http://git-wip-us.apache.org/repos/asf/hbase-site/blob/75eda567/apidocs/src-html/org/apache/hadoop/hbase/mapreduce/ImportTsv.html
----------------------------------------------------------------------
diff --git a/apidocs/src-html/org/apache/hadoop/hbase/mapreduce/ImportTsv.html
b/apidocs/src-html/org/apache/hadoop/hbase/mapreduce/ImportTsv.html
index 7e1985c..4b7d070 100644
--- a/apidocs/src-html/org/apache/hadoop/hbase/mapreduce/ImportTsv.html
+++ b/apidocs/src-html/org/apache/hadoop/hbase/mapreduce/ImportTsv.html
@@ -115,7 +115,7 @@
<span class="sourceLineNo">107</span> * If table didn't exist and was
created in dry-run mode, this flag is<a name="line.107"></a>
<span class="sourceLineNo">108</span> * flipped to delete it when MR ends.<a
name="line.108"></a>
<span class="sourceLineNo">109</span> */<a name="line.109"></a>
-<span class="sourceLineNo">110</span> private static boolean
dryRunTableCreated;<a name="line.110"></a>
+<span class="sourceLineNo">110</span> private static boolean
DRY_RUN_TABLE_CREATED;<a name="line.110"></a>
<span class="sourceLineNo">111</span><a name="line.111"></a>
<span class="sourceLineNo">112</span> public static class TsvParser {<a
name="line.112"></a>
<span class="sourceLineNo">113</span> /**<a name="line.113"></a>
@@ -483,312 +483,320 @@
<span class="sourceLineNo">475</span><a name="line.475"></a>
<span class="sourceLineNo">476</span> // See if a non-default Mapper
was set<a name="line.476"></a>
<span class="sourceLineNo">477</span> String mapperClassName =
conf.get(MAPPER_CONF_KEY);<a name="line.477"></a>
-<span class="sourceLineNo">478</span> Class mapperClass =<a
name="line.478"></a>
-<span class="sourceLineNo">479</span> mapperClassName != null ?
Class.forName(mapperClassName) : DEFAULT_MAPPER;<a name="line.479"></a>
-<span class="sourceLineNo">480</span><a name="line.480"></a>
-<span class="sourceLineNo">481</span> TableName tableName =
TableName.valueOf(args[0]);<a name="line.481"></a>
-<span class="sourceLineNo">482</span> Path inputDir = new
Path(args[1]);<a name="line.482"></a>
-<span class="sourceLineNo">483</span> String jobName =
conf.get(JOB_NAME_CONF_KEY,NAME + "_" + tableName.getNameAsString());<a
name="line.483"></a>
-<span class="sourceLineNo">484</span> job = Job.getInstance(conf,
jobName);<a name="line.484"></a>
-<span class="sourceLineNo">485</span>
job.setJarByClass(mapperClass);<a name="line.485"></a>
-<span class="sourceLineNo">486</span>
FileInputFormat.setInputPaths(job, inputDir);<a name="line.486"></a>
-<span class="sourceLineNo">487</span>
job.setInputFormatClass(TextInputFormat.class);<a name="line.487"></a>
-<span class="sourceLineNo">488</span>
job.setMapperClass(mapperClass);<a name="line.488"></a>
-<span class="sourceLineNo">489</span>
job.setMapOutputKeyClass(ImmutableBytesWritable.class);<a name="line.489"></a>
-<span class="sourceLineNo">490</span> String hfileOutPath =
conf.get(BULK_OUTPUT_CONF_KEY);<a name="line.490"></a>
-<span class="sourceLineNo">491</span> String[] columns =
conf.getStrings(COLUMNS_CONF_KEY);<a name="line.491"></a>
-<span class="sourceLineNo">492</span>
if(StringUtils.isNotEmpty(conf.get(CREDENTIALS_LOCATION))) {<a
name="line.492"></a>
-<span class="sourceLineNo">493</span> String fileLoc =
conf.get(CREDENTIALS_LOCATION);<a name="line.493"></a>
-<span class="sourceLineNo">494</span> Credentials cred =
Credentials.readTokenStorageFile(new File(fileLoc), conf);<a
name="line.494"></a>
-<span class="sourceLineNo">495</span>
job.getCredentials().addAll(cred);<a name="line.495"></a>
-<span class="sourceLineNo">496</span> }<a name="line.496"></a>
-<span class="sourceLineNo">497</span><a name="line.497"></a>
-<span class="sourceLineNo">498</span> if (hfileOutPath != null) {<a
name="line.498"></a>
-<span class="sourceLineNo">499</span> if
(!admin.tableExists(tableName)) {<a name="line.499"></a>
-<span class="sourceLineNo">500</span> LOG.warn(format("Table '%s'
does not exist.", tableName));<a name="line.500"></a>
-<span class="sourceLineNo">501</span> if
("yes".equalsIgnoreCase(conf.get(CREATE_TABLE_CONF_KEY, "yes"))) {<a
name="line.501"></a>
-<span class="sourceLineNo">502</span> // TODO: this is
backwards. Instead of depending on the existence of a table,<a
name="line.502"></a>
-<span class="sourceLineNo">503</span> // create a sane splits
file for HFileOutputFormat based on data sampling.<a name="line.503"></a>
-<span class="sourceLineNo">504</span> createTable(admin,
tableName, columns);<a name="line.504"></a>
-<span class="sourceLineNo">505</span> if (isDryRun) {<a
name="line.505"></a>
-<span class="sourceLineNo">506</span> LOG.warn("Dry run:
Table will be deleted at end of dry run.");<a name="line.506"></a>
-<span class="sourceLineNo">507</span> dryRunTableCreated =
true;<a name="line.507"></a>
+<span class="sourceLineNo">478</span> Class mapperClass =
mapperClassName != null? Class.forName(mapperClassName): DEFAULT_MAPPER;<a
name="line.478"></a>
+<span class="sourceLineNo">479</span><a name="line.479"></a>
+<span class="sourceLineNo">480</span> TableName tableName =
TableName.valueOf(args[0]);<a name="line.480"></a>
+<span class="sourceLineNo">481</span> Path inputDir = new
Path(args[1]);<a name="line.481"></a>
+<span class="sourceLineNo">482</span> String jobName =
conf.get(JOB_NAME_CONF_KEY,NAME + "_" + tableName.getNameAsString());<a
name="line.482"></a>
+<span class="sourceLineNo">483</span> job = Job.getInstance(conf,
jobName);<a name="line.483"></a>
+<span class="sourceLineNo">484</span> job.setJarByClass(mapperClass);<a
name="line.484"></a>
+<span class="sourceLineNo">485</span>
FileInputFormat.setInputPaths(job, inputDir);<a name="line.485"></a>
+<span class="sourceLineNo">486</span>
job.setInputFormatClass(TextInputFormat.class);<a name="line.486"></a>
+<span class="sourceLineNo">487</span>
job.setMapperClass(mapperClass);<a name="line.487"></a>
+<span class="sourceLineNo">488</span>
job.setMapOutputKeyClass(ImmutableBytesWritable.class);<a name="line.488"></a>
+<span class="sourceLineNo">489</span> String hfileOutPath =
conf.get(BULK_OUTPUT_CONF_KEY);<a name="line.489"></a>
+<span class="sourceLineNo">490</span> String[] columns =
conf.getStrings(COLUMNS_CONF_KEY);<a name="line.490"></a>
+<span class="sourceLineNo">491</span>
if(StringUtils.isNotEmpty(conf.get(CREDENTIALS_LOCATION))) {<a
name="line.491"></a>
+<span class="sourceLineNo">492</span> String fileLoc =
conf.get(CREDENTIALS_LOCATION);<a name="line.492"></a>
+<span class="sourceLineNo">493</span> Credentials cred =
Credentials.readTokenStorageFile(new File(fileLoc), conf);<a
name="line.493"></a>
+<span class="sourceLineNo">494</span>
job.getCredentials().addAll(cred);<a name="line.494"></a>
+<span class="sourceLineNo">495</span> }<a name="line.495"></a>
+<span class="sourceLineNo">496</span><a name="line.496"></a>
+<span class="sourceLineNo">497</span> if (hfileOutPath != null) {<a
name="line.497"></a>
+<span class="sourceLineNo">498</span> if
(!admin.tableExists(tableName)) {<a name="line.498"></a>
+<span class="sourceLineNo">499</span> LOG.warn(format("Table '%s'
does not exist.", tableName));<a name="line.499"></a>
+<span class="sourceLineNo">500</span> if
("yes".equalsIgnoreCase(conf.get(CREATE_TABLE_CONF_KEY, "yes"))) {<a
name="line.500"></a>
+<span class="sourceLineNo">501</span> // TODO: this is backwards.
