Re: low-hanging SRFI fruit (SRFI-98)

[Neil Jerram]

Julian Graham <address@hidden> writes:

>> Looks good to me, please go ahead.
>
> Find attached a patch that includes the implementation previously
> attached, plus unit tests and additions to the NEWS and modules.texi
> files.
>
>
> Regards,
> Julian

Hi Julian,

Thanks for doing this.  The code and test all look fine to me; I just
have a few comments on the commit message and the documentation.

Am I correct in thinking that you can commit this yourself? - subject
to Savannah coming back up, of course.  If so, please do that after
incorporating the following points.

> From a1e4bcb30b80199de6f2cd620d950d04e0734592 Mon Sep 17 00:00:00 2001
> From: Julian Graham <address@hidden(none)>
> Date: Thu, 28 May 2009 18:15:05 -0400
> Subject: [PATCH] Implementation of SRFI-98 (An interface to environment 
> variables)

When you commit, the commit message should include ChangeLog-style
entries for the changes.

> ---
>  NEWS                          |    1 +
>  doc/ref/srfi-modules.texi     |   19 +++++++++++++++++
>  module/srfi/srfi-98.scm       |   44 
> +++++++++++++++++++++++++++++++++++++++++
>  test-suite/tests/srfi-98.test |   38 +++++++++++++++++++++++++++++++++++
>  4 files changed, 102 insertions(+), 0 deletions(-)
>  create mode 100644 module/srfi/srfi-98.scm
>  create mode 100644 test-suite/tests/srfi-98.test
>
> diff --git a/NEWS b/NEWS
> index 1785fe8..94d6be7 100644
> --- a/NEWS
> +++ b/NEWS
> @@ -10,6 +10,7 @@ Changes in 1.9.0:
>  * New modules (see the manual for details)
>  
>  ** `(srfi srfi-18)', multithreading support
> +** `(srfi srfi-98)', an interface to environment variables
>  ** The `(ice-9 i18n)' module provides internationalization support
>  
>  * Changes to the distribution
> diff --git a/doc/ref/srfi-modules.texi b/doc/ref/srfi-modules.texi
> index 1fa50b2..638aa46 100644
> --- a/doc/ref/srfi-modules.texi
> +++ b/doc/ref/srfi-modules.texi
> @@ -47,6 +47,7 @@ get the relevant SRFI documents from the SRFI home page
>  * SRFI-61::                     A more general `cond' clause
>  * SRFI-69::                     Basic hash tables.
>  * SRFI-88::                     Keyword objects.
> +* SRFI-98::                     An interface to access environment variables.

That's not quite idiomatic English, is it?  I know it's the title of
the SRFI, but I think that `Accessing environment variables' would be
simpler and better.

>  @end menu
>  
>  
> @@ -3608,6 +3609,24 @@ Return the keyword object whose name is @var{str}.
>  @end example
>  @end deffn
>  
> address@hidden SRFI-98
> address@hidden SRFI-98 An interface to access environment variables.

Same again here.

> address@hidden SRFI-98
> address@hidden environment variables
> +
> +This is a portable wrapper around Guile's built-in support for 
> +interacting with the current environment, @xref{Runtime Environment}.
> +
> address@hidden {Scheme Procedure} get-environment-variable name
> +Returns the value of the named environment variable as a string, or 
> address@hidden if the named environment variable is not found.  This is 
> +equivalent to @code{(getenv name)}.

Looking at this from the point of view of a developer wanting to use
this procedure...  Should name be a symbol, or a string; or can it be
either?

> address@hidden deffn
> +
> address@hidden {Scheme Procedure} get-environment-variables
> +Returns names and values of all the environment variables as an
> +association list.

In which the keys and values are strings?  Or are the keys symbols?

Regards,
      Neil