Can anyone confirm or deny this Savannah problem? I haven't ever seen it.
---------- Forwarded message ---------- From: Max <[email protected]> Date: 13 July 2011 15:14 Subject: Re: [OctDev] Bug in __getlegenddata__.m To: Søren Hauberg <[email protected]> Cc: [email protected] On 12.07.2011 22:55, Søren Hauberg wrote: > tir, 12 07 2011 kl. 21:18 +0200, skrev Max: >> Hello, >> >> I think, there is a bug in __getlegenddata__.m (3.4.2): >> >> 35 while (k > 0 >> 36 && ! (strcmp (typ, "line") || strcmp (typ, "surface") >> 37 || strcmp (typ, "patch") || strcmp (typ, "hggroup"))) >> 38 typ = get (kids(--k), "type"); >> 39 endwhile >> >> k=1 leads to "error: subscript indices must be either positive integers >> or logicals". It should be k-- in line 38 (or k>1 in line 35?). >> >> Best, >> Max > > This looks like a bug in core Octave rather than a specific package. > Could you file a report at http://bugs.octave.org ? > > Thanks > Søren > I really would like to file a bug report but http://savannah.gnu.org gives me "Failed to connect to database: Lost connection to MySQL server at 'reading initial communication packet', system error: 113" whenever I try to connect to the site (is this only a problem for me?). The error can be reproduced by the command sequence figure(1) clf(gcf()) hold on plot(1,1,'*k;Data;'); text(1,1,' Data 1'); plot(2,2,'*k'); hold off line 6 (2nd plot command) then leads to error: __getlegenddata__: subscript indices must be either positive integers or logicals error: called from: error: /usr/local/share/octave/3.4.2/m/plot/private/__getlegenddata__.m at line 38, column 11 error: /usr/local/share/octave/3.4.2/m/plot/private/__plt__.m at line 57, column 23 error: /usr/local/share/octave/3.4.2/m/plot/plot.m at line 194, column 9 error: /home/lilge/test.m at line 6, column 1 Best, Max _______________________________________________ Octave-dev mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/octave-dev
