Hi Johannes. > The report is now generated. All the intrinsics are executed. > > Just some notes: the line 1643 of pdf-fp-func.c is not executed by > the tests. PDF_EOVERFLOW and PDF_EIMPLIMIT are not exercised > neither. The unexercised line (1643) is obsolete anyway, I removed it. I added tests to exercise PDF_EOVERFLOW and PDF_EIMPLIMIT. I will run the tests on a 64 bit host tomorrow to be sure, but I suppose it should work.
I also added an entry in gnupdf.texi for the defines PDF_REAL_[MAX,MIN] that I introduced, but I am not sure if I added it at the right position. Could you please have a look at this? I pushed your patch after moving the documentation for the constants to the "Floating Point Constants" section. Regarding the definition of PDF_REAL_MAX: #define PDF_REAL_MAX (340282346638528859811704183484516925440.000000) I think that it would be better to use the POSIX FLT_MAX constant instead of the raw number. Do you agree? -- Jose E. Marchesi <[email protected]> http://www.jemarch.net GNU Project http://www.gnu.org
