Fixes #1091 - Support gzip encoded requests

[orf]

I've never written any C++ nor used Bazel before, so please don't laugh if this is terrible!

I've had a go at supporting gzipped request bodies inside the tensorflow-serving http server: #1091

I've got no idea if the ticket will be accepted, but I thought it would be fun to have a go at fixing it.

I've also got no idea how to test this, I copied an existing test and made it used zlib (compatible with gzip) to send encoded data.

[googlebot]

Thanks for your pull request. It looks like this may be your first contribution to a Google open source project (if not, look below for help). Before we can look at your pull request, you'll need to sign a Contributor License Agreement (CLA).

📝 Please visit https://cla.developers.google.com/ to sign.

Once you've signed (or fixed any issues), please reply here (e.g. I signed it!) and we'll verify it.

---

What to do if you already signed the CLA

Individual signers

• It's possible we don't have your GitHub username or you're using a different email address on your commit. Check your existing CLA data and verify that your email is set on your git commits.

Corporate signers

• Your company has a Point of Contact who decides which employees are authorized to participate. Ask your POC to be added to the group of authorized contributors. If you don't know who your Point of Contact is, direct the Google project maintainer to go/cla#troubleshoot (Public version).
• The email used to register you as an authorized contributor must be the email used for the Git commit. Check your existing CLA data and verify that your email is set on your git commits.
• The email used to register you as an authorized contributor must also be attached to your GitHub account.

[orf]

I signed it!

[googlebot]

CLAs look good, thanks!

[netfs]

Thanks for your effort and the PR!

[wenbozhu]

Sorry for the delayed response. I am working on the tensorflow-serving http server and the proper fix will require a bit more work (to be spec compliant etc). We may not want to depend on the gzip wrapper library either.

Do you mind filing an issue (with a link to this PR) and I try to get the feature supported soon?

Thanks.

[orf]

Sorry for the delayed response. I am working on the tensorflow-serving http server and the proper fix will require a bit more work (to be spec compliant etc). We may not want to depend on the gzip wrapper library either.

Do you mind filing an issue (with a link to this PR) and I try to get the feature supported soon?

Thanks.

I've filed a bug here: #1091

Yeah this will take more work, I guess this was just a POC more than anything else. I saw that Tensorflow itself has some gzip/zlib decoders, but they seem tightly coupled to files and hard to adapt here.

[gautamvasudevan]

With the issue filed, we'll track the work there. Thanks again!