This archive contains answers to questions sent to Unidata support through mid-2025. Note that the archive is no longer being updated. We provide the archive for reference; many of the answers presented here remain technically correct, even if somewhat outdated. For the most up-to-date information on the use of NSF Unidata software and data services, please consult the Software Documentation first.
>From: Dana Quinn <address@hidden> >Organization: College of DuPage >Keywords: 199907091643.KAA02410 ldm-mcidas Dana, >Hi! I'm trying to build ldm-mcidas 7.4 with mcidas 7.5 and ldm 5.0.8. >First of all, is this combination of different versions possible? I don't know. I havn't tried this combination. >It seems like ldm 5.0.8 didn't generate a libregexp in the same way the >older versions did. Soooo, I used the libregexp.a from my ldm 5.0.7. >Probably problem #1 right there. I'm also using the netcdf stuff from >the mcidas distribution - is that the correct thing to do? When I build here at the UPC, I use a build of the netCDF done from the latest release. The netCDF bundled with McIDAS is not the latest release, but I think that it should work anyway. I will have to try this out in the future. >Anyway, this is the problem I'm seeing - I start with a 'make all', as >the instructions tell me (after running configure, of course), and get >this output (pasted below after this paragraph). I can't really tell >what's going on with this. Any ideas? No, not really. I will have to get together with Steve Emmerson later to see if he has any ideas. >After the error output, I'll >include what I've set the environment variables to. Since the majority of the ldm-mcidas decoders are now obsolete (the non-imagery products were dropped from the Unidata-Wisconsin datastream on July 1), the only decoders you need to have for use with the LDM are lwtoa3, nldn2md, and possibly nids2area. You can get all of these from a binary release for Linux in: machine: ftp.unidata.ucar.edu user: anonymous pass: your_full_email_address directory: pub/binary/linux_2.0-i586 file: ldm-mcidas.tar.Z I think that it would be easiest for both of us if you went to the binary distribution (I am running as hard as I can to get the McIDAS-X 7.60 distribution packaged for release this month, so I would rather not divert attention to this when a solution already exists). >Oh, one more question - where can I find the newest version of fort77? The latest version of fort77 that I have tested with McIDAS is available in the unix/740/fortran/linux directory. I want to point out, however, that the latest 7.5 (and 7.6) release of McIDAS no longer uses fort77. Instead, a Bourne shell script, mcfc, is created by the build process. 'mcfc' can be used in place of fort77 in building ldm-mcidas, so I recommend that you use it IF you choose to rebuild ldm-mcidas. Is there some pressing reason that you would rather build a new ldm-mcidas distribution instead of using the binary release? >Dana > >kamin ldm-4> make all >+ make decoders >make[1]: Entering directory `/home/ldm/ldm-mcidas-7.4.0/src' > >making `all' in directory /home/ldm/ldm-mcidas-7.4.0/src/port > >make[2]: Entering directory `/home/ldm/ldm-mcidas-7.4.0/src/port' > >making `all' in directory /home/ldm/ldm-mcidas-7.4.0/src/port/misc > >make[3]: Entering directory `/home/ldm/ldm-mcidas-7.4.0/src/port/misc' >/usr/bin/cc -c -O -I. -I/home/ldm/ldm-mcidas-7.4.0/../include -D_REENTRANT >-Df2cFortran regex.c >/usr/include/regexp.h: In function `compile': >In file included from regex.h:25, > from regex.c:14: >/usr/include/regexp.h:112: `INIT' undeclared (first use this function) >/usr/include/regexp.h:112: (Each undeclared identifier is reported only once >/usr/include/regexp.h:112: for each function it appears in.) >/usr/include/regexp.h:112: parse error before `regex_t' >/usr/include/regexp.h:114: `regex_t' undeclared (first use this function) >/usr/include/regexp.h:114: parse error before `)' >/usr/include/regexp.h:126: `__expr_ptr' undeclared (first use this function) >/usr/include/regexp.h:126: parse error before `)' >/usr/include/regexp.h:165: `REG_NEWLINE' undeclared (first use this