rawreissuance rpc #475

instagibbs · 2018-12-03T15:54:20Z

Follows a similar transaction creation flow as #474 . Tests demonstrate a multi-sig spend of a reissuance token to issue additional assets.

Also includes a number of fixes I discovered along the way.

stevenroose

Looks good!

stevenroose · 2018-12-03T16:37:37Z

qa/rpc-tests/feature_issuance.py

+        self.nodes[0].sendtoaddress(blind_addr, total_amount, "", "", True)
+        self.nodes[1].generate(1)
+
+        # Next we used convinience function for transaction flows that will work


convenience*?

removed in other pr

stevenroose · 2018-12-03T16:38:44Z

qa/rpc-tests/feature_issuance.py

+        # Next we used convinience function for transaction flows that will work
+
+        # Start with single issuance input, unblinded (makes 5 outputs for later larger issuances)
+        # NOTE: the "blind" argument must should for all entries for `blindrawtransaction` step


This sentence doesn't seem to make sense to me.

removed in other PR

stevenroose · 2018-12-03T16:47:24Z

qa/rpc-tests/feature_issuance.py

+        assert_equal(len(id_set), 1)
+
+        # Random contract string should again differ
+        issued_tx = self.nodes[2].rawissueasset(funded_tx, [{"asset_amount":1, "asset_address":nonblind_addr, "contract":"deadbeef"}])[0]["hex"]


Reminder: contract value 32-bytes.

fixed in other PR, will rebased post-merge

stevenroose · 2018-12-03T16:56:24Z

qa/rpc-tests/feature_issuance.py

+        self.nodes[0].generate(1)
+        assert_equal(self.nodes[0].gettransaction(tx_id)["confirmations"], 1)
+
+        # TODO test raw issuance for non-empty contract hash


I think you did that.. Or do you mean raw REissuance with non-empty contract hash?

instagibbs · 2018-12-03T18:06:24Z

I'll rebase this PR post-merge of parent PR.

instagibbs · 2018-12-03T18:07:17Z

src/rpc/rawtransaction.cpp

+        }
+
+        uint256 asset_blinder;
+        asset_blinder.SetHex(asset_blinder_str);


Can use ParseHashV for error handling and parsing of bitcoin hash-like objects.

instagibbs · 2018-12-03T18:07:24Z

src/rpc/rawtransaction.cpp

+        }
+
+        uint256 entropy;
+        entropy.SetHex(entropy_str);


Can use ParseHashV for error handling and parsing of bitcoin hash-like objects.

instagibbs · 2018-12-03T18:09:02Z

src/rpc/rawtransaction.cpp

+            if (!IsHex(contract_hash_str) || contract_hash_str.size() != 64) {
+                throw JSONRPCError(RPC_INVALID_PARAMETER, "Contract hash must be 64 character hex strings.");
+            }
+            contract_hash = uint256(ParseHex(contract_hash_str));


We should be using ParseHashV to make sure we're parsing it like a hash like everywhere else in Bitcoin Core and to do argument checking.

instagibbs · 2018-12-04T17:45:46Z

src/rpc/rawtransaction.cpp

+    mtx.vin[issuance_input_index].assetIssuance.nAmount = asset_amount;
+
+    // Place assets into randomly placed output slots, before change output, inserted in place
+    assert(mtx.vout.size() >= 1);


need to validate user input in callee, or just return a useful error

instagibbs · 2018-12-12T17:01:38Z

rebased onto tip post #474 merge

instagibbs · 2018-12-12T17:06:56Z

READY FOR REVIEW

stevenroose

Doc nit. API is a bit awkward with the amount of data that needs to be kept from the first issuance to the second, but I guess that's just how it is. I don't think it makes sense to try and improve the UX by bundling all that into a single blob. This is for advanced users only, I guess.

tACK a1d6a91

stevenroose · 2018-12-12T19:04:47Z

src/rpc/rawtransaction.cpp

+            "rawreissueasset transaction {\"vin\":\"n\", \"asset_amount\":x.xxx, \"asset_address\":\"address\", \"asset_blinder\":<hex>, \"entropy\":<hex>, ( \"contract_hash\":<hex> )}\n"
+            "\nRe-issue an asset by attaching pseudo-inputs to transaction inputs, revealing the underlying reissuance token of the input. Returns the transaction hex.\n"
+            "\nArguments:\n"
+            "1. \"transaction\"           (string, required) Transaction in hex in which to include a peg-in input.\n"


You mean issuance here instead of pegin?

instagibbs · 2018-12-12T20:19:33Z

It's pretty much impossible to carry around a blob, since as soon as you spend the initial output, you'll have to update it. Would be nice to have something more straight forward, but indeed this is for power users.

