From f2afdf857a58aa00fa7572d6a348b85d2f1cabdb Mon Sep 17 00:00:00 2001
From: Maximilian Moser <maximilian.moser@tuwien.ac.at>
Date: Fri, 23 Feb 2024 15:56:40 +0100
Subject: [PATCH] Store list of results for results per format
* building up this list while loading results reduces the amount of
potential work for filtering later on
* also enable passing a list of already known formats to
`load_results(...)`
* new formats will be added to the supplied list as they're encountered
---
formatscaper/core/models.py | 2 ++
formatscaper/core/utils.py | 23 ++++++++++++++++++-----
2 files changed, 20 insertions(+), 5 deletions(-)
diff --git a/formatscaper/core/models.py b/formatscaper/core/models.py
index 28a54f4..15749ff 100644
--- a/formatscaper/core/models.py
+++ b/formatscaper/core/models.py
@@ -1,4 +1,5 @@
import dataclasses
+from typing import List
@dataclasses.dataclass
@@ -18,6 +19,7 @@ class Format:
name: str
mime: str
endangered: bool = False
+ results: List["Result"] = dataclasses.field(default_factory=list)
def as_dict(self):
"""Dump the data as dictionary."""
diff --git a/formatscaper/core/utils.py b/formatscaper/core/utils.py
index 4742385..e8d60f3 100644
--- a/formatscaper/core/utils.py
+++ b/formatscaper/core/utils.py
@@ -88,13 +88,19 @@ def store_results(results: List[Result], file_name: str, file_format: str) -> bo
def load_results(
- file_name: str, file_format: Optional[str] = None, strict: bool = True
+ file_name: str,
+ file_format: Optional[str] = None,
+ strict: bool = True,
+ formats: Optional[List[Format]] = None,
) -> Optional[List[Result]]:
"""Load the results from the given file.
In case the ``file_format`` isn't specified, auto-detection is attempted.
If ``strict`` is set, then the result loading will fail if it cannot parse
the format for a result.
+ Optionally, a list of already known ``formats`` can be supplied to avoid
+ creating duplicate ``Format`` instances.
+ Newly encountered formats will be appended to the supplied list.
"""
if file_format is None:
if re.search(r"\.ya?ml$", file_name, re.IGNORECASE):
@@ -116,12 +122,17 @@ def load_results(
# note: we deduplicate formats so that manipulation of one entry updates all entries
results = []
- known_formats = {}
+ formats = formats or []
+ known_formats = {format.puid: format for format in formats}
+
for res in raw_results:
format = None
try:
- f = Format(**res["format"])
- format = known_formats.setdefault(res["format"]["puid"], f)
+ format = known_formats.get(res["format"]["puid"], None)
+ if format is None:
+ format = Format(**res["format"])
+ known_formats[format.puid] = format
+ formats.append(format)
except (TypeError, KeyError) as e:
# TypeError: the result doesn't have all required parts for Format()
# KeyError: either the result doesn't have a format or it lacks the PUID
@@ -129,6 +140,8 @@ def load_results(
raise e
res.pop("format", None)
- results.append(Result(**res, format=format))
+ result = Result(**res, format=format)
+ format.results.append(result)
+ results.append(result)
return results
--
GitLab