Skip to content

Commit

Permalink
Truncate token uri in logging (#99)
Browse files Browse the repository at this point in the history
Some URLs are very long (e.g. data urls), so we truncate them to make them easier to read in logs and use less space.

This PR also fixes the foundation parser, which was broken, because the API it relied on doesn't exist anymore. It's now falling back to contract calls.
  • Loading branch information
ligustah authored Mar 20, 2024
1 parent d497fa3 commit d25fb44
Show file tree
Hide file tree
Showing 7 changed files with 87 additions and 37 deletions.
5 changes: 5 additions & 0 deletions docs/changelog.md
Original file line number Diff line number Diff line change
@@ -1,5 +1,10 @@
# Changelog

## v0.3.1

- Trim token_uri in some log outputs, this is mainly useful for data uris that are too long and make logs unreadable
- Fix `FoundationParser`, the API it relied on doesn't exist anymore, so we are falling back to contract calls to get the metadata

## v0.3.0

- Upgrade web3 to 6.11.3
Expand Down
2 changes: 1 addition & 1 deletion docs/index.md
Original file line number Diff line number Diff line change
@@ -1,6 +1,6 @@
# Getting Started

Documentation for version: **v0.3.0**
Documentation for version: **v0.3.1**

## Overview

Expand Down
75 changes: 60 additions & 15 deletions offchain/metadata/parsers/collection/foundation.py
Original file line number Diff line number Diff line change
Expand Up @@ -12,24 +12,69 @@
class FoundationParser(CollectionParser):
_COLLECTION_ADDRESSES: list[str] = [CollectionAddress.FOUNDATION]

def parse_metadata(self, token: Token, raw_data: Optional[dict], *args, **kwargs) -> Optional[Metadata]: # type: ignore[no-untyped-def, type-arg] # noqa: E501
if token.uri is None or raw_data is None:
token.uri = f"https://api.foundation.app/opensea/{token.token_id}"
raw_data = self.fetcher.fetch_content(token.uri) # type: ignore[assignment]
metadata = DefaultCatchallParser(self.fetcher).parse_metadata(token=token, raw_data=raw_data) # type: ignore[arg-type] # noqa: E501
def _normalize_metadata(self, metadata: Optional[Metadata]) -> Optional[Metadata]:
if metadata is None:
return None

metadata.standard = None # type: ignore[union-attr]
if metadata.content.uri.endswith("glb"): # type: ignore[union-attr]
metadata.content.mime_type = "model/gltf-binary" # type: ignore[union-attr]
if (
metadata
and metadata.image
and metadata.image.uri
and metadata.image.uri.endswith("glb")
):
metadata.image.mime_type = "model/gltf-binary"

return metadata

async def _gen_parse_metadata_impl(self, token: Token, raw_data: Optional[dict], *args, **kwargs) -> Optional[Metadata]: # type: ignore[no-untyped-def, type-arg] # noqa: E501
def parse_metadata(
self, token: Token, raw_data: Optional[dict], *args, **kwargs
) -> Optional[Metadata]:
if token.uri is None or raw_data is None:
token.uri = f"https://api.foundation.app/opensea/{token.token_id}"
raw_data = await self.fetcher.gen_fetch_content(token.uri) # type: ignore[assignment]
metadata = await DefaultCatchallParser(self.fetcher).gen_parse_metadata(token=token, raw_data=raw_data) # type: ignore[arg-type] # noqa: E501
metadata.standard = None # type: ignore[union-attr]
if metadata.content.uri.endswith("glb"): # type: ignore[union-attr]
metadata.content.mime_type = "model/gltf-binary" # type: ignore[union-attr]
token.uri = self.contract_caller.single_address_single_fn_many_args(
token.collection_address,
"tokenURI(uint256)",
["string"],
[[token.token_id]],
)[0]
if token.uri is None:
return None

return metadata
content = self.fetcher.fetch_content(token.uri)
if content and isinstance(content, dict):
raw_data = content

if raw_data is None:
return None

metadata = DefaultCatchallParser(self.fetcher).parse_metadata(
token=token, raw_data=raw_data
)

