checking for MySQL prerequisites... external MySQL 251 checking for mysql_config... /usr/bin/mysql_config 252 checking MySQL version... OK 253 checking for MySQL Client library... includes -I/usr/include/mysql, libraries -Wl,-O1 -Wl,--as-needed -rdynamic -L/usr/lib64/mysql -lmysqlclient -L/usr//lib64 -lz -lcrypt -lnsl -lm -L/usr/lib64/ -lssl -lcrypto 254 checking MySQL Connector/C++... external 255 checking mysql_driver.h usability... no 256 checking mysql_driver.h presence... no 257 checking for mysql_driver.h... no 258 configure: error: mysql_driver.h not found. install MySQL C++ Connectivity 259 260 !!! Please attach the following file when seeking support: 261 !!! /var/tmp/portage/app-office/libreoffice-3.6.2.2/work/libreoffice-core-3.6.2.2/config.log 262 * ERROR: app-office/libreoffice-3.6.2.2 failed (configure phase): 263 * econf failed 264 * 265 * Call stack: 266 * ebuild.sh, line 85: Called src_configure Reproducible: Always
Created attachment 325134 [details] /var/log/portage/app-office:libreoffice-3.6.2.2:20120927-173646.log
It asks you for file config.log please provide it.
Created attachment 325184 [details] config.tar here is the config.log. thanks for the fast support
In file included from conftest.cpp:36:0: /usr/include/mysql_driver.h:28:32: fatal error: boost/scoped_ptr.hpp: No such file or directory compilation terminated. So I suspect yoru boost install is hosed or you are using the fancy new 1.50 boost wich is really creative in ways how to f**k up your system.
i have this installed. [I] dev-libs/boost Available versions: (0) 1.35.0-r2 1.35.0-r5 (1.37) (~)1.37.0-r1 (1.39) (~)1.39.0 (1.41) 1.41.0-r3 (1.42) (~)1.42.0 (~)1.42.0-r1 1.42.0-r2 (1.45) (~)1.45.0 (1.46) 1.46.1-r1 (1.47) (~)1.47.0 (~)1.47.0-r1 (1.48) 1.48.0-r2 (1.49) 1.49.0-r1 (1.50) [M](~)1.50.0 [M](~)1.50.0-r2 (1.51) [M](~)1.51.0_pre^m[1] {{debug doc +eselect expat icu mpi python static-libs test tools}} Installed versions: 1.49.0-r1(1.49)(14:02:06 12.08.2012)(eselect icu mpi python -debug -doc -static-libs -test -tools) Homepage: http://www.boost.org/ Description: Boost Libraries for C++ so it must be sth else.
Hmm, after the boost moves maybe it forgets to find the boost headers in fancy locations like we in gentoo have...
Yup the header check is just simple header.h lookup AC_CHECK_HEADER(mysql_driver.h, [], [AC_MSG_ERROR(mysql_driver.h not found. install MySQL C++ Connectivity)], []) So I suppose we could try to add there the include path for boost so it won't fail.
well if someone make update, i will make test compilation.
Boost was unslotted now, so it should work again by default. Could you try it and close the bug if it is correct?
*** Bug 433485 has been marked as a duplicate of this bug. ***
(In reply to comment #9) > Boost was unslotted now, so it should work again by default. > > Could you try it and close the bug if it is correct? unfortunatelly i cant test it now. my gentoo box got a hardware failure and i hope i get a replace component. :(.
Thursday I could compile app-office/libreoffice-3.6.2.2 with dev-libs/boost-1.51.0-r1 without further tweaking. Anyhow, I don't know if really related to this bug today app-office/libreoffice-3.6.3.2 fails with an error concerning dev-db/mysql-connector-c++: checking version... configure: error: not suitable, we need >= 1.0.6 The version installed is 1.1.0-r1 and worked fine with the previous libreoffice.
Created attachment 328214 [details] emerge --info
Created attachment 328216 [details] config.log
(In reply to comment #14) > Created attachment 328216 [details] > config.log If you take look on the config.log file you will se dozens of undefined refference error there. So it seems like you changed mysql and didnt rebuild the connector. revdep-rebuild or @preserved-rebuild are your friends :-)
Thanks, just saw it after pasting the logfile. Interestingly portage did not complain about @preserved-libs. Just rebuilt the connector manually and now the office compilation passed the configure phase. I will report later if the compilation works out.
Compiling finished successful with boost-1.51.0-r1. I think the bug can be closed.