Instead of depending on the existence of a table,<a name="line.501"></a>
+<span class="sourceLineNo">502</span> // create a sane splits
file for HFileOutputFormat based on data sampling.<a name="line.502"></a>
+<span class="sourceLineNo">503</span> createTable(admin,
tableName, columns);<a name="line.503"></a>
+<span class="sourceLineNo">504</span> if (isDryRun) {<a
name="line.504"></a>
+<span class="sourceLineNo">505</span> LOG.warn("Dry run: Table
will be deleted at end of dry run.");<a name="line.505"></a>
+<span class="sourceLineNo">506</span> synchronized
(ImportTsv.class) {<a name="line.506"></a>
+<span class="sourceLineNo">507</span> DRY_RUN_TABLE_CREATED =
true;<a name="line.507"></a>
<span class="sourceLineNo">508</span> }<a name="line.508"></a>
-<span class="sourceLineNo">509</span> } else {<a
name="line.509"></a>
-<span class="sourceLineNo">510</span> String errorMsg =<a
name="line.510"></a>
-<span class="sourceLineNo">511</span> format("Table '%s'
does not exist and '%s' is set to no.", tableName,<a name="line.511"></a>
-<span class="sourceLineNo">512</span>
CREATE_TABLE_CONF_KEY);<a name="line.512"></a>
-<span class="sourceLineNo">513</span> LOG.error(errorMsg);<a
name="line.513"></a>
-<span class="sourceLineNo">514</span> throw new
TableNotFoundException(errorMsg);<a name="line.514"></a>
-<span class="sourceLineNo">515</span> }<a name="line.515"></a>
+<span class="sourceLineNo">509</span> }<a name="line.509"></a>
+<span class="sourceLineNo">510</span> } else {<a
name="line.510"></a>
+<span class="sourceLineNo">511</span> String errorMsg =<a
name="line.511"></a>
+<span class="sourceLineNo">512</span> format("Table '%s' does
not exist and '%s' is set to no.", tableName,<a name="line.512"></a>
+<span class="sourceLineNo">513</span>
CREATE_TABLE_CONF_KEY);<a name="line.513"></a>
+<span class="sourceLineNo">514</span> LOG.error(errorMsg);<a
name="line.514"></a>
+<span class="sourceLineNo">515</span> throw new
TableNotFoundException(errorMsg);<a name="line.515"></a>
<span class="sourceLineNo">516</span> }<a name="line.516"></a>
-<span class="sourceLineNo">517</span> try (Table table =
connection.getTable(tableName);<a name="line.517"></a>
-<span class="sourceLineNo">518</span> RegionLocator
regionLocator = connection.getRegionLocator(tableName)) {<a name="line.518"></a>
-<span class="sourceLineNo">519</span> boolean noStrict =
conf.getBoolean(NO_STRICT_COL_FAMILY, false);<a name="line.519"></a>
-<span class="sourceLineNo">520</span> // if no.strict is false
then check column family<a name="line.520"></a>
-<span class="sourceLineNo">521</span> if(!noStrict) {<a
name="line.521"></a>
-<span class="sourceLineNo">522</span> ArrayList<String>
unmatchedFamilies = new ArrayList<String>();<a name="line.522"></a>
-<span class="sourceLineNo">523</span> Set<String> cfSet =
getColumnFamilies(columns);<a name="line.523"></a>
-<span class="sourceLineNo">524</span> HTableDescriptor tDesc =
table.getTableDescriptor();<a name="line.524"></a>
-<span class="sourceLineNo">525</span> for (String cf : cfSet)
{<a name="line.525"></a>
-<span class="sourceLineNo">526</span>
if(tDesc.getFamily(Bytes.toBytes(cf)) == null) {<a name="line.526"></a>
-<span class="sourceLineNo">527</span>
unmatchedFamilies.add(cf);<a name="line.527"></a>
-<span class="sourceLineNo">528</span> }<a name="line.528"></a>
+<span class="sourceLineNo">517</span> }<a name="line.517"></a>
+<span class="sourceLineNo">518</span> try (Table table =
connection.getTable(tableName);<a name="line.518"></a>
+<span class="sourceLineNo">519</span> RegionLocator regionLocator
= connection.getRegionLocator(tableName)) {<a name="line.519"></a>
+<span class="sourceLineNo">520</span> boolean noStrict =
conf.getBoolean(NO_STRICT_COL_FAMILY, false);<a name="line.520"></a>
+<span class="sourceLineNo">521</span> // if no.strict is false then
check column family<a name="line.521"></a>
+<span class="sourceLineNo">522</span> if(!noStrict) {<a
name="line.522"></a>
+<span class="sourceLineNo">523</span> ArrayList<String>
unmatchedFamilies = new ArrayList<String>();<a name="line.523"></a>
+<span class="sourceLineNo">524</span> Set<String> cfSet =
getColumnFamilies(columns);<a name="line.524"></a>
+<span class="sourceLineNo">525</span> HTableDescriptor tDesc =
table.getTableDescriptor();<a name="line.525"></a>
+<span class="sourceLineNo">526</span> for (String cf : cfSet) {<a
name="line.526"></a>
+<span class="sourceLineNo">527</span>
if(tDesc.getFamily(Bytes.toBytes(cf)) == null) {<a name="line.527"></a>
+<span class="sourceLineNo">528</span>
unmatchedFamilies.add(cf);<a name="line.528"></a>
<span class="sourceLineNo">529</span> }<a name="line.529"></a>
-<span class="sourceLineNo">530</span>
if(unmatchedFamilies.size() > 0) {<a name="line.530"></a>
-<span class="sourceLineNo">531</span> ArrayList<String>
familyNames = new ArrayList<String>();<a name="line.531"></a>
-<span class="sourceLineNo">532</span> for (HColumnDescriptor
family : table.getTableDescriptor().getFamilies()) {<a name="line.532"></a>
-<span class="sourceLineNo">533</span>
familyNames.add(family.getNameAsString());<a name="line.533"></a>
-<span class="sourceLineNo">534</span> }<a name="line.534"></a>
-<span class="sourceLineNo">535</span> String msg =<a
name="line.535"></a>
-<span class="sourceLineNo">536</span> "Column Families "
+ unmatchedFamilies + " specified in " + COLUMNS_CONF_KEY<a name="line.536"></a>
-<span class="sourceLineNo">537</span> + " does not match
with any of the table " + tableName<a name="line.537"></a>
-<span class="sourceLineNo">538</span> + " column families
" + familyNames + ".\n"<a name="line.538"></a>
-<span class="sourceLineNo">539</span> + "To disable
column family check, use -D" + NO_STRICT_COL_FAMILY<a name="line.539"></a>
-<span class="sourceLineNo">540</span> + "=true.\n";<a
name="line.540"></a>
-<span class="sourceLineNo">541</span> usage(msg);<a
name="line.541"></a>
-<span class="sourceLineNo">542</span> System.exit(-1);<a
name="line.542"></a>
-<span class="sourceLineNo">543</span> }<a name="line.543"></a>
+<span class="sourceLineNo">530</span> }<a name="line.530"></a>
+<span class="sourceLineNo">531</span> if(unmatchedFamilies.size()
> 0) {<a name="line.531"></a>
+<span class="sourceLineNo">532</span> ArrayList<String>
familyNames = new ArrayList<String>();<a name="line.532"></a>
+<span class="sourceLineNo">533</span> for (HColumnDescriptor
family : table.getTableDescriptor().getFamilies()) {<a name="line.533"></a>
+<span class="sourceLineNo">534</span>
familyNames.add(family.getNameAsString());<a name="line.534"></a>
+<span class="sourceLineNo">535</span> }<a name="line.535"></a>
+<span class="sourceLineNo">536</span> String msg =<a
name="line.536"></a>
+<span class="sourceLineNo">537</span> "Column Families " +
unmatchedFamilies + " specified in " + COLUMNS_CONF_KEY<a name="line.537"></a>
+<span class="sourceLineNo">538</span> + " does not match
with any of the table " + tableName<a name="line.538"></a>
+<span class="sourceLineNo">539</span> + " column families "
+ familyNames + ".\n"<a name="line.539"></a>
+<span class="sourceLineNo">540</span> + "To disable column
family check, use -D" + NO_STRICT_COL_FAMILY<a name="line.540"></a>
+<span class="sourceLineNo">541</span> + "=true.\n";<a
name="line.541"></a>
+<span class="sourceLineNo">542</span> usage(msg);<a
name="line.542"></a>
+<span class="sourceLineNo">543</span> System.exit(-1);<a
name="line.543"></a>
<span class="sourceLineNo">544</span> }<a name="line.544"></a>
-<span class="sourceLineNo">545</span> if
(mapperClass.equals(TsvImporterTextMapper.class)) {<a name="line.545"></a>
-<span class="sourceLineNo">546</span>
job.setMapOutputValueClass(Text.class);<a name="line.546"></a>
-<span class="sourceLineNo">547</span>
job.setReducerClass(TextSortReducer.class);<a name="line.547"></a>
-<span class="sourceLineNo">548</span> } else {<a
name="line.548"></a>
-<span class="sourceLineNo">549</span>
job.setMapOutputValueClass(Put.class);<a name="line.549"></a>
-<span class="sourceLineNo">550</span>
job.setCombinerClass(PutCombiner.class);<a name="line.550"></a>
-<span class="sourceLineNo">551</span>
job.setReducerClass(PutSortReducer.class);<a name="line.551"></a>
-<span class="sourceLineNo">552</span> }<a name="line.552"></a>
-<span class="sourceLineNo">553</span> if (!isDryRun) {<a
name="line.553"></a>
