Dear all,
the MAPI & JDBC tests implemented in C and Java in MonetDB/Clients and
MonetDB/Java (installed in <prefix>/lib/MonetDB/Tests/ &
<prefix>/share/MonetDB/Tests/, respectively) are not yet distributed in any
binary package.
For convenience of running Mtest.py on binary installations, should we
concider to package the C-MAPI tests from MonetDB/Clients in, say, a
MonetDB-client-testing RPM and the Java-JDBC tests from MonetDB/Java in, say,
a MonetDB-java-testing.jar?
Opinions and/or (other) suggestions?
Stefan
--
| Dr. Stefan Manegold | mailto:Stefan.Manegold@cwi.nl |
| CWI, P.O.Box 94079 | http://www.cwi.nl/~manegold/ |
| 1090 GB Amsterdam | Tel.: +31 (20) 592-4212 |
| The Netherlands | Fax : +31 (20) 592-4312 |
Dear all,
Mtest.oy and several tests (see below) (still) rely on monetdb-java-config to
find out where MonetDB/Java (i.e., the *.jar's and the Test_*.class
examples/tests) are installed.
However, we do not distribute monetdb-java-config in any (binary) package
(only with the sources).
I see the followin options:
1) Leave everything as it is.
2) Keep monetdb-java-config in the source (for conveniencce with Mtest.py on
source-base installations), but extend Mtest.py with command line options
as an alternative manner to provide the requires information in case of
binary installation (i.e., w/o monetdb-java-config).
3) Abandon monetdb-java-config completely and have Mtest.py rely only in the
(to be added) command line options
Any preferences and/or other suggestions?
Stefan
========
./MonetDB/src/testing/Mtest.py.in: if CheckExec('monetdb-java-config'):
./MonetDB/src/testing/Mtest.py.in: CONDITIONALS['HAVE_MONETDB_JAVAJDBC'] = GetConfig('monetdb-java-config', '--HAVE_MONETDB_JAVAJDBC')
./MonetDB/src/testing/Mtest.py.in: CONDITIONALS['HAVE_MONETDB_JAVAXRPC'] = GetConfig('monetdb-java-config', '--HAVE_MONETDB_JAVAXRPC')
./pathfinder/tests/BugTracker/Tests/JDBC_0_results.SF-1886326.XQUERY.sh:pdd="`monetdb-java-config --pkgdatadir`"
./pathfinder/tests/BugTracker/Tests/JDBC_250_results.SF-1730556.XQUERY.sh:pdd="`monetdb-java-config --pkgdatadir`"
./pf_rox/tests/BugTracker/Tests/JDBC_0_results.SF-1886326.XQUERY.sh:pdd="`monetdb-java-config --pkgdatadir`"
./pf_rox/tests/BugTracker/Tests/JDBC_250_results.SF-1730556.XQUERY.sh:pdd="`monetdb-java-config --pkgdatadir`"
./sql/src/jdbc/tests/Tests/Test_JdbcClient.SQL.sh:pdd="`monetdb-java-config --pkgdatadir`"
./sql/src/jdbc/tests/Tests/Test.SQL.sh:pdd="`monetdb-java-config --pkgdatadir`"
./sql/src/test/snodgrass/Tests/cast_select.SQL.sh:pdd="`monetdb-java-config --pkgdatadir`"
========
--
| Dr. Stefan Manegold | mailto:Stefan.Manegold@cwi.nl |
| CWI, P.O.Box 94079 | http://www.cwi.nl/~manegold/ |
| 1090 GB Amsterdam | Tel.: +31 (20) 592-4212 |
| The Netherlands | Fax : +31 (20) 592-4312 |
Dear MonetDB users and developers,
The MonetDB team at CWI is pleased to announce the new "Jun2008" release of
the MonetDB suite of programs.
For MonetDB/XQuery, this release marks the roll forward from the old
"milprint_summer"-based Pathfinder compiler to the new "Algebra"-based
Pathfinder compiler.
MonetDB/SQL has been extended with initial support for OpenGIS via the
"Geom[etry]" extension module, aka. MonetDB/SQL/GIS.