function) >/usr/include/regexp.h:170: `REG_BADPAT' undeclared (first use this function) >/usr/include/regexp.h:171: `REG_ECOLLATE' undeclared (first use this function) >/usr/include/regexp.h:172: `REG_ECTYPE' undeclared (first use this function) >/usr/include/regexp.h:173: `REG_EESCAPE' undeclared (first use this function) >/usr/include/regexp.h:174: `REG_BADRPT' undeclared (first use this function) >/usr/include/regexp.h:175: `REG_EEND' undeclared (first use this function) >/usr/include/regexp.h:176: `REG_ERPAREN' undeclared (first use this function) >/usr/include/regexp.h:180: `REG_ESUBREG' undeclared (first use this function) >/usr/include/regexp.h:182: `REG_EBRACK' undeclared (first use this function) >/usr/include/regexp.h:184: `REG_EPAREN' undeclared (first use this function) >/usr/include/regexp.h:186: `REG_EBRACE' undeclared (first use this function) >/usr/include/regexp.h:188: `REG_BADBR' undeclared (first use this function) >/usr/include/regexp.h:190: `REG_ERANGE' undeclared (first use this function) >/usr/include/regexp.h:192: `REG_ESPACE' undeclared (first use this function) >/usr/include/regexp.h:193: `REG_ESIZE' undeclared (first use this function) >regex.h: At top level: >In file included from regex.c:14: >regex.h:28: parse error before `regexp' >regex.h:28: warning: no semicolon at end of struct or union >regex.h:31: parse error before `}' >regex.h:31: `regex_t' used prior to declaration >regex.h:31: warning: data definition has no type or storage class >regex.h:42: parse error before `*' >regex.h:50: parse error before `*' >regex.h:58: parse error before `*' >regex.h:63: parse error before `*' >regex.c:75: parse error before `regex_t' >regex.c:75: warning: data definition has no type or storage class >regex.c:78: parse error before `{' >regex.c:102: parse error before `regex_t' >regex.c:102: warning: data definition has no type or storage class >regex.c:105: parse error before `{' >regex.c: In function `execute': >regex.c:132: parse error before `*' >regex.c:132: declaration for parameter `regex_t' but no such parameter >regex.c:139: invalid type argument of `->' >regex.c:142: invalid type argument of `->' >regex.c:150: `NSUBEXP' undeclared (first use this function) >regex.c:150: invalid type argument of `->' >regex.c:151: invalid type argument of `->' >regex.c:152: invalid type argument of `->' >regex.c: In function `ldm_regexec': >regex.c:169: parse error before `*' >regex.c:169: declaration for parameter `regex_t' but no such parameter >regex.c:177: warning: comparison between pointer and integer >regex.c:180: invalid type argument of `->' >regex.c: In function `ldm_regerror': >regex.c:203: parse error before `*' >regex.c:203: declaration for parameter `regex_t' but no such parameter >regex.c:219: warning: passing arg 1 of `strncpy' makes pointer from integer >without a cast >regex.c:220: subscripted value is neither array nor pointer >regex.c: At top level: >regex.c:234: parse error before `regex_t' >regex.c:234: warning: data definition has no type or storage class >regex.c:235: parse error before `{' >make[3]: *** [regex.o] Error 1 >make[3]: Leaving directory `/home/ldm/ldm-mcidas-7.4.0/src/port/misc' >make[2]: *** [misc/all] Error 1 >make[2]: Leaving directory `/home/ldm/ldm-mcidas-7.4.0/src/port' >make[1]: *** [port/all] Error 1 >make[1]: Leaving directory `/home/ldm/ldm-mcidas-7.4.0/src' >+ set +x > >CPPFLAGS=-Df2cFortran >CC=/usr/bin/cc >CFLAGS=-O >FC=fort77 >FFLAGS=-O >CPP_MCIDAS=-I/home/mcidas/mcidas7.5/src >LD_MCIDAS=-L/home/mcidas/lib -lmcidas >LD_FORTRAN=-static -lf2c >LD_LDM=-L/home/ldm/lib -lldm >LD_LDMREGEXP=-L/home/ldm/oldstuff/ldm-5.0.7internal/lib -lregexp >LD_X11=-L/usr/lib/X11 -lX11 >LD_NETCDF=-L/home/mcidas/mcidas7.5/netcdf/libsrc -lnetcdf >CPP_NETCDF=-I/home/mcidas/mcidas7.5/netcdf/libsrc > >kamin ldm-6> uname -a >Linux kamin.cod.edu 2.0.36 #1 Tue Dec 29 13:11:13 EST 1998 i686 unknown Tom >From address@hidden Fri Jul 9 12:48:13 1999 Tom- You're right. This is the best way to do it. I usually build from source reflexively, I didn't think to check for a binary distribution. Good idea. That will be the quickest way for me to get things up. Thanks for your help! Dana