Hi Simon, On Mon, Feb 05, 2018 at 12:36:06PM +0100, Simon McVittie wrote: > On Sun, 04 Feb 2018 at 16:16:14 +0200, Juhani Numminen wrote: > > (gweled:2987): GLib-WARNING **: GError set over the top of a previous > > GError or uninitialized memory. > > This indicates a bug in someone's code. You must ensure an error is NULL > > before it's set. > > The overwriting error message was: Operation not supported > > Regardless of whether there's been an incompatible behaviour change, > this warning indicates a bug in either gweled or a library that it > uses. GErrors are like exceptions; after calling some function that > fails, you must either propagate the error to your caller, or clear the > error indicator (usually with g_clear_error()) to indicate that it has > been handled. > > G_DEBUG=fatal-warnings is sometimes useful to pin down the cause for > warnings like this by turning them into fatal errors that leave a core > dump.
Yes, thanks, I was aware of that ;-). There are a number of problems in gweled around the troublesome RSVG call, notably forgetting to initialise the error pointer to NULL, and then calling g_free on it, which crashes gweled when an error actually occurs... I’m fixing all that in gweled, which will still leave the RSVG regression. Regards, Stephen
signature.asc
Description: PGP signature
