swh.storage.cassandra.model module

Classes representing tables in the Cassandra database.

They are very close to classes found in swh.model.model, but most of them are subtly different:

• Large objects are split into other classes (eg. RevisionRow has no ‘parents’ field, because parents are stored in a different table, represented by RevisionParentRow)

• They have a “cols” field, which returns the list of column names of the table

• They only use types that map directly to Cassandra’s schema (ie. no enums)

Therefore, this model doesn’t reuse swh.model.model, except for types that can be mapped to UDTs (Person and TimestampWithTimezone).

swh.storage.cassandra.model.MAGIC_NULL_PK = b'<null>'

NULLs (or all-empty blobs) are not allowed in primary keys; instead we use a special value that can’t possibly be a valid hash.

swh.storage.cassandra.model.content_index_table_name(algo: str, skipped_content: bool) → str

Given an algorithm name, returns the name of one of the ‘content_by_*’ and ‘skipped_content_by_*’ tables that serve as index for the ‘content’ and ‘skipped_content’ tables based on this algorithm’s hashes.

For now it is a simple substitution, but future versions may append a version number to it, if needed for schema updates.

class swh.storage.cassandra.model.BaseRow

Bases: object

TABLE: ClassVar[str]

PARTITION_KEY: ClassVar[Tuple[str, ...]]

CLUSTERING_KEY: ClassVar[Tuple[str, ...]] = ()

classmethod from_dict(d: Dict[str, Any]) → T

classmethod cols() → List[str]

to_dict() → Dict[str, Any]

class swh.storage.cassandra.model.ContentRow(sha1: bytes, sha1_git: bytes, sha256: bytes, blake2s256: bytes, length: int, ctime: datetime.datetime, status: str)

Bases: BaseRow

TABLE: ClassVar[str] = 'content'

PARTITION_KEY: ClassVar[Tuple[str, ...]] = ('sha256',)

CLUSTERING_KEY: ClassVar[Tuple[str, ...]] = ('sha1', 'sha1_git', 'blake2s256')

sha1: bytes

sha1_git: bytes

sha256: bytes

blake2s256: bytes

length: int

ctime: datetime

creation time, i.e. time of (first) injection into the storage

status: str

class swh.storage.cassandra.model.SkippedContentRow(sha1: bytes | None, sha1_git: bytes | None, sha256: bytes | None, blake2s256: bytes | None, length: int | None, ctime: datetime.datetime | None, status: str, reason: str, origin: str)

Bases: BaseRow

TABLE: ClassVar[str] = 'skipped_content'

PARTITION_KEY: ClassVar[Tuple[str, ...]] = ('sha1', 'sha1_git', 'sha256', 'blake2s256')

sha1: bytes | None

sha1_git: bytes | None

sha256: bytes | None

blake2s256: bytes | None

length: int | None

ctime: datetime | None

creation time, i.e. time of (first) injection into the storage

status: str

reason: str

origin: str

classmethod from_dict(d: Dict[str, Any]) → SkippedContentRow

class swh.storage.cassandra.model.DirectoryRow(id: bytes, raw_manifest: bytes | None)

Bases: BaseRow

TABLE: ClassVar[str] = 'directory'

PARTITION_KEY: ClassVar[Tuple[str, ...]] = ('id',)

id: bytes

raw_manifest: bytes | None

NULL if the object can be rebuild from (sorted) entries

class swh.storage.cassandra.model.DirectoryEntryRow(directory_id: bytes, name: bytes, target: bytes, perms: int, type: str)

Bases: BaseRow

TABLE: ClassVar[str] = 'directory_entry'

PARTITION_KEY: ClassVar[Tuple[str, ...]] = ('directory_id',)

CLUSTERING_KEY: ClassVar[Tuple[str, ...]] = ('name',)

directory_id: bytes

name: bytes

path name, relative to containing dir

target: bytes

perms: int

unix-like permissions

type: str

target type

class swh.storage.cassandra.model.RevisionRow(id: bytes, date: swh.model.model.TimestampWithTimezone | None, committer_date: swh.model.model.TimestampWithTimezone | None, type: str, directory: bytes, message: bytes, author: swh.model.model.Person, committer: swh.model.model.Person, synthetic: bool, metadata: str, extra_headers: dict, raw_manifest: bytes | None)

Bases: BaseRow

TABLE: ClassVar[str] = 'revision'

