Thanks, Carl. Yes, I’ve read about this. Part of what I’m trying to do is avoid going down useless alleys. We’ll see if this helps.
I also suspect Han-Wen’s advice (ly_display_scm()) will help as well… -d > On Mar 30, 2016, at 12:10 PM, Carl Sorensen <[email protected]> wrote: > > On 3/30/16 10:12 AM, "[email protected] on > behalf of Redwood (Daniel)" > <[email protected] on behalf of > [email protected]> wrote: > >> >> 2) I try debugging from within C++ >> >> (gdb) b Performance::output >> Breakpoint 1 at 0x82d6750: file performance.cc, line 72. >> (gdb) r dant.ly >> [Š] >> Breakpoint 1, Performance::output (this=0x87a74f0, midi_stream=..., >> performance_name="") at performance.cc:72 >> (gdb) p *this >> $1 = {<Music_output> = {<Smob<Music_output>> = {<Smob_core> = { >> self_scm_ = 0xb2915d30}, <Smob_base<Music_output>> = { >> static smob_tag_ = 14975, static scm_init_ = {static list_ = >> 0x84c3edc <Smob_base<Scale>::scm_init_>, fun_ = 0x80e12b0 >> <Smob_base<Music_output>::init()>, >> next_ = 0x84bd090 <Smob_base<Paper_outputter>::scm_init_>}, >> static smob_name_ = "Music_output"}, protection_cons_ = 0x404}, >> _vptr.Music_output = 0x83cc308 <vtable for Performance+8>, >> static type_p_name_ = "ly:music-output?"}, >> audio_staffs_ = std::vector of length 2, capacity 2 = {0x87a6a18, >> 0x87a6c88}, audio_elements_ = std::vector of length 18, capacity 32 = >> { >> 0x87a6a18, 0x865a798, 0x865a7b0, 0x865a900, 0x865c3b0, 0x865c4b0, >> 0x865c4f0, 0x865a8e8, 0x87a4838, 0x87a6cf8, 0x87a6c88, 0x865caf0, >> 0x865cd38, 0x865c910, 0x865c700, 0x865c788, 0x865ce30, 0x865c660}, >> midi_ = >> 0x87a4ff0, ports_ = false, header_ = 0x404} >> >> My guess is that most of these raw numbers are actually scheme objects, >> which I don¹t know how to debug. > > Please look at section 10.7.2 of the Contributor's Guide. It has a > gdbinit file that should allow you to do something like > > ps audio_elements_[0] > > or > > pmusic *this > > at a debugger prompt to see the scheme object. > > Warning -- I haven't tried this. And I'm just shooting in the dark. But > the gdbinit file certainly helped me in the past. > > HTH, > > Carl > _______________________________________________ lilypond-devel mailing list [email protected] https://lists.gnu.org/mailman/listinfo/lilypond-devel
