source_patch_id
source_patch_id(
format_name: str ,
version: str ,
path: str ,
key: <class ‘object = None,
size_bytes: int | None[int, None] = None,
mtime_ns: int | None[int, None] = None,
)-> ’str’
Return the id of data read from a file.
The deterministic ID lets repeated reads of the same source agree and preserves provenance across processes.
Parameters
| Parameter | Description |
|---|---|
| format_name | The format the reader was using. |
| version | The version of that format the file was read at. |
| path | The path or URI the data came from, as the archive spells it. |
| key |
What names this patch within the file, when a file holds more than one. The reader’s own key if it has one, else the ordinal. |
| size_bytes | The size of the source, when it can be had. |
| mtime_ns | When the source was last written, when it can be had. |
Note
Every field already exists in the index, so no filesystem access is required. Size and modification time distinguish replaced sources; unavailable values may be None. Changing modification time changes the ID even if contents are unchanged. Path-derived IDs depend on archive layout and are not stable across hosts. IDs stored by a format remain authoritative and portable.