Skip to content
This repository has been archived by the owner on Sep 24, 2023. It is now read-only.

Latest commit

 

History

History
132 lines (89 loc) · 6.19 KB

README.md

File metadata and controls

132 lines (89 loc) · 6.19 KB
Raw

This repository is archived. The current code lives here

homepagebanner_comp1

eth flower nft 🌼

TO-DOs

  • check all addresses and parameters in deploy.zsh
  • run make deploy-production-ic-full
  • check if all assets uploaded correctly by calling the canisters getTokens
  • top canister up with cycles
  • add canister to tip jar
  • run off chain backup script with mainnet canister id
  • run disburse script with mainnet canister id
  • call endAuction after auction ends
  • call shuffleAssets 24 hours after endAuction

structure

  • the main contains the logic that is currently deployed on mainnet
  • the development branch is the one all the rewrite branches are merged into
  • the local-development branch can be used for local deployment to avoid accidentially deploying the canister to mainnet

quick deploy 🏃‍♀️

  • To quickly reinstall (WIPES ALL STATE) the NFT staging canister locally run dfx deploy staging --mode reinstall

sophisticated deploy 📚

  • use make to run the standard local deploy, use make deploy-staging-ic to deploy the staging canister to the mainnet, by default it deploys the NFT staging canister locally and uses assets/output.mp4 and metadata.json as file paths
    • metadata.json MUST NOT contain a mint number! (use cat btcflower.json| sed '/mint/ d' > metadata.json to remove the mint number)
    • note that you need ext installed
  • The deploy.zsh adds another oracle to the NFT canister because the script in the source SVG won't be executed the way it's currently structured. Make sure you use the correct API endpoint there as well!
    • note: this script is not allowed to contain any & or > characters!
    • make sure you change the asset canister url and the currency fetched from the oracle
  • make sure you create and assets folder and provide the seed.mp4 file and the metadata.json file and specify their names in the script accordingly
  • the weird looking sed when uploading the metadata is escaping " characters and the variable $j is needed for the correct index (j=$i-1)

caveats 🕳

  • The canister code is written in a way that the seed animation ALWAYS has to be the first asset uploaded to the canister.
  • The seed animation video needs to be encoded in a way that it can be played on iOS devices, use HandBrake for that or ffmpeg

vessel 🚢

  • Run vessel verify --version 0.6.21 to verify everything still builds correctly after adding a new depdenceny
  • To use vesselss moc version when deploying, use DFX_MOC_PATH="$(vessel bin)/moc" dfx deploy

shuffle 🔀

  • The shuffle uses the random beacon to derive a random seed for the PRNG
  • It basically shuffles all the assets in the assets stable variable
  • The link inside the canister is
tokenIndex -> assetIndex
assetIndex -> NFT
  • initially the tokenIndex matches the assetIndex (assetIndex = tokenIndex+1) and the assetIndex matches the NFT (NFT = assetIndex+1)
  • but after the shuffle the assetIndex and the NFT mint number no longer match
  • so so token at tokenIndex still points to the same asset at assetIndex, but this asset no longer has the same NFT mint number
  • we can always retrieve the NFT mint number from the _asset[index].name property which we specify when adding an asset to the canister

off-chain backup ⛓

We use the getRegistry (tokenIndex -> AccountIdentifier) and getTokens (tokenIndex -> NFT) canister methods to backup state offchain. Therefore we simply use a script that queries the afore mentioned methods every 60 minutes and saves the responses on a server. You can find the script in state_backup. We are also submitting every transaction to CAP, which again offers off-chain backups of their data.

Note that the indices of the json outputs represent the indices of the internal storage. E.g. index 0 means it is the first item in the array. In the UI (entrepot or stoic wallet) those indices are incremented by one, so they start with 1 and not with 0.

To have the same token identifiers for the same tokens, it is important to keep the order of the minting when reinstantiating the canister.

So when executing mintNFT, the to address is taken from registry.json and the asset is taken from tokens.json. It's important here that the uploading of the assets is on order (start with flower 1, end with flower 2009) and that the assets index 0 is used by something other than an NFT asset (before it was the seed animation)! It's also crucial to remove shuffleAssets functionality from the canister!

testing 🧪

deploy the canister with

dfx deploy

use the following command to upload an asset that fits into a single message

dfx canister call btcflower addAsset '(record {name = "privat";payload = record {ctype = "text/html"; data = vec {blob "hello world!"} } })'

use the following command to mint a token

dfx canister call btcflower mintNFT '(record {to = "75c52c5ee26d10c7c3da77ec7bc2b4c75e1fdc2b92e01d3da6986ba67cfa1703"; asset = 0 : nat32})'

run icx-proxy to be able to user query parameters locally

$(dfx cache show)/icx-proxy --address 127.0.0.1:8453 -vv

NOTE

you can also use http://127.0.0.1:8000/?canisterId=rrkah-fqaaa-aaaaa-aaaaq-cai&asset=0 or http://127.0.0.1:8000/1.svg?canisterId=rrkah-fqaaa-aaaaa-aaaaq-cai&asset=0 locally

add the following line to /etc/hosts if on mac

127.0.0.1       rrkah-fqaaa-aaaaa-aaaaq-cai.localhost

the canister can now be accesed with

http://rwlgt-iiaaa-aaaaa-aaaaa-cai.localhost:8453/?tokenid=rwlgt-iiaaa-aaaaa-aaaaa-cai

or via command line with

curl "rwlgt-iiaaa-aaaaa-aaaaa-cai.localhost:8453/?tokenid=rwlgt-iiaaa-aaaaa-aaaaa-cai"

ext-cli 🔌

to get the tokenid from the canister and index do the following

  1. clone https://github.com/Toniq-Labs/ext-cli and https://github.com/Toniq-Labs/ext-js in the same directory
  2. run npm i -g from within ext-cli
  3. run ext token <canister_id> <index>