From 963d2f538d66bab0ef147f52c03b0697e151fcbf Mon Sep 17 00:00:00 2001
From: Florian Spreckelsen <f.spreckelsen@indiscale.com>
Date: Mon, 25 Mar 2024 12:50:41 +0100
Subject: [PATCH] WIP: Move to separate converter
---
src/caoscrawler/converters.py | 26 ++++++++++++++++++++++++--
src/caoscrawler/default_converters.yml | 3 +++
2 files changed, 27 insertions(+), 2 deletions(-)
diff --git a/src/caoscrawler/converters.py b/src/caoscrawler/converters.py
index bc30c7d4..7d994377 100644
--- a/src/caoscrawler/converters.py
+++ b/src/caoscrawler/converters.py
@@ -810,7 +810,29 @@ class DictElementConverter(Converter):
raise RuntimeError("Element must be a DictElement.")
return match_name_and_value(self.definition, element.name, element.value)
- def create_records(self, values: GeneralStore, records: RecordStore, element: StructureElement):
+
+class HeuristicDictConverter(DictElementConverter):
+ """Extend the :py:class:`DictElementConverter` by a heuristic to set
+ property values from the dictionary keys.
+
+ """
+ def _validate_definition(definition: dict, name: str):
+
+ if "record_from_dict" not in definition or definition["record_from_dict"] is None:
+
+ raise ValueError(
+ "You need to specify the (root) record, the properties of "
+ f"which will be set from the dict in converter {name}."
+ )
+
+ def __init__(self, definition: dict, name: str, converter_registry: dict):
+
+ _validate_definition(definition)
+ super().__init__(definition, name, converter_registry)
+
+ def create_records(self, values: GeneralStore, records: RecordStore,
+ element: StructureElement, referenced_record_callback:
+ Optional[callable] = None):
keys_modified = []
if "record_from_dict" in self.definition and self.definition["record_from_dict"] is not None:
@@ -826,7 +848,7 @@ class DictElementConverter(Converter):
class DictConverter(DictElementConverter):
def __init__(self, *args, **kwargs):
warnings.warn(DeprecationWarning(
- "This class is depricated. Please use DictConverter."))
+ "This class is depricated. Please use DictElementConverter."))
super().__init__(*args, **kwargs)
diff --git a/src/caoscrawler/default_converters.yml b/src/caoscrawler/default_converters.yml
index e192ab1b..98c7345d 100644
--- a/src/caoscrawler/default_converters.yml
+++ b/src/caoscrawler/default_converters.yml
@@ -11,6 +11,9 @@ Date:
Dict:
converter: DictElementConverter
package: caoscrawler.converters
+DictHeuristic:
+ converter: HeuristicDictConverter
+ package: caoscrawler.converters
FloatElement:
converter: FloatElementConverter
package: caoscrawler.converters
--
GitLab