-<span class="sourceLineNo">554</span> Path outputDir = new
Path(hfileOutPath);<a name="line.554"></a>
-<span class="sourceLineNo">555</span>
FileOutputFormat.setOutputPath(job, outputDir);<a name="line.555"></a>
-<span class="sourceLineNo">556</span>
HFileOutputFormat2.configureIncrementalLoad(job, table.getTableDescriptor(),<a
name="line.556"></a>
-<span class="sourceLineNo">557</span> regionLocator);<a
name="line.557"></a>
-<span class="sourceLineNo">558</span> }<a name="line.558"></a>
+<span class="sourceLineNo">545</span> }<a name="line.545"></a>
+<span class="sourceLineNo">546</span> if
(mapperClass.equals(TsvImporterTextMapper.class)) {<a name="line.546"></a>
+<span class="sourceLineNo">547</span>
job.setMapOutputValueClass(Text.class);<a name="line.547"></a>
+<span class="sourceLineNo">548</span>
job.setReducerClass(TextSortReducer.class);<a name="line.548"></a>
+<span class="sourceLineNo">549</span> } else {<a
name="line.549"></a>
+<span class="sourceLineNo">550</span>
job.setMapOutputValueClass(Put.class);<a name="line.550"></a>
+<span class="sourceLineNo">551</span>
job.setCombinerClass(PutCombiner.class);<a name="line.551"></a>
+<span class="sourceLineNo">552</span>
job.setReducerClass(PutSortReducer.class);<a name="line.552"></a>
+<span class="sourceLineNo">553</span> }<a name="line.553"></a>
+<span class="sourceLineNo">554</span> if (!isDryRun) {<a
name="line.554"></a>
+<span class="sourceLineNo">555</span> Path outputDir = new
Path(hfileOutPath);<a name="line.555"></a>
+<span class="sourceLineNo">556</span>
FileOutputFormat.setOutputPath(job, outputDir);<a name="line.556"></a>
+<span class="sourceLineNo">557</span>
HFileOutputFormat2.configureIncrementalLoad(job, table.getTableDescriptor(),<a
name="line.557"></a>
+<span class="sourceLineNo">558</span> regionLocator);<a
name="line.558"></a>
<span class="sourceLineNo">559</span> }<a name="line.559"></a>
-<span class="sourceLineNo">560</span> } else {<a name="line.560"></a>
-<span class="sourceLineNo">561</span> if
(!admin.tableExists(tableName)) {<a name="line.561"></a>
-<span class="sourceLineNo">562</span> String errorMsg =
format("Table '%s' does not exist.", tableName);<a name="line.562"></a>
-<span class="sourceLineNo">563</span> LOG.error(errorMsg);<a
name="line.563"></a>
-<span class="sourceLineNo">564</span> throw new
TableNotFoundException(errorMsg);<a name="line.564"></a>
-<span class="sourceLineNo">565</span> }<a name="line.565"></a>
-<span class="sourceLineNo">566</span> if
(mapperClass.equals(TsvImporterTextMapper.class)) {<a name="line.566"></a>
-<span class="sourceLineNo">567</span>
usage(TsvImporterTextMapper.class.toString()<a name="line.567"></a>
-<span class="sourceLineNo">568</span> + " should not be used
for non bulkloading case. use "<a name="line.568"></a>
-<span class="sourceLineNo">569</span> +
TsvImporterMapper.class.toString()<a name="line.569"></a>
-<span class="sourceLineNo">570</span> + " or custom mapper
whose value type is Put.");<a name="line.570"></a>
-<span class="sourceLineNo">571</span> System.exit(-1);<a
name="line.571"></a>
-<span class="sourceLineNo">572</span> }<a name="line.572"></a>
-<span class="sourceLineNo">573</span> if (!isDryRun) {<a
name="line.573"></a>
-<span class="sourceLineNo">574</span> // No reducers. Just write
straight to table. Call initTableReducerJob<a name="line.574"></a>
-<span class="sourceLineNo">575</span> // to set up the
TableOutputFormat.<a name="line.575"></a>
-<span class="sourceLineNo">576</span>
TableMapReduceUtil.initTableReducerJob(tableName.getNameAsString(), null,
job);<a name="line.576"></a>
-<span class="sourceLineNo">577</span> }<a name="line.577"></a>
-<span class="sourceLineNo">578</span> job.setNumReduceTasks(0);<a
name="line.578"></a>
-<span class="sourceLineNo">579</span> }<a name="line.579"></a>
-<span class="sourceLineNo">580</span> if (isDryRun) {<a
name="line.580"></a>
-<span class="sourceLineNo">581</span>
job.setOutputFormatClass(NullOutputFormat.class);<a name="line.581"></a>
-<span class="sourceLineNo">582</span>
job.getConfiguration().setStrings("io.serializations",<a name="line.582"></a>
-<span class="sourceLineNo">583</span>
job.getConfiguration().get("io.serializations"),<a name="line.583"></a>
-<span class="sourceLineNo">584</span>
MutationSerialization.class.getName(), ResultSerialization.class.getName(),<a
name="line.584"></a>
-<span class="sourceLineNo">585</span>
KeyValueSerialization.class.getName());<a name="line.585"></a>
-<span class="sourceLineNo">586</span> }<a name="line.586"></a>
-<span class="sourceLineNo">587</span>
TableMapReduceUtil.addDependencyJars(job);<a name="line.587"></a>
-<span class="sourceLineNo">588</span>
TableMapReduceUtil.addDependencyJars(job.getConfiguration(),<a
name="line.588"></a>
-<span class="sourceLineNo">589</span>
com.google.common.base.Function.class /* Guava used by TsvParser */);<a
name="line.589"></a>
-<span class="sourceLineNo">590</span> }<a name="line.590"></a>
-<span class="sourceLineNo">591</span> }<a name="line.591"></a>
-<span class="sourceLineNo">592</span> return job;<a name="line.592"></a>
-<span class="sourceLineNo">593</span> }<a name="line.593"></a>
-<span class="sourceLineNo">594</span><a name="line.594"></a>
-<span class="sourceLineNo">595</span> private static void createTable(Admin
admin, TableName tableName, String[] columns)<a name="line.595"></a>
-<span class="sourceLineNo">596</span> throws IOException {<a
name="line.596"></a>
-<span class="sourceLineNo">597</span> HTableDescriptor htd = new
HTableDescriptor(tableName);<a name="line.597"></a>
-<span class="sourceLineNo">598</span> Set<String> cfSet =
getColumnFamilies(columns);<a name="line.598"></a>
-<span class="sourceLineNo">599</span> for (String cf : cfSet) {<a
name="line.599"></a>
-<span class="sourceLineNo">600</span> HColumnDescriptor hcd = new
HColumnDescriptor(Bytes.toBytes(cf));<a name="line.600"></a>
-<span class="sourceLineNo">601</span> htd.addFamily(hcd);<a
name="line.601"></a>
-<span class="sourceLineNo">602</span> }<a name="line.602"></a>
-<span class="sourceLineNo">603</span> LOG.warn(format("Creating table '%s'
with '%s' columns and default descriptors.",<a name="line.603"></a>
-<span class="sourceLineNo">604</span> tableName, cfSet));<a
name="line.604"></a>
-<span class="sourceLineNo">605</span> admin.createTable(htd);<a
name="line.605"></a>
-<span class="sourceLineNo">606</span> }<a name="line.606"></a>
-<span class="sourceLineNo">607</span><a name="line.607"></a>
-<span class="sourceLineNo">608</span> private static void
deleteTable(Configuration conf, String[] args) {<a name="line.608"></a>
-<span class="sourceLineNo">609</span> TableName tableName =
TableName.valueOf(args[0]);<a name="line.609"></a>
-<span class="sourceLineNo">610</span> try (Connection connection =
ConnectionFactory.createConnection(conf);<a name="line.610"></a>
-<span class="sourceLineNo">611</span> Admin admin =
connection.getAdmin()) {<a name="line.611"></a>
-<span class="sourceLineNo">612</span> try {<a name="line.612"></a>
-<span class="sourceLineNo">613</span> admin.disableTable(tableName);<a
name="line.613"></a>
-<span class="sourceLineNo">614</span> } catch (TableNotEnabledException
e) {<a name="line.614"></a>
-<span class="sourceLineNo">615</span> LOG.debug("Dry mode: Table: " +
tableName + " already disabled, so just deleting it.");<a name="line.615"></a>
-<span class="sourceLineNo">616</span> }<a name="line.616"></a>
-<span class="sourceLineNo">617</span> admin.deleteTable(tableName);<a
name="line.617"></a>
-<span class="sourceLineNo">618</span> } catch (IOException e) {<a
name="line.618"></a>
-<span class="sourceLineNo">619</span> LOG.error(format("***Dry run:
Failed to delete table '%s'.***\n%s", tableName, e.toString()));<a
name="line.619"></a>
-<span class="sourceLineNo">620</span> return;<a name="line.620"></a>
-<span class="sourceLineNo">621</span> }<a name="line.621"></a>
-<span class="sourceLineNo">622</span> LOG.info(format("Dry run: Deleted
table '%s'.", tableName));<a name="line.622"></a>
-<span class="sourceLineNo">623</span> }<a name="line.623"></a>
-<span class="sourceLineNo">624</span><a name="line.624"></a>
-<span class="sourceLineNo">625</span> private static Set<String>
getColumnFamilies(String[] columns) {<a name="line.625"></a>
-<span class="sourceLineNo">626</span> Set<String> cfSet = new