PARTITION_KEY: ClassVar[Tuple[str, ...]] = ('id',)

id: bytes

date: TimestampWithTimezone | None

committer_date: TimestampWithTimezone | None

type: str

directory: bytes

source code “root” directory

message: bytes

author: Person

committer: Person

synthetic: bool

true iff revision has been created by Software Heritage

metadata: str

extra metadata as JSON(tarball checksums, etc…)

extra_headers: dict

extra commit information as (tuple(key, value), …)

raw_manifest: bytes | None

NULL if the object can be rebuild from other cells and revision_parent.

class swh.storage.cassandra.model.RevisionParentRow(id: bytes, parent_rank: int, parent_id: bytes)

Bases: BaseRow

TABLE: ClassVar[str] = 'revision_parent'

PARTITION_KEY: ClassVar[Tuple[str, ...]] = ('id',)

CLUSTERING_KEY: ClassVar[Tuple[str, ...]] = ('parent_rank',)

id: bytes

parent_rank: int

parent position in merge commits, 0-based

parent_id: bytes

class swh.storage.cassandra.model.ReleaseRow(id: bytes, target_type: str, target: bytes, date: swh.model.model.TimestampWithTimezone, name: bytes, message: bytes, author: swh.model.model.Person, synthetic: bool, raw_manifest: bytes | None)

Bases: BaseRow

TABLE: ClassVar[str] = 'release'

PARTITION_KEY: ClassVar[Tuple[str, ...]] = ('id',)

id: bytes

target_type: str

target: bytes

date: TimestampWithTimezone

name: bytes

message: bytes

author: Person

synthetic: bool

true iff release has been created by Software Heritage

raw_manifest: bytes | None

NULL if the object can be rebuild from other cells

class swh.storage.cassandra.model.SnapshotRow(id: bytes)

Bases: BaseRow

TABLE: ClassVar[str] = 'snapshot'

PARTITION_KEY: ClassVar[Tuple[str, ...]] = ('id',)

id: bytes

class swh.storage.cassandra.model.SnapshotBranchRow(snapshot_id: bytes, name: bytes, target_type: str | None, target: bytes | None)

Bases: BaseRow

For a given snapshot_id, branches are sorted by their name, allowing easy pagination.

TABLE: ClassVar[str] = 'snapshot_branch'

PARTITION_KEY: ClassVar[Tuple[str, ...]] = ('snapshot_id',)

CLUSTERING_KEY: ClassVar[Tuple[str, ...]] = ('name',)

snapshot_id: bytes

name: bytes

target_type: str | None

target: bytes | None

class swh.storage.cassandra.model.OriginVisitRow(origin: str, visit: int, date: datetime.datetime, type: str)

Bases: BaseRow

TABLE: ClassVar[str] = 'origin_visit'

PARTITION_KEY: ClassVar[Tuple[str, ...]] = ('origin',)

CLUSTERING_KEY: ClassVar[Tuple[str, ...]] = ('visit',)

origin: str

visit: int

date: datetime

type: str

class swh.storage.cassandra.model.OriginVisitStatusRow(origin: str, visit: int, date: datetime.datetime, type: str, status: str, metadata: str, snapshot: bytes)

Bases: BaseRow

TABLE: ClassVar[str] = 'origin_visit_status'

PARTITION_KEY: ClassVar[Tuple[str, ...]] = ('origin',)

CLUSTERING_KEY: ClassVar[Tuple[str, ...]] = ('visit', 'date')

origin: str

visit: int

date: datetime

type: str

status: str

metadata: str

snapshot: bytes

classmethod from_dict(d: Dict[str, Any]) → T

class swh.storage.cassandra.model.OriginRow(sha1: bytes, url: str, next_visit_id: int)

Bases: BaseRow

TABLE: ClassVar[str] = 'origin'

PARTITION_KEY: ClassVar[Tuple[str, ...]] = ('sha1',)

sha1: bytes

url: str

next_visit_id: int

We need integer visit ids for compatibility with the pgsql storage, so we’re using lightweight transactions with this trick: https://stackoverflow.com/a/29391877/539465

class swh.storage.cassandra.model.MetadataAuthorityRow(url: str, type: str)

Bases: BaseRow

TABLE: ClassVar[str] = 'metadata_authority'

PARTITION_KEY: ClassVar[Tuple[str, ...]] = ('url',)

CLUSTERING_KEY: ClassVar[Tuple[str, ...]] = ('type',)

url: str

type: str

class swh.storage.cassandra.model.MetadataFetcherRow(name: str, version: str)

Bases: BaseRow

TABLE: ClassVar[str] = 'metadata_fetcher'

