sometimes scel will get stuck in a state where sclang-start reaches the point of
*** Welcome to SuperCollider 3.13.0. *** For help type C-c C-y.
but then nothing happens - no "Index help topics" or "Built symbol table" messages appear, and no SC code can be evaluated.
i am not sure what circumstances lead to this exactly, but after some debugging, it seems the problem relates to the sclang-command-process-previous variable. i checked its value while in this stuck state, and it was an extremely long string (didnt save it, sorry) which leads me to wonder whether some part of the string processing in sclang-command-process-filter just fails to deal with the amount of data.
regardless, it seems to me that sclang-command-process-previous should be cleared when sclang is stopped. after doing this manually, sclang-start works fine again. i can make a PR adding this to sclang-stop-command-process.
sometimes scel will get stuck in a state where
sclang-startreaches the point of*** Welcome to SuperCollider 3.13.0. *** For help type C-c C-y.but then nothing happens - no "Index help topics" or "Built symbol table" messages appear, and no SC code can be evaluated.
i am not sure what circumstances lead to this exactly, but after some debugging, it seems the problem relates to the
sclang-command-process-previousvariable. i checked its value while in this stuck state, and it was an extremely long string (didnt save it, sorry) which leads me to wonder whether some part of the string processing insclang-command-process-filterjust fails to deal with the amount of data.regardless, it seems to me that
sclang-command-process-previousshould be cleared when sclang is stopped. after doing this manually,sclang-startworks fine again. i can make a PR adding this tosclang-stop-command-process.