HashSet<String>();<a name="line.626"></a>
-<span class="sourceLineNo">627</span> for (String aColumn : columns) {<a
name="line.627"></a>
-<span class="sourceLineNo">628</span> if
(TsvParser.ROWKEY_COLUMN_SPEC.equals(aColumn)<a name="line.628"></a>
-<span class="sourceLineNo">629</span> ||
TsvParser.TIMESTAMPKEY_COLUMN_SPEC.equals(aColumn)<a name="line.629"></a>
-<span class="sourceLineNo">630</span> ||
TsvParser.CELL_VISIBILITY_COLUMN_SPEC.equals(aColumn)<a name="line.630"></a>
-<span class="sourceLineNo">631</span> ||
TsvParser.CELL_TTL_COLUMN_SPEC.equals(aColumn)<a name="line.631"></a>
-<span class="sourceLineNo">632</span> ||
TsvParser.ATTRIBUTES_COLUMN_SPEC.equals(aColumn))<a name="line.632"></a>
-<span class="sourceLineNo">633</span> continue;<a name="line.633"></a>
-<span class="sourceLineNo">634</span> // we are only concerned with the
first one (in case this is a cf:cq)<a name="line.634"></a>
-<span class="sourceLineNo">635</span> cfSet.add(aColumn.split(":",
2)[0]);<a name="line.635"></a>
-<span class="sourceLineNo">636</span> }<a name="line.636"></a>
-<span class="sourceLineNo">637</span> return cfSet;<a name="line.637"></a>
-<span class="sourceLineNo">638</span> }<a name="line.638"></a>
-<span class="sourceLineNo">639</span><a name="line.639"></a>
-<span class="sourceLineNo">640</span> /*<a name="line.640"></a>
-<span class="sourceLineNo">641</span> * @param errorMsg Error message. Can
be null.<a name="line.641"></a>
-<span class="sourceLineNo">642</span> */<a name="line.642"></a>
-<span class="sourceLineNo">643</span> private static void usage(final String
errorMsg) {<a name="line.643"></a>
-<span class="sourceLineNo">644</span> if (errorMsg != null &&
errorMsg.length() > 0) {<a name="line.644"></a>
-<span class="sourceLineNo">645</span> System.err.println("ERROR: " +
errorMsg);<a name="line.645"></a>
-<span class="sourceLineNo">646</span> }<a name="line.646"></a>
-<span class="sourceLineNo">647</span> String usage =<a name="line.647"></a>
-<span class="sourceLineNo">648</span> "Usage: " + NAME + " -D"+
COLUMNS_CONF_KEY + "=a,b,c <tablename> <inputdir>\n" +<a
name="line.648"></a>
-<span class="sourceLineNo">649</span> "\n" +<a name="line.649"></a>
-<span class="sourceLineNo">650</span> "Imports the given input directory
of TSV data into the specified table.\n" +<a name="line.650"></a>
+<span class="sourceLineNo">560</span> }<a name="line.560"></a>
+<span class="sourceLineNo">561</span> } else {<a name="line.561"></a>
+<span class="sourceLineNo">562</span> if
(!admin.tableExists(tableName)) {<a name="line.562"></a>
+<span class="sourceLineNo">563</span> String errorMsg =
format("Table '%s' does not exist.", tableName);<a name="line.563"></a>
+<span class="sourceLineNo">564</span> LOG.error(errorMsg);<a
name="line.564"></a>
+<span class="sourceLineNo">565</span> throw new
TableNotFoundException(errorMsg);<a name="line.565"></a>
+<span class="sourceLineNo">566</span> }<a name="line.566"></a>
+<span class="sourceLineNo">567</span> if
(mapperClass.equals(TsvImporterTextMapper.class)) {<a name="line.567"></a>
+<span class="sourceLineNo">568</span>
usage(TsvImporterTextMapper.class.toString()<a name="line.568"></a>
+<span class="sourceLineNo">569</span> + " should not be used
for non bulkloading case. use "<a name="line.569"></a>
+<span class="sourceLineNo">570</span> +
TsvImporterMapper.class.toString()<a name="line.570"></a>
+<span class="sourceLineNo">571</span> + " or custom mapper
whose value type is Put.");<a name="line.571"></a>
+<span class="sourceLineNo">572</span> System.exit(-1);<a
name="line.572"></a>
+<span class="sourceLineNo">573</span> }<a name="line.573"></a>
+<span class="sourceLineNo">574</span> if (!isDryRun) {<a
name="line.574"></a>
+<span class="sourceLineNo">575</span> // No reducers. Just write
straight to table. Call initTableReducerJob<a name="line.575"></a>
+<span class="sourceLineNo">576</span> // to set up the
TableOutputFormat.<a name="line.576"></a>
+<span class="sourceLineNo">577</span>
TableMapReduceUtil.initTableReducerJob(tableName.getNameAsString(), null,
job);<a name="line.577"></a>
+<span class="sourceLineNo">578</span> }<a name="line.578"></a>
+<span class="sourceLineNo">579</span> job.setNumReduceTasks(0);<a
name="line.579"></a>
+<span class="sourceLineNo">580</span> }<a name="line.580"></a>
+<span class="sourceLineNo">581</span> if (isDryRun) {<a
name="line.581"></a>
+<span class="sourceLineNo">582</span>
job.setOutputFormatClass(NullOutputFormat.class);<a name="line.582"></a>
+<span class="sourceLineNo">583</span>
job.getConfiguration().setStrings("io.serializations",<a name="line.583"></a>
+<span class="sourceLineNo">584</span>
job.getConfiguration().get("io.serializations"),<a name="line.584"></a>
+<span class="sourceLineNo">585</span>
MutationSerialization.class.getName(), ResultSerialization.class.getName(),<a
name="line.585"></a>
+<span class="sourceLineNo">586</span>
KeyValueSerialization.class.getName());<a name="line.586"></a>
+<span class="sourceLineNo">587</span> }<a name="line.587"></a>
+<span class="sourceLineNo">588</span>
TableMapReduceUtil.addDependencyJars(job);<a name="line.588"></a>
+<span class="sourceLineNo">589</span>
TableMapReduceUtil.addDependencyJars(job.getConfiguration(),<a
name="line.589"></a>
+<span class="sourceLineNo">590</span>
com.google.common.base.Function.class /* Guava used by TsvParser */);<a
name="line.590"></a>
+<span class="sourceLineNo">591</span> }<a name="line.591"></a>
+<span class="sourceLineNo">592</span> }<a name="line.592"></a>
+<span class="sourceLineNo">593</span> return job;<a name="line.593"></a>
+<span class="sourceLineNo">594</span> }<a name="line.594"></a>
+<span class="sourceLineNo">595</span><a name="line.595"></a>
+<span class="sourceLineNo">596</span> private static void createTable(Admin
admin, TableName tableName, String[] columns)<a name="line.596"></a>
+<span class="sourceLineNo">597</span> throws IOException {<a
name="line.597"></a>
+<span class="sourceLineNo">598</span> HTableDescriptor htd = new
HTableDescriptor(tableName);<a name="line.598"></a>
+<span class="sourceLineNo">599</span> Set<String> cfSet =
getColumnFamilies(columns);<a name="line.599"></a>
+<span class="sourceLineNo">600</span> for (String cf : cfSet) {<a
name="line.600"></a>
+<span class="sourceLineNo">601</span> HColumnDescriptor hcd = new
HColumnDescriptor(Bytes.toBytes(cf));<a name="line.601"></a>
+<span class="sourceLineNo">602</span> htd.addFamily(hcd);<a
name="line.602"></a>
+<span class="sourceLineNo">603</span> }<a name="line.603"></a>
+<span class="sourceLineNo">604</span> LOG.warn(format("Creating table '%s'
with '%s' columns and default descriptors.",<a name="line.604"></a>
+<span class="sourceLineNo">605</span> tableName, cfSet));<a
name="line.605"></a>
+<span class="sourceLineNo">606</span> admin.createTable(htd);<a
name="line.606"></a>
+<span class="sourceLineNo">607</span> }<a name="line.607"></a>
+<span class="sourceLineNo">608</span><a name="line.608"></a>
+<span class="sourceLineNo">609</span> private static void
deleteTable(Configuration conf, String[] args) {<a name="line.609"></a>
+<span class="sourceLineNo">610</span> TableName tableName =
TableName.valueOf(args[0]);<a name="line.610"></a>
+<span class="sourceLineNo">611</span> try (Connection connection =
ConnectionFactory.createConnection(conf);<a name="line.611"></a>
+<span class="sourceLineNo">612</span> Admin admin =
connection.getAdmin()) {<a name="line.612"></a>
+<span class="sourceLineNo">613</span> try {<a name="line.613"></a>
+<span class="sourceLineNo">614</span> admin.disableTable(tableName);<a
name="line.614"></a>
+<span class="sourceLineNo">615</span> } catch (TableNotEnabledException
e) {<a name="line.615"></a>
+<span class="sourceLineNo">616</span> LOG.debug("Dry mode: Table: " +
tableName + " already disabled, so just deleting it.");<a name="line.616"></a>
+<span class="sourceLineNo">617</span> }<a name="line.617"></a>
+<span class="sourceLineNo">618</span> admin.deleteTable(tableName);<a
name="line.618"></a>
+<span class="sourceLineNo">619</span> } catch (IOException e) {<a
name="line.619"></a>
+<span class="sourceLineNo">620</span> LOG.error(format("***Dry run:
Failed to delete table '%s'.***%n%s", tableName,<a name="line.620"></a>
+<span class="sourceLineNo">621</span> e.toString()));<a
name="line.621"></a>
