Merge pull request #102 from apdavison:prepare-release

* fix warning about invalid escape

* Release 0.6.0

Update CHANGELOG and set the version identifier to 0.6.0.

This is a minor rather than a patch release because #100 and #103 add new options to `by_name()`, and because #95 and #101 both change behaviour in ways that may require changes in downstream code, even though they are bug fixes.

Author: apdavison

CHANGELOG.md

@@ -96,3 +96,21 @@ For more detail see #29.
 ## Release 0.5.1 (2026-04-02)
 
 - Changed `embed_linked_nodes` from boolean to LinkedNodeEmbedding enum [#92](https://github.com/openMetadataInitiative/openMINDS_Python/pull/92). This adds an option "if necessary", which embeds linked nodes inline when they lack an @id, and otherwise uses a reference — useful for mixed scenarios where some nodes don't yet have identifiers. This is backwards compatible: True is accepted in place of "always" and False in place of "never.
+
+
+## Release 0.6.0 (2026-08-20)
+
+Note that this release contains two changes in behaviour which, although they are bug fixes, may require changes in downstream code. See "Changes in behaviour" below.
+
+- New options for the `by_name()` method [#100](https://github.com/openMetadataInitiative/openMINDS_Python/pull/100), [#103](https://github.com/openMetadataInitiative/openMINDS_Python/pull/103):
+    - `case_sensitive` (default `True`); set it to `False` to ignore case when matching.
+    - a third `match` mode, `"within"`, which finds instances whose name-like property is contained in the string you provide (the mirror image of `"contains"`).
+    - `ignore_accents` (default `False`); set it to `True` to ignore accents and other diacritical marks when matching, and to treat special letters such as ß, æ, ø and ł as their closest plain-letter equivalents. For example, `SovereignState.by_name("Republique francaise", ignore_accents=True)` finds "France". Combine it with `case_sensitive=False` to ignore differences in case as well.
+- Bug fixes in `by_name()` [#100](https://github.com/openMetadataInitiative/openMINDS_Python/pull/100):
+    - a crash when an instance did not have one of the name-like properties set.
+    - `by_name(..., all=True)` could return the same instance more than once, where it matched through several keys (e.g. a name that is also listed as one of its own synonyms).
+
+### Changes in behaviour
+
+- Cross-references between instance library instances are now resolved to typed openMINDS objects instead of being left as raw dicts [#95](https://github.com/openMetadataInitiative/openMINDS_Python/pull/95), fixing [#94](https://github.com/openMetadataInitiative/openMINDS_Python/issues/94). For example, `Accessibility.direct_virtual_open_access.payment_models[0]` is now a `PaymentModelType` object, where previously it was a dict. Code that treated such values as dicts will need updating. As part of this fix, `Node._resolve_links()` now keeps an unresolvable `Link` in a list-valued property rather than raising `KeyError`.
+- IRI validation has moved from the `IRI` constructor to the `validate()` method [#101](https://github.com/openMetadataInitiative/openMINDS_Python/pull/101), fixing [#93](https://github.com/openMetadataInitiative/openMINDS_Python/issues/93). Creating an `IRI` with an invalid value no longer raises `ValueError`; the problem is instead reported as a validation failure, with a specific hint to replace spaces with "%20" where that would make the IRI valid.

build.py

@@ -154,7 +154,7 @@
         fp.write(f"from . import ({', '.join(sorted(module_list))})\n")
 
 context = {
-    "version": "0.5.1",
+    "version": "0.6.0",
 }
 
 with open("target/pyproject.toml", "w") as fp:

pipeline/src/codemeta.json

@@ -58,6 +58,12 @@
         "@id": "https://orcid.org/0000-0003-4509-6969",
         "givenName": "Reema",
         "familyName": "Gupta"
+      },
+      {
+        "@type": "Person",
+        "@id": "https://orcid.org/0000-0002-2247-2418",
+        "givenName": "Geneviève",
+        "familyName": "Fleury"
       }
     ]
   }

pipeline/src/pyproject_template.toml.txt

@@ -9,7 +9,8 @@ authors = [
     { name="Louisa Köhnen" },
     { name="Peyman Najafi" },
     { name="Raphaël Gazzotti" },
-    { name="Reema Gupta" }
+    { name="Reema Gupta" },
+    { name="Geneviève Fleury" }
 ]
 requires-python = ">=3.9"
 license = "MIT"