End-of-file character

Hey Guys,

Does anyone know of a cross-platform way of generating an ‘end-of-file’
character? Alternatively, does anyone know how to figure out what the
‘end-of-file’ character is for a particular OS?

Sonny.

On 4/15/07, Sonny C. [email protected] wrote:

Hey Guys,

Does anyone know of a cross-platform way of generating an ‘end-of-file’
character? Alternatively, does anyone know how to figure out what the
‘end-of-file’ character is for a particular OS?

I had no idea it was platform dependent. I always assumed it was ^D
for all platforms.

“\004” should be it.

Blessings,
TwP

On 4/15/07, Tim P. [email protected] wrote:

On 4/15/07, Sonny C. [email protected] wrote:

Does anyone know of a cross-platform way of generating an ‘end-of-file’
character? Alternatively, does anyone know how to figure out what the
‘end-of-file’ character is for a particular OS?
I had no idea it was platform dependent. I always assumed it was ^D
for all platforms.

“\004” should be it.

This is not correct. There is, in fact, no “end-of-file” character on
Unix, and Ctrl-Z (0x26) is the end-of-file character on Windows only
when the file is not opened in binary mode. It is conventional for a
terminal to accept Ctrl-D (0x04) as an end of file character, but this
is convention, not standard.

If you want to close a file, just close it. The underlying routines
will do the appropriate work necessary.

-austin

Thanks for the clarification Austin.

I’m actually to trying to solve the following problem. I’ve been
redirecting the contents of a_file to an os_command, ie:

os_command < a_file

Usually a_file is very small and I have to create this file before
invoking the os_command. Is it possible to avoid the creation of a_file
and just pass its contents directly to os_command?

Sonny.

There is, in fact, no “end-of-file” character on
Unix, and Ctrl-Z (0x26) is the end-of-file character on Windows only
when the file is not opened in binary mode. It is conventional for a
terminal to accept Ctrl-D (0x04) as an end of file character, but this
is convention, not standard.

If you want to close a file, just close it. The underlying routines
will do the appropriate work necessary.

-austin

Austin Z. wrote:

This is not correct. There is, in fact, no “end-of-file” character on
Unix, and Ctrl-Z (0x26) is the end-of-file character on Windows only
when the file is not opened in binary mode. It is conventional for a
terminal to accept Ctrl-D (0x04) as an end of file character, but this
is convention, not standard.

And, except on terminals, ^Z has been obsolescent on DOS and Windows
since DOS 1.1, twenty-five years ago.


John W. Kennedy
“Give up vows and dogmas, and fixed things, and you may grow like That.
…you may come to think a blow bad, because it hurts, and not because
it humiliates. You may come to think murder wrong, because it is
violent, and not because it is unjust.”
– G. K. Chesterton. “The Ball and the Cross”

On Apr 15, 2:41 pm, Sonny C. [email protected] wrote:

Does anyone know of a cross-platform way of generating an ‘end-of-file’
character? Alternatively, does anyone know how to figure out what the
‘end-of-file’ character is for a particular OS?

I’ve never heard of an end-of-file character. EOF happens when a file
handle reaches the end of a file - not when it hits a specific
character. Many binary file types use null padding at the end
(character \0) although this is not an EOF character per se. To
generate a null character, you could do this “\0”.

Hey Guys,

Thanks for all the answers. I just found the answer with a little
googling. It looks like I want to use the pipe operator… as in:

echo a_file_contents | os_command

Thanks, again.

Sonny.

I’m actually to trying to solve the following problem. I’ve been
redirecting the contents of a_file to an os_command, ie:

os_command < a_file

Usually a_file is very small and I have to create this file before
invoking the os_command. Is it possible to avoid the creation of a_file
and just pass its contents directly to os_command?

Sonny.

Awesome, thanks Brian. Great idea!

Sonny.

Brian C. wrote:

On Mon, Apr 16, 2007 at 11:56:48AM +0900, Sonny C. wrote:

I’m actually to trying to solve the following problem. I’ve been
redirecting the contents of a_file to an os_command, ie:

os_command < a_file

Usually a_file is very small and I have to create this file before
invoking the os_command. Is it possible to avoid the creation of a_file
and just pass its contents directly to os_command?

If you are running os_command from Ruby, use IO.popen to run it. You
then
get an IO object for a pipe to squirt data at it.

On Mon, Apr 16, 2007 at 11:56:48AM +0900, Sonny C. wrote:

I’m actually to trying to solve the following problem. I’ve been
redirecting the contents of a_file to an os_command, ie:

os_command < a_file

Usually a_file is very small and I have to create this file before
invoking the os_command. Is it possible to avoid the creation of a_file
and just pass its contents directly to os_command?

If you are running os_command from Ruby, use IO.popen to run it. You
then
get an IO object for a pipe to squirt data at it.