+<span class="sourceLineNo">622</span> return;<a name="line.622"></a>
+<span class="sourceLineNo">623</span> }<a name="line.623"></a>
+<span class="sourceLineNo">624</span> LOG.info(format("Dry run: Deleted
table '%s'.", tableName));<a name="line.624"></a>
+<span class="sourceLineNo">625</span> }<a name="line.625"></a>
+<span class="sourceLineNo">626</span><a name="line.626"></a>
+<span class="sourceLineNo">627</span> private static Set<String>
getColumnFamilies(String[] columns) {<a name="line.627"></a>
+<span class="sourceLineNo">628</span> Set<String> cfSet = new
HashSet<String>();<a name="line.628"></a>
+<span class="sourceLineNo">629</span> for (String aColumn : columns) {<a
name="line.629"></a>
+<span class="sourceLineNo">630</span> if
(TsvParser.ROWKEY_COLUMN_SPEC.equals(aColumn)<a name="line.630"></a>
+<span class="sourceLineNo">631</span> ||
TsvParser.TIMESTAMPKEY_COLUMN_SPEC.equals(aColumn)<a name="line.631"></a>
+<span class="sourceLineNo">632</span> ||
TsvParser.CELL_VISIBILITY_COLUMN_SPEC.equals(aColumn)<a name="line.632"></a>
+<span class="sourceLineNo">633</span> ||
TsvParser.CELL_TTL_COLUMN_SPEC.equals(aColumn)<a name="line.633"></a>
+<span class="sourceLineNo">634</span> ||
TsvParser.ATTRIBUTES_COLUMN_SPEC.equals(aColumn))<a name="line.634"></a>
+<span class="sourceLineNo">635</span> continue;<a name="line.635"></a>
+<span class="sourceLineNo">636</span> // we are only concerned with the
first one (in case this is a cf:cq)<a name="line.636"></a>
+<span class="sourceLineNo">637</span> cfSet.add(aColumn.split(":",
2)[0]);<a name="line.637"></a>
+<span class="sourceLineNo">638</span> }<a name="line.638"></a>
+<span class="sourceLineNo">639</span> return cfSet;<a name="line.639"></a>
+<span class="sourceLineNo">640</span> }<a name="line.640"></a>
+<span class="sourceLineNo">641</span><a name="line.641"></a>
+<span class="sourceLineNo">642</span> /*<a name="line.642"></a>
+<span class="sourceLineNo">643</span> * @param errorMsg Error message. Can
be null.<a name="line.643"></a>
+<span class="sourceLineNo">644</span> */<a name="line.644"></a>
+<span class="sourceLineNo">645</span> private static void usage(final String
errorMsg) {<a name="line.645"></a>
+<span class="sourceLineNo">646</span> if (errorMsg != null &&
errorMsg.length() > 0) {<a name="line.646"></a>
+<span class="sourceLineNo">647</span> System.err.println("ERROR: " +
errorMsg);<a name="line.647"></a>
+<span class="sourceLineNo">648</span> }<a name="line.648"></a>
+<span class="sourceLineNo">649</span> String usage =<a name="line.649"></a>
+<span class="sourceLineNo">650</span> "Usage: " + NAME + " -D"+
COLUMNS_CONF_KEY + "=a,b,c <tablename> <inputdir>\n" +<a
name="line.650"></a>
<span class="sourceLineNo">651</span> "\n" +<a name="line.651"></a>
-<span class="sourceLineNo">652</span> "The column names of the TSV data
must be specified using the -D" + COLUMNS_CONF_KEY + "\n" +<a
name="line.652"></a>
-<span class="sourceLineNo">653</span> "option. This option takes the form
of comma-separated column names, where each\n" +<a name="line.653"></a>
-<span class="sourceLineNo">654</span> "column name is either a simple
column family, or a columnfamily:qualifier. The special\n" +<a
name="line.654"></a>
-<span class="sourceLineNo">655</span> "column name " +
TsvParser.ROWKEY_COLUMN_SPEC + " is used to designate that this column should
be used\n" +<a name="line.655"></a>
-<span class="sourceLineNo">656</span> "as the row key for each imported
record. You must specify exactly one column\n" +<a name="line.656"></a>
-<span class="sourceLineNo">657</span> "to be the row key, and you must
specify a column name for every column that exists in the\n" +<a
name="line.657"></a>
-<span class="sourceLineNo">658</span> "input data. Another special
column" + TsvParser.TIMESTAMPKEY_COLUMN_SPEC +<a name="line.658"></a>
-<span class="sourceLineNo">659</span> " designates that this column
should be\n" +<a name="line.659"></a>
-<span class="sourceLineNo">660</span> "used as timestamp for each record.
Unlike " + TsvParser.ROWKEY_COLUMN_SPEC + ", " +<a name="line.660"></a>
-<span class="sourceLineNo">661</span> TsvParser.TIMESTAMPKEY_COLUMN_SPEC
+ " is optional.\n" +<a name="line.661"></a>
-<span class="sourceLineNo">662</span> "You must specify at most one
column as timestamp key for each imported record.\n" +<a name="line.662"></a>
-<span class="sourceLineNo">663</span> "Record with invalid timestamps
(blank, non-numeric) will be treated as bad record.\n" +<a name="line.663"></a>
-<span class="sourceLineNo">664</span> "Note: if you use this option, then
'" + TIMESTAMP_CONF_KEY + "' option will be ignored.\n" +<a name="line.664"></a>
-<span class="sourceLineNo">665</span> "\n" +<a name="line.665"></a>
-<span class="sourceLineNo">666</span> "Other special columns that can be
specified are " + TsvParser.CELL_TTL_COLUMN_SPEC +<a name="line.666"></a>
-<span class="sourceLineNo">667</span> " and " +
TsvParser.CELL_VISIBILITY_COLUMN_SPEC + ".\n" +<a name="line.667"></a>
-<span class="sourceLineNo">668</span> TsvParser.CELL_TTL_COLUMN_SPEC + "
designates that this column will be used " +<a name="line.668"></a>
-<span class="sourceLineNo">669</span> "as a Cell's Time To Live (TTL)
attribute.\n" +<a name="line.669"></a>
-<span class="sourceLineNo">670</span>
TsvParser.CELL_VISIBILITY_COLUMN_SPEC + " designates that this column contains
the " +<a name="line.670"></a>
-<span class="sourceLineNo">671</span> "visibility label expression.\n"
+<a name="line.671"></a>
-<span class="sourceLineNo">672</span> "\n" +<a name="line.672"></a>
-<span class="sourceLineNo">673</span> TsvParser.ATTRIBUTES_COLUMN_SPEC+"
can be used to specify Operation Attributes per record.\n"+<a
name="line.673"></a>
-<span class="sourceLineNo">674</span> " Should be specified as
key=>value where "+TsvParser.DEFAULT_ATTRIBUTES_COLUMN_INDEX+ " is used
\n"+<a name="line.674"></a>
-<span class="sourceLineNo">675</span> " as the seperator. Note that more
than one OperationAttributes can be specified.\n"+<a name="line.675"></a>
-<span class="sourceLineNo">676</span> "By default importtsv will load
data directly into HBase. To instead generate\n" +<a name="line.676"></a>
-<span class="sourceLineNo">677</span> "HFiles of data to prepare for a
bulk data load, pass the option:\n" +<a name="line.677"></a>
-<span class="sourceLineNo">678</span> " -D" + BULK_OUTPUT_CONF_KEY +
"=/path/for/output\n" +<a name="line.678"></a>
-<span class="sourceLineNo">679</span> " Note: if you do not use this
option, then the target table must already exist in HBase\n" +<a
name="line.679"></a>
-<span class="sourceLineNo">680</span> "\n" +<a name="line.680"></a>
-<span class="sourceLineNo">681</span> "Other options that may be
specified with -D include:\n" +<a name="line.681"></a>
-<span class="sourceLineNo">682</span> " -D" + DRY_RUN_CONF_KEY + "=true
- Dry run mode. Data is not actually populated into" +<a name="line.682"></a>
-<span class="sourceLineNo">683</span> " table. If table does not exist,
it is created but deleted in the end.\n" +<a name="line.683"></a>
-<span class="sourceLineNo">684</span> " -D" + SKIP_LINES_CONF_KEY +
"=false - fail if encountering an invalid line\n" +<a name="line.684"></a>
-<span class="sourceLineNo">685</span> " -D" + LOG_BAD_LINES_CONF_KEY +
"=true - logs invalid lines to stderr\n" +<a name="line.685"></a>
-<span class="sourceLineNo">686</span> " '-D" + SEPARATOR_CONF_KEY + "=|'
- eg separate on pipes instead of tabs\n" +<a name="line.686"></a>
-<span class="sourceLineNo">687</span> " -D" + TIMESTAMP_CONF_KEY +
"=currentTimeAsLong - use the specified timestamp for the import\n" +<a
name="line.687"></a>
-<span class="sourceLineNo">688</span> " -D" + MAPPER_CONF_KEY +
"=my.Mapper - A user-defined Mapper to use instead of " +<a name="line.688"></a>
-<span class="sourceLineNo">689</span> DEFAULT_MAPPER.getName() + "\n" +<a
name="line.689"></a>
-<span class="sourceLineNo">690</span> " -D" + JOB_NAME_CONF_KEY +
"=jobName - use the specified mapreduce job name for the import\n" +<a
name="line.690"></a>
-<span class="sourceLineNo">691</span> " -D" + CREATE_TABLE_CONF_KEY +
"=no - can be used to avoid creation of table by this tool\n" +<a
name="line.691"></a>
-<span class="sourceLineNo">692</span> " Note: if you set this to 'no',
then the target table must already exist in HBase\n" +<a name="line.692"></a>
