On 26/05/2023 16:28, Alejandro Colomar wrote:
Hi Ian!
On 5/26/23 17:19, Ian Abbott wrote:
As per the C standard, calling ungetc() with the character parameter
equal to EOF causes it to fail, returning EOF.
Signed-off-by: Ian Abbott <abbotti@xxxxxxxxx>
---
man3/fgetc.3 | 8 +++++++-
1 file changed, 7 insertions(+), 1 deletion(-)
diff --git a/man3/fgetc.3 b/man3/fgetc.3
index 75dcaeaf6..d6bf62327 100644
--- a/man3/fgetc.3
+++ b/man3/fgetc.3
@@ -62,7 +62,13 @@ A terminating null byte (\[aq]\e0\[aq])
is stored after the last character in the buffer.
.PP
.BR ungetc ()
-pushes
+returns
+.B EOF
+if the value of
+.I c
+equals that of the macro
+.BR EOF ,
+otherwise it pushes
I would put that detail at the end of the description, rather than
the beginning. In C code, that kind of short-cutting can help
reading, but in the manual, it's better left as a detail at the end
of it, to give it less importance.
OK, but I think it would still need some sort of remark in the first
sentence to indicate that it does not always apply. Maybe something
along the lines of:
normally pushes c back to stream, ...
Then add the c==EOF exception as a new paragraph along the lines of:
If the value of c equals that of the macro EOF, nothing is
pushed back to the stream and an error is returned.
--
-=( Ian Abbott <abbotti@xxxxxxxxx> || MEV Ltd. is a company )=-
-=( registered in England & Wales. Regd. number: 02862268. )=-
-=( Regd. addr.: S11 & 12 Building 67, Europa Business Park, )=-
-=( Bird Hall Lane, STOCKPORT, SK3 0XA, UK. || www.mev.co.uk )=-
