Skip to contentSkip to conversation

binfmt_script and ^M

DISCUSSION Jesse Pollard 2 participants 2 messages Mar 5, 2001 – Mar 6, 2001

lore
Share & tools
lore
#0 lore raw

John Kodis <[email protected]>:

Quoted 11 lines · Richard B. Johnson
Quoted 3 lines

Somebody must have missed the boat entirely. Unix does not, never has, and never will end a text line with '\r'.

Unix does not, never has, and never will end a text line with ' ' (a space character) or with \t (a tab character). Yet if I begin a shell script with '#!/bin/sh ' or '#!/bin/sh\t', the training white space is striped and /bin/sh gets exec'd. Since \r has no special significance to Unix, I'd expect it to be treated the same as any other whitespace character -- it should be striped, and /bin/sh should get exec'd.

Actually it does have some significance - it causes a return, then the following text overwrites the current text. Granted, this is only used occasionally for generating bold/underline/...

This is used in some formatters (troff) occasionally, though it tends to use backspace now.

\r is not considered whitespace, though it should be possible to define it that way. A line terminator is always \n.

Another point, is that the "#!/bin/sh" can have options added: it can be "#!/bin/sh -vx" and the option -vx is passed to the shell. The space is not just "stripped". It is used as a parameter separator. As such, the "stripping" is only because the first parameter is separated from the command by whitespace.

-------------------------------------------------------------------------
Jesse I Pollard, II
Email: [email protected]

Any opinions expressed are solely my own.

#1 lore raw

----- Original Message -----
From: "Jesse Pollard" <[email protected]>
To: <[email protected]>; "Richard B. Johnson" <[email protected]>
Cc: <[email protected]>; <[email protected]>
Sent: Monday, 5. March 2001 19:14
Subject: Re: binfmt_script and ^M

Quoted 13 lines

John Kodis <[email protected]>:

Quoted 11 lines · Richard B. Johnson
Quoted 3 lines

Somebody must have missed the boat entirely. Unix does not, never has, and never will end a text line with '\r'.

Unix does not, never has, and never will end a text line with ' ' (a space character) or with \t (a tab character). Yet if I begin a shell script with '#!/bin/sh ' or '#!/bin/sh\t', the training white space is striped and /bin/sh gets exec'd. Since \r has no special significance to Unix, I'd expect it to be treated the same as any other whitespace character -- it should be striped, and /bin/sh should get exec'd.

Quoted 6 lines

Actually it does have some significance - it causes a return, then the following text overwrites the current text. Granted, this is only used occasionally for generating bold/underline/...

This is used in some formatters (troff) occasionally, though it tends to use backspace now.

Less supports it, but ^H is quite more oftenly used. ISO_646.irv:1991 aka ISO-IR-6 aka US-ASCII-7 _also_ defines it, and we're going to be not ASCII-compatible any longer if we aren't going to support CRLF line endings.
I also oftenly have the other problem round: LF endings in files which are to be viewed under DOS. I use a 15-year-old text editor from Digital Research (yes, DOS 3.41) which still is fine under W** and DOSEMU, it looks like jstar only that I miss find and replace. IMHO those problems could be solved with programmes/kernels/libs accepting LF as line ending and CRLF (and possibly CRCRLF ...) as a synonyme for LF, but treat CR non-LF differently. I have seen this behaviour quite often in the past and am using it for myself, too (except for native assembly progs).

Quoted 2 lines

\r is not considered whitespace, though it should be possible to define it that way. A line terminator is always \n.

ACK

Quoted 5 lines

Another point, is that the "#!/bin/sh" can have options added: it can be "#!/bin/sh -vx" and the option -vx is passed to the shell. The space is not just "stripped". It is used as a parameter separator. As such, the "stripping" is only because the first parameter is separated from the command by whitespace.

That's why I suggest treating CRLF (and only CR only-LF) as LF.

-mirabilos