Can I safely ignore these cmake compiler warnings?
I'm learning to compile packages from source and practicing on MySQL.
Should I be searching for and installing dev libraries when I see "notices" like this (referencing specific "not found" files):
$ cmake . -LA
...
-- Looking for include file cxxabi.h
-- Looking for include file cxxabi.h - not found.
-- Looking for include file dirent.h
-- Looking for include file dirent.h - found
-- Looking for include file dlfcn.h
-- Looking for include file dlfcn.h - found
And what should I do about notices referencing these "not found" messages:
-- Looking for bmove
-- Looking for bmove - not found
-- Looking for bsearch
-- Looking for bsearch - found
-- Looking for index
-- Looking for index - found
For example, cxxabi.h can be found in libstdc++6-4.7-dev on Debian. Do I need to install libstdc++6-4.7-dev to have a proper compile of MySQL?
I also have some (constant?) warnings that I'm unsure of:
-- Performing Test TIME_T_UNSIGNED
-- Performing Test TIME_T_UNSIGNED - Failed
-- Performing Test HAVE_GETADDRINFO
-- Performing Test HAVE_GETADDRINFO - Success
Overall, my build seems to work good, but I want to be sure.
If CMake configuration process doesn't fail, it means these headers are optional and there are workarounds in the MySQL code for these cases.
It might be also, that when some headers aren't present some features are silently turned off. It make sense to provide MySQL as many optional headers as you can.
Beware that some headers are OS-specific, so you can't and don't have to provide them.
Related
I follow the basically solutions to solve it, but I already had the problem.
In my configure.ac file I have a check for mysql:
AC_CHECK_HEADER([mysql/mysql.h], ,AC_MSG_ERROR([Could not find mysql headers !]))
and of course it complain because, as explain here:
If the header files are installed in a nonstandard location, such as
/opt/include, and CPPFLAGS doesn't refer to that directory-for
example, as -I/opt/include-the AC_CHECK_HEADER macro will fail, even
though the files do exist on the system. However, this is an issue for
the system's administrator. Part of the convenience of autoconf is
that you, as the developer, don't need to worry about these details.
So, as developer, what's the way to go to solve it properly ?
I also put the path of real location in Makefile with -I/usr/include/mysql, but it continues to complain.
EDIT: as suggestd I post the configure.ac (the main parts):
useMysql=no
AC_MSG_CHECKING([whether to use mysql])
AC_ARG_ENABLE(mysql,
[ --enable-mysql Enable mysql support],
[MYSQL="$enableval"]
useMysql=yes,
[MYSQL="no"]
)
AC_MSG_RESULT([$MYSQL])
AC_SUBST([MYSQL])
[...]
if test "$MYSQL" = "yes"; then
AC_CHECKING([for MYSQL Library and Header files])
AC_CHECK_HEADER([mysql/mysql.h], ,AC_MSG_ERROR([Could not find mysql headers !]))
AC_CHECK_LIB(mysqlclient, mysql_init, [ MYSQL_LIBS="-lmysqlclient" ], [AC_MSG_ERROR([$PACKAGE_NAME requires but cannot find mysqlclient])])
AC_DEFINE(USE_MYSQL, 1, [Use MYSQL library])
AC_SUBST(MYSQL_LIBS)
fi
then I use the MYSQL_LIBS in the Makefile:
AM_CFLAGS = -g -fPIC -rdynamic -I$(top_srcdir)/include -I/usr/include/mysql
I am trying to connect a database to Weka 3.6.13 in Linux Elementary OS.
First, I had a problem with JDBC connection, solved by this answer changing the /usr/bin/weka file.
Now, when I load the database, this error comes:
Unknown data type: INT. Add entry in weka/experiment/DatabaseUtils.props.
However, I am trying to use explorer only, this file doesn't even exists in my installation.
I installed via sudo apt install weka.
What should I do?
Look inside the directory where your weka.jar file resides, and check if there exists a file called DatabaseUtils.props.
The Weka wiki says:
Weka only looks for the DatabaseUtils.props file. If you take one of
the example files listed above, you need to rename it first.
My file is different I think the actual name does not really matter, it's the filename extension that matters.
In my version of this file there is a section that looks like this:
... (snip...
