mesa/src/compiler/glsl/glcpp
Kenneth Graunke ea9fa5ff05 glcpp: Remove empty mid-rule action which changes test behavior.
Apparently this causes a slight difference in the parser's token
expectations, leading to a different error message.

It seems harmless, but I wanted to be cautious and separate it out.

Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
2016-03-07 23:02:05 -08:00
..
tests glcpp: Remove empty mid-rule action which changes test behavior. 2016-03-07 23:02:05 -08:00
.gitignore glsl: move to compiler/ 2016-01-26 16:08:33 +00:00
glcpp-lex.l glcpp: Implicitly resolve version after the first non-space/hash token. 2016-03-07 23:01:43 -08:00
glcpp-parse.y glcpp: Remove empty mid-rule action which changes test behavior. 2016-03-07 23:02:05 -08:00
glcpp.c glsl: move to compiler/ 2016-01-26 16:08:33 +00:00
glcpp.h glcpp: Implicitly resolve version after the first non-space/hash token. 2016-03-07 23:01:43 -08:00
pp.c glsl: move to compiler/ 2016-01-26 16:08:33 +00:00
README glsl: move to compiler/ 2016-01-26 16:08:33 +00:00

glcpp -- GLSL "C" preprocessor

This is a simple preprocessor designed to provide the preprocessing
needs of the GLSL language. The requirements for this preprocessor are
specified in the GLSL 1.30 specification availble from:

http://www.opengl.org/registry/doc/GLSLangSpec.Full.1.30.10.pdf

This specification is not precise on some semantics, (for example,
#define and #if), defining these merely "as is standard for C++
preprocessors". To fill in these details, I've been using a draft of
the C99 standard as available from:

http://www.open-std.org/jtc1/sc22/wg14/www/docs/n1256.pdf

Any downstream compiler accepting output from glcpp should be prepared
to encounter and deal with the following preprocessor macros:

	#line
	#pragma
	#extension

All other macros will be handled according to the GLSL specification
and will not appear in the output.

Known limitations
-----------------
A file that ends with a function-like macro name as the last
non-whitespace token will result in a parse error, (where it should be
passed through as is).