Charly Coste

> Using libraries.io, bundler has 24.7k dependent repos but pip has 96 and setuptools has 159? how does that make sense? setuptools and pip usually don't appear in the list...

> hmm. that makes it hard to use libraries.io's dependencies metrics as an apples-to-apples comparison of importance then, right? Dependencies metrics should be pretty reliable for everything besides build tools,...

> OpenSSL has no replacement in the wings. There were at least two alternatives even before Heartbleed, and new ones since then.

I don't have a simple way of reproducing the problem. I discovered it when running [liberapay](https://github.com/liberapay/liberapay.com)'s test suite with the experimental coverage plugins, and I confirmed that the changes in...

I've rebased this branch on master, replacing `frame->f_lineno` with `PyFrame_GetLineNumber(frame)` like in 7e5e28f1aba87c10b96d0ae1244352f4c520aedc.

I was thinking users would choose depending on the kind of data they're loading. Basically, `CBORDecoder` could have an `intern` option with four possible values: `'all'` would intern all strings,...

The possibilities of significant slowdowns I can think of are: - if the decoder tries to intern very long strings, because the hashing time is proportional to the string length,...

I'm not convinced by the proposed API. A stream of CBOR objects is officially called a CBOR Sequence ([draft-ietf-cbor-sequence-02](https://tools.ietf.org/html/draft-ietf-cbor-sequence-02)), so we should use the word "sequence" instead of "multiple". The...

Maybe `load_sequence` and `loads_sequence`? Or `load_seq` and `loads_seq`? Alternatively we could add a `sequence=False` keyword argument to the existing functions. The impact on performance of a single boolean check should...

I'm not a CPython expert either, but I managed to get it work. I've opened a PR on @toravir's fork to merge my changes into this branch: .