# mysql-conversion / type-mappings
CHAR=0
TEXT=0
VARCHAR=0
STRING=0
LONGVARCHAR=9
BINARY=0
VARBINARY=0
LONGVARBINARY=9
BIT=1
BOOL=1
NUMERIC=2
DECIMAL=2
FLOAT=2
DOUBLE=2
TINYINT=3
SMALLINT=4
#SHORT=4
SHORT=5
INTEGER=5
INT=5
BIGINT=6
LONG=6
REAL=7
DATE=8
TIME=10
TIMESTAMP=11
#mappings for table creation
CREATE_STRING=TEXT
CREATE_INT=INT
CREATE_DOUBLE=DOUBLE
CREATE_DATE=DATETIME
DateFormat=yyyy-MM-dd HH:mm:ss
#database flags
checkUpperCaseNames=false
checkLowerCaseNames=false
checkForTable=true
setAutoCommit=true
createIndex=false
# All the reserved keywords for this database
Keywords=\
AND,\
ASC,\
BY,\
DESC,\
FROM,\
GROUP,\
INSERT,\
ORDER,\
SELECT,\
UPDATE,\
WHERE
# The character to append to attribute names to avoid exceptions due to
# clashes between keywords and attribute names
KeywordsMaskChar=_
#flags for loading and saving instances using DatabaseLoader/Saver
nominalToStringLimit=50
idColumn=auto_generated_id
If you do a google search for this file, another guy has posted his on github. The weka Wiki or SVN/Git-Repo might also list an offfical version somewhere (cannot find it right now), or you can open your weka.jar file as a zip file and extract the .props file (/src/main/java/weka/experiment/DatabaseUtils.props.mysql).
In any case, Mysql exists in many different versions, and I think you can even switch the query engine inside mysql. So I cannot express any guarantees that any of these 2 .props files shown here really work for you. You should experiment a bit.
The OpenGrok wrapper script has an update option, but when I run it without any options (as echoed in the usage), I get
Loading the default instance configuration ...
FATAL ERROR: OpenGrok Source Path /var/opengrok/src doesn't exist - Aborting!
I have also tried specifying the SRC_ROOT, but continue to get the same error.
This might not be the right answer, but I have been able to update by re-running the index job itself. It doesn't take as long as the initial indexing.
from https://github.com/OpenGrok/OpenGrok
E.g. if opengrok data directory is /tank/opengrok and source root is
in /tank/source then to get more verbosity run the indexer as:
$ OPENGROK_VERBOSE=true OPENGROK_INSTANCE_BASE=/tank/opengrok \
./OpenGrok index /tank/source
SRC_ROOT is a variable in the OpenGrok wrapper (normally in /usr/opengrok/bin/OpenGrok), this variable is to say OpenGrok where you have your src code to be indexed. So you need to edit it:
SRC_ROOT="your/src/path"
Also, after that maybe you'll se an error for the Data location... and you have to set also the variable DATA_ROOT (index location)
DATA_ROOT="you/data"
I'm compiling a deb package and when I run dpkg-buildpackage I get:
dpkg-shlibdeps: error: no dependency information found for /usr/local/lib/libopencv_highgui.so.2.3
...
make: *** [binary-arch] Error 2
This happens because I installed the dependency manually. I know that the problem will be fixed if I install the dependency (or use checkinstall), and I want to generate the package anyway because I'm not interested on dependency checking. I know that I can give to dpkg-shlibdeps the option --ignore-missing-info which prevents a fail if dependency information can't be found. But I don't know how to pass this option to dpkg-shlibdeps since I'm using dpkg-buildpackage and dpkg-buildpackage calls dpkg-shlibdeps...
I have already tried:
sudo dpkg-buildpackage -rfakeroot -d -B
And with:
export DEB_DH_MAKESHLIBS_ARG=--ignore-missing-info
as root.
Any ideas?
use:
override_dh_shlibdeps:
dh_shlibdeps --dpkg-shlibdeps-params=--ignore-missing-info
if your rule file hasn't the dh_shlibdeps call in it. That's usually the case if you've
%:
dh $#
as only rule in it ... in above you must use a tab and not spaces in front of the dh_shlibdeps
If you want it to just ignore that flag, change the debian/rules line from:
dh_shlibdeps
to:
dh_shlibdeps --dpkg-shlibdeps-params=--ignore-missing-info
Yet another way, without modifying build scripts, just creating one file.
