Use absolute import for ease of use outside baseband.

Also fix remaining reference to ASP.

baseband/kotekan/base.py

@@ -1,11 +1,11 @@
 # Licensed under the GPLv3 - see LICENSE
 import astropy.units as u
 
-from ..helpers import sequentialfile as sf
-from ..base.base import (
+from baseband.helpers import sequentialfile as sf
+from baseband.base.base import (
     FileBase,
     FileOpener, FileInfo)
-from ..base.file_info import FileReaderInfo
+from baseband.base.file_info import FileReaderInfo
 from .header import KotekanHeader
 from .frame import KotekanFrame
 

baseband/kotekan/frame.py

@@ -1,5 +1,5 @@
 # Licensed under the GPLv3 - see LICENSE
-from ..base.frame import FrameBase
+from baseband.base.frame import FrameBase
 from .header import KotekanHeader
 from .payload import KotekanPayload
 

baseband/kotekan/header.py

@@ -2,8 +2,8 @@
 from astropy.time import Time
 from astropy import units as u
 
-from ..base.header import ParsedHeaderBase
-from ..base.utils import fixedvalue
+from baseband.base.header import ParsedHeaderBase
+from baseband.base.utils import fixedvalue
 
 
 __all__ = ['KotekanHeader']
@@ -84,7 +84,7 @@ def fromfile(cls, fh, verify=True, **kwargs):
         """
         s = fh.read(cls._dtype.itemsize)
         if len(s) < cls._dtype.itemsize:
-            raise EOFError('reached EOF while reading ASPFileHeader')
+            raise EOFError('reached EOF while reading frame header')
         words = np.ndarray(buffer=s, shape=(), dtype=cls._dtype)
         self = cls(words, verify=verify, **kwargs)
         self.mutable = False

baseband/kotekan/payload.py

@@ -4,7 +4,7 @@
 
 import numpy as np
 
-from ..base.payload import PayloadBase
+from baseband.base.payload import PayloadBase
 
 
 __all__ = ['KotekanPayload']
@@ -19,5 +19,6 @@ def decode_4bit(words):
 
 
 class KotekanPayload(PayloadBase):
+    _dtype_word = np.dtype('u1')
     _decoders = {4: decode_4bit}
     _sample_shape_maker = namedtuple('SampleShape', 'npol')