes, the term 'Web 2.0' is ridiculously overhyped.)

Let's say you have a Web application which takes two optional GET variables v1 and v2. Let's assume that v1 is always an integer and that v2 is always a character string. We'll also assume that a DirectoryIndex index.php directive has been applied:

http://www.example.com/somedir/index.php?v1=27&v2=hello

And let's say your application has an administrative interface in an 'admin' subdirectory:

http://www.example.com/somedir/admin/admin.php

Try putting the following in your Apache configuration (this assumes that mod_rewrite is enabled):

RewriteEngine on
RewriteBase /somedir
RewriteRule ^admin - [L]
RewriteRule ^(\d+)/([a-z]+)/?$ /somedir/index.php?v1=$1&v2=$2 [L]
RewriteRule ^([a-z]+)/(\d+)/?$ /somedir/index.php?v1=$2&v2=$1 [L]
RewriteRule ^(\d+)/?$ /somedir/index.php?v1=$1 [L]
RewriteRule ^([a-z]+)/?$ /somedir/index.php?v2=$1 [L]

(You could put this in an .htaccess file if your Apache configuration applies an AllowOverride FileInfo directive to this part of your content area.)

You application should now be accessible by any of the following URLs:

http://www.example.com/somedir/hello/27/
http://www.example.com/somedir/27/hello/
http://www.example.com/somedir/hello/
http://www.example.com/somedir/27/
http://www.example.com/somedir/

The first and second URLs should set both variables, the third URL should set v2 only, the fourth should set v1 only, and the fifth should set neither (the fifth URL should still run the application because of the DirectoryIndex index.php directive).

Of course, it doesn't have to be '27' and 'hello'--it can be '42' and 'foobar', or whatever.

And your admin interfaces should still be accessible as before (the dash in the 'admin' RewriteRule means 'no alteration').

In the example I've used PHP as the hypothetical Web application, but the language doesn't matter--this is all Apache magic. But if you want to test this with some PHP code, try this in index.php:

header('Content-type: text/plain');
$v1 = '';
$v2 = '';

if ( isset($_GET['v1']) && $_GET['v1'] ) {
   $v1 = $_GET['v1'];
   printf("v1 is %s\n", $_GET['v1']);
}
else {
   printf("v1 not set\n");
}

if ( isset($_GET['v2']) && $_GET['v2'] ) {
   $v2 = $_GET['v2'];
   printf("v2 is %s\n", $_GET['v2']);
}
else {
   printf("v2 not set\n");
}

_X-Men: The Last Stand_

I got to see a preview screening of the new X-Men movie last night. It was awesome. Overall I think I enjoyed the first two films more. I think they tried to do too much and introduce too many characters in this film. But it was still awesome. I even liked Kelsey Grammer as Beast (I didn't think I would).

Make sure you stay until the bitter end. Sit through the credits, and don't leave the theater until the projector turns off.

Perl programming in joe

I do a lot of Perl programming, and I use joe (Joe's Own Editor). I've defined a few macros that I found pretty useful when scripting. Just add the following macro definitions to your .joerc file (do a text search in .joerc for 'Macros:', and add the lines there-ish):

:def comment filt,"sed -e \'s/^/#/g\'",rtn,tomarkb,markk,prevpos
:def uncomment filt,"sed -e \'s/^#//g\'",rtn,tomarkb,markk,prevpos
:def perltidy bof,markb,eof,markk,filt,"perltidy",rtn,bof,markk

comment ^K 1
uncomment ^K 2
perltidy ^K 3

The first macro comments out a section of text by putting a '#' at the beginning of each line of the section. To use it, just go the the beginning of the section you want to comment out and do 'Ctrl-K b', move to the end of the section and do 'Ctrl-K k', and then do 'Ctrl-1'. It'll comment out those lines, disengage the block (so that the section of text will no longer be highlighted), and return the cursor to the previous position (probably the end of the section of text).

The second macro does just the opposite by uncommenting a section of text.

(You could probably replace the '#' in the macro definitions with a pair of forward slashes for programming in PHP or JavaScript.)

