From: Levin A. <ma...@le...> - 2010-04-08 00:07:44
|
Hi, I have a problem with mdbtools on OSX 10.6 (installed via macports). I have this JET3 database that seems to be parsed correctly on my ubuntu box; but fails to be read correctly on OSX with the same version of mdbtools. For example, mdb-tables on OSX returns no tables: levin@rhea:~ $ uname -a Darwin rhea.local 10.3.0 Darwin Kernel Version 10.3.0: Fri Feb 26 11:58:09 PST 2010; root:xnu 1504.3.12~1/RELEASE_I386 i386 levin@rhea:~ $ port list mdbtools mdbtools @0.6pre1 databases/mdbtools levin@rhea:~ $ md5 db.mdb MD5 (db.mdb) = c1be560bede6c55e53651f7c6e7ff2b8 levin@rhea:~ $ mdb-ver -M db.mdb mdbtools v0.6pre1 JET3 levin@rhea:~ $ mdb-tables db.mdb levin@rhea:~ $ For comparison, here is the (correct) output from ubuntu: levin@q:~$ uname -a Linux q 2.6.24-26-generic #1 SMP Tue Dec 1 17:55:03 UTC 2009 x86_64 GNU/Linux levin@q:~$ aptitude show mdbtools | grep Version Version: 0.5.99.0.6pre1.0.20051109-4 levin@q:~$ md5sum db.mdb c1be560bede6c55e53651f7c6e7ff2b8 db.mdb levin@q:~$ mdb-ver -M db.mdb mdbtools v0.6pre1 JET3 levin@q:~$ mdb-tables db.mdb BE BG EBM20R EBM20RA EBM20Z EBM87R EBM87Z EBM96R GOA88R GOA88Z GOA96R GOA96RAS GOA96Z KAPITEL KEYS EBM20RAS EBM96Z GOA96RA KEYTABS levin@q:~$ Something is clearly wrong, (Further testing shows that the current mdbtools from sourceforge compiled from source on the ubuntu box also gives the incorrect results. The problem therefore seems to be fixed by one of the patches to the debian mdbtools package [1] but nothing jumps out) Anyone have seen this kind of problem before or have an idea how to debug this? Thank you, Levin Alexander [1] http://packages.debian.org/changelogs/pool/main/m/mdbtools/mdbtools_0.5.99.0.6pre1.0.20051109-6/changelog |
From: Sam M. <pa...@gm...> - 2010-04-08 00:54:52
|
On your mac can you run the following command for me: lipo -info `which mdb-tables` And return what the output is? Cheers, Sam Moffatt http://pasamio.id.au On Thu, Apr 8, 2010 at 10:01 AM, Levin Alexander <ma...@le...> wrote: > Hi, > > I have a problem with mdbtools on OSX 10.6 (installed via macports). I > have this JET3 database that seems to be parsed correctly on my ubuntu > box; but fails to be read correctly on OSX with the same version of > mdbtools. > > For example, mdb-tables on OSX returns no tables: > > levin@rhea:~ $ uname -a > Darwin rhea.local 10.3.0 Darwin Kernel Version 10.3.0: Fri Feb 26 > 11:58:09 PST 2010; root:xnu 1504.3.12~1/RELEASE_I386 i386 > levin@rhea:~ $ port list mdbtools > mdbtools @0.6pre1 databases/mdbtools > levin@rhea:~ $ md5 db.mdb > MD5 (db.mdb) = c1be560bede6c55e53651f7c6e7ff2b8 > levin@rhea:~ $ mdb-ver -M db.mdb > mdbtools v0.6pre1 > JET3 > levin@rhea:~ $ mdb-tables db.mdb > > levin@rhea:~ $ > > For comparison, here is the (correct) output from ubuntu: > > levin@q:~$ uname -a > Linux q 2.6.24-26-generic #1 SMP Tue Dec 1 17:55:03 UTC 2009 > x86_64 GNU/Linux > levin@q:~$ aptitude show mdbtools | grep Version > Version: 0.5.99.0.6pre1.0.20051109-4 > levin@q:~$ md5sum db.mdb > c1be560bede6c55e53651f7c6e7ff2b8 db.mdb > levin@q:~$ mdb-ver -M db.mdb > mdbtools v0.6pre1 > JET3 > levin@q:~$ mdb-tables db.mdb > BE BG EBM20R EBM20RA EBM20Z EBM87R EBM87Z EBM96R GOA88R GOA88Z > GOA96R GOA96RAS GOA96Z KAPITEL KEYS EBM20RAS EBM96Z GOA96RA KEYTABS > levin@q:~$ > > Something is clearly wrong, > > (Further testing shows that the current mdbtools from sourceforge > compiled from source on the ubuntu box also gives the incorrect > results. The problem therefore seems to be fixed by one of the patches > to the debian mdbtools package [1] but nothing