Yair K. wrote: > On Sunday 03 February 2008 03:58:56 Yair K. wrote: >> On Sunday 03 February 2008 03:36:29 Clive Wright wrote: >>> Re. 4.B >>> No need to create an extra variable (aval) just move set_value function >>> call between val assignments. >> The problem is that set_values can detect a mixer structure change, so it >> calls reload_gui (). After that, we don't know that srec exists, so ossxmix >> could crash. > > To be more accurate, we can be almost certain that srec doesn't exist, since > reload_gui () frees all these variables and creates new records. Even if srec > accidently matches something, srec->left/right probably don't match a slider. > >> (That's why set_value was moved to be always after >> gtk_adjustment_... in the previous patch, and why ossxmix now checks for -1 >> return value of get_value). >> > > Another option would be to let set_value return -1 and check for that as in > get_value, but I found that to be mostly unneccesary since moving set_value > to be after gtk_adjustment_.. was an easy solution. > Thats ok. Leave it as it is. I had identified a problem with the gang_change function and proposed a patch which Hannu did not implement fully as he preferred max level to mean level. I had assumed that the call to set_value had moved as a consequence of partial implementation of my patch. I had not taken your change to reload the gui instead of restarting ossxmix into account.
Slightly off topic, why does vmix0-vol mono-slider not get updated in ossxmix when its value gets changed as a result of an external event? Is it because widgets that have their parent equal to zero get handled differently to other widgets? If so is there a good reason for this as it also appears to effect the width of gtkvu widgets? Kind regards, Clive _______________________________________________ oss-devel mailing list oss-devel@mailman.opensound.com http://mailman.opensound.com/mailman/listinfo/oss-devel