return self._normalize_metadata(metadata)

async def _gen_parse_metadata_impl(
self, token: Token, raw_data: Optional[dict], *args, **kwargs
) -> Optional[Metadata]:
if token.uri is None or raw_data is None:
token.uri = await self.contract_caller.rpc.async_reader.call_function(
token.collection_address,
"tokenURI(uint256)",
["string"],
[token.token_id],
)
if token.uri is None:
return None

content = await self.fetcher.gen_fetch_content(token.uri)
if content and isinstance(content, dict):
raw_data = content

if raw_data is None:
return None

metadata = await DefaultCatchallParser(self.fetcher).gen_parse_metadata(
token=token, raw_data=raw_data
)

return self._normalize_metadata(metadata)
19 changes: 14 additions & 5 deletions offchain/metadata/pipelines/metadata_pipeline.py
Original file line number Diff line number Diff line change
Expand Up @@ -25,6 +25,15 @@
)


def _truncate_uri(uri: str, max_length: int = 100) -> str:
if len(uri) <= max_length:
return uri

keep_length = (max_length - 3) // 2 # 3 is for the '...'

return uri[:keep_length] + "..." + uri[-keep_length:]


class MetadataPipeline(BasePipeline):
"""Pipeline for processing NFT metadata.
Expand Down Expand Up @@ -161,7 +170,7 @@ def fetch_token_metadata(
try:
raw_data = self.fetcher.fetch_content(token.uri)
except Exception as e:
error_message = f"({token.chain_identifier}-{token.collection_address}-{token.token_id}) Failed to parse token uri: {token.uri}. {str(e)}" # noqa: E501
error_message = f"({token.chain_identifier}-{token.collection_address}-{token.token_id}) Failed to parse token uri: {_truncate_uri(token.uri)}. {str(e)}" # noqa: E501
logger.error(error_message)
possible_metadatas_or_errors.append(
MetadataProcessingError.from_token_and_error(
Expand Down Expand Up @@ -216,9 +225,9 @@ async def gen_fetch_token_metadata(
Union[Metadata, MetadataProcessingError]: returns either a Metadata
or a MetadataProcessingError if unable to parse.
"""
possible_metadatas_or_errors: list[
Union[Metadata, MetadataProcessingError]
] = []
possible_metadatas_or_errors: list[Union[Metadata, MetadataProcessingError]] = (
[]
)

if not token.uri:
return MetadataProcessingError.from_token_and_error(
Expand All @@ -230,7 +239,7 @@ async def gen_fetch_token_metadata(
try:
raw_data = await self.fetcher.gen_fetch_content(token.uri)
except Exception as e:
error_message = f"({token.chain_identifier}-{token.collection_address}-{token.token_id}) Failed to parse token uri: {token.uri}. {str(e)}" # noqa: E501
error_message = f"({token.chain_identifier}-{token.collection_address}-{token.token_id}) Failed to parse token uri: {_truncate_uri(token.uri)}. {str(e)}" # noqa: E501
logger.error(error_message)
possible_metadatas_or_errors.append(
MetadataProcessingError.from_token_and_error(
Expand Down
2 changes: 1 addition & 1 deletion pyproject.toml
Original file line number Diff line number Diff line change
@@ -1,6 +1,6 @@
[tool.poetry]
name = "offchain"
version = "0.3.0"
version = "0.3.1"
description = "Open source metadata processing framework"
authors = ["Zora eng <[email protected]>"]
readme = "README.md"
Expand Down
2 changes: 1 addition & 1 deletion tests/metadata/fetchers/test_metadata_fetcher.py
Original file line number Diff line number Diff line change
Expand Up @@ -54,7 +54,7 @@ async def test_gen_fetch_base_adapter(self): # type: ignore[no-untyped-def]
async def test_gen_fetch_mime_type_and_size(self): # type: ignore[no-untyped-def]
fetcher = MetadataFetcher()
result = await fetcher.gen_fetch_mime_type_and_size(
"https://ipfs.io/ipfs/QmQaYaf3Q2oCBaUfUvV6mBP58EjbUTbMk6dC1o4YGjeWCo"
"https://ipfs.decentralized-content.com/ipfs/QmQaYaf3Q2oCBaUfUvV6mBP58EjbUTbMk6dC1o4YGjeWCo"
)
assert result == ("image/png", "2887641") # type: ignore[comparison-overlap]
print(result)
Expand Down
19 changes: 5 additions & 14 deletions tests/metadata/parsers/test_foundation_parser.py
Original file line number Diff line number Diff line change
Expand Up @@ -22,9 +22,7 @@ class TestFoundationParser:
raw_data = {
"name": "Experiment #0004",
"description": "They rise again!",
"image": "https://d1hiserqh6k9o1.cloudfront.net/Ax/kk/QmWwB2LXk7VKu5KtDrtUYdwpHK1NvJ49XrQvFRJxqiAxkk/nft.png",
"animation_url": "ipfs://QmWwB2LXk7VKu5KtDrtUYdwpHK1NvJ49XrQvFRJxqiAxkk/nft.glb",
"external_url": "https://foundation.app/@pw_3Dlab/foundation/113384",
"image": "ipfs://QmWwB2LXk7VKu5KtDrtUYdwpHK1NvJ49XrQvFRJxqiAxkk/nft.glb",
}

