Fix uploaded to -proposed, pending SRU team review.
** Description changed:
libcommons-csv-java depends on default-jre, which pulls a full JRE. It
is used by solr, which is installed on server environments. Fixing this
library to only depend on "default-jre-headless | java2-runtime-
headless" would avoid installing X libraries when installing solr.
+
+ == SRU report ==
+
+ Impact:
+ Due to this bug, Apache Solr (enterprise search) cannot be installed via the
repo packages on Ubuntu Server without installing a huge number of desktop
dependencies, including a wide array of media codecs and other cruft.
+
+ Dev branch fix:
+ This was fixed in Maverick by depending on headless runtimes instead.
+
+ Minimal patch:
+ -Depends: openjdk-6-jre | java2-runtime
+ +Depends: openjdk-6-jre-headless | java2-runtime-headless
+
+ TEST CASE:
+ On a new Lucid server, try:
+ apt-get install libcommons-csv-java
+ On an affected setup, it tries to install 70+ packages, including
"libpango1.0-0"
+ On a fixed setup, it should install less than 20 packages.
+
+ Regression potential;
+ This only affects reverse dependencies of libcommons-csv-java (i.e.
solr-common) and does not affect already-installed setups, so the regression
potential is very limited.
** Changed in: commons-csv (Ubuntu Lucid)
Status: In Progress => Confirmed
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/598017
Title:
libcommons-csv-java should depend on a headless runtime
To manage notifications about this bug go to:
https://bugs.launchpad.net/server-papercuts/+bug/598017/+subscriptions
--
ubuntu-bugs mailing list
[email protected]
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs