empty chararrays (ticket 1948)
Hi, In ticket 1948 a backwards compatibility issue with chararray is reported. Indexing a chararray with [] or a bool array of False used to return [] in numpy 1.2.1 (consistent with ndarray behavior), but now returns an empty string. Unfortunately this changed behavior has been present for the 1.5.x and 1.6.x releases. So the question is if this should be changed back or not? The change was likely unintentional; there's no test for it. Ralf
On Sun, Mar 25, 2012 at 10:12 AM, Ralf Gommers <ralf.gommers@googlemail.com>wrote:
Hi,
In ticket 1948 a backwards compatibility issue with chararray is reported. Indexing a chararray with [] or a bool array of False used to return [] in numpy 1.2.1 (consistent with ndarray behavior), but now returns an empty string. Unfortunately this changed behavior has been present for the 1.5.x and 1.6.x releases.
So the question is if this should be changed back or not? The change was likely unintentional; there's no test for it.
I believe the stsci folks were maintaining chararray, although I don't see anyone from there with commit permissions. Hmm... I'd be inclined to reinstate the old behavior, but the stsci folks may have deliberately made the change, I'd like to hear from them first. Chuck
On Sun, Mar 25, 2012 at 7:03 PM, Charles R Harris <charlesr.harris@gmail.com
wrote:
On Sun, Mar 25, 2012 at 10:12 AM, Ralf Gommers < ralf.gommers@googlemail.com> wrote:
Hi,
In ticket 1948 a backwards compatibility issue with chararray is reported. Indexing a chararray with [] or a bool array of False used to return [] in numpy 1.2.1 (consistent with ndarray behavior), but now returns an empty string. Unfortunately this changed behavior has been present for the 1.5.x and 1.6.x releases.
So the question is if this should be changed back or not? The change was likely unintentional; there's no test for it.
I believe the stsci folks were maintaining chararray, although I don't see anyone from there with commit permissions. Hmm... I'd be inclined to reinstate the old behavior, but the stsci folks may have deliberately made the change, I'd like to hear from them first.
The change was made by Michael Droettboom (CC'd), who did have commit permissions for this. They got lost with the Github move it seems. Ralf
On Sun, Mar 25, 2012 at 7:09 PM, Ralf Gommers <ralf.gommers@googlemail.com>wrote:
On Sun, Mar 25, 2012 at 7:03 PM, Charles R Harris < charlesr.harris@gmail.com> wrote:
On Sun, Mar 25, 2012 at 10:12 AM, Ralf Gommers < ralf.gommers@googlemail.com> wrote:
Hi,
In ticket 1948 a backwards compatibility issue with chararray is reported. Indexing a chararray with [] or a bool array of False used to return [] in numpy 1.2.1 (consistent with ndarray behavior), but now returns an empty string. Unfortunately this changed behavior has been present for the 1.5.x and 1.6.x releases.
So the question is if this should be changed back or not? The change was likely unintentional; there's no test for it.
I believe the stsci folks were maintaining chararray, although I don't see anyone from there with commit permissions. Hmm... I'd be inclined to reinstate the old behavior, but the stsci folks may have deliberately made the change, I'd like to hear from them first.
The change was made by Michael Droettboom (CC'd), who did have commit permissions for this. They got lost with the Github move it seems.
I've sent a PR for this, https://github.com/numpy/numpy/pull/247, which changes chararray to return an empty chararray for indexing with an empty array/list. Ralf
participants (2)
-
Charles R Harris -
Ralf Gommers