From: "Jouk Jansen" <jo...@hrem.nano.tudelft.nl> > Parse Style: Extended > > Case Lookup: Blind > > So I assumed it was because of the extended parse style.
Never assume. ( http://en.wikipedia.org/wiki/Desk_Set Everyone needs to see this movie.) My next guess would be that you have some CRTL feature logical name(s) defined. SHOW LOGICAL DECC$* Regarding your suggested changes, I would do things a little differently. For example, in "test/tcrl.com" Original: cmd := mcr 'exe_dir'openssl crl Yours: cmd := "mcr ''exe_dir'openssl crl" Better: cmd = "mcr ''exe_dir'openssl crl" Mine (still better, I claim): cmd = "$ ''exe_dir'openssl crl" I did run into a related problem a few months ago, but I didn't go through all the (or at least all the "test") command procedures to suck out all the instances of ":=", which _really_ should be done. In an environment where case may actually need to be preserved, ":=" is just a problem waiting to happen. Ideally (I claim), OPENSSL.EXE would be built with a LIB$INITIALIZE function which would set things like "DECC$ARGV_PARSE_STYLE", so that everyone would have these problems (or not, but in a consistent way). There seems to be some existing FIPS-related LIB$INITIALIZE set-up in the 0.9.8 trees (fips/fips_premain.c), but I don't see anything in the 1.0.0 code, so it might be easy (easier?) to add one there. My favorites are: DECC$ARGV_PARSE_STYLE, DECC$EFS_CASE_PRESERVE, and DECC$EFS_CHARSET. Around here, enabling DECC$ARGV_PARSE_STYLE will cause this problem: ALP $ @ tests.com test_x509 test normal x509v1 certificate testing X509 conversions p -> d p -> n p -> p d -> d n -> d p -> d d -> n n -> n p -> n d -> p n -> p p -> p test first x509v3 certificate testing X509 conversions [... all happy ...] ALP $ define DECC$ARGV_PARSE_STYLE enable ALP $ @ tests.com test_x509 test normal x509v1 certificate testing X509 conversions p -> d p -> n p -> p d -> d n -> d p -> d d -> n n -> n p -> n d -> p n -> p p -> p %BACKUP-E-OPENIN, error opening ALP$DKA100:[UTILITY.SOURCE.OPENSSL.openssl-1^.0^ .0-stable-SNAP-20100414.test]f.p;1 as input -RMS-E-FNF, file not found Apparently, we're not both totally crazy. So, the test command procedures should definitely be fixed (":=" -> "=", and I'd do "mcr" -> "$"). Then, I'd vote for setting my favorite DECC feature stuff in the main program using LIB$INITIALIZE. (I assume that this will get the usual lack of discussion and/or action here, but I'm willing to pretend for a while.) And for you, leaving those "DECC$*" things defined in the general environment is a dangerous thing to do, as seen here. And, by the way, what are you using for "tar"? ALP $ vmstar tfv openssl-1_0_0_vmspatch.tar -rw-r--r-- 192/192 102 Apr 15 07:22:51 2010 openssl-1.0.0/AXPMAKE.COM drwxr-xr-x 192/192 0 Apr 27 03:42:28 2009 openssl-1/0/0/test -rw-r--r-- 192/192 2579 Apr 15 07:23:23 2010 openssl-1.0.0/test/tcrl.com [...] "openssl-1/0/0/test"? Yow. If you're not already using it, try: http://antinode.info/ftp/vmstar/v3r5_pre8_src.zip (And, if you are using it, describe the situation (off this list), so that I can fix it.) ------------------------------------------------------------------------ Steven M. Schweda s...@antinode-info 382 South Warwick Street (+1) 651-699-9818 Saint Paul MN 55105-2547 ______________________________________________________________________ OpenSSL Project http://www.openssl.org Development Mailing List openssl-dev@openssl.org Automated List Manager majord...@openssl.org