jumps out) > > Anyone have seen this kind of problem before or have an idea how to debug this? > > > Thank you, > Levin Alexander > > [1] http://packages.debian.org/changelogs/pool/main/m/mdbtools/mdbtools_0.5.99.0.6pre1.0.20051109-6/changelog > > ------------------------------------------------------------------------------ > Download Intel® Parallel Studio Eval > Try the new software tools for yourself. Speed compiling, find bugs > proactively, and fine-tune applications for parallel performance. > See why Intel Parallel Studio got high marks during beta. > http://p.sf.net/sfu/intel-sw-dev > _______________________________________________ > mdbtools-dev mailing list > mdb...@li... > https://lists.sourceforge.net/lists/listinfo/mdbtools-dev > |
From: Levin A. <ma...@le...> - 2010-04-08 12:47:12
|
On Thu, Apr 8, 2010 at 05:05, Sam Moffatt <pa...@gm...> wrote: (sorry the reply should have gone to the list) > On your mac can you run the following command for me: > > lipo -info `which mdb-tables` > > And return what the output is? levin@rhea:~ $ lipo -info `which mdb-tables` Non-fat file: /opt/local/bin/mdb-tables is architecture: x86_64 > So it is using 64-bit integers which might break it, try convincing > macports to compile it as i386 and see if that makes a difference. I was unable to figure out how to convince MacPorts to build 32bit-binaries without recompiling everything. :( But I tried building mdbtools from source on the latest 32 bit Ubuntu 10.4 beta (i686) It shows the same problems as the 64 bit Ubuntu. (config.log: https://gist.github.com/83c8ac9b0892eaa9d6a5) > Is your Linux install 32 bit or 64-bit? It's broken on both. What I'll try next is to selectively apply the debian patches to the sources and see which patch, if any, makes it work. Is the source on the mdbtools-sourceforge page really the current repository? It seems hard to believe that the last commit was 4 years ago. -Levin |
From: Sam M. <pa...@gm...> - 2010-04-08 13:51:41
|
There hasn't been that much activity around. Strange that it is broken now on both platforms with that. Ubuntu must have some trick they use or similar. Sam Moffatt http://pasamio.id.au On Thu, Apr 8, 2010 at 10:46 PM, Levin Alexander <ma...@le...> wrote: > On Thu, Apr 8, 2010 at 05:05, Sam Moffatt <pa...@gm...> wrote: > > (sorry the reply should have gone to the list) > >> On your mac can you run the following command for me: >> >> lipo -info `which mdb-tables` >> >> And return what the output is? > > levin@rhea:~ $ lipo -info `which mdb-tables` > Non-fat file: /opt/local/bin/mdb-tables is architecture: x86_64 > >> So it is using 64-bit integers which might break it, try convincing >> macports to compile it as i386 and see if that makes a difference. > > I was unable to figure out how to convince MacPorts to build > 32bit-binaries without recompiling everything. :( > > But I tried building mdbtools from source on the latest 32 bit Ubuntu > 10.4 beta (i686) It shows the same problems as the 64 bit Ubuntu. > > (config.log: https://gist.github.com/83c8ac9b0892eaa9d6a5) > >> Is your Linux install 32 bit or 64-bit? > > It's broken on both. > > > What I'll try next is to selectively apply the debian patches to the > sources and see which patch, if any, makes it work. > > Is the source on the mdbtools-sourceforge page really the current > repository? It seems hard to believe that the last commit was 4 years > ago. > > -Levin > > ------------------------------------------------------------------------------ > Download Intel® Parallel Studio Eval > Try the new software tools for yourself. Speed compiling, find bugs > proactively, and fine-tune applications for parallel performance. > See why Intel Parallel Studio got high marks during beta. > http://p.sf.net/sfu/intel-sw-dev > _______________________________________________ > mdbtools-dev mailing list > mdb...@li... > https://lists.sourceforge.net/lists/listinfo/mdbtools-dev > |
