This is the mail archive of the
newlib@sourceware.org
mailing list for the newlib project.
Re: Problem with building PDFs
- From: Freddie Chopin <freddie_chopin at op dot pl>
- To: newlib at sourceware dot org
- Date: Sun, 28 Apr 2013 19:06:33 +0200
- Subject: Re: Problem with building PDFs
- References: <50FB02B2 dot 9070304 at op dot pl> <50FBAF51 dot 501 at op dot pl> <50FBDA23 dot 1040106 at op dot pl> <50FC4D5B dot 3030900 at oarcorp dot com> <50FC5DEF dot 4030704 at op dot pl> <50FFB08E dot 3090805 at op dot pl> <5100142F dot 9040603 at oarcorp dot com> <510037E2 dot 3050702 at op dot pl> <51003B76 dot 6080702 at oarcorp dot com> <51004245 dot 3030503 at op dot pl> <51004453 dot 4020503 at oarcorp dot com>
Hi!
A small info for future generations (and myself if I forget). The
problem I had is found in 5 places where exactly this sequence is present:
o-
RETURNS
(there may be multiple blank lines between "o-" and "RETURNS")
I don't know what the right solution is, but I found that putting a dot
in between these two solves the issue for me, so sth like this works:
o-
.
RETURNS
The dot is not visible in the PDF file (it's not there), so it changes
nothing.
I believe that this may have sth to do with texinfo 5 - to do my tests I
had to do "make info pdf" and it fails (with multiple errors), but
eventually it creates .def files needed for make pdf. Many projects
(gcc, binutils) had many problems with recent upgrade to texinfo 5...
Actually there's one thing I had to fix in newlib for the rest of
process to work - I'll send that in another message.
I attach a patch with the "fixes", but it's just for reference - I don't
think it's the right solution.
4\/3!!
diff --git a/newlib/libc/stdio/sscanf.c b/newlib/libc/stdio/sscanf.c
index 974ae81..d3d3211 100644
--- a/newlib/libc/stdio/sscanf.c
+++ b/newlib/libc/stdio/sscanf.c
@@ -317,7 +317,7 @@ DESCRIPTION
where objects inclosed in square brackets are optional, and <<ddd>>
represents decimal, octal, or hexadecimal digits.
o-
-
+.
RETURNS
<<scanf>> returns the number of input fields successfully
scanned, converted and stored; the return value does
diff --git a/newlib/libc/stdio/swscanf.c b/newlib/libc/stdio/swscanf.c
index 273f59b..b385ffa 100644
--- a/newlib/libc/stdio/swscanf.c
+++ b/newlib/libc/stdio/swscanf.c
@@ -302,7 +302,7 @@ DESCRIPTION
where objects inclosed in square brackets are optional, and <<ddd>>
represents decimal, octal, or hexadecimal digits.
o-
-
+.
RETURNS
<<wscanf>> returns the number of input fields successfully
scanned, converted and stored; the return value does
diff --git a/newlib/libc/time/strftime.c b/newlib/libc/time/strftime.c
index 2bcedb2..22c68fe 100644
--- a/newlib/libc/time/strftime.c
+++ b/newlib/libc/time/strftime.c
@@ -234,7 +234,7 @@ as if by calling tzset(). [tm_isdst]
o %%
A single character, `<<%>>'.
o-
-
+.
RETURNS
When the formatted time takes up no more than <[maxsize]> characters,
the result is the length of the formatted string. Otherwise, if the
diff --git a/newlib/libm/common/isgreater.c b/newlib/libm/common/isgreater.c
index bd4e95e..f7d2377 100644
--- a/newlib/libm/common/isgreater.c
+++ b/newlib/libm/common/isgreater.c
@@ -58,7 +58,7 @@ o isless(NaN*0., 1.0)
causes an exception due to the "NaN*0.", but not from the
resultant reduced comparison of isless(NaN, 1.0).
o-
-
+.
RETURNS
@comment Formatting note: "$@" forces a new line
No floating-point exceptions are raised for any of the macros.@*
diff --git a/newlib/libm/common/s_matherr.c b/newlib/libm/common/s_matherr.c
index 58e2428..3e2da5f 100644
--- a/newlib/libm/common/s_matherr.c
+++ b/newlib/libm/common/s_matherr.c
@@ -92,8 +92,7 @@ e.g. <<sin(10e70)>>.
o PLOSS
Partial loss of precision.
o-
-
-
+.
RETURNS
The library definition for <<matherr>> returns <<0>> in all cases.