The third macro applies the 'perltidy' program to the entire file (not just a highlighed section). Of course, you'll need perltidy installed for this to work. It's in the Fedora Core extras, and the project home page is http://perltidy.sourceforge.net/.

the Golden Ratio and the Fibonacci series (Perl program)

I've been re-reading Dan Brown's The Da Vinci Code lately, and I was intrigued by the Golden Ratio and how it relates to the Fibonacci series. The Fibonacci series is such that each term is the sum of the previous two terms:
1, 1, 2, 3, 5, 8, 13, ...

The ratio of consecutive terms in the Fibonacci series converges on the Golden Ratio, which has a value of approximately 1.618. It also shows up in a geometric analysis of the pentacle. These properties are discussed in mathematical detail on the following Web pages:

the Golden Ratio
pentacle geometry

I wanted to see this from a computational point of view, so I wrote a quick Perl program to compute the Fibonacci numbers and ratios out to around 100 terms. The program prints out deviations from the Golden Ratio, and the output shows a near-zero deviation after less than 40 terms.

#!/usr/bin/perl -w

use strict;
use diagnostics;

my $NUM_ITERATIONS = 100;
my $GOLDEN_RATIO = 0.5 * (1.0 + sqrt(5));

my ($previous_term, $current_term) = (1, 1);
my $iteration_number = 1;
while ( $iteration_number <= $NUM_ITERATIONS ) {
    my $new_term = $previous_term + $current_term;
    my $ratio = $new_term / $current_term;
    my $deviation = abs($GOLDEN_RATIO-$ratio) / $GOLDEN_RATIO;
    printf "% 3d: %e\n", $iteration_number, $deviation;
    $previous_term = $current_term;
    $current_term = $new_term;
    $iteration_number++;
}

seeing 'svn diff' output while editing the commit log in joe

I use subversion and I use the text editor joe (http://sourceforge.net/projects/joe-editor/). I've got the following in my ~/.bashrc file, so that joe is the editor that's launched when I type 'svn commit':

export VISUAL="/usr/bin/joe"

(export SVN_EDITOR="/usr/bin/joe" also works, if you've already got VISUAL set to something else.)

I like being able to look at the output of 'svn diff' while I edit the commit log entry, and I've found two ways of doing this (I'm still figuring out which I like better).

The first way is to redirect 'svn diff' to a file, edit the file (putting the log entry in the file), remove the 'svn diff' output from the file, and then use that file as the log entry when committing:

$ svn diff > ~/commit.txt
$ joe ~/commit.txt
(compose the log entry at the top of the file,
    and delete the 'svn diff' output)
$ svn commit --file ~/commit.txt
$ rm ~/commit.txt

The second way is to pull the 'svn diff' output into the editor below the 'This line, and those below, will be ignored' line. Then I just compose the log entry at the top of the file and then quit-and-save. In fact, this can be done with a joe macro. Try adding this to the 'Macros' section of ~/.joerc:

:def svndiff eof,insf,"!svn diff",rtn,bof
svndiff ^K 4

Then, next time you want to commit, just type 'svn commit' and do a '^K 4' in joe. That'll add the 'svn diff' output at the bottom of the file and return the cursor to the top of the file.

The second approach requires fewer keystrokes. The only drawback I've noticed is that if you decide you want to abort the commit and you've already saved your edits, you need to be sure to delete the svn-commit.tmp file before exiting joe, or the commit will proceed with whatever's saved in the svn-commit.tmp file.

grabbing HTTP headers

Sometimes it's useful to inspect the HTTP response headers from a Webserver (for example, to know if the Webserver is running Apache, IIS, or something else). Most Web browsers have some sort of 'Page Info' feature which will display the response headers. But it's often more convenient to do it from the command line.

The text-only browser lynx has a nice feature for this. Typing the following command will dump the HTTP response headers to the screen without displaying the content of the www.example.com homepage:

lynx -dump -head http://www.example.com/

wget can also do this. wget ordinarily downloads the Web content to a local file without displaying response headers. The following will show the headers and discard the content (-S displays the headers, and -O diverts the output here to /dev/null):

wget -S -O /dev/null http://www.example.com/

The curl utility can do much the same thing (note that this is a lower-case o to specify the output destination, and there a bare hyphen after the -D, indicating that the headers should be written to stdout):

curl -D - -o /dev/null http://www.example.com/

netcat offers a fourth way of getting the headers by allowing you to hurl a custom HTTP request at port 80 on the Webserver:

printf "HEAD / HTTP/1.0\n\n" | nc www.example.com 80

The previous example would only work for HTTP. For HTTPS, you can do a similar trick using the s_client mode for the openssl utility (this example uses an HTTP v1.1 request, which requires a host request header):

printf "HEAD / HTTP/1.1\nhost: www.example.com\n\n" \
| openssl s_client -ign_eof -connect www.example.com:443