PARTITION_KEY: ClassVar[Tuple[str, ...]] = ('name',)

CLUSTERING_KEY: ClassVar[Tuple[str, ...]] = ('version',)

name: str

version: str

class swh.storage.cassandra.model.RawExtrinsicMetadataRow(id: bytes, type: str, target: str, authority_type: str, authority_url: str, discovery_date: datetime, fetcher_name: str, fetcher_version: str, format: str, metadata: bytes, origin: str | None, visit: int | None, snapshot: str | None, release: str | None, revision: str | None, path: bytes | None, directory: str | None)

Bases: BaseRow

An explanation is in order for the primary key:

Intuitively, the primary key should only be ‘id’, because two metadata entries are the same iff the id is the same; and ‘id’ is used for deduplication.

However, we also want to query by (target, authority_type, authority_url, discovery_date) The naive solution to this would be an extra table, to use as index; but it means 1. extra code to keep them in sync 2. overhead when writing 3. overhead + random reads (instead of linear) when reading.

Therefore, we use a single table for both, by adding the column we want to query with before the id. It solves both a) the query/order issues and b) the uniqueness issue because:

1. adding the id at the end of the primary key does not change the rows’ order: for two different rows, id1 != id2, so (target1, …, date1) < (target2, …, date2) <=> (target1, …, date1, id1) < (target2, …, date2, id2)

2. the id is a hash of all the columns, so: rows are the same <=> id1 == id2 <=> (target1, …, date1, id1) == (target2, …, date2, id2)

TABLE: ClassVar[str] = 'raw_extrinsic_metadata'

PARTITION_KEY: ClassVar[Tuple[str, ...]] = ('target',)

CLUSTERING_KEY: ClassVar[Tuple[str, ...]] = ('authority_type', 'authority_url', 'discovery_date', 'id')

id: bytes

type: str

target: str

authority_type: str

authority_url: str

discovery_date: datetime

fetcher_name: str

fetcher_version: str

format: str

metadata: bytes

origin: str | None

visit: int | None

snapshot: str | None

release: str | None

revision: str | None

path: bytes | None

directory: str | None

class swh.storage.cassandra.model.RawExtrinsicMetadataByIdRow(id: bytes, target: str, authority_type: str, authority_url: str)

Bases: BaseRow

TABLE: ClassVar[str] = 'raw_extrinsic_metadata_by_id'

PARTITION_KEY: ClassVar[Tuple[str, ...]] = ('id',)

CLUSTERING_KEY: ClassVar[Tuple[str, ...]] = ()

id: bytes

target: str

authority_type: str

authority_url: str

class swh.storage.cassandra.model.ObjectCountRow(partition_key: int, object_type: str, count: int)

Bases: BaseRow

TABLE: ClassVar[str] = 'object_count'

PARTITION_KEY: ClassVar[Tuple[str, ...]] = ('partition_key',)

CLUSTERING_KEY: ClassVar[Tuple[str, ...]] = ('object_type',)

partition_key: int

object_type: str

count: int

class swh.storage.cassandra.model.ExtIDRow(extid_type: str, extid: bytes, extid_version: int, target_type: str, target: bytes)

Bases: BaseRow

TABLE: ClassVar[str] = 'extid'

PARTITION_KEY: ClassVar[Tuple[str, ...]] = ('extid_type', 'extid')

CLUSTERING_KEY: ClassVar[Tuple[str, ...]] = ('extid_version', 'target_type', 'target')

extid_type: str

extid: bytes

extid_version: int

target_type: str

target: bytes

class swh.storage.cassandra.model.ExtIDByTargetRow(target_type: str, target: bytes, target_token: int)

Bases: BaseRow

TABLE: ClassVar[str] = 'extid_by_target'

PARTITION_KEY: ClassVar[Tuple[str, ...]] = ('target_type', 'target')

CLUSTERING_KEY: ClassVar[Tuple[str, ...]] = ('target_token',)

target_type: str

target: bytes

target_token: int

value of token(pk) on the “primary” table

class swh.storage.cassandra.model.ObjectReferenceRow(target_type: str, target: bytes, source_type: str, source: bytes)

Bases: BaseRow

TABLE: ClassVar[str] = 'object_references'

PARTITION_KEY: ClassVar[Tuple[str, ...]] = ('target_type', 'target')

CLUSTERING_KEY: ClassVar[Tuple[str, ...]] = ('source_type', 'source')

target_type: str

target: bytes

source_type: str

source: bytes