labath added inline comments.
================ Comment at: lldb/lit/SymbolFile/NativePDB/local-variables.cpp:31-37 +// CHECK-NEXT: 14 } +// CHECK-NEXT: 15 +// CHECK-NEXT: 16 int main(int argc, char **argv) { +// CHECK-NEXT: -> 17 int SomeLocal = argc * 2; +// CHECK-NEXT: 18 return Function(SomeLocal, 'a'); +// CHECK-NEXT: 19 } +// CHECK-NEXT: 20 ---------------- zturner wrote: > labath wrote: > > I'm not thrilled by all of the hard-coded information (line numbers, the > > default format of stop-information printing) here, which is irrelevant for > > the tested functionality here. This will make it very hard to update this > > test if there is ever a need for (perhaps one would like to remove the line > > break in the RUN: command once the `env LLDB_USE_NATIVE_PDB_READER=1` part > > disappears), or the default way of printing stop information in lldb > > changes. It doesn't seem like a maintainable long term strategy. > I felt the same way TBH, but wanted to see what the feedback would be like. > What do you think about just removing all the backtrace lines from the check > output and only checking the print commands and their output? I think that would be **much** better. CHANGES SINCE LAST ACTION https://reviews.llvm.org/D55575/new/ https://reviews.llvm.org/D55575 _______________________________________________ lldb-commits mailing list lldb-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits