[CMake] xcodeproj internal reference lottery

Bill Hoffman bill.hoffman at kitware.com
Wed Mar 3 12:12:03 EST 2010


hauns at mac.com wrote:
>> Those id's are all over the place in projects.  But, sounds like the 
>> only ones that need to persist are the ones that represent files.  I 
>> suppose you could load in the existing project if there is one, and 
>> populate a map of object id's for the files.  You would have to write a 
>> parser for the crazy Xcode file format.  That would be the cleanest way 
>> to do it, as the information is already on the disk.   As Dave 
>> mentioned, we don't have any funding for this right now, but you can 
>> clone CMake, and give it a try.  I can provide assistance if you want to 
>> attempt this.
>>
>> Thanks.
>>
>> -Bill
> 
> isn't it easier to just query hash values from the file system( for
files that exist on disk) as the xcodeproj is generated? for nonexistent
files (build products) one has to find unique new ids - using the
current method. that would save us the xcodeproj parsing bit.
> 
> what expert time/funding are we talking? 
>

That might work as well.  You would have to come up with a hash that was 
exactly this long: 218CE70218CE70218CE70000, and was always unique to 
the file.  But, if you did that it would work as well.

I am not sure what you are asking with the time/funding question?

-Bill



More information about the CMake mailing list