NFT-event file append pulls back events that are not associated with the asset-identifier

[timstackblock]

Describe the bug NFT-event file append pulls back events that are not associated with the asset-identifier asset_identifier": "ST1PQHQKV0RJXZFY1DGX8MNSNYVE3VGZJSRTPGZGM.redeeem-nft-v0",

I ran this predicate and only wanted to return mint events for "ST1PQHQKV0RJXZFY1DGX8MNSNYVE3VGZJSRTPGZGM.redeeem-nft-v0"

Here is the predicate file

I successfully got records for that asset identifier which is great but the file is very large and there are a lot of records for other NFTs that do not mat ch the asset identifier "ST1PQHQKV0RJXZFY1DGX8MNSNYVE3VGZJSRTPGZGM.redeeem-nft-v0",

Here are screen shots this is the GOOD data

Here is the bad data the event type also seems to not filter for "MINT:" events and is pulling in data that is outside of the scope. You will see transfer data and you will see asset identifiers for "asset_class_identifier":"ST39HFKW38EPPPRQ1R52GK02WCN8314DQAP1P6AYE.test-wn::wolfs-pack-academy" the filtering seems to be broken.

Please let me know if you need more information

[lgalabru]

Thanks for the time spent on this @timstackblock, I think it's time for us to take on https://github.com/hirosystems/chainhook/issues/287. @timstackblock in future bug reports, could you please provide the predicate in json format, instead of screenshots? Also when reporting the results, providing a copy paste (full block or full transaction) would speed up debugging! Thank you again for your halp 🙏

[timstackblock]

Thanks for the time spent on this @timstackblock, I think it's time for us to take on #287. @timstackblock in future bug reports, could you please provide the predicate in json format, instead of screenshots? Also when reporting the results, providing a copy paste (full block or full transaction) would speed up debugging! Thank you again for your halp 🙏

sure no problem @lgalabru will do

[MicaiahReid]

@timstackblock This issue should be fixed currently in the develop branch. Can you confirm?

[timstackblock]

sure let me take a look

[timstackblock]

Please scroll to the bottom of the explorer page in the link below to see the MINT NFT events chainhooks should be pulling back

https://explorer.hiro.so/txid/ST1PQHQKV0RJXZFY1DGX8MNSNYVE3VGZJSRTPGZGM.redeeem-nft-v0?chain=testnet

Here is the error I am getting when I run the predicate

Chainhooks seems like it cannot pull back the mint events for the NFT it is getting 0 occurrences but the NFT events are present. There is also a strange new error Aug 03 16:19:25.089 ERRO unable to parse stacks block_header EOF while parsing a string at line 1 column 3616 Please let me know if there is an issue with the NFT I have selected

I have pasted the predicate below so you can run it

{ "chain": "stacks", "uuid": "c3c3bf08-2035-40bf-a5eb-0e4295911f8a", "name": "nft event file", "version": 1, "networks": { "testnet": { "start_block": 20835, "end_block": 38357, "if_this": { "scope": "nft_event", "asset_identifier": "ST1PQHQKV0RJXZFY1DGX8MNSNYVE3VGZJSRTPGZGM.redeeem-nft-v0", "actions": ["mint"] }, "then_that": { "file_append": { "path": "automate/tests/stacks-predicates/nft-event/nft-event-file-result.json" } } } } }

[timstackblock]

Here is the transaction

The first MINT NFT_EVENT is in block 20835 that is why we start the block there and the last one is in 38357 there are many in between

https://explorer.hiro.so/txid/0xf3fd7d7278a104ddafaf4517d371bdbf3d75b8ecdbb4cc4ac51488f4240f76de?chain=testnet

[MicaiahReid]

Hey @timstackblock, I've found what's going on here. The asset id you are using is deployer_address.contract_name, but it should be deployer_address.contract_name::nft_name. For example, rather than "asset_identifier": "ST1PQHQKV0RJXZFY1DGX8MNSNYVE3VGZJSRTPGZGM.redeeem-nft-v0", it should be "asset_identifier": "ST1PQHQKV0RJXZFY1DGX8MNSNYVE3VGZJSRTPGZGM.redeeem-nft-v0::redeeem-nft-v0"

[MicaiahReid]

This predicate worked for me:

{
  "chain": "stacks",
  "uuid": "c3c3bf08-2035-40bf-a5eb-0e4295911f8a",
  "name": "nft event file",
  "version": 1,
  "networks": {
    "testnet": {
      "start_block": 20835,
      "end_block": 20835,
      "if_this": {
        "scope": "nft_event",
        "asset_identifier": "ST1PQHQKV0RJXZFY1DGX8MNSNYVE3VGZJSRTPGZGM.redeeem-nft-v0::redeeem-nft-v0",
        "actions": ["mint"]
      },
      "then_that": {
        "file_append": {
          "path": "automate/tests/stacks-predicates/nft-event/nft-event-file-result.json"
        }
      }
    }
  }
}

[timstackblock]

Thanks for looking into this. One last question using this updated asset identifier I got 22 occurrences but there are 55 mint function calls in the explorer between 20835 and 38357. Is there additional filtering criteria in there somewhere? It looks like its skipping multiple function calls in the same block, when I look at the output file I see all unique blocks listed but in explorer I can see there are multiple MINT events in the same block and they are not being pulled back.

{
  "chain": "stacks",
  "uuid": "c3c3bf08-2035-40bf-a5eb-0e4295911f8a",
  "name": "nft event file",
  "version": 1,
  "networks": {
    "testnet": {
      "start_block": 20835,
      "end_block": 38357,
      "if_this": {
        "scope": "nft_event",
        "asset_identifier": "ST1PQHQKV0RJXZFY1DGX8MNSNYVE3VGZJSRTPGZGM.redeeem-nft-v0::redeeem-nft-v0",
        "actions": ["mint"]
      },
      "then_that": {
        "file_append": {
          "path": "automate/tests/stacks-predicates/nft-event/nft-event-file-result.json"
        }
      }
    }
  }
}

If you look at block 20836 only 1 record is pulled back for that block but if you look into the explorer multiple mints occurred in that block, below is a screenshot in explorer of all the mints in the block we are not pulling back.

[MicaiahReid]

If you look at block 20836 only 1 record is pulled back for that block but if you look into the explorer multiple mints occurred in that block

What are you using to gauge how many records are "pulled back"? As I mentioned in this comment, when an "occurrence" is listed in the Chainhook output, this means that a block has at least on transaction that matches a predicate's criteria. Because of that one match, the whole block will be sent to the output file.

So, for block 20836, since there is at least one event that match the specified predicate, the whole block is output to the file. Inside of that block is all of the transactions in the block with all of that transaction's events. So, that one block counts as one "occurrence", but the data within that occurrence has multiple matching events.

Does that make sense?

Maybe what is confusing is that we output X occurrences found and instead it should be X blocks found that contain transactions or events matching predicate conditions, but that is a little wordy, haha

[timstackblock]

Got it makes sense I was expecting to count events but the block for the event is being pulled back

[MicaiahReid]

Awesome! Closing this one as being fixed by #364!