diff options
author | Reynold Xin <rxin@databricks.com> | 2015-02-13 23:03:22 -0800 |
---|---|---|
committer | Reynold Xin <rxin@databricks.com> | 2015-02-13 23:03:22 -0800 |
commit | e98dfe627c5d0201464cdd0f363f391ea84c389a (patch) | |
tree | 794beea739eb04bf2e0926f9b0e19ffacb94ba08 /python/pyspark/sql/context.py | |
parent | 0ce4e430a81532dc317136f968f28742e087d840 (diff) | |
download | spark-e98dfe627c5d0201464cdd0f363f391ea84c389a.tar.gz spark-e98dfe627c5d0201464cdd0f363f391ea84c389a.tar.bz2 spark-e98dfe627c5d0201464cdd0f363f391ea84c389a.zip |
[SPARK-5752][SQL] Don't implicitly convert RDDs directly to DataFrames
- The old implicit would convert RDDs directly to DataFrames, and that added too many methods.
- toDataFrame -> toDF
- Dsl -> functions
- implicits moved into SQLContext.implicits
- addColumn -> withColumn
- renameColumn -> withColumnRenamed
Python changes:
- toDataFrame -> toDF
- Dsl -> functions package
- addColumn -> withColumn
- renameColumn -> withColumnRenamed
- add toDF functions to RDD on SQLContext init
- add flatMap to DataFrame
Author: Reynold Xin <rxin@databricks.com>
Author: Davies Liu <davies@databricks.com>
Closes #4556 from rxin/SPARK-5752 and squashes the following commits:
5ef9910 [Reynold Xin] More fix
61d3fca [Reynold Xin] Merge branch 'df5' of github.com:davies/spark into SPARK-5752
ff5832c [Reynold Xin] Fix python
749c675 [Reynold Xin] count(*) fixes.
5806df0 [Reynold Xin] Fix build break again.
d941f3d [Reynold Xin] Fixed explode compilation break.
fe1267a [Davies Liu] flatMap
c4afb8e [Reynold Xin] style
d9de47f [Davies Liu] add comment
b783994 [Davies Liu] add comment for toDF
e2154e5 [Davies Liu] schema() -> schema
3a1004f [Davies Liu] Dsl -> functions, toDF()
fb256af [Reynold Xin] - toDataFrame -> toDF - Dsl -> functions - implicits moved into SQLContext.implicits - addColumn -> withColumn - renameColumn -> withColumnRenamed
0dd74eb [Reynold Xin] [SPARK-5752][SQL] Don't implicitly convert RDDs directly to DataFrames
97dd47c [Davies Liu] fix mistake
6168f74 [Davies Liu] fix test
1fc0199 [Davies Liu] fix test
a075cd5 [Davies Liu] clean up, toPandas
663d314 [Davies Liu] add test for agg('*')
9e214d5 [Reynold Xin] count(*) fixes.
1ed7136 [Reynold Xin] Fix build break again.
921b2e3 [Reynold Xin] Fixed explode compilation break.
