cmn-display not generating .eps file

[terrymullett]

Hi all,

I've spent some time googling this and poring over the source code, and decided I'd better ask for help. None of my calls to cmn-display (either for a whole slippery-chicken object or for a set-palette or rthm-seq-palette) produces an .eps file. I appreciate any help :-)

I'm on Linux (Unbuntu 24.04) with SBCL 2.2.9 running in Emacs/Slime, though I've tried these things at the SBCL prompt in a terminal as well. I've set
(sc:set-sc-config 'default-dir "/home/terry/Desktop/")
and although it didn't look like it would matter I also set
(setf cmn::cmn-output-pathname "/home/terry/Desktop")

Both of those are in my .sbclrc and checking those variables shows they have the expected values.

Also, running straight cmn works, as in
(cmn staff treble c4 q)
produces the file I expect (aaa.eps) in the folder indicated above.

When I try to run any of the examples in the manual that have (cmn-display), I see a "file not found" in my inferior-lisp output for the expected file. Or to be clear, that's what I see when I run it on a slippery-chicken object. For a set-palette, I see no errors, but the method returns NIL.

No other errors or warnings. MIDI, Lilypond and MusicXML files all generate properly in the expected location.

Sorry if this is obvious or whatever, but I'm a little stumped. Having both Lilypond and Musescore on the machine, I'm not in general hurting for notation output, but it would be very helpful I think to be able to use the palette displays.

Oh, running cmn-display on a set-palette returns NIL. If I understand the source docs correctly, it would be true on success, right? But I get no errors or warnings in my inferior-lisp buffer for that ...

Thanks in advance!
Terry

[terrymullett]

oh, one correction to the above, sorry, that should have been
(setf cmn::cmn-output-pathname "/home/terry/Desktop/aaa.eps")
for the one I thought probably doesn't matter :-)

[mdedwards]

Hello, Terry, could it be that you loaded/compiled slippery chicken before CMN was loaded/compiled? If so that would mean all cmn methods in slippery would be essentially left out. Try deleting all the .fasl (compilation files) for slippery — (asdf:clear-system "slippery-chicken”) — load the packages you want, including cmn, then try again with slippery. If that doesn’t help, please send along your sbcl init file and the code you’re trying to run. I’m off for more than two weeks later today and I’m going to be maxed out with workshops/performances in Bangkok—wifi will be patchy too—so apologies if I don’t respond again until after April 8th. Best, Michael

[terrymullett]

Michael, thanks for the fast reply. I hope everything goes well for you in Bangkok!

This is of course no big emergency, more of a puzzle that I'm having a hard time letting go of, and I didn't make it a bug report since it seems like something that would have come up earlier if it were a bug.

I tried clearing the FASLs but get the same results, so I'm attaching my .sbclrc, one example of the code I've tried, and what appears in the REPL during the run. All of the Lilypond files, MIDI, and MusicXML do get generated, just not the .eps.

Thanks for looking at this.
Best,
Terry
cmn-display test.zip

[terrymullett]

Solved, with
rm -rf ~/.cache/common-lisp
and then restarting SBCL.

[mdedwards]

Ah that’s really good. Thanks for letting me know Terry. So (asdf:clear-system "slippery-chicken”) didn’t work, rather you had to delete the .fasl files with an rm -f? Rings a bell to be honest. And so I’m guessing that it was a compilation order thing. I’ll have to work in some mechanism to detect this problem and issue a suitable warning. All the best, Michael

[mdedwards]

PS I’ve added some eval-when magic to produce a hopefully helpful error message when this occurs. It’s a rare (first time I’ve seen it!) situation but when you’re getting started it’s not obvious why it occurs. Thanks again for the very helpful error reporting.

…

On 25. Mar 2026, at 11:53, Michael Edwards ***@***.***> wrote: Ah that’s really good. Thanks for letting me know Terry. So (asdf:clear-system "slippery-chicken”) didn’t work, rather you had to delete the .fasl files with an rm -f? Rings a bell to be honest. And so I’m guessing that it was a compilation order thing. I’ll have to work in some mechanism to detect this problem and issue a suitable warning. All the best, Michael

[terrymullett]

All of that pretty much sounds right to me, though I'd add that removing all the FASLs on the system isn't exactly surgical :-D I might have narrowed that down some. But what really took me there was suspecting that my Lisp environment might not be entirely sane, and could probably use a good spring cleaning anyway. Thanks for helping with this, though, your pointing to the FASLs solved it. Best regards Terry

…

On Tue, Mar 24, 2026 at 10:54 PM Michael Edwards ***@***.***> wrote: Ah that’s really good. Thanks for letting me know Terry. So (asdf:clear-system "slippery-chicken”) didn’t work, rather you had to delete the .fasl files with an rm -f? Rings a bell to be honest. And so I’m guessing that it was a compilation order thing. I’ll have to work in some mechanism to detect this problem and issue a suitable warning. All the best, Michael — Reply to this email directly, view it on GitHub <#109?email_source=notifications&email_token=ABNZNAYHCKWUZIOLE5OYB5T4SNQ6XA5CNFSNUABIM5UWIORPF5TWS5BNNB2WEL2ENFZWG5LTONUW63SDN5WW2ZLOOQXTCNRTGA2DENRZUZZGKYLTN5XKMYLVORUG64VFMV3GK3TUVRTG633UMVZF6Y3MNFRWW#discussioncomment-16304269>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ABNZNA2ERRT6EK4N4F5ZCST4SNQ6XAVCNFSM6AAAAACW27KKJOVHI2DSMVQWIX3LMV43URDJONRXK43TNFXW4Q3PNVWWK3TUHMYTMMZQGQZDMOI> . You are receiving this because you authored the thread.Message ID: ***@***.*** com>

[mdedwards]

Hi Terry,

All of that pretty much sounds right to me, though I'd add that removing all the FASLs on the system isn't exactly surgical :-D I might have narrowed that down some.

It’s a big hammer for a small job, for sure, but you don’t lose anything by deleting them all except for a little compilation time.

But what really took me there was suspecting that my Lisp environment might not be entirely sane, and could probably use a good spring cleaning anyway.

:-) Mine too, to be honest.

Thanks for helping with this, though, your pointing to the FASLs solved it.

You’re welcome. Glad we found the solution so easily. All the best, Michael