fix #3; allow for blank lines in input file

HISTORY.rst

@@ -1,6 +1,11 @@
 History
 =======
 
+v0.4.3 (2018-04-10)
+-------------------
+* Allow for blank lines at beginning of input file [fixes #3]
+
+
 v0.4.2 (2017-05-29)
 -------------------
 * parser saves unknown tags into an ``unknown_tag`` key in dict

RISparser/__init__.py

@@ -1,6 +1,6 @@
 """Parse WOK and Refman's RIS files"""
 
-__version__ = '0.4.2'
+__version__ = '0.4.3'
 
 from .config import LIST_TYPE_TAGS, TAG_KEY_MAPPING # noqa
 from .parser import * # noqa

RISparser/parser.py

@@ -167,7 +167,7 @@ def readris(file_, mapping=None, wok=False):
  # Corrects for BOM in utf-8 encodings while keeping an 8-bit
  # string representation
  st = filelines[0]
- if (st[0], st[1], st[2]) == ('\xef', '\xbb', '\xbf'):
+ if len(st) >= 3 and (st[0], st[1], st[2]) == ('\xef', '\xbb', '\xbf'):
  filelines[0] = st[3:]
 
  return read(filelines, mapping, wok)

tests/example_starting_newlines.ris

@@ -0,0 +1,11 @@
+
+
+TY - JOUR
+AU - Shannon,Claude E.
+PY - 1948/07//
+TI - A Mathematical Theory of Communication
+JF - Bell System Technical Journal
+SP - 379
+EP - 423
+VL - 27
+ER - 

tests/test_parser.py

@@ -167,3 +167,9 @@ def test_parse_multiple_unknown_tags_ris(self):
  with open(filepath, 'r') as bibliography_file:
  entries = list(readris(bibliography_file))
  self.compare([result_entry], entries)
+
+ def test_starting_newline(self):
+ fn = os.path.join(CURRENT_DIR, 'example_starting_newlines.ris')
+ with open(fn, 'r') as f:
+ entries = list(readris(f))
+ assert len(entries) == 1