Some of the tools i'm using in day-to-day development - Residing in ~/bin, along with some installations, as follows:
colcount
- Converts standard input lines to their respective column countsq
- See https://github.com/harelba/qsecs2date
- Converts timestamps to human date-time stringssigma
- Sums up all the numbers from stdinxvsy
- Draws an XY graph of x values vs y values - Getting pairs of x,y lines from standard inputcategoryvsy
- Draws a bar graph for category names vs y values - Getting pairs of category-name,y lines from standard inputdisplay
- Uses disper in order to control multiple monitor setups for the same machine (docked, undocked,home etc.)disable-company-settings/enable-company-settings
- Allow quick switching between a company provided maven settings.xml and a default onesvn-diff-meld
- Wrapper so svn would work with meldmvn-test-multiple
- Run a test using maven multiple times, each time saving the log file and the test results to separate filesgroupby
- Count lines per distinct data value in column number colnumber.hive-table-fields-to-pig-load-command
- Converts the output of a hive DESCRIBE command into a Pig LOAD command
zkfind
- get a zookeeper path and lists all nodes recursively, similar to linux findzkget
- gets the data of a specific node in zookeeperzkrmr
- Performs a recusrive delete of zookeeper nodes (this capability is included in zkCli if you're using zookeeper >3.4)zkgrep
- Finds a substring in the data of zookeeper nodes
- apache-maven-2.2/
- apache-maven-3/
- apache-maven/
- apache-tomcat-2nd-instance/
- apache-tomcat/
- eclipse-plugins/
- eclipse/
- hadoop-0.20.2-cdh3u3/
- hive-0.8.1/
- sqoop-1.3.0-cdh3u3/
- storm-0.7.2/
- kafka-0.7.0-incubating-src/
- pig-0.10.0/
- apache-cassandra-1.1.0/
- apache-activemq-5.4.3/
- zookeeper-3.3.5/