diff options
Diffstat (limited to 'numpy/lib/io.py')
| -rw-r--r-- | numpy/lib/io.py | 12 |
1 files changed, 11 insertions, 1 deletions
diff --git a/numpy/lib/io.py b/numpy/lib/io.py index 4ff2a3945..12765e17c 100644 --- a/numpy/lib/io.py +++ b/numpy/lib/io.py @@ -837,6 +837,7 @@ def genfromtxt(fname, dtype=float, comments='#', delimiter=None, skiprows=0, missing_values = [_.missing_values for _ in converters] # Update the converters to use the user-defined ones + uc_update = [] for (i, conv) in user_converters.iteritems(): # If the converter is specified by column names, use the index instead if _is_string_like(i): @@ -850,6 +851,9 @@ def genfromtxt(fname, dtype=float, comments='#', delimiter=None, skiprows=0, converters[i].update(conv, default=None, missing_values=missing_values[i], locked=True) + uc_update.append((i, conv)) + # Make sure we have the corrected keys in user_converters... + user_converters.update(uc_update) # Reset the names to match the usecols if (not first_line) and usecols: @@ -960,8 +964,14 @@ def genfromtxt(fname, dtype=float, comments='#', delimiter=None, skiprows=0, descr.append(('', ttype)) else: descr.append(('', dtype)) + # So we changed the dtype ? if not ishomogeneous: - dtype = np.dtype(descr) + # We have more than one field + if len(descr) > 1: + dtype = np.dtype(descr) + # We have only one field: drop the name if not needed. + else: + dtype = np.dtype(ttype) # output = np.array(data, dtype) if usemask: |