Merge pull request #18025 from jdavcs/24.0_tag_regex

[24.0] Fix tag regex pattern

client/src/components/Tags/model.js

@@ -7,7 +7,7 @@ import { keyedColorScheme } from "utils/color";
 
 // Valid tag regex. The basic format here is a tag name with optional subtags
 // separated by a period, and then an optional value after a colon.
-export const VALID_TAG_RE = /^([^\s.:])+(.[^\s.:]+)*(:[^\s.:]+)?$/;
+export const VALID_TAG_RE = /^([^\s.:])+(\.[^\s.:]+)*(:\S+)?$/;
 
 export class TagModel {
     /**

lib/galaxy/model/tags.py

@@ -295,7 +295,14 @@ def get_tag_by_name(self, tag_name):
         return None
 
     def _create_tag(self, tag_str: str):
-        """Create a Tag object from a tag string."""
+        """
+        Create or retrieve one or more Tag objects from a tag string. If there are multiple
+        hierarchical tags in the tag string, the string will be split along `self.hierarchy_separator` chars.
+        A Tag instance will be created for each non-empty prefix. If a prefix corresponds to the
+        name of an existing tag, that tag will be retrieved; otherwise, a new Tag object will be created.
+        For example, for the tag string `a.b.c` 3 Tag instances will be created: `a`, `a.b`, `a.b.c`.
+        Return the last tag created (`a.b.c`).
+        """
         tag_hierarchy = tag_str.split(self.hierarchy_separator)
         tag_prefix = ""
         parent_tag = None

lib/galaxy/schema/schema.py

@@ -66,6 +66,8 @@
 
 OptionalNumberT = Annotated[Optional[Union[int, float]], Field(None)]
 
+TAG_ITEM_PATTERN = r"^([^\s.:])+(\.[^\s.:]+)*(:\S+)?$"
+
 
 class DatasetState(str, Enum):
     NEW = "new"
@@ -527,7 +529,7 @@ class HistoryContentSource(str, Enum):
 DatasetCollectionInstanceType = Literal["history", "library"]
 
 
-TagItem = Annotated[str, Field(..., pattern=r"^([^\s.:])+(.[^\s.:]+)*(:[^\s.:]+)?$")]
+TagItem = Annotated[str, Field(..., pattern=TAG_ITEM_PATTERN)]
 
 
 class TagCollection(RootModel):

test/unit/app/managers/test_TagHandler.py

@@ -112,3 +112,15 @@ def test_item_has_tag(self):
         # Tag
         assert self.tag_handler.item_has_tag(self.user, item=hda, tag=hda.tags[0].tag)
         assert not self.tag_handler.item_has_tag(self.user, item=hda, tag="tag2")
+
+    def test_get_name_value_pair(self):
+        """Verify that parsing a single tag string correctly splits it into name/value pairs."""
+        assert self.tag_handler.parse_tags("a") == [("a", None)]
+        assert self.tag_handler.parse_tags("a.b") == [("a.b", None)]
+        assert self.tag_handler.parse_tags("a.b:c") == [("a.b", "c")]
+        assert self.tag_handler.parse_tags("a.b:c.d") == [("a.b", "c.d")]
+        assert self.tag_handler.parse_tags("a.b:c.d:e.f") == [("a.b", "c.d:e.f")]
+        assert self.tag_handler.parse_tags("a.b:c.d:e.f.") == [("a.b", "c.d:e.f.")]
+        assert self.tag_handler.parse_tags("a.b:c.d:e.f..") == [("a.b", "c.d:e.f..")]
+        assert self.tag_handler.parse_tags("a.b:c.d:e.f:") == [("a.b", "c.d:e.f:")]
+        assert self.tag_handler.parse_tags("a.b:c.d:e.f::") == [("a.b", "c.d:e.f::")]

test/unit/schema/test_schema.py

@@ -1,8 +1,12 @@
+import re
 from uuid import uuid4
 
 from pydantic import BaseModel
 
-from galaxy.schema.schema import DatasetStateField
+from galaxy.schema.schema import (
+    DatasetStateField,
+    TAG_ITEM_PATTERN,
+)
 from galaxy.schema.tasks import (
     GenerateInvocationDownload,
     RequestUser,
@@ -34,3 +38,43 @@ class StateModel(BaseModel):
 def test_dataset_state_coercion():
     assert StateModel(state="ok").state == "ok"
     assert StateModel(state="deleted").state == "discarded"
+
+
+class TestTagPattern:
+
+    def test_valid(self):
+        tag_strings = [
+            "a",
+            "aa",
+            "aa.aa",
+            "aa.aa.aa",
+            "~!@#$%^&*()_+`-=[]{};'\",./<>?",
+            "a.b:c",
+            "a.b:c.d:e.f",
+            "a.b:c.d:e..f",
+            "a.b:c.d:e.f:g",
+            "a.b:c.d:e.f::g",
+            "a.b:c.d:e.f::g:h",
+            "a::a",  # leading colon for tag value
+            "a:.a",  # leading period for tag value
+            "a:a:",  # trailing colon OK for tag value
+            "a:a.",  # trailing period OK for tag value
+        ]
+        for t in tag_strings:
+            assert re.match(TAG_ITEM_PATTERN, t)
+
+    def test_invalid(self):
+        tag_strings = [
+            " a",  # leading space for tag name
+            ":a",  # leading colon for tag name
+            ".a",  # leading period for tag name
+            "a ",  # trailing space for tag name
+            "a a",  # space inside tag name
+            "a: a",  # leading space for tag value
+            "a:a a",  # space inside tag value
+            "a:",  # trailing colon for tag name
+            "a.",  # trailing period for tag name
+            "a:b ",  # trailing space for tag value
+        ]
+        for t in tag_strings:
+            assert not re.match(TAG_ITEM_PATTERN, t)