You can specify local shlib overrides by creating debian/shlibs.local with the following format: library-name soname-version dependencies
For example, given the following (trimmed) ldd /path/to/binary output
libevent-2.0.so.5 => /usr/lib/libevent-2.0.so.5 (0x00007fc9e47aa000)
libgcrypt.so.20 => /usr/lib/libgcrypt.so.20 (0x00007fc9e4161000)
libpthread.so.0 => /usr/lib/libpthread.so.0 (0x00007fc9e3b1a000)
The contents of debian/shlibs.local would be:
libevent-2.0 5 libevent-2.0
libgcrypt 20 libgcrypt
libpthread 0 libpthread
The "dependencies" list (third column) doesn't need to be 100% accurate - I just use the library name itself again.
Of course this isn't needed in a sane debian system which has this stuff defined in /var/lib/dpkg/info (which can be used as inspiration for these overrides). Mine isn't a sane debian system.
Instead of merely ignoring the error, you might also want to fix the source of the error, which is usually either a missing or an incorrect package.shlibs or package.symbols file in package which contains the shared library triggering the error.
[1] documents how dpkg-shlibdeps uses the package.shlibs resp. package.symbols, files, [2] documents the format of the package.shlibs and package.symbols files.
[1] https://manpages.debian.org/jessie/dpkg-dev/dpkg-shlibdeps.1.en.html
[2] https://www.debian.org/doc/debian-policy/ch-sharedlibs.html
You've just misspelled your export. It should be like this:
export DEB_DH_SHLIBDEPS_ARGS_ALL=--dpkg-shlibdeps-params=--ignore-missing-info
dpkg-buildpackage uses make to process debian/rules. in this process, dpkg-buildpackage it might call dpkg-shlibdeps.
thus, the proper way to pass modify a part of the package building process is to edit debian/rules.
it's hard to give you any more hints, without seeing the actual debian/rules.
Finally I did it in the brute way:
I edited the script /usr/bin/dpkg-shlibdeps, changing this :
my $ignore_missing_info = 0;
to
my $ignore_missing_info = 1;
You can use this:
dh_makeshlibs -a -n
exactly after dh_install
This is weird and I'm not sure who the culprit really is.
I'm doing some scripting, on FreeBSD (6.2)? which makes extensive use of the following ***bash***ism:
do_something <(mysql --skip-column-names -B -e 'select ... from ... where ...;')
... where "do_something is a somewhat crufty utility (in Perl) that won't read from a pipeline. If I use a regular file it works fine. My bash script using things like exec 4< <(...) with these sorts of queries (following by loops of the form while read x y z <&4; do ... never seem to have any issues.
However, Perl (5.8.x) seems to periodically block (apparently forever). I tried changing out the chomp(my $data = <MYDATA>); with a routine that used sysread and I wrote some test cases in Python for comparison. These seem to block far less often than the idiomatic Perl code, but they still do it sometimes. (The Python code using f.read() or os.read(f.fileno()...) seems to behave about equally in this issue).
I've tried reproducing the issue using ... <(cat ...) (where I'm cating the regular file) and that never seems to reproduce that stall.
I've glanced at some ktrace/kdump data ... but I'm far more familiar with Linux strace or even Solaris truss ... so I haven't figured out what's going from there yet, either.
I suppose we can mostly rule out Perl, because I've reproduced the same issue using Python ... I don't see how the bash could be doing anything wrong here (it's just creating a named pipe in /var/tmp/sh-np-xxx and wiring the processes up to that).
What could the mysql shell/utility be doing that might cause this? I don't think I've seen it from anything else (such as cat or dd). I haven't tested this scenario under Linux ... but I've used <(...) (process substitution) for years under Linux and don't recall ever seeing this.
Is it a FreeBSD issue?
Sure I can work around the issue using temporary files ... but I'd sure rather understand why it's doing this (and avoid some of the races and clean-up messiness that temporary files entail).
Any suggestions?
The big difference between operating on the output of mysql and directly on a file is timing. When the perl process is stalled, the big question is: "why is it not making forward progress"? You can use the "l" option to ps to see the wait channel for the perl process; that way you can see if it blocked on a read, or if something else is going on. If it is really blocked on pipe input, I expect the MWCHAN entry for perl to be "piperd".
The same information would be interesting for the mysql process.
What does your Python test code look like?
Another way of writing this while avoiding the bashism is this; that would allow you to rule out bash:
mysql --skip-column-names -B -e 'select ... from ... where ...;' | do_something /dev/stdin
Other interesting questions:
Does the --unbuffered option to mysql change anything?
Does piping the mysql output through dd change anything? (eg. "perlscript <(mysql ... | dd)
Summary: Need more information.