Steven Fackler

Hmm, should this STT_NOTYPE actually be here? https://github.com/gimli-rs/object/blob/master/src/read/elf/symbol.rs#L476 It looks like the backtrace crate limits itself to STT_FUNC and STT_OBJECT: https://github.com/rust-lang/backtrace-rs/blob/master/src/symbolize/gimli/elf.rs#L113

Alternatively, the logic could try to pick the "best" name for a given address. Beyond the weird `$x.224` symbols, this could help more generally. For example, in Ubuntu 20.04's libpthread.so.0,...

> As an aside, we don't currently include versions in the map, but maybe that would be useful. It appears that the version is directly embedded in the symbol name,...

Updated to filter out zero sized symbols instead.

This is the logic I ended up using, which also incorporates that heuristic to pick the "best" symbol: ```rust fn symbol_map(file: &File) -> SymbolMap>::new(); let symbols = file .symbol_table() //...

Your terminal controls what the colors look like: https://en.wikipedia.org/wiki/ANSI_escape_code#Colors

Ah, gotcha. I pulled the color selection from https://logback.qos.ch/manual/layouts.html#coloring but don't feel particularly strongly about the specifics.

I license past and future contributions under the dual MIT/Apache-2.0 license, allowing licensees to chose either at their option

The spec seems pretty unambiguous to me: > Otherwise, clients must use `application/json`. It's breaking a Conjure server implementation that was built with an understanding that "must use application/json" meant...

Sure - I think a little toy HTTP server would make sense as a decent example.