-<span class="sourceLineNo">693</span> " -D" + NO_STRICT_COL_FAMILY +
"=true - ignore column family check in hbase table. " +<a name="line.693"></a>
-<span class="sourceLineNo">694</span> "Default is false\n\n" +<a
name="line.694"></a>
-<span class="sourceLineNo">695</span> "For performance consider the
following options:\n" +<a name="line.695"></a>
-<span class="sourceLineNo">696</span> "
-Dmapreduce.map.speculative=false\n" +<a name="line.696"></a>
-<span class="sourceLineNo">697</span> "
-Dmapreduce.reduce.speculative=false";<a name="line.697"></a>
-<span class="sourceLineNo">698</span><a name="line.698"></a>
-<span class="sourceLineNo">699</span> System.err.println(usage);<a
name="line.699"></a>
-<span class="sourceLineNo">700</span> }<a name="line.700"></a>
-<span class="sourceLineNo">701</span><a name="line.701"></a>
-<span class="sourceLineNo">702</span> @Override<a name="line.702"></a>
-<span class="sourceLineNo">703</span> public int run(String[] args) throws
Exception {<a name="line.703"></a>
-<span class="sourceLineNo">704</span> if (args.length < 2) {<a
name="line.704"></a>
-<span class="sourceLineNo">705</span> usage("Wrong number of arguments: "
+ args.length);<a name="line.705"></a>
-<span class="sourceLineNo">706</span> return -1;<a name="line.706"></a>
-<span class="sourceLineNo">707</span> }<a name="line.707"></a>
-<span class="sourceLineNo">708</span><a name="line.708"></a>
-<span class="sourceLineNo">709</span> // When MAPPER_CONF_KEY is null, the
user wants to use the provided TsvImporterMapper, so<a name="line.709"></a>
-<span class="sourceLineNo">710</span> // perform validation on these
additional args. When it's not null, user has provided their<a
name="line.710"></a>
-<span class="sourceLineNo">711</span> // own mapper, thus these validation
are not relevant.<a name="line.711"></a>
-<span class="sourceLineNo">712</span> // TODO: validation for
TsvImporterMapper, not this tool. Move elsewhere.<a name="line.712"></a>
-<span class="sourceLineNo">713</span> if (null ==
getConf().get(MAPPER_CONF_KEY)) {<a name="line.713"></a>
-<span class="sourceLineNo">714</span> // Make sure columns are
specified<a name="line.714"></a>
-<span class="sourceLineNo">715</span> String[] columns =
getConf().getStrings(COLUMNS_CONF_KEY);<a name="line.715"></a>
-<span class="sourceLineNo">716</span> if (columns == null) {<a
name="line.716"></a>
-<span class="sourceLineNo">717</span> usage("No columns specified.
Please specify with -D" +<a name="line.717"></a>
-<span class="sourceLineNo">718</span> COLUMNS_CONF_KEY+"=...");<a
name="line.718"></a>
-<span class="sourceLineNo">719</span> return -1;<a name="line.719"></a>
-<span class="sourceLineNo">720</span> }<a name="line.720"></a>
-<span class="sourceLineNo">721</span><a name="line.721"></a>
-<span class="sourceLineNo">722</span> // Make sure they specify exactly
one column as the row key<a name="line.722"></a>
-<span class="sourceLineNo">723</span> int rowkeysFound = 0;<a
name="line.723"></a>
-<span class="sourceLineNo">724</span> for (String col : columns) {<a
name="line.724"></a>
-<span class="sourceLineNo">725</span> if
(col.equals(TsvParser.ROWKEY_COLUMN_SPEC)) rowkeysFound++;<a
name="line.725"></a>
-<span class="sourceLineNo">726</span> }<a name="line.726"></a>
-<span class="sourceLineNo">727</span> if (rowkeysFound != 1) {<a
name="line.727"></a>
-<span class="sourceLineNo">728</span> usage("Must specify exactly one
column as " + TsvParser.ROWKEY_COLUMN_SPEC);<a name="line.728"></a>
-<span class="sourceLineNo">729</span> return -1;<a name="line.729"></a>
-<span class="sourceLineNo">730</span> }<a name="line.730"></a>
-<span class="sourceLineNo">731</span><a name="line.731"></a>
-<span class="sourceLineNo">732</span> // Make sure we have at most one
column as the timestamp key<a name="line.732"></a>
-<span class="sourceLineNo">733</span> int tskeysFound = 0;<a
name="line.733"></a>
-<span class="sourceLineNo">734</span> for (String col : columns) {<a
name="line.734"></a>
-<span class="sourceLineNo">735</span> if
(col.equals(TsvParser.TIMESTAMPKEY_COLUMN_SPEC))<a name="line.735"></a>
-<span class="sourceLineNo">736</span> tskeysFound++;<a
name="line.736"></a>
-<span class="sourceLineNo">737</span> }<a name="line.737"></a>
-<span class="sourceLineNo">738</span> if (tskeysFound > 1) {<a
name="line.738"></a>
-<span class="sourceLineNo">739</span> usage("Must specify at most one
column as "<a name="line.739"></a>
-<span class="sourceLineNo">740</span> +
TsvParser.TIMESTAMPKEY_COLUMN_SPEC);<a name="line.740"></a>
-<span class="sourceLineNo">741</span> return -1;<a name="line.741"></a>
-<span class="sourceLineNo">742</span> }<a name="line.742"></a>
-<span class="sourceLineNo">743</span><a name="line.743"></a>
-<span class="sourceLineNo">744</span> int attrKeysFound = 0;<a
name="line.744"></a>
-<span class="sourceLineNo">745</span> for (String col : columns) {<a
name="line.745"></a>
-<span class="sourceLineNo">746</span> if
(col.equals(TsvParser.ATTRIBUTES_COLUMN_SPEC))<a name="line.746"></a>
-<span class="sourceLineNo">747</span> attrKeysFound++;<a
name="line.747"></a>
-<span class="sourceLineNo">748</span> }<a name="line.748"></a>
-<span class="sourceLineNo">749</span> if (attrKeysFound > 1) {<a
name="line.749"></a>
-<span class="sourceLineNo">750</span> usage("Must specify at most one
column as "<a name="line.750"></a>
-<span class="sourceLineNo">751</span> +
TsvParser.ATTRIBUTES_COLUMN_SPEC);<a name="line.751"></a>
-<span class="sourceLineNo">752</span> return -1;<a name="line.752"></a>
-<span class="sourceLineNo">753</span> }<a name="line.753"></a>
-<span class="sourceLineNo">754</span><a name="line.754"></a>
-<span class="sourceLineNo">755</span> // Make sure one or more columns
are specified excluding rowkey and<a name="line.755"></a>
-<span class="sourceLineNo">756</span> // timestamp key<a
name="line.756"></a>
-<span class="sourceLineNo">757</span> if (columns.length - (rowkeysFound
+ tskeysFound + attrKeysFound) < 1) {<a name="line.757"></a>
-<span class="sourceLineNo">758</span> usage("One or more columns in
addition to the row key and timestamp(optional) are required");<a
name="line.758"></a>
-<span class="sourceLineNo">759</span> return -1;<a name="line.759"></a>
-<span class="sourceLineNo">760</span> }<a name="line.760"></a>
-<span class="sourceLineNo">761</span> }<a name="line.761"></a>
-<span class="sourceLineNo">762</span><a name="line.762"></a>
-<span class="sourceLineNo">763</span> // If timestamp option is not
specified, use current system time.<a name="line.763"></a>
-<span class="sourceLineNo">764</span> long timstamp =
getConf().getLong(TIMESTAMP_CONF_KEY, System.currentTimeMillis());<a
name="line.764"></a>
-<span class="sourceLineNo">765</span><a name="line.765"></a>
-<span class="sourceLineNo">766</span> // Set it back to replace invalid
timestamp (non-numeric) with current<a name="line.766"></a>
-<span class="sourceLineNo">767</span> // system time<a name="line.767"></a>
-<span class="sourceLineNo">768</span> getConf().setLong(TIMESTAMP_CONF_KEY,
timstamp);<a name="line.768"></a>
-<span class="sourceLineNo">769</span><a name="line.769"></a>
-<span class="sourceLineNo">770</span> dryRunTableCreated = false;<a
name="line.770"></a>
-<span class="sourceLineNo">771</span> Job job =
createSubmittableJob(getConf(), args);<a name="line.771"></a>
-<span class="sourceLineNo">772</span> boolean success =
job.waitForCompletion(true);<a name="line.772"></a>
-<span class="sourceLineNo">773</span> if (dryRunTableCreated) {<a
name="line.773"></a>
-<span class="sourceLineNo">774</span> deleteTable(getConf(), args);<a
name="line.774"></a>
-<span class="sourceLineNo">775</span> }<a name="line.775"></a>
-<span class="sourceLineNo">776</span> return success ? 0 : 1;<a
name="line.776"></a>
-<span class="sourceLineNo">777</span> }<a name="line.777"></a>
-<span class="sourceLineNo">778</span><a name="line.778"></a>
-<span class="sourceLineNo">779</span> public static void main(String[] args)
throws Exception {<a name="line.779"></a>
-<span class="sourceLineNo">780</span> int status =
ToolRunner.run(HBaseConfiguration.create(), new ImportTsv(), args);<a
name="line.780"></a>
-<span class="sourceLineNo">781</span> System.exit(status);<a
name="line.781"></a>
-<span class="sourceLineNo">782</span> }<a name="line.782"></a>
