API: add reads_json, reads_msgpack, reads_csv, reads_pickle as convience functions which accept a string

pandas/io/api.py

@@ -2,14 +2,14 @@
 Data IO api
 """
 
-from pandas.io.parsers import read_csv, read_table, read_fwf
+from pandas.io.parsers import read_csv, reads_csv, read_table, read_fwf
 from pandas.io.clipboard import read_clipboard
 from pandas.io.excel import ExcelFile, ExcelWriter, read_excel
 from pandas.io.pytables import HDFStore, Term, get_store, read_hdf
-from pandas.io.json import read_json
+from pandas.io.json import read_json, reads_json
 from pandas.io.html import read_html
 from pandas.io.sql import read_sql
 from pandas.io.stata import read_stata
-from pandas.io.pickle import read_pickle, to_pickle
-from pandas.io.packers import read_msgpack, to_msgpack
+from pandas.io.pickle import read_pickle, reads_pickle, to_pickle
+from pandas.io.packers import read_msgpack, reads_msgpack, to_msgpack
 from pandas.io.gbq import read_gbq

pandas/io/common.py

@@ -6,7 +6,7 @@
 
 from pandas.compat import StringIO
 from pandas import compat
-
+from functools import wraps
 
 if compat.PY3:
     from urllib.request import urlopen, pathname2url
@@ -45,6 +45,37 @@ class PerformanceWarning(Warning):
 class DtypeWarning(Warning):
     pass
 
+def _create_string_file_reader(func):
+    """
+    create and return a new function that takes string input and
+    passed to a file-like reader function,
+    e.g. read_json
+
+    Parameters
+    ----------
+    func : the function with a file-like interface
+
+    Returns
+    -------
+    new function that transform input to file-like
+    """
+
+    @wraps(func)
+    def f(path_or_buf, *args, **kwargs):
+        if  not hasattr(path_or_buf,'read'):
+            if isinstance(path_or_buf, compat.string_types):
+                path_or_buf = StringIO(path_or_buf)
+            elif isinstance(path_or_buf, compat.binary_type):
+                path_or_buf = compat.BytesIO(path_or_buf)
+            try:
+                return func(path_or_buf, *args, **kwargs)
+            finally:
+                if not ('iterator' in kwargs or 'chunksize' in kwargs):
+                    path_or_buf.close()
+
+        return func(path_or_buf, *args, **kwargs)
+
+    return f
 
 def _is_url(url):
     """Check to see if a URL has a valid protocol.

pandas/io/json.py

@@ -10,7 +10,7 @@
 from pandas.compat import long, u
 from pandas import compat, isnull
 from pandas import Series, DataFrame, to_datetime
-from pandas.io.common import get_filepath_or_buffer
+from pandas.io.common import get_filepath_or_buffer, _create_string_file_reader
 import pandas.core.common as com
 
 loads = _json.loads
@@ -109,7 +109,7 @@ def read_json(path_or_buf=None, orient=None, typ='frame', dtype=True,
 
     Parameters
     ----------
-    filepath_or_buffer : a valid JSON string or file-like
+    path_or_buffer : a valid file-like
         The string could be a URL. Valid URL schemes include http, ftp, s3, and
         file. For file URLs, a host is expected. For instance, a local file
         could be ``file://localhost/path/to/table.json``
@@ -171,25 +171,14 @@ def read_json(path_or_buf=None, orient=None, typ='frame', dtype=True,
     result : Series or DataFrame
     """
 
