Re: [fpc-pascal] Why do string indices start at 1, but array indices start at 0?

2011-10-21 Thread Marco van de Voort 
In our previous episode, Jonas Maebe said:
> > During compilation, by statically checking the indices used to access the
> > string, the compiler could fire a warning (or error?) if a  string[0] is
> > found.
> 
> The compiler already does that (except for shortstrings, where string[0] is 
> valid).

Isn't the warning mode dependent? Delphi warns, or at least it did in older
versions.

___
fpc-pascal maillist  -  fpc-pascal@lists.freepascal.org
http://lists.freepascal.org/mailman/listinfo/fpc-pascal

Re: [fpc-pascal] Why do string indices start at 1, but array indices start at 0?

2011-10-21 Thread Jonas Maebe 

On 21 Oct 2011, at 08:25, Roberto P. wrote:

> During compilation, by statically checking the indices used to access the
> string, the compiler could fire a warning (or error?) if a  string[0] is
> found.

The compiler already does that (except for shortstrings, where string[0] is 
valid).


Jonas

___
fpc-pascal maillist  -  fpc-pascal@lists.freepascal.org
http://lists.freepascal.org/mailman/listinfo/fpc-pascal

Re: [fpc-pascal] Why do string indices start at 1, but array indices start at 0?

2011-10-20 Thread Roberto P. 
2011/10/21 Tomas Hajny 

> On 20 Oct 11, at 17:43, Andrew Pennebaker wrote:
>
> > It's inconsistent and ripe for bugs.
>
> Array indices may start at any ordinal value (including e.g.
> characters, values of enumerated types, etc.), not just 0. Only
> dynamic arrays always start at 0 because that is how they have been
> "imported" from other languages.
>
> Tomas
>
>
As said by Thomas, I think we will have to get along with string[1].
However, since I stumbled upon the string[0] mistake more than once, I take
the chance to suggest a way to reduce the likelihood of it.

During compilation, by statically checking the indices used to access the
string, the compiler could fire a warning (or error?) if a  string[0] is
found.
I think that most of the times the string[0] happens inside a for loop or
similar, so a good percentage of these mistakes can be caught.

R#


> ___
> fpc-pascal maillist  -  fpc-pascal@lists.freepascal.org
> http://lists.freepascal.org/mailman/listinfo/fpc-pascal
>
___
fpc-pascal maillist  -  fpc-pascal@lists.freepascal.org
http://lists.freepascal.org/mailman/listinfo/fpc-pascal

Re: [fpc-pascal] Why do string indices start at 1, but array indices start at 0?

2011-10-20 Thread waldo kitty 

On 10/20/2011 17:43, Andrew Pennebaker wrote:

It's inconsistent and ripe for bugs.


funny thing, that... i thought the same thing when looking at the proposals and 
requests :LOL:

___
fpc-pascal maillist  -  fpc-pascal@lists.freepascal.org
http://lists.freepascal.org/mailman/listinfo/fpc-pascal

Re: [fpc-pascal] Why do string indices start at 1, but array indices start at 0?

2011-10-20 Thread Tomas Hajny 
On 20 Oct 11, at 17:43, Andrew Pennebaker wrote:

> It's inconsistent and ripe for bugs.

Array indices may start at any ordinal value (including e.g. 
characters, values of enumerated types, etc.), not just 0. Only 
dynamic arrays always start at 0 because that is how they have been 
"imported" from other languages.

Tomas

___
fpc-pascal maillist  -  fpc-pascal@lists.freepascal.org
http://lists.freepascal.org/mailman/listinfo/fpc-pascal

Re: [fpc-pascal] Why do string indices start at 1, but array indices start at 0?

2011-10-20 Thread Andrew Haines 
On 10/20/11 17:43, Andrew Pennebaker wrote:
> It's inconsistent and ripe for bugs.
> 
> Cheers,
> 
> Andrew Pennebaker
> www.yellosoft.us 
> 
> 
> ___
> fpc-pascal maillist  -  fpc-pascal@lists.freepascal.org
> http://lists.freepascal.org/mailman/listinfo/fpc-pascal
As far as I know it's historical. String[0] used to contain the string
length (255 max). Now it may be something else.

Regards,

Andrew
___
fpc-pascal maillist  -  fpc-pascal@lists.freepascal.org
http://lists.freepascal.org/mailman/listinfo/fpc-pascal

[fpc-pascal] Why do string indices start at 1, but array indices start at 0?

2011-10-20 Thread Andrew Pennebaker 
It's inconsistent and ripe for bugs.

Cheers,

Andrew Pennebaker
www.yellosoft.us
___
fpc-pascal maillist  -  fpc-pascal@lists.freepascal.org
http://lists.freepascal.org/mailman/listinfo/fpc-pascal