The -ign_eof keeps the connection open so that the s_client will see the printf output: this also requires manually closing the connection (Control-C should do it). Additionally, you may get certificate verification errors from openssl. If so, try specifying your system's certificate authority bundle (which contains the public keys of a list of trusted certificate authorities, and which may be in a different location that this example):

printf "HEAD / HTTP/1.1\nhost: www.example.com\n\n" \
| openssl s_client -ign_eof \
-CAfile /etc/pki/tls/certs/ca-bundle.crt \
-connect www.example.com:443 

sorting files by modification date

At times I've found that I need to sort the contents of a directory by modification date. If all the files are in a single directory, 'ls -lt' will do the trick. But it's not so easy if the files are scattered through an arbitrarily complicated directory structure. So I've saved the following shell script in ~/bin/file_epoch.sh (and made it executable w/ 'chmod 700 ~/bin/file_epoch.sh'):

#!/bin/bash
file_name="$1"
mod_time="`/bin/ls --full-time \"${file_name}\" \
    | awk '{ print $6,$7; }' | cut -d'.' -f1`"
epoch="`date -d \"${mod_time}\" +%Y%m%d%H%M%S`"
echo "${epoch} ${file_name}"

The script takes a single filename as its argument, and the output is something like this:

20060324152639 /path/to/some/file

Then when I need to sort the files, I just do this:

find /path/to/directory/structure -type f \
    -exec ~/bin/file_epoch.sh {} \; | sort -rn

This runs the script on every file in /path/to/directory/structure, and piping the output to 'sort -rn' sorts the files in chronological order (from newest to oldest).

current year (e.g., for copyright) in TT2 templates

If you are using the Perl Template Toolkit and want the template always to display a copyright notice giving the current year, just include the following in your template:

© [% USE date %][% date.format(date.now, '%Y') %]

It uses the Template::Plugin::Date plugin.

Using GPG as a password wallet

I have too many passwords. In fact, I make many of my less-used passwords by grabbing the first 10 or 12 characters of output from something like this:

dd if=/dev/urandom bs=1k count=1 | md5sum

So I get passwords like '1758dbed4331'--no way I'll remember that.

For a while I tried using a nice text-based password wallet program called the Password Management System (PMS--yes, it's a very unfortunate acronym). I learned about this from a Linux Journal article by Marcel Gagne. I built it from source and used it on FC3 for months.

Then I upgraded to FC4. I rebuilt PMS from source and tried reading the previous PMS data files (two data files in my home directory, as I recall). I kept getting segmentation faults. PMS on FC4 worked OK on new data, but it wouldn't read the old data. Some weird incompatibility in the libraries, I guess. So I had to build PMS on another FC3 box, run it, and copy-and-paste all my passwords out of PMS and into a text file. Lame. Hella lame.

So now I just stick with that text file, but it's encrypted by gpg:

gpg -c passwords.txt

When prompted, I gave it a good password. Now, when I need to look up one of my passwords, I just do this:

cat passwords.txt.gpg | gpg | less

Because the output is in 'less', I can even do text searches (each username/password pair is accompanied by some text describing where the password is used). When I've looked up whatever password I need, I just hit 'q', and I'm done.

trick for hosting large DVD ISOs on Apache

I recently needed to make a large file (a DVD ISO) available from a Webserver running Apache. The ISO was around 3.1GB, and Apache wouldn't serve it (the file wouldn't even show up in the mod_autoindex listing).

So I split up the original file into smaller chunks:

split -b 1073741824 huge_DVD.iso

This created 4 files called xaa, xab, xac, and xad. The first three were 1073741824 bytes each, and xad was around 31MB. I renamed the files (in order) to huge_DVD.chunk1, huge_DVD.chunk2, huge_DVD.chunk3, and huge_DVD.chunk4.

I posted the 'chunks' to the Webserver along with a text file containing the SHA1 checksum of the original file. The original file can be reassembled and its checksum computed in a single read by running the following command (after downloading the chunks):

cat huge_DVD.chunk* | tee huge_DVD.iso | openssl dgst -sha1