From: Levin A. <ma...@le...> - 2010-04-08 14:57:49
|
On Thu, Apr 8, 2010 at 15:51, Sam Moffatt <pa...@gm...> wrote: > There hasn't been that much activity around. Strange that it is broken > now on both platforms with that. Ubuntu must have some trick they use > or similar. I tried building the latest CVS version. This version works. (after `aptitude uninstall flex-old; aptitude install flex`) Why hasn't there been a release after 0.6pre1? Thanks for your help, Levin |
From: Kurt S. <ku...@cc...> - 2010-04-23 17:55:19
|
Hi Levin, I've got mdbtools compiling in fink on 10.6. I'm using a version I checked out from sourceforge back in 2008. -kurt (goatbar) http://schwehr.org |
From: Levin A. <ma...@le...> - 2010-04-26 15:10:13
|
On Fri, Apr 23, 2010 at 19:51, Kurt Schwehr <ku...@cc...> wrote: > Hi Levin, > > I've got mdbtools compiling in fink on 10.6. I'm using a version I checked out > from sourceforge back in 2008. I ended up making a git mirror from the CVS repository (didn't want to deal with CVS) and built that. That worked. $ sudo port install libtool automake autoconf bison flex $ git clone http://github.com/levinalex/mdbtools.git $ cd mdbtools $ ./autogen.sh && make # this fails [1] $ make distclean && make # but compiling again succeeds $ sudo make install -Levin [1] /bin/sh ../../libtool --tag=CC --mode=compile gcc -DHAVE_CONFIG_H -I. -I../../include -I../../include -I/opt/local/include/glib-2.0 -I/opt/local/lib/glib-2.0/include -I/opt/local/include -g -O2 -DSQL -MT parser.lo -MD -MP -MF .deps/parser.Tpo -c -o parser.lo parser.c libtool: compile: gcc -DHAVE_CONFIG_H -I. -I../../include -I../../include -I/opt/local/include/glib-2.0 -I/opt/local/lib/glib-2.0/include -I/opt/local/include -g -O2 -DSQL -MT parser.lo -MD -MP -MF .deps/parser.Tpo -c parser.c -fno-common -DPIC -o .libs/parser.o libtool: compile: gcc -DHAVE_CONFIG_H -I. -I../../include -I../../include -I/opt/local/include/glib-2.0 -I/opt/local/lib/glib-2.0/include -I/opt/local/include -g -O2 -DSQL -MT parser.lo -MD -MP -MF .deps/parser.Tpo -c parser.c -o parser.o >/dev/null 2>&1 mv -f .deps/parser.Tpo .deps/parser.Plo /bin/sh ../../ylwrap lexer.l lex.yy.c lexer.c -- flex -i /bin/sh ../../libtool --tag=CC --mode=compile gcc -DHAVE_CONFIG_H -I. -I../../include -I../../include -I/opt/local/include/glib-2.0 -I/opt/local/lib/glib-2.0/include -I/opt/local/include -g -O2 -DSQL -MT lexer.lo -MD -MP -MF .deps/lexer.Tpo -c -o lexer.lo lexer.c libtool: compile: gcc -DHAVE_CONFIG_H -I. -I../../include -I../../include -I/opt/local/include/glib-2.0 -I/opt/local/lib/glib-2.0/include -I/opt/local/include -g -O2 -DSQL -MT lexer.lo -MD -MP -MF .deps/lexer.Tpo -c lexer.c -fno-common -DPIC -o .libs/lexer.o lexer.l: In function ‘yylex’: lexer.l:38: error: ‘OR’ undeclared (first use in this function) lexer.l:38: error: (Each undeclared identifier is reported only once lexer.l:38: error: for each function it appears in.) lexer.l:39: error: ‘NOT’ undeclared (first use in this function) lexer.l:40: error: ‘IS’ undeclared (first use in this function) lexer.l:41: error: ‘NUL’ undeclared (first use in this function) lexer.l:63: error: ‘IDENT’ undeclared (first use in this function) make[2]: *** [lexer.lo] Error 1 make[1]: *** [all-recursive] Error 1 make: *** [all-recursive] Error 1 |