-<span class="sourceLineNo">783</span>}<a name="line.783"></a>
+<span class="sourceLineNo">652</span> "Imports the given input directory
of TSV data into the specified table.\n" +<a name="line.652"></a>
+<span class="sourceLineNo">653</span> "\n" +<a name="line.653"></a>
+<span class="sourceLineNo">654</span> "The column names of the TSV data
must be specified using the -D" + COLUMNS_CONF_KEY + "\n" +<a
name="line.654"></a>
+<span class="sourceLineNo">655</span> "option. This option takes the form
of comma-separated column names, where each\n" +<a name="line.655"></a>
+<span class="sourceLineNo">656</span> "column name is either a simple
column family, or a columnfamily:qualifier. The special\n" +<a
name="line.656"></a>
+<span class="sourceLineNo">657</span> "column name " +
TsvParser.ROWKEY_COLUMN_SPEC + " is used to designate that this column should
be used\n" +<a name="line.657"></a>
+<span class="sourceLineNo">658</span> "as the row key for each imported
record. You must specify exactly one column\n" +<a name="line.658"></a>
+<span class="sourceLineNo">659</span> "to be the row key, and you must
specify a column name for every column that exists in the\n" +<a
name="line.659"></a>
+<span class="sourceLineNo">660</span> "input data. Another special
column" + TsvParser.TIMESTAMPKEY_COLUMN_SPEC +<a name="line.660"></a>
+<span class="sourceLineNo">661</span> " designates that this column
should be\n" +<a name="line.661"></a>
+<span class="sourceLineNo">662</span> "used as timestamp for each record.
Unlike " + TsvParser.ROWKEY_COLUMN_SPEC + ", " +<a name="line.662"></a>
+<span class="sourceLineNo">663</span> TsvParser.TIMESTAMPKEY_COLUMN_SPEC
+ " is optional." + "\n" +<a name="line.663"></a>
+<span class="sourceLineNo">664</span> "You must specify at most one
column as timestamp key for each imported record.\n" +<a name="line.664"></a>
+<span class="sourceLineNo">665</span> "Record with invalid timestamps
(blank, non-numeric) will be treated as bad record.\n" +<a name="line.665"></a>
+<span class="sourceLineNo">666</span> "Note: if you use this option, then
'" + TIMESTAMP_CONF_KEY + "' option will be ignored.\n" +<a name="line.666"></a>
+<span class="sourceLineNo">667</span> "\n" +<a name="line.667"></a>
+<span class="sourceLineNo">668</span> "Other special columns that can be
specified are " + TsvParser.CELL_TTL_COLUMN_SPEC +<a name="line.668"></a>
+<span class="sourceLineNo">669</span> " and " +
TsvParser.CELL_VISIBILITY_COLUMN_SPEC + ".\n" +<a name="line.669"></a>
+<span class="sourceLineNo">670</span> TsvParser.CELL_TTL_COLUMN_SPEC + "
designates that this column will be used " +<a name="line.670"></a>
+<span class="sourceLineNo">671</span> "as a Cell's Time To Live (TTL)
attribute.\n" +<a name="line.671"></a>
+<span class="sourceLineNo">672</span>
TsvParser.CELL_VISIBILITY_COLUMN_SPEC + " designates that this column contains
the " +<a name="line.672"></a>
+<span class="sourceLineNo">673</span> "visibility label expression.\n"
+<a name="line.673"></a>
+<span class="sourceLineNo">674</span> "\n" +<a name="line.674"></a>
+<span class="sourceLineNo">675</span> TsvParser.ATTRIBUTES_COLUMN_SPEC+"
can be used to specify Operation Attributes per record.\n"+<a
name="line.675"></a>
+<span class="sourceLineNo">676</span> " Should be specified as
key=>value where "+TsvParser.DEFAULT_ATTRIBUTES_COLUMN_INDEX+ " is used
\n"+<a name="line.676"></a>
+<span class="sourceLineNo">677</span> " as the seperator. Note that more
than one OperationAttributes can be specified.\n"+<a name="line.677"></a>
+<span class="sourceLineNo">678</span> "By default importtsv will load
data directly into HBase. To instead generate\n" +<a name="line.678"></a>
+<span class="sourceLineNo">679</span> "HFiles of data to prepare for a
bulk data load, pass the option:\n" +<a name="line.679"></a>
+<span class="sourceLineNo">680</span> " -D" + BULK_OUTPUT_CONF_KEY +
"=/path/for/output\n" +<a name="line.680"></a>
+<span class="sourceLineNo">681</span> " Note: if you do not use this
option, then the target table must already exist in HBase\n" +<a
name="line.681"></a>
+<span class="sourceLineNo">682</span> "\n" +<a name="line.682"></a>
+<span class="sourceLineNo">683</span> "Other options that may be
specified with -D include:\n" +<a name="line.683"></a>
+<span class="sourceLineNo">684</span> " -D" + DRY_RUN_CONF_KEY + "=true
- Dry run mode. Data is not actually populated into" +<a name="line.684"></a>
+<span class="sourceLineNo">685</span> " table. If table does not exist,
it is created but deleted in the end.\n" +<a name="line.685"></a>
+<span class="sourceLineNo">686</span> " -D" + SKIP_LINES_CONF_KEY +
"=false - fail if encountering an invalid line\n" +<a name="line.686"></a>
+<span class="sourceLineNo">687</span> " -D" + LOG_BAD_LINES_CONF_KEY +
"=true - logs invalid lines to stderr\n" +<a name="line.687"></a>
+<span class="sourceLineNo">688</span> " '-D" + SEPARATOR_CONF_KEY + "=|'
- eg separate on pipes instead of tabs\n" +<a name="line.688"></a>
+<span class="sourceLineNo">689</span> " -D" + TIMESTAMP_CONF_KEY +
"=currentTimeAsLong - use the specified timestamp for the import\n" +<a
name="line.689"></a>
+<span class="sourceLineNo">690</span> " -D" + MAPPER_CONF_KEY +
"=my.Mapper - A user-defined Mapper to use instead of " +<a name="line.690"></a>
+<span class="sourceLineNo">691</span> DEFAULT_MAPPER.getName() + "\n" +<a
name="line.691"></a>
+<span class="sourceLineNo">692</span> " -D" + JOB_NAME_CONF_KEY +
"=jobName - use the specified mapreduce job name for the import\n" +<a
name="line.692"></a>
+<span class="sourceLineNo">693</span> " -D" + CREATE_TABLE_CONF_KEY +
"=no - can be used to avoid creation of table by this tool\n" +<a
name="line.693"></a>
+<span class="sourceLineNo">694</span> " Note: if you set this to 'no',
then the target table must already exist in HBase\n" +<a name="line.694"></a>
+<span class="sourceLineNo">695</span> " -D" + NO_STRICT_COL_FAMILY +
"=true - ignore column family check in hbase table. " +<a name="line.695"></a>
+<span class="sourceLineNo">696</span> "Default is false\n\n" +<a
name="line.696"></a>
+<span class="sourceLineNo">697</span> "For performance consider the
following options:\n" +<a name="line.697"></a>
+<span class="sourceLineNo">698</span> "
-Dmapreduce.map.speculative=false\n" +<a name="line.698"></a>
+<span class="sourceLineNo">699</span> "
-Dmapreduce.reduce.speculative=false";<a name="line.699"></a>
+<span class="sourceLineNo">700</span><a name="line.700"></a>
+<span class="sourceLineNo">701</span> System.err.println(usage);<a
name="line.701"></a>
+<span class="sourceLineNo">702</span> }<a name="line.702"></a>
+<span class="sourceLineNo">703</span><a name="line.703"></a>
+<span class="sourceLineNo">704</span> @Override<a name="line.704"></a>
+<span class="sourceLineNo">705</span> public int run(String[] args) throws
Exception {<a name="line.705"></a>
+<span class="sourceLineNo">706</span> if (args.length < 2) {<a
name="line.706"></a>
+<span class="sourceLineNo">707</span> usage("Wrong number of arguments: "
+ args.length);<a name="line.707"></a>
+<span class="sourceLineNo">708</span> return -1;<a name="line.708"></a>
+<span class="sourceLineNo">709</span> }<a name="line.709"></a>
+<span class="sourceLineNo">710</span><a name="line.710"></a>
+<span class="sourceLineNo">711</span> // When MAPPER_CONF_KEY is null, the
user wants to use the provided TsvImporterMapper, so<a name="line.711"></a>
+<span class="sourceLineNo">712</span> // perform validation on these
additional args. When it's not null, user has provided their<a
name="line.712"></a>
+<span class="sourceLineNo">713</span> // own mapper, thus these validation
are not relevant.<a name="line.713"></a>
+<span class="sourceLineNo">714</span> // TODO: validation for
TsvImporterMapper, not this tool. Move elsewhere.<a name="line.714"></a>
+<span class="sourceLineNo">715</span> if (null ==
getConf().get(MAPPER_CONF_KEY)) {<a name="line.715"></a>
+<span class="sourceLineNo">716</span> // Make sure columns are
specified<a name="line.716"></a>
+<span class="sourceLineNo">717</span> String[] columns =
getConf().getStrings(COLUMNS_CONF_KEY);<a name="line.717"></a>
+<span class="sourceLineNo">718</span> if (columns == null) {<a
name="line.718"></a>
+<span class="sourceLineNo">719</span> usage("No columns specified.
