Stig Bjørlykke <stig@...> writes: > I think this sort of warnings are important to an user trying to > configure the installed version of Wireshark (which usually is only > one) with some custom configurations, because he/she may not be able > to spot an error if not given a warning. They are less important to > developers/advanced users having multiple instances of Wireshark, > because we know what we are doing (well, most of the time).
Developers aside and focusing on normal users, in this case (the case of an invalid color filter coming from a saved filter), the user would have had to manually edit the colorfilters file (the one that says at the top, "DO NOT EDIT THIS FILE"), as Wireshark would not have allowed them to create an invalid color filter through the GUI. I don't think that's very likely to happen, so the only ones who probably see these specific color filter warnings are the developers doing things like what I'm doing. > My personal opinion is that we should focus on the user (give > warnings), but have an option for developers/advanced users to disable > warnings window(s). Maybe display all warnings (fetched from > statusbar_push_temporary_msg?) in one window at startup, with an > option to disable? This window can also be available from the menu > for later analysis of all warnings. Well, that's essentially what we've got. The warning would now be printed in the console window, assuming the user has configured his preferences such that "Open a console window" is not set to "Never". Of course, this is currently a "Windows-only" preference, but is there any reason why it couldn't be added for all platforms? That would give all users the choice of whether or not they wanted to see these warnings or not. ___________________________________________________________________________ Sent via: Wireshark-dev mailing list <wireshark-dev@wireshark.org> Archives: http://www.wireshark.org/lists/wireshark-dev Unsubscribe: https://wireshark.org/mailman/options/wireshark-dev mailto:wireshark-dev-requ...@wireshark.org?subject=unsubscribe