Additionally, the configure defaults for compilation from sources on
Unix(-like) systems as well as the Windows installers have been cleaned-up.
For more information, please see the detailed release notes at
http://monetdb.cwi.nl/Development/Releases/Version4.24/
IMPORTANT NOTE:
Before upgrading to the latest "Jun2008" release from any previous release
of MonetDB, please make a dump of your database, and rename/move/backup the
database directories (dbfarm, sql_logs, xquery_logs). After the upgrade,
restore your database, again. Once you have checked and confirmed the
correctness of the restore, you can delete the above renamed/moved/backed-up
old database directories.
For more information, please see the detailed release notes at
http://monetdb.cwi.nl/Development/Releases/Version4.24/
The new release consists of the following packages:
MonetDB Common 1.24.0
MonetDB Clients 1.24.0
MonetDB4 Server 4.24.0
MonetDB4 XQuery 0.24.0
MonetDB5 Server 5.6.0
MonetDB5 SQL 2.24.0
MonetDB5 Geom 0.4.0
monetdb-1.8-jdbc.jar
There is of course also a new "Super Source Ball", a tarball containing the
combined sources.
All this is available on our SourceForge project page. See
https://sourceforge.net/project/showfiles.php?group_id=56967 for the
downloadable files.
Enjoy!
The MonetDB development team.
--
| Dr. Stefan Manegold | mailto:Stefan.Manegold@cwi.nl |
| CWI, P.O.Box 94079 | http://www.cwi.nl/~manegold/ |
| 1090 GB Amsterdam | Tel.: +31 (20) 592-4212 |
| The Netherlands | Fax : +31 (20) 592-4312 |
Hello,
I am trying to write a subquery that can serve as a poor man's median function. However I am running into issues every angle I try and cannot find any details about this in the docs or list archives.
Question 1:
Currently I can calculate the median manually as two steps:
sql> select count(*)/2 from mytable where myIntVal is not null;
+----------------+
| sql_div_count_ |
+================+
| 10 |
+----------------+
sql> select myIntVal as median from mytable order by myIntVal limit 1 offset 10;
+--------+
| median |
+========+
| 1172 |
+--------+
However I cannot figure out how to combine these two queries into one query so that the count/2 can be done together with the query, so that I don't have to manually first get the count and then hard code it into the second query that does the median. How can these be combined?
For instance I tried this which does not work:
sql> select myIntVal as median from mytable order by myIntVal limit 1 offset (select count(*)/2 from mytable where myIntVal is not null);
but this gives me:
!syntax error, unexpected '(', expecting IDENT or sqlINT
So I assume you can not dynamically substitute a subquery for an OFFSET value. But is there then another way to combine these statements into one statement?
Question 2:
Once I get the above resolved, I need to be able to get the median as an aggregate function, so that I can get medians in a group by query for each element in the group. For instance, if there really was a median() aggregate function I would do this:
select median(myIntval), customerid from mytable group by customerid;
However of course there is no median function. So I understand the work around for not having a median function is to get the total rows, sort the rows and then take the row in the middle.
But how can I do that if it is in a group by statement like the above?
What I am trying desperately to avoid is having to make a query to get all the group by results, and then for each result having to do a separate individual query to get its median separately. Please tell me this is possible! :) And if so, how?
Thank you!!
Hello,
I have a start up script as part of a server image that runs the first time a new instance of the server is started.
It performs the following:
# start the monetdb server
/usr/bin/merovingian
# create and start the db
monetdb create mydb
monetdb start mydb
# create tables
mclient -lsql --time --database=mydb < /db/schemas.sql
However the mclient fails to execute the schemas. The consoles shows the following:
successfully created database 'mydb'
database 'bhtg' does not allow connections
However, if I execute the same exact commands from a bash shell once the system is booted, then it works fine and the mclient executes the .sql script and creates the tables as expected.
What could then be causing the error about not allowing connections? The startup script is running with this:
# chkconfig: 345 15 95
So I think the network is fully initialized and operational by the time this runs.
Any advice please? Thank you.