On 26 April 2012 11:12, Dodji Seketeli wrote: > > So maybe it'd be better to canonicalize the _cpp_file::path when it's > first build? One drawback of that approach would be that > _cpp_file::path will then permanently loose the information about the > current directory, that is indirectly encoded into the way the file path > is originally built. But do we really need that information?
I don't know, but I certainly don't need it in diagnostic output. It occurred to me last night (while debugging some deeply nested template instantiations in libstdc++ headers) that the same target/lib/../../../.. junk appears in file paths shown by GDB. Does this change also affect the paths in the debug info? I hope it does, as it will make stack traces involving libstdc++ headers easier to read.