14698d4 [Davies Liu] flatMap
ba3e12d [Reynold Xin] style
d08c92d [Davies Liu] add comment
5c8b524 [Davies Liu] add comment for toDF
a4e5e66 [Davies Liu] schema() -> schema
d377fc9 [Davies Liu] Dsl -> functions, toDF()
6b3086c [Reynold Xin] - toDataFrame -> toDF - Dsl -> functions - implicits moved into SQLContext.implicits - addColumn -> withColumn - renameColumn -> withColumnRenamed
807e8b1 [Reynold Xin] [SPARK-5752][SQL] Don't implicitly convert RDDs directly to DataFrames
Diffstat (limited to 'python/pyspark/sql/context.py')
-rw-r--r-- | python/pyspark/sql/context.py | 34 |
1 files changed, 30 insertions, 4 deletions
diff --git a/python/pyspark/sql/context.py b/python/pyspark/sql/context.py index 082f1b691b..7683c1b4df 100644 --- a/python/pyspark/sql/context.py +++ b/python/pyspark/sql/context.py @@ -38,6 +38,25 @@ except ImportError: __all__ = ["SQLContext", "HiveContext"] +def _monkey_patch_RDD(sqlCtx): + def toDF(self, schema=None, sampleRatio=None): + """ + Convert current :class:`RDD` into a :class:`DataFrame` + + This is a shorthand for `sqlCtx.createDataFrame(rdd, schema, sampleRatio)` + + :param schema: a StructType or list of names of columns + :param samplingRatio: the sample ratio of rows used for inferring + :return: a DataFrame + + >>> rdd.toDF().collect() + [Row(name=u'Alice', age=1)] + """ + return sqlCtx.createDataFrame(self, schema, sampleRatio) + + RDD.toDF = toDF + + class SQLContext(object): """Main entry point for Spark SQL functionality. @@ -49,15 +68,20 @@ class SQLContext(object): def __init__(self, sparkContext, sqlContext=None): """Create a new SQLContext. + It will add a method called `toDF` to :class:`RDD`, which could be + used to convert an RDD into a DataFrame, it's a shorthand for + :func:`SQLContext.createDataFrame`. + :param sparkContext: The SparkContext to wrap. :param sqlContext: An optional JVM Scala SQLContext. If set, we do not instatiate a new SQLContext in the JVM, instead we make all calls to this object. >>> from datetime import datetime + >>> sqlCtx = SQLContext(sc) >>> allTypes = sc.parallelize([Row(i=1, s="string", d=1.0, l=1L, ... b=True, list=[1, 2, 3], dict={"s": 0}, row=Row(a=1), ... time=datetime(2014, 8, 1, 14, 1, 5))]) - >>> df = sqlCtx.createDataFrame(allTypes) + >>> df = allTypes.toDF() >>> df.registerTempTable("allTypes") >>> sqlCtx.sql('select i+1, d+1, not b, list[1], dict["s"], time, row.a ' ... 'from allTypes where b and i > 0').collect() @@ -70,6 +94,7 @@ class SQLContext(object): self._jsc = self._sc._jsc self._jvm = self._sc._jvm self._scala_SQLContext = sqlContext + _monkey_patch_RDD(self) @property def _ssql_ctx(self): @@ -442,7 +467,7 @@ class SQLContext(object): Row(f1=2, f2=None, f3=Row(field4=22,..., f4=[Row(field7=u'row2')]) Row(f1=None, f2=u'row3', f3=Row(field4=33, field5=[]), f4=None) - >>> df3 = sqlCtx.jsonFile(jsonFile, df1.schema()) + >>> df3 = sqlCtx.jsonFile(jsonFile, df1.schema) >>> sqlCtx.registerRDDAsTable(df3, "table2") >>> df4 = sqlCtx.sql( ... "SELECT field1 AS f1, field2 as f2, field3 as f3, " @@ -495,7 +520,7 @@ class SQLContext(object): Row(f1=2, f2=None, f3=Row(field4=22..., f4=[Row(field7=u'row2')]) Row(f1=None, f2=u'row3', f3=Row(field4=33, field5=[]), f4=None) - >>> df3 = sqlCtx.jsonRDD(json, df1.schema()) + >>> df3 = sqlCtx.jsonRDD(json, df1.schema) >>> sqlCtx.registerRDDAsTable(df3, "table2") >>> df4 = sqlCtx.sql( ... "SELECT field1 AS f1, field2 as f2, field3 as f3, " @@ -800,7 +825,8 @@ def _test(): Row(field1=2, field2="row2"), Row(field1=3, field2="row3")] ) - globs['df'] = sqlCtx.createDataFrame(rdd) + _monkey_patch_RDD(sqlCtx) + globs['df'] = rdd.toDF() jsonStrings = [ '{"field1": 1, "field2": "row1", "field3":{"field4":11}}', '{"field1" : 2, "field3":{"field4":22, "field5": [10, 11]},' |