[racket] srfi usage within Racket

From: Ray Racine (ray.racine at gmail.com)
Date: Wed Apr 4 16:02:58 EDT 2012

On Wed, Apr 4, 2012 at 12:39 PM, John Clements <clements at brinckerhoff.org>wrote:

> A quick scan of your repository suggests that you were also focused on
> julian<->date conversions and on formatting. Is this a fair summary?
> John

The primary goal was to position everything where I had a top level suite
predefined  DateTime <-> String formats as well as an Instant in time <->
String formats.

OK... now here is what really happened.  It started with the letter 'Z' ...

... the initial point of ignition was Amazon's spanking new DynamoDB did
not understand 'Z,' zulu UTC despite EVERY OTHER HTTP API ON THE PLANET and
therefore the known universe, being able to do so, including the remainder
of the AWS apis.  And yes, their doc does claim the appropriate ISO/RFC
support which does specifically include 'Z' in the spec.  So I'd had this
little 'Z' issue ...

Faced with attempting to raise a bug report with Amazon versus making a one
line change in the SRFI src I went with option #2.

Of course then I had to copy and rename the file outside the standard
collection to prevent collision, and of course started to change a few
identifier names, and of course figured I'd wrap it as a module(s), and
heck, split the source into a few logical files while I was in there, and
of course the pièce de résistance Typed Racket annotations.   And thus boys
and girls, do yaks get shaved.

One more anecdote I also seem to recall Typed Racket uncovering a bug in
SRFI-19 string formatting code.  I just can't say enough about the value of
Typed Racket.    If it had a finger I'd put a ring on it.[1]


[1] No honeymoon jokes, or commentary of what our kids would look like
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.racket-lang.org/users/archive/attachments/20120404/2dc09495/attachment.html>

Posted on the users mailing list.