Re: [Success, 1% lisp-tests fail] XEmacs 21.5-b0 "alfalfa", i586-pc-win32

>>>>> "Martin" == Martin Buchholz <martin(a)xemacs.org&gt; writes: APA> These test fails because my system prints 3 digits for the exponent: APA> (format "%e" 100) APA> "1.000000e+002" APA> MSDN actually documents that exponents are always printed with three APA> digits. APA> What's the preferable way to fix these tests? APA> 1. APA> (cond APA> ((equal system-type 'windows-nt) APA> (Assert (string= (format "%e" 100) "1.000000e+002"))) APA> (t APA> (Assert (string= (format "%e" 100) "1.000000e+02")))) APA> 2. APA> (Assert (string-match "1\\.000000e\\+0?02" (format "%e" 100) 0)) Martin> I vote for 2. Except that the 3rd arg to string-match Martin> seems unnecessary. Martin> (Assert (string-match "1\\.000000e\\+0?02" (format "%e" 100))) APA> Martin, for maintainability (regexps are pretty easily screwed up, I APA> have used the fix of perl, grep, XEmacs for too long) I would prefer APA> 1. Would it be OK with you if generated a patch grouping all %[eEgG] APA> together in on (cond ...) and adding your comment from below? That is fine with me. But I have some concerns that the solution you come up with isn't portable enough and will need to be fixed again in the future. 1. For example, MS might actually fix the bug. 2. (equal system-type 'windows-nt) is better written (eq system-type 'windows-nt) 3. Which systems does this really apply to? Only windows-nt? What about windows-98? What about if microsoft renames their OS again? What about if we're on a windows system, but we're using a cygwin libc? Doesn't Borland ship a libc without this bug? etc... This bug is *not* a bug of the system, but of the C library. The principle is to avoid system-level ifdefs (usually in favor of feature tests). Time to add an autoconf test? 4. Instead of (cond), how about (if) or (case)?