Please specify with -D" +<a name="line.719"></a>
+<span class="sourceLineNo">720</span> COLUMNS_CONF_KEY+"=...");<a
name="line.720"></a>
+<span class="sourceLineNo">721</span> return -1;<a name="line.721"></a>
+<span class="sourceLineNo">722</span> }<a name="line.722"></a>
+<span class="sourceLineNo">723</span><a name="line.723"></a>
+<span class="sourceLineNo">724</span> // Make sure they specify exactly
one column as the row key<a name="line.724"></a>
+<span class="sourceLineNo">725</span> int rowkeysFound = 0;<a
name="line.725"></a>
+<span class="sourceLineNo">726</span> for (String col : columns) {<a
name="line.726"></a>
+<span class="sourceLineNo">727</span> if
(col.equals(TsvParser.ROWKEY_COLUMN_SPEC)) rowkeysFound++;<a
name="line.727"></a>
+<span class="sourceLineNo">728</span> }<a name="line.728"></a>
+<span class="sourceLineNo">729</span> if (rowkeysFound != 1) {<a
name="line.729"></a>
+<span class="sourceLineNo">730</span> usage("Must specify exactly one
column as " + TsvParser.ROWKEY_COLUMN_SPEC);<a name="line.730"></a>
+<span class="sourceLineNo">731</span> return -1;<a name="line.731"></a>
+<span class="sourceLineNo">732</span> }<a name="line.732"></a>
+<span class="sourceLineNo">733</span><a name="line.733"></a>
+<span class="sourceLineNo">734</span> // Make sure we have at most one
column as the timestamp key<a name="line.734"></a>
+<span class="sourceLineNo">735</span> int tskeysFound = 0;<a
name="line.735"></a>
+<span class="sourceLineNo">736</span> for (String col : columns) {<a
name="line.736"></a>
+<span class="sourceLineNo">737</span> if
(col.equals(TsvParser.TIMESTAMPKEY_COLUMN_SPEC))<a name="line.737"></a>
+<span class="sourceLineNo">738</span> tskeysFound++;<a
name="line.738"></a>
+<span class="sourceLineNo">739</span> }<a name="line.739"></a>
+<span class="sourceLineNo">740</span> if (tskeysFound > 1) {<a
name="line.740"></a>
+<span class="sourceLineNo">741</span> usage("Must specify at most one
column as "<a name="line.741"></a>
+<span class="sourceLineNo">742</span> +
TsvParser.TIMESTAMPKEY_COLUMN_SPEC);<a name="line.742"></a>
+<span class="sourceLineNo">743</span> return -1;<a name="line.743"></a>
+<span class="sourceLineNo">744</span> }<a name="line.744"></a>
+<span class="sourceLineNo">745</span><a name="line.745"></a>
+<span class="sourceLineNo">746</span> int attrKeysFound = 0;<a
name="line.746"></a>
+<span class="sourceLineNo">747</span> for (String col : columns) {<a
name="line.747"></a>
+<span class="sourceLineNo">748</span> if
(col.equals(TsvParser.ATTRIBUTES_COLUMN_SPEC))<a name="line.748"></a>
+<span class="sourceLineNo">749</span> attrKeysFound++;<a
name="line.749"></a>
+<span class="sourceLineNo">750</span> }<a name="line.750"></a>
+<span class="sourceLineNo">751</span> if (attrKeysFound > 1) {<a
name="line.751"></a>
+<span class="sourceLineNo">752</span> usage("Must specify at most one
column as "<a name="line.752"></a>
+<span class="sourceLineNo">753</span> +
TsvParser.ATTRIBUTES_COLUMN_SPEC);<a name="line.753"></a>
+<span class="sourceLineNo">754</span> return -1;<a name="line.754"></a>
+<span class="sourceLineNo">755</span> }<a name="line.755"></a>
+<span class="sourceLineNo">756</span><a name="line.756"></a>
+<span class="sourceLineNo">757</span> // Make sure one or more columns
are specified excluding rowkey and<a name="line.757"></a>
+<span class="sourceLineNo">758</span> // timestamp key<a
name="line.758"></a>
+<span class="sourceLineNo">759</span> if (columns.length - (rowkeysFound
+ tskeysFound + attrKeysFound) < 1) {<a name="line.759"></a>
+<span class="sourceLineNo">760</span> usage("One or more columns in
addition to the row key and timestamp(optional) are required");<a
name="line.760"></a>
+<span class="sourceLineNo">761</span> return -1;<a name="line.761"></a>
+<span class="sourceLineNo">762</span> }<a name="line.762"></a>
+<span class="sourceLineNo">763</span> }<a name="line.763"></a>
+<span class="sourceLineNo">764</span><a name="line.764"></a>
+<span class="sourceLineNo">765</span> // If timestamp option is not
specified, use current system time.<a name="line.765"></a>
+<span class="sourceLineNo">766</span> long timstamp =
getConf().getLong(TIMESTAMP_CONF_KEY, System.currentTimeMillis());<a
name="line.766"></a>
+<span class="sourceLineNo">767</span><a name="line.767"></a>
+<span class="sourceLineNo">768</span> // Set it back to replace invalid
timestamp (non-numeric) with current<a name="line.768"></a>
+<span class="sourceLineNo">769</span> // system time<a name="line.769"></a>
+<span class="sourceLineNo">770</span> getConf().setLong(TIMESTAMP_CONF_KEY,
timstamp);<a name="line.770"></a>
+<span class="sourceLineNo">771</span><a name="line.771"></a>
+<span class="sourceLineNo">772</span> synchronized (ImportTsv.class) {<a
name="line.772"></a>
+<span class="sourceLineNo">773</span> DRY_RUN_TABLE_CREATED = false;<a
name="line.773"></a>
+<span class="sourceLineNo">774</span> }<a name="line.774"></a>
+<span class="sourceLineNo">775</span> Job job =
createSubmittableJob(getConf(), args);<a name="line.775"></a>
+<span class="sourceLineNo">776</span> boolean success =
job.waitForCompletion(true);<a name="line.776"></a>
+<span class="sourceLineNo">777</span> boolean delete = false;<a
name="line.777"></a>
+<span class="sourceLineNo">778</span> synchronized (ImportTsv.class) {<a
name="line.778"></a>
+<span class="sourceLineNo">779</span> delete = DRY_RUN_TABLE_CREATED;<a
name="line.779"></a>
+<span class="sourceLineNo">780</span> }<a name="line.780"></a>
+<span class="sourceLineNo">781</span> if (delete) {<a name="line.781"></a>
+<span class="sourceLineNo">782</span> deleteTable(getConf(), args);<a
name="line.782"></a>
+<span class="sourceLineNo">783</span> }<a name="line.783"></a>
+<span class="sourceLineNo">784</span> return success ? 0 : 1;<a
name="line.784"></a>
+<span class="sourceLineNo">785</span> }<a name="line.785"></a>
+<span class="sourceLineNo">786</span><a name="line.786"></a>
+<span class="sourceLineNo">787</span> public static void main(String[] args)
throws Exception {<a name="line.787"></a>
+<span class="sourceLineNo">788</span> int status =
ToolRunner.run(HBaseConfiguration.create(), new ImportTsv(), args);<a
name="line.788"></a>
+<span class="sourceLineNo">789</span> System.exit(status);<a
name="line.789"></a>
+<span class="sourceLineNo">790</span> }<a name="line.790"></a>
+<span class="sourceLineNo">791</span>}<a name="line.791"></a>
