octave-bug-tracker
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[Octave-bug-tracker] [bug #56047] dlmread with NaN as range parameter re


From: Mike Miller
Subject: [Octave-bug-tracker] [bug #56047] dlmread with NaN as range parameter returns an empty array
Date: Mon, 1 Apr 2019 12:26:42 -0400 (EDT)
User-agent: Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/73.0.3683.86 Safari/537.36

Update of bug #56047 (project octave):

                Severity:              3 - Normal => 2 - Minor              
              Item Group:                   Other => Incorrect Result       
                  Status:                    None => Need Info              
                 Summary: dlmread() - NaN / different file length => dlmread
with NaN as range parameter returns an empty array

    _______________________________________________________

Follow-up Comment #1:

Thank you for your report. I don't know if this is a bug or not.

Since you didn't explicitly show the difference, in Octave 4.2.1 and earlier I
see


>> data = dlmread ("foo.txt", ",")
data =

   0.7246355   0.9038270   0.2715474   0.0537131   0.0062652   0.6589647
   0.5013773   0.7448110   0.9847691   0.6116570   0.3918950   0.8894601
   0.1572770   0.2567405   0.1690453   0.5567996   0.0110273   0.2038035
   0.1387583   0.5858817   0.5020694   0.4169618   0.8551791   0.0294324
   0.5909749   0.9383733   0.8427086   0.4166940   0.3021889   0.8196946
   0.8825450   0.8234280   0.3286623   0.2107583   0.9080787   0.1634514

>> size (data)
ans =

   6   6

>> data = dlmread ("foo.txt", ",", [1, 1, NaN, 2])
data =

   0.74481   0.98477
   0.25674   0.16905
   0.58588   0.50207
   0.93837   0.84271
   0.82343   0.32866



That is, NaN is a synonym for the last row or column.

In Octave 4.2.2


>> data = dlmread ("foo.txt", ",", [1, 1, NaN, 2])
data = [](0x0)


But, in Octave 4.2.2 and current versions of Octave (highly recommended to
upgrade), you can use Inf instead of NaN to get the behavior that you want.
Using a parameter of Inf makes more intuitive sense to me than NaN.

What does Matlab do with a range parameter of NaN?

    _______________________________________________________

Reply to this item at:

  <https://savannah.gnu.org/bugs/?56047>

_______________________________________________
  Message sent via Savannah
  https://savannah.gnu.org/




reply via email to

[Prev in Thread] Current Thread [Next in Thread]