Permalink
Switch branches/tags
Commits on Nov 21, 2013
  1. bump to snapshot

    alexott committed Nov 21, 2013
  2. release 1.4.4

    alexott committed Nov 21, 2013
Commits on Nov 20, 2013
  1. Merge pull request #18 from vadali/master

    alexott committed Nov 20, 2013
    added type annotations
  2. added type annotations

    shlomiv
    shlomiv committed Nov 20, 2013
Commits on Aug 9, 2013
  1. Merge pull request #17 from va1en0k/allow_calculated_params_in_defjob

    alexott committed Aug 9, 2013
    Allow calculated params in defjob (relax args checking)
Commits on Aug 3, 2013
Commits on Jul 14, 2013
  1. release 1.4.3

    alexott committed Jul 14, 2013
  2. Merge pull request #16 from methylene/master

    alexott committed Jul 14, 2013
    bump hadoop version
Commits on Jul 13, 2013
Commits on Mar 18, 2013
  1. Merge pull request #13 from Roxxi/master

    alexott committed Mar 18, 2013
    Hadoop 1.0.3 ~> 1.0.4; Added test dir to compilation path.
Commits on Mar 17, 2013
  1. Merge https://github.com/alexott/clojure-hadoop

    Alex Bahouth
    Alex Bahouth committed Mar 17, 2013
    Conflicts:
    	README.txt
    	project.clj
  2. Hadoop 1.0.3 ~> 1.0.4; Added test dir to compilation path.

    Alex Bahouth
    Alex Bahouth committed Mar 17, 2013
    1.0.4 is the latest version of hadoop. Bumped it up.
    
    If someone were to download and run lein test, it would fail because the test cases are not compiled. Added the directive in the project.clj to compile them so lein test passes.
Commits on Mar 16, 2013
  1. Add files and archives to the DistributedCache via configuraiton.

    Roxxi authored and alexott committed Mar 12, 2013
    This patch enables you to add files and archives to the Distributed Cache via commandline specification and configuration within a job. This is important when mappers and reducers need to refer to configuration or relatively small data sets as part of their execution.
    
    The new file "wordcount7.clj" demonstrates both of these two methods, and provides unit testing. The commandline usage can be tested manually, and the current additional test in the file tests specifying the added test files to the cache for use in the mapper.
    
    The commandline help has been updated, as well as the README.txt with regards to wordcount7.clj.
    
    Conflicts:
    	README.txt
    	project.clj
    	src/clojure_hadoop/config.clj
  2. Merge pull request #11 from Roxxi/arbitrary-command-line-args

    alexott committed Mar 16, 2013
    Specifying key-value pairs for the JobConf via the commandline
Commits on Mar 12, 2013
  1. Merge branch 'adding-files-to-distributed-cache'

    Roxxi committed Mar 12, 2013
    Conflicts:
    	README.txt
    	project.clj
    	src/clojure_hadoop/config.clj
  2. Add files and archives to the DistributedCache via configuraiton.

    Roxxi committed Mar 12, 2013
    This patch enables you to add files and archives to the Distributed Cache via commandline specification and configuration within a job. This is important when mappers and reducers need to refer to configuration or relatively small data sets as part of their execution.
    
    The new file "wordcount7.clj" demonstrates both of these two methods, and provides unit testing. The commandline usage can be tested manually, and the current additional test in the file tests specifying the added test files to the cache for use in the mapper.
    
    The commandline help has been updated, as well as the README.txt with regards to wordcount7.clj.
Commits on Mar 10, 2013
  1. Specifying key-value pairs for the JobConf via the commandline

    Roxxi committed Mar 10, 2013
    With this patch, a user can specify key-value pairs
    via the commandline to be configured in the JobConf.
    Parameters can be specified by placing an 'X' infront
    of the key, followed by a value that will be treated
    as a String in the configuration.
    
    For example the -X parameters below:
    java -cp examples.jar clojure_hadoop.job \
         -job clojure-hadoop.examples.wordcount6/job \
         -input README.txt -output out6 \
         -Xmy.custom.repeat.string hello \
         -Xmy.custom.multiplier 11
    
    will be set as key-value pairs in the JobConf such that
    jobconf.getConfiguration().get("my.custom.repeat.string")
      => "hello"
    jobconf.getConfiguration().get("my.custom.multiplier")
      => "11"
    
    wordcount6.clj provides an written example of such usage, and
    also outlines the use of the :params param that can be specified
    in the job mapping which, to my knowledge, was previously undocumented.
    
    I've also verified that in the case where :params are specified, and
    conflicting key-value pairs are specified over the commandline, the
    commandline will override the values specified in :params, effectively
    allowing the programmer to specify default values for their application.
    
    Added -X param usage description when printing command line help.
Commits on Jun 15, 2012
  1. Merge pull request #6 from mtnygard/master

    alexott committed Jun 15, 2012
    Rev Hadoop dependency to 1.0.3
Commits on Apr 24, 2012
  1. use Hadoop 1.0.2

    alexott committed Apr 24, 2012
Commits on Mar 29, 2012
  1. bump Hadoop version to 1.0.1

    alexott committed Mar 29, 2012
  2. add link to my article

    alexott committed Mar 29, 2012
  3. update README

    alexott committed Mar 29, 2012
Commits on Feb 7, 2012
  1. Merge pull request #4 from davelambert/hadoop-catchup

    alexott committed Feb 7, 2012
    Hadoop 1.0.0 catchup
  2. Update dependencies: Hadoop → 1.0.0, swank-clojure → 1.4.0

    davelambert committed Feb 7, 2012
    Removed the dependency on jackson-mapper-asl, because the actual user
    (Hadoop 1.0.0) brings it in itself.  Dropped the explicit Apache
    repository reference: Hadoop 1.0.0 is available from maven.org.
  3. Update docstrings for Hadoop 0.20 changes

    davelambert committed Feb 7, 2012
    Hadoop's 0.20 API replaced Collectors with Contexts, JobConf with Job.
Commits on Jan 24, 2012
  1. Make working after merge

    alexott committed Jan 24, 2012
    Hadoop 0.20.205.0 exists only in Apache's maven repository
  2. minor changes after merge

    alexott committed Jan 24, 2012
    - change repo URL
    - increase swank-clojure version
  3. Merge pull request #3 from cmiles74/master

    alexott committed Jan 24, 2012
    Many Minor Changes: Works with Clojure 1.3 + Hadoop 0.20.205.0
Commits on Jan 23, 2012
  1. Merged "project.clj" from eslick.

    cmiles74 committed Jan 23, 2012
    Increased the version number to 1.4.
    Changed Hadoop dependency from 0.20.205.0 to 0.20.2. This let me
    remove the jackson dependency.
    Added development dependency on swank-clojure.
  2. Merged changes from eslick.

    cmiles74 committed Jan 23, 2012
  3. Made formerly "defvar-" vars private.

    cmiles74 committed Jan 23, 2012
    Added ":private true" to the meta-data on vars that had been declared
    with "defvar-".