The more I look at how the database backends are compiled in and
tested, the more puzzled I get.

1. Question:

When compiling with:

$ ../heimdal-7.1.0/configure --with-libintl 
--with-libintl-include=/usr/local/include --with-libintl-lib=/usr/local/lib 
--prefix=/usr/heimdal-7.1.0-lmdb --disable-kcm --with-openssl 
--with-openssl-include=/usr/include --with-openssl-lib=/usr/lib --disable-otp 
--enable-pthread-support --with-readline=/usr/local --enable-digest --with-ipv6 
--enable-kx509 --without-openldap --enable-pk-init --with-sqlite3=/usr/local 
--with-x --x-libraries=/usr/local/lib --x-includes=/usr/local/include 
--localstatedir=/var --disable-silent-rules --disable-ndbm-db --enable-lmdb-db 
--enable-mdb-db CFLAGS=-I/usr/local/include LDFLAGS=-L/usr/local/lib -Wl,-rpath 
-Wl,/usr/local/lib -lintl

Is the following the expected result?

# /usr/local/src/heimdal-7.1.0-build-lmdb/kdc/kdc --builtin-hdb
builtin hdb backends: db:, db1:, mit-db:, mdb:, lmdb:, keytab:, sqlite:

What's the difference between mdb and lmdb?

2. Question: 

What database backends should be tested? 
My guess would be "all compiled in". So how do I best get the list
of db backends that should be compiled in?

3. Question:

When specifying a database backend as an argument to a test script,
should it be like "lmdb" or "lmdb:" or "lmdb:,"? The have-db script
is btw broken. IPS?!

oldIFS="$IFS"
IPS=,
set - ${list}
IFS="$oldIFS"

In my current compile all tests in db PASS. But then of course only
lmdb is tested. 

4. Question:

The test which loads the text-dump-0.7 has a sed command that removes
stuff at the end of the output before the compare:


${kadmin} dump | sort | sed 's/[0-9]* -$//' > out-text-dump-0.7
sort < ${srcdir}/text-dump-0.7 | sed 's/[0-9]*$//' > out-text-dump-0.7-orig

if the output comes from the sqlite database, the lines are (before
the sed is taking place):

defa...@example.org 
0::3:3B2A671585E93D6B:3/"EXAMPLE.ORGdefault"::2:3B2A671585E93D6B:3/"EXAMPLE.ORGdefault"::1:3B2A671585E93D6B:3/"EXAMPLE.ORGdefault"::18:AF401411D3F29C204611A9BA1EF54AEDEC43A01B0123C57B994B2EE104E7F127:3/"EXAMPLE.ORGdefault"::16:02401CAD7A92760E464025760BCD3BE5DF616DD5A798C719:3/"EXAMPLE.ORGdefault"::23:31D6CFE0D16AE931B73C59D7E0C089C0:3/"EXAMPLE.ORGdefault"
 20050728203748:kadmin/ad...@example.org - - - - 86400 604800 254 
20050728203748:863727:0

defa...@example.org 
0::3:3B2A671585E93D6B:3/"EXAMPLE.ORGdefault"::2:3B2A671585E93D6B:3/"EXAMPLE.ORGdefault"::1:3B2A671585E93D6B:3/"EXAMPLE.ORGdefault"::18:AF401411D3F29C204611A9BA1EF54AEDEC43A01B0123C57B994B2EE104E7F127:3/"EXAMPLE.ORGdefault"::16:02401CAD7A92760E464025760BCD3BE5DF616DD5A798C719:3/"EXAMPLE.ORGdefault"::23:31D6CFE0D16AE931B73C59D7E0C089C0:3/"EXAMPLE.ORGdefault"
 20050728203748:kadmin/ad...@example.org - - - - 86400 604800 254 
20050728203748:863727:-1 -


What value is in the end?
Now the sed removes the "0" and the other sed the "1 -". Should the whole "-1 
-" be removed?

Harald.

Reply via email to