Cannot create patterns

[GoogleCodeExporter]

After upgrading to Klystrack 1.6.0 r1162M, I cannot make patterns.

If I make a new file, typing '0' in the sequence editor on any row after 0000 
will show "00+0", as expected.  When I type '0' on row 0000 though, the pattern 
is not visible unless it is moved down using Alt+F12.  Pressing enter on any 
pattern in the sequence editor takes me to the pattern editor, where the track 
appears to be blank, except for the cursor. No dashes or numbers.  Pressing 
Alt+F12 does not do anything. I cannot enter notes even in edit mode.

Maybe I am missing something?

Original issue reported on code.google.com by erdav...@gmail.com on 3 May 2012 at 6:33

[GoogleCodeExporter]

I can't reproduce this. One possible cause I can think of is that the pattern 
is zero steps long - if you put pattern 00 somewhere where it is visible, does 
it have at least one step?

Original comment by kometb...@gmail.com on 10 May 2012 at 2:43

[GoogleCodeExporter]

No, it seems like there are no steps in any pattern.  Maybe I messed up the new 
file template when I compiled it for OS X.

Original comment by erdav...@gmail.com on 10 May 2012 at 3:52

[GoogleCodeExporter]

Can you check if you can use the sequence editor normally if you load an 
example song (that has correct patterns) and put those patterns in the 
sequence? Maybe for some reason patterns are created with a length of zero...

Original comment by kometb...@gmail.com on 12 May 2012 at 6:48

[GoogleCodeExporter]

Yes, if I load a song I can edit its patterns just fine, and it displays 
correctly in the sequencer window. I still have the problem when I try to 
create new patterns, but duplicating works.

Original comment by erdav...@gmail.com on 1 Jun 2012 at 9:33

[GoogleCodeExporter]

I have the same list of symptoms on 64-bit Archlinux.

I tried compiling klystrack for 32-bit architecture (I installed the 32-bit 
versions of the dependencies and passed the option -m32 to gcc), and the 
pattern editor finally showed up. So, this problem is very likely related to 
64-bit compatibility.

Original comment by guevel.e...@gmail.com on 27 Apr 2013 at 12:46

Attachments:

• screenshot01.png

[GoogleCodeExporter]

That's strange. :o I guess 64-bit compilation is simply out of question, for 
now. I'm unable to compile 64-bit for the moment so thanks for finding this out 
for me.

Original comment by kometb...@gmail.com on 28 Apr 2013 at 8:43

[GoogleCodeExporter]

I was also running 64-bit OS X 10.7.5 when I experienced this bug. Sorry I 
forgot to write that in the original post.

Original comment by erdav...@gmail.com on 28 Apr 2013 at 12:25

[GoogleCodeExporter]

New user. Experiencing this symptom EXACTLY as described on Win7 x64 with the 
1.6.0 "Xmas preview" December 2012 release and all later nightly-based compiles 
provided here. 

Klystrack 1.5.4 works fine. Might want to post a warning somewhere conspicuous 
for x64 users...!

Original comment by ledrobs...@gmail.com on 13 Dec 2013 at 11:08

[GoogleCodeExporter]

Original comment by kometb...@gmail.com on 19 Dec 2013 at 12:19

• Changed state: Accepted

[GoogleCodeExporter]

To fix the problem with the sequnce editor not displaying the first row, all I 
had to do was change line 64 in src/view/sequence.c from this:

if (sp->position + len <= top) continue;

to this:

if (sp->position + len < top) continue;

(from '<=' to '<')

Original comment by megalole...@gmail.com on 2 Nov 2014 at 9:43

[GoogleCodeExporter]

[deleted comment]

[GoogleCodeExporter]

Also, as a quick fix for the other problem, I added this line to src/event.c at 
line 694:

if(mused.song.pattern[pattern].num_steps == 0) 
resize_pattern(&mused.song.pattern[pattern], mused.sequenceview_steps);

as the first line inside the add_sequence function.  There's probably a nicer 
solution, but at least this works for now.

Original comment by megalole...@gmail.com on 2 Nov 2014 at 12:10

[GoogleCodeExporter]

Original comment by kometb...@gmail.com on 15 Dec 2014 at 2:39

• Changed state: Fixed

[GoogleCodeExporter]

This issue was closed by revision r1289.

Original comment by kometb...@gmail.com on 15 Dec 2014 at 2:40