Subject: | |
From: | |
Reply To: | Borgia, William M. |
Date: | Wed, 3 Mar 1999 11:27:35 -0500 |
Content-Type: | text/plain |
Parts/Attachments: |
|
|
> -----Original Message-----
> From: Borgia, William M.
> Sent: Wednesday, March 03, 1999 11:27 AM
> To: [log in to unmask]
> Subject: RE: The March of Progress
>
>
> If you want to make the contrast more dramatic, add some plain text, a
> variable-length string, and an integer in hexadecimal.
>
> How many lines does it take to do this in Ada, C++, or Java:
>
> printf("The result for %-30s is %10.2f (%08x)\n", x, y, z);
>
> My question: Has anyone written good Ada equivalents to printf and
> scanf?
>
> Mike
>
> I think that this is exactly the point. Most developers acknowledge that
> printf and scanf are some of the most flexible, powerful library methods
> in any language. Because of this, however, they are quite difficult for
> the average guy maintaining or reviewing the software to understand. To
> put it mildly, they rely on a large set of cryptic formatting characters.
> They also rely on being able to send a variable number of parameters,
> which Ada does not allow -- this is why C/C++ places parameters on the
> stack in backwards order.
>
> Without looking at a reference, I can't understand what the line of code
> above attempts to do. I do understand what Wes accomplishes because he
> uses a simple function and the parameter semantics are meaningful because
> of named parameter association.
>
> In short, I'd rather see a few lines of code that I can understand at a
> glance rather than one that requires a more knowledge or a reference.
>
> Bill
|
|
|