On 1 Nov 2007, at 1:37 PM, Adam R. Maxwell wrote: > > On Nov 1, 2007, at 2:45 AM, Christiaan Hofman wrote: > >> >> On 1 Nov 2007, at 1:33 AM, Adam R. Maxwell wrote: >> >>> According to http://developer.apple.com/releasenotes/ >>> GraphicsImaging/RN-CoreImage/index.html the CILinearGradient should >>> only be reversed on 10.5 if the app was linked on 10.5. Were >>> gradients reversed on Skim on 10.5 when it was compiled on 10.4? >>> That's probably worth checking before release. >>> >> >> Can you see with a current nightly? If they are reversed now, then >> they were OK earlier. > > Last nightly has them correct, so I guess the docs are wrong. That's > actually easier to deal with, I think; changing behavior at compile > time is less fun. This makes me wonder about the other things they've > documented as changing when linked on 10.5. >
Do you mean the nightly of yesterday, or of today? Certainly check today's, because I changed it back just before that one. >>> Types are also pretty broken if we override NSDocumentController >>> methods and use LSItemContentTypes in the Info.plist (but I think >>> that's only when linking against the 10.5 SDK). I have some hacks >>> in BD to work around that, but the export stuff for templates still >>> doesn't work. The easy solution is to remove the UTI from the >>> plist and just use extensions. Alternately, we can use the new >>> export key NSExportableTypes for Info.plist that accepts a UTI, and >>> then declare a UTI for each of the types we can export. >>> >> >> But UITs won't work for export types, right? So perhaps we should >> just remove the UTIs? > > UTIs will work for export types if we use the NSExportableTypes plist > key for each document that supports export...I think. ISTR there's > some custom code in the document controller and document classes for > templates that may be problematic. The sad part is that I really > prefer UTIs over extensions and NSDocument's mishmash of named type/ > extension. > I'm talking about custom export types, those are not included in the plist. I agree that UTIs is generally better. However recently I had a problem that in Skim the dvi was not recognized because it checked the UTI, and no program on my system seems to bother declaring that UTI. >>> Any thoughts? The main problem now is for people compiling on >>> 10.5, since the trunk version won't open or save files. The branch >>> will work. >>> Is the relevant change for the branch that you removed the UTIs? >>> -- >>> adam >> >> And that's because of the UTI problem? > > That and the Omni frameworks won't compile. I don't think my > NSRecordAllocationEvent() change on the branch is correct for Tiger, > so that'll require a framework upgrade. > > -- > adam That's annoying. Do they already have an upgrade for Leopard? Christiaan ------------------------------------------------------------------------- This SF.net email is sponsored by: Splunk Inc. Still grepping through log files to find problems? Stop. Now Search log events and configuration files using AJAX and a browser. Download your FREE copy of Splunk now >> http://get.splunk.com/ _______________________________________________ Bibdesk-develop mailing list Bibdesk-develop@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bibdesk-develop
