Serialization improvements

[dtebbs]

• standard way to export variable assignments
• some remaining interface inconsistencies and composability fixes
• modified binary format with maximum compatibility with other implemenations. serializers for all relevant objects.

[AntoineRondelet]

@dtebbs - I see that you modified the submodule here. Was that necessary for the CI to pass?
I'm asking in the context of #361 for which the CI fails with:

npm WARN The package scrypt is included as both a dev and production dependency.

npm ERR! code 1
npm ERR! Command failed: git submodule update -q --init --recursive
npm ERR! warning: templates not found in /tmp/pacote-git-template-tmp/git-clone-1caab2ca
npm ERR! fatal: remote error: upload-pack: not our ref 1710792e1b6da5905dadda19167475f458f2b2ba
npm ERR! Fetched in submodule path 'depends/libff', but it did not contain 1710792e1b6da5905dadda19167475f458f2b2ba. Direct fetching of that commit failed.
npm ERR! 

npm ERR! A complete log of this run can be found in:
npm ERR!     /home/runner/.npm/_logs/2021-03-31T16_54_31_814Z-debug.log
npm ERR! code ELIFECYCLE
npm ERR! errno 1
npm ERR! zeth-contracts@1.0.0 install: `ln -s ../../depends/ganache-cli node_modules/ganache-cli && cd node_modules/ganache-cli && npm install`
npm ERR! Exit status 1
npm ERR! 
npm ERR! Failed at the zeth-contracts@1.0.0 install script.
npm ERR! This is probably not a problem with npm. There is likely additional logging output above.

npm ERR! A complete log of this run can be found in:
npm ERR!     /home/runner/.npm/_logs/2021-03-31T16_54_33_344Z-debug.log

So I'm curious to know if you had the same here.

Also, I saw you requested a review here, so I changed the status of the PR from Draft to Ready to review. Please feel free to revert if my review was asked by mistake. Finally, I can see some unchecked boxes in your PR description, can I let you check them all if this PR is ready to go?

[dtebbs]

@dtebbs - I see that you modified the submodule here. Was that necessary for the CI to pass?
I'm asking in the context of #361 for which the CI fails with:

So I'm curious to know if you had the same here.

Yes, that looks like the same issue. While fixing the libff.js reference to a purged commit I hit other version related issues (stale version numbers in some package-lock.json files etc.). I'll put just the submodule update in a new PR so we can merge it independently. That should fix your PR too.

Also, I saw you requested a review here, so I changed the status of the PR from Draft to Ready to review. Please feel free to revert if my review was asked by mistake. Finally, I can see some unchecked boxes in your PR description, can I let you check them all if this PR is ready to go?

Yes, no problem. Thanks.

[AntoineRondelet]

Yes, that looks like the same issue. While fixing the libff.js reference to a purged commit I hit other version related issues (stale version numbers in some package-lock.json files etc.). I'll put just the submodule update in a new PR so we can merge it independently. That should fix your PR too.

Great, thanks for confirming and moving the submodule update in another PR @dtebbs . Do you plan to rebase this one on top of #362?

[EDIT] Merged #362 since this PR is already rebased on top of it

[dtebbs]

Thanks for the comments @AntoineRondelet . Hopefully all addressed now. Note I added one small commit to end of this PR (to write out full assignments, required to generate test data for external provers), so you may want to review that in isolation. Thanks.

[AntoineRondelet]

LGTM overall - thanks @dtebbs
The CI failed though, so I just restarted the jobs - let's see if that passes.

[dtebbs]

LGTM overall - thanks @dtebbs
The CI failed though, so I just restarted the jobs - let's see if that passes.

Ah, looks like I broke something in the final commit. I'll check.

[dtebbs]

LGTM overall - thanks @dtebbs
The CI failed though, so I just restarted the jobs - let's see if that passes.

Ah, looks like I broke something in the final commit. I'll check.

@AntoineRondelet Should be fixed now. Sorry about that.

[AntoineRondelet]

Great, thanks @dtebbs - LGTM