On Fri, Sep 09, 2005 at 02:07:01AM -0400, Ove Kaaven wrote: > Steve Langasek wrote:
> >>The wine-safe script displays no warning before running an msdos > >>program. > >>This is probably due to the fact that on line 57, XMESSAGE is used > >>before having been defined. > >>Therefore, any MSDOS executable will be run from a browser with *no* > >>confirmation from the user. > >Why should wine-safe display a warning before running the program? Is > >there any reason for this other than the second bug that you filed, > >citing incorrect mime handlers? > It should, that's why it's using "wine-safe" and not "wine", to allow > the user to control whether a potential windoze mail virus should be > allowed to run or not. So if wine-safe doesn't display a message, then > that's a bug, which I suppose I'll have to look into. > It's that it shouldn't register itself in mailcap that I'd disagree with: > > The wine package should not register wine-safe in mailcap. I *never* > > expect to run a program when running: > > see foobar.bat > I've discussed this before, at which time people gave the opinion that > wine *should* register itself in mailcap. Most users *do* want the likes > of "see funnyanimationfromafriend.exe" and "see cdtraycupholder.exe", at > least the MUA's equivalent, to run the program. At least with suitable > warning first, given that the program *may* be harmful (albeit much less > so than on a real Windows system). If you have the kinds of friends > normal people do, though, then most of the time they do want to run > whatever amusements they get sent. Wow, well I certainly disagree with that. I use wine, but would never want something like this on my systems; I don't see any reason why "view an attachment" should be converted into "run an attachment" under any circumstances, whether the executable is a Windows executable or a Linux executable. And if there *is* demand for such a feature, I think it ought to be handled consistently for all executable types, using binfmt in the kernel instead of giving mailers that extra-special virusy goodness for purposes of Windows compatibility. :) Anyway, I think there's certainly only one RC bug here, and the other represents a bug of lower severity... it seems there's disagreement about which should be which, though. -- Steve Langasek Give me a lever long enough and a Free OS Debian Developer to set it on, and I can move the world. [EMAIL PROTECTED] http://www.debian.org/
signature.asc
Description: Digital signature