-    filepath_or_buffer, _ = get_filepath_or_buffer(path_or_buf)
-    if isinstance(filepath_or_buffer, compat.string_types):
-        try:
-            exists = os.path.exists(filepath_or_buffer)
-
-        # if the filepath is too long will raise here
-        # 5874
-        except (TypeError,ValueError):
-            exists = False
-
-        if exists:
-            with open(filepath_or_buffer, 'r') as fh:
-                json = fh.read()
-        else:
-            json = filepath_or_buffer
-    elif hasattr(filepath_or_buffer, 'read'):
-        json = filepath_or_buffer.read()
+    path_or_buffer, _ = get_filepath_or_buffer(path_or_buf)
+    if isinstance(path_or_buffer, compat.string_types):
+        with open(path_or_buffer, 'r') as fh:
+            json = fh.read()
+    elif hasattr(path_or_buffer, 'read'):
+        json = path_or_buffer.read()
     else:
-        json = filepath_or_buffer
+        raise ValueError("path_or_buffer must be a file or file-like buffer")
 
     obj = None
     if typ == 'frame':
@@ -206,6 +195,7 @@ def read_json(path_or_buf=None, orient=None, typ='frame', dtype=True,
 
     return obj
 
+reads_json = _create_string_file_reader(read_json)
 
 class Parser(object):
 

pandas/io/packers.py

@@ -56,7 +56,7 @@
 from pandas.sparse.array import BlockIndex, IntIndex
 from pandas.core.generic import NDFrame
 from pandas.core.common import needs_i8_conversion
-from pandas.io.common import get_filepath_or_buffer
+from pandas.io.common import get_filepath_or_buffer, _create_string_file_reader
 from pandas.core.internals import BlockManager, make_block
 import pandas.core.internals as internals
 
@@ -124,7 +124,7 @@ def read_msgpack(path_or_buf, iterator=False, **kwargs):
 
     Parameters
     ----------
-    path_or_buf : string File path, BytesIO like or string
+    path_or_buf : string File path or buffer
     iterator : boolean, if True, return an iterator to the unpacker
                (default is False)
 
@@ -146,26 +146,15 @@ def read(fh):
     # see if we have an actual file
     if isinstance(path_or_buf, compat.string_types):
 
-        try:
-            exists = os.path.exists(path_or_buf)
-        except (TypeError,ValueError):
-            exists = False
-
-        if exists:
-            with open(path_or_buf, 'rb') as fh:
-                return read(fh)
-
-    # treat as a string-like
-    if not hasattr(path_or_buf, 'read'):
-
-        try:
-            fh = compat.BytesIO(path_or_buf)
+        with open(path_or_buf, 'rb') as fh:
             return read(fh)
-        finally:
-            fh.close()
 
-    # a buffer like
-    return read(path_or_buf)
+    elif hasattr(path_or_buf, 'read'):
+        return read(path_or_buf)
+    else:
+        raise ValueError("path_or_buffer must be a file or file-like buffer")
+
+reads_msgpack = _create_string_file_reader(read_msgpack)
 
 dtype_dict = {21: np.dtype('M8[ns]'),
               u('datetime64[ns]'): np.dtype('M8[ns]'),

pandas/io/parsers.py

@@ -15,7 +15,7 @@
 import pandas.core.common as com
 from pandas.core.config import get_option
 from pandas.io.date_converters import generic_parser
-from pandas.io.common import get_filepath_or_buffer
+from pandas.io.common import get_filepath_or_buffer, _create_string_file_reader
 
 from pandas.util.decorators import Appender
 
@@ -417,6 +417,7 @@ def parser_f(filepath_or_buffer,
 
 read_csv = _make_parser_function('read_csv', sep=',')
 read_csv = Appender(_read_csv_doc)(read_csv)
+reads_csv = _create_string_file_reader(read_csv)
 
 read_table = _make_parser_function('read_table', sep='\t')
 read_table = Appender(_read_table_doc)(read_table)

pandas/io/pickle.py

@@ -1,5 +1,5 @@
 from pandas.compat import cPickle as pkl, pickle_compat as pc, PY3
-
+from pandas.io.common import _create_string_file_reader
 
 def to_pickle(obj, path):
     """
@@ -51,3 +51,5 @@ def try_read(path, encoding=None):
         if PY3:
             return try_read(path, encoding='latin1')
         raise
+
+reads_pickle= _create_string_file_reader(read_pickle)