def test_foundation_parser_should_parse_token(self): # type: ignore[no-untyped-def]
Expand All @@ -37,40 +35,33 @@ def test_foundation_parser_parses_metadata(self): # type: ignore[no-untyped-def
fetcher = MetadataFetcher()
contract_caller = ContractCaller()
fetcher.fetch_mime_type_and_size = MagicMock(return_value=("application/json", 0)) # type: ignore[assignment]
fetcher.fetch_content = MagicMock(return_value=self.raw_data) # type: ignore[assignment]
fetcher.fetch_content = MagicMock(return_value=None) # type: ignore[assignment]
parser = FoundationParser(fetcher=fetcher, contract_caller=contract_caller) # type: ignore[abstract]
metadata = parser.parse_metadata(token=self.token, raw_data=self.raw_data)
assert metadata == Metadata(
token=Token(
chain_identifier="ETHEREUM-MAINNET",
collection_address="0x3b3ee1931dc30c1957379fac9aba94d1c48a5405",
token_id=113384,
uri="https://api.foundation.app/opensea/113384",
uri="ipfs://QmRxAiR7FsT78mLcyMiE1p86a77gBQVJGWfGRADtMwqyEe/metadata.json",
),
raw_data={
"name": "Experiment #0004",
"description": "They rise again!",
"image": "https://d1hiserqh6k9o1.cloudfront.net/Ax/kk/QmWwB2LXk7VKu5KtDrtUYdwpHK1NvJ49XrQvFRJxqiAxkk/nft.png",
"animation_url": "ipfs://QmWwB2LXk7VKu5KtDrtUYdwpHK1NvJ49XrQvFRJxqiAxkk/nft.glb",
"external_url": "https://foundation.app/@pw_3Dlab/foundation/113384",
"image": "ipfs://QmWwB2LXk7VKu5KtDrtUYdwpHK1NvJ49XrQvFRJxqiAxkk/nft.glb",
},
attributes=[],
standard=None,
name="Experiment #0004",
description="They rise again!",
mime_type="application/json",
image=MediaDetails(
size=0,
sha256=None,
uri="https://d1hiserqh6k9o1.cloudfront.net/Ax/kk/QmWwB2LXk7VKu5KtDrtUYdwpHK1NvJ49XrQvFRJxqiAxkk/nft.png",
mime_type="application/json",
),
content=MediaDetails(
size=0,
sha256=None,
uri="ipfs://QmWwB2LXk7VKu5KtDrtUYdwpHK1NvJ49XrQvFRJxqiAxkk/nft.glb",
mime_type="model/gltf-binary",
),
content=None,
additional_fields=[],
)

Expand Down

0 comments on commit d25fb44

Please sign in to comment.