Merge pull request #29997 Add a ReadFromTsv example.

sdks/python/apache_beam/dataframe/io.py

@@ -86,7 +86,7 @@ def read_gbq(
 
 
 @frame_base.with_docs_from(pd)
-def read_csv(path, *args, splittable=False, **kwargs):
+def read_csv(path, *args, splittable=False, binary=True, **kwargs):
   """If your files are large and records do not contain quoted newlines, you may
   pass the extra argument ``splittable=True`` to enable dynamic splitting for
   this read on newlines. Using this option for records that do contain quoted
@@ -99,6 +99,7 @@ def read_csv(path, *args, splittable=False, **kwargs):
       args,
       kwargs,
       incremental=True,
+      binary=binary,
       splitter=_TextFileSplitter(args, kwargs) if splittable else None)
 
 

sdks/python/apache_beam/yaml/inline_python.md

@@ -59,6 +59,28 @@ as an input, e.g.
         output_type: integer
 ```
 
+This can be used to call arbitrary transforms in the Beam SDK, e.g.
+
+```
+pipeline:
+  transforms:
+    - type: PyTransform
+      name: ReadFromTsv
+      input: {}
+      config:
+        constructor: apache_beam.io.ReadFromCsv
+        kwargs:
+           path: '/path/to/*.tsv'
+           sep: '\t'
+           skip_blank_lines: True
+           true_values: ['yes']
+           false_values: ['no']
+           comment: '#'
+           on_bad_lines: 'skip'
+           binary: False
+           splittable: False
+```
+
 
 ## Defining a transform inline using `__constructor__`
 

sdks/python/apache_beam/yaml/readme_test.py

@@ -154,6 +154,9 @@ def input_file(self, name, content):
   def input_csv(self):
     return self.input_file('input.csv', 'col1,col2,col3\nabc,1,2.5\n')
 
+  def input_tsv(self):
+    return self.input_file('input.tsv', 'col1\tcol2\tcol3\nabc\t1\t2.5\n')
+
   def input_json(self):
     return self.input_file(
         'input.json', '{"col1": "abc", "col2": 1, "col3": 2.5"}\n')
@@ -192,6 +195,8 @@ def create_test_method(test_type, test_name, test_yaml):
 
   def test(self):
     with TestEnvironment() as env:
+      nonlocal test_yaml
+      test_yaml = test_yaml.replace('/path/to/*.tsv', env.input_tsv())
       spec = yaml.load(test_yaml, Loader=SafeLoader)
       if test_type == 'PARSE':
         return