…d line

… on blinding keys

instagibbs · 2018-12-12T21:25:03Z

While adding the final TODO test, I ran into:

A couple blinding bugs in the wallet where it was looking at the wrong field, or didn't push the right values when amounts were 0 in certain initial issuance cases. I had forgotten how some of this issuance consensus logic worked and wrote it wrong, and current tests weren't quite catching it.
contract_hash for reissuance was unused. Heh. entropy field is enough to re-calculate everything in a reissuance. So the final test has a non-null contract test, though it ended up mostly testing (1) instead.

stevenroose · 2018-12-17T20:24:42Z

qa/rpc-tests/feature_issuance.py

+        issued_tx = self.nodes[2].rawissueasset(issued_tx, [{"asset_amount":1, "asset_address":nonblind_addr, "token_address":nonblind_addr, "contract":"deadbeee"*8}])[0]["hex"]
        decode_tx = self.nodes[0].decoderawtransaction(issued_tx)
        id_set.add(decode_tx["vin"][1]["issuance"]["asset"])
+        non_null_contract_token = decode_tx["vin"][1]["issuance"]["token"]


I got confused you weren't using this variable and you got through the python linter. But forgot this is 0.14. In 0.17 the linter doesn't allow you to not use variables.

stevenroose · 2018-12-17T20:36:20Z

retACK 181d099

That was a subtle change, good catch!

-    // Don't issue stuff or set values unless non-zero (both are against consensus)
+    // Explicit 0 is represented by a null value, don't set to non-null in that case
+    if (blind_issuance || asset_amount != 0) {
+        mtx.vin[issuance_input_index].assetIssuance.nAmount = asset_amount;
+    }
+    // Don't make zero value output(impossible by consensus)
     if (asset_amount > 0) {
         mtx.vout.insert(mtx.vout.begin()+asset_place, asset_out);
-        mtx.vin[issuance_input_index].assetIssuance.nAmount = asset_amount;
     }

instagibbs · 2018-12-17T20:53:39Z

removed the unused variable in the test

bc257db f'rawreissueasset functional tests' (Gregory Sanders) 181d099 rawreissueasset functional tests (Gregory Sanders) 21cb8aa rawreissuance rpc (Gregory Sanders) 1251ecd fixup rawissueasset help comment (Gregory Sanders) 23371af add confidential_key explanation in validateaddress help (Gregory Sanders) c579f4b Correct how wallet blinding logic detects which token id to use based on blinding keys (Gregory Sanders) fff2dcc blindrawtransaction: Have blind_issuances parse correctly from command line (Gregory Sanders) 756bdc0 issueasset_base: Push 0-value value to issuance field if blinding (Gregory Sanders)

stevenroose reviewed Dec 3, 2018

View reviewed changes

instagibbs commented Dec 3, 2018

View reviewed changes

instagibbs commented Dec 4, 2018

View reviewed changes

instagibbs force-pushed the rawreissuance_rpc branch from b4ece6e to a1d6a91 Compare December 12, 2018 17:01

stevenroose reviewed Dec 12, 2018

View reviewed changes

instagibbs added 4 commits December 12, 2018 16:22

issueasset_base: Push 0-value value to issuance field if blinding

756bdc0

blindrawtransaction: Have blind_issuances parse correctly from comman…

fff2dcc

…d line

Correct how wallet blinding logic detects which token id to use based…

c579f4b

… on blinding keys

add confidential_key explanation in validateaddress help

23371af

instagibbs force-pushed the rawreissuance_rpc branch from a1d6a91 to bf19b1a Compare December 12, 2018 21:22

instagibbs added 3 commits December 12, 2018 16:27

fixup rawissueasset help comment

1251ecd

rawreissuance rpc

21cb8aa

rawreissueasset functional tests

181d099

instagibbs force-pushed the rawreissuance_rpc branch from bf19b1a to 181d099 Compare December 12, 2018 21:28

stevenroose reviewed Dec 17, 2018

View reviewed changes

f'rawreissueasset functional tests'

bc257db

instagibbs merged commit bc257db into ElementsProject:elements-0.14.1 Dec 17, 2018

instagibbs mentioned this pull request Dec 18, 2018

raw issuance RPC Blockstream/liquid#12

Merged

instagibbs mentioned this pull request Jan 2, 2019

Need more raw createrawpegin #476

Closed

rawreissuance rpc #475

rawreissuance rpc #475

Uh oh!

Conversation

instagibbs commented Dec 3, 2018 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

stevenroose left a comment

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

instagibbs commented Dec 3, 2018

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

instagibbs commented Dec 12, 2018

Uh oh!

instagibbs commented Dec 12, 2018

Uh oh!

stevenroose left a comment • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

instagibbs commented Dec 12, 2018

Uh oh!

instagibbs commented Dec 12, 2018 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

stevenroose commented Dec 17, 2018

Uh oh!

instagibbs commented Dec 17, 2018

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

instagibbs commented Dec 3, 2018 •

edited

Loading

stevenroose left a comment •

edited

Loading

instagibbs commented Dec 12, 2018 •

edited

Loading