At 12:05 PM -0700 7/31/03, Mark Berryman wrote: >System Info: > >Compaq C V6.5-001 on OpenVMS Alpha V7.3-1 with Multinet V4.4 > >All released patches to VMS and Multinet applied. > >Build was done on an ODS-2 disk (the build procedure aborts when it >starts to build the extensions on an ODS-5 disk - should this be added >to the README file?).
I do most of my builds on ODS-5 disks and have no trouble. Do you have whichever DECC$... logical name defined that enforces case sensitivity in filenames (not case preservation)? That's just a thoght and may not be the problem. I also build on ODS-2 without trouble, so there is something about your environment that we should fix, work around, or at least document if we can figure out what it is. Ah, I had a thought. If the top-level directory name has dots in it on an ODS-5 disk, then you're sunk (and yes, we should document this). You need to either unpack using $ vmstar -xovf where the -o (or /ODS2) option forces ODS-2 safe names, or after unpacking $ set proc /parse=ext $ rename perl-5^.8^.1-RC3.dir perl-5_8_1-RC3.dir Perl itself has ODS-2 safe names throughout, but the top-level directory will cause problems if it has dots in it. > >I used @configure "-des" and MMK (also tried MMS, no difference). > >The procedure works fine until it tries to build the Encode extension. >At this point it complaints that several *_t.h files are missing: > >%CC-F-NOINCLFILEF, Cannot find file "byte_t.h" >%CC-F-NOINCLFILEF, Cannot find file "ma_05_t.h" >%CC-F-NOINCLFILEF, Cannot find file "ebcdic_t.h" >%CC-F-NOINCLFILEF, Cannot find file "ma_05_t.h" >%CC-F-NOINCLFILEF, Cannot find file "ma_04_t.h" >%CC-F-NOINCLFILEF, Cannot find file "symbol_t.h" >%CC-F-NOINCLFILEF, Cannot find file "cp_02_t.h" >%CC-F-NOINCLFILEF, Cannot find file "def_t.h" > >Does this mean that the kit is incomplete or has my system failed to >build some files on the fly during the build procedure? I believe it's the latter. I'm pretty sure these files are built on the fly by the Encode build process. I'm not certain and haven't had time to check, but I'd suspect they are created by the enc2xs script that gets run like so: Reading gb12345-raw (gb12345-raw) Writing compiled form 34800 bytes in string tables 2786 bytes (92.6%) saved spotting duplicates MCR [---]miniperl.exe "-I[---.lib]" [-.bin]enc2xs -"Q" -o MA_05_T.C -f ma_05_t.fnm >Note that this >is the first time I have tried to build any version of 5.8.1 so I do not >know if this is specific to RC3 or not. It's probably not but there is a slightly later kit at: http://www.iki.fi/jhi/[EMAIL PROTECTED] -- ________________________________________ Craig A. Berry mailto:[EMAIL PROTECTED] "... getting out of a sonnet is much more difficult than getting in." Brad Leithauser
