The grammar for the lexical analysis contains:

EndOfFile:
physical end of the file
\u0000
\u001A

I don't understand, what's a \u001A? Substitution? How does that work? Wouldn't you want something like \u0019, end of medium?

In article <e0s3ct$1pk6$1@digitaldaemon.com>, Luís Marques says...
>
>The grammar for the lexical analysis contains:
>
>EndOfFile:
>physical end of the file
>\u0000
>\u001A
>
>I don't understand, what's a \u001A? Substitution? How does that work? Wouldn't you want something like \u0019, end of medium?
>
>

0x1A (or CTRL-Z or ASCII 26) is a holdover from CP/M, which Windows inherited
via DOS:

http://blogs.msdn.com/oldnewthing/archive/2004/03/16/90448.aspx

The 'copy' command that the article mentions works something like this:

c:\> copy con > foobar.txt

This will copy every keypress to the text file until you press CTRL-Z.  Its actually pretty handy if you're repairing a system with a paperclip, a battery and some duct-tape.

- EricAnderton at yahoo

In article <e0s5nm$1rpr$1@digitaldaemon.com>, pragma says...
>0x1A (or CTRL-Z or ASCII 26) is a holdover from CP/M, which Windows inherited
>via DOS:
>
>http://blogs.msdn.com/oldnewthing/archive/2004/03/16/90448.aspx
>
>The 'copy' command that the article mentions works something like this:
>
>c:\> copy con > foobar.txt
>
>This will copy every keypress to the text file until you press CTRL-Z.  Its actually pretty handy if you're repairing a system with a paperclip, a battery and some duct-tape.

That's very interesting. I use CTRL-Z in several console programs but I didn't
know programs read that
as \u001A nor from where that convention came.

I still don't really grok the relationship between CTRL-[X] codes and
ASCII/Unicode, on unix and
windows, but I guess that's another story.

Which also makes me wonder, if it wasn't a type afterall, what's the purpose of
characters like \u0019.
The sites listing Unicode characters don't generally have much semantic
information on them.

Luís

Luís Marques wrote:
> In article <e0s5nm$1rpr$1@digitaldaemon.com>, pragma says...
> 
>>0x1A (or CTRL-Z or ASCII 26) is a holdover from CP/M, which Windows inherited
>>via DOS:
>>
>>http://blogs.msdn.com/oldnewthing/archive/2004/03/16/90448.aspx
>>
>>The 'copy' command that the article mentions works something like this:
>>
>>c:\> copy con > foobar.txt
>>
>>This will copy every keypress to the text file until you press CTRL-Z.  Its
>>actually pretty handy if you're repairing a system with a paperclip, a battery
>>and some duct-tape.
> 
> 
> That's very interesting. I use CTRL-Z in several console programs but I didn't
> know programs read that as \u001A nor from where that convention came.
> 
> I still don't really grok the relationship between CTRL-[X] codes and
> ASCII/Unicode, on unix and windows, but I guess that's another story.
> 
> Which also makes me wonder, if it wasn't a type afterall, what's the purpose of
> characters like \u0019. The sites listing Unicode characters don't generally have much semantic
> information on them.
> 
> Luís


IIRC CTRL-[x] ends up, in ASCII, masking out a single bit from [x], I think it is the 64s place but I may be wrong.

On Mon, 03 Apr 2006 16:11:04 -0700, BCS <BCS_member@pathlink.com> wrote:
> Luís Marques wrote:
>> In article <e0s5nm$1rpr$1@digitaldaemon.com>, pragma says...
>>
>>> 0x1A (or CTRL-Z or ASCII 26) is a holdover from CP/M, which Windows inherited
>>> via DOS:
>>>
>>> http://blogs.msdn.com/oldnewthing/archive/2004/03/16/90448.aspx
>>>
>>> The 'copy' command that the article mentions works something like this:
>>>
>>> c:\> copy con > foobar.txt
>>>
>>> This will copy every keypress to the text file until you press CTRL-Z.  Its
>>> actually pretty handy if you're repairing a system with a paperclip, a battery
>>> and some duct-tape.
>>   That's very interesting. I use CTRL-Z in several console programs but I didn't
>> know programs read that as \u001A nor from where that convention came.
>>  I still don't really grok the relationship between CTRL-[X] codes and
>> ASCII/Unicode, on unix and windows, but I guess that's another story.
>>  Which also makes me wonder, if it wasn't a type afterall, what's the purpose of
>> characters like \u0019. The sites listing Unicode characters don't generally have much semantic
>> information on them.
>>  Luís
>
>
> IIRC CTRL-[x] ends up, in ASCII, masking out a single bit from [x], I think it is the 64s place but I may be wrong.

I think you're correct. If you look at an ASCII character table, find the @ character, you'll notice that following it in sequence are the capital A thru Z characters, then [ \ ] ^ _

CTRL+[X] where [X] is one of those characters results in the value of that character with the 64s place masked, eg.

@ == 0x40, CTRL+@ == 0x00
A == 0x41, CTRL+A == 0x01
..
Z == 0x5A, CTRL+Z == 0x1A
..
_ == 0x5F, CTRL+_ == 0x1F

Regan

Luís Marques wrote:
> The grammar for the lexical analysis contains:
> 
> EndOfFile:
> physical end of the file
> \u0000
> \u001A
> 
> I don't understand, what's a \u001A? Substitution? How does that work?
> Wouldn't you want something like \u0019, end of medium?

Some text editors mark the end of the text with a 0x1A (control Z) character. It's a holdover from the DOS days when files got padded out to the next sector boundary.