From 12cb071fe2829005ad2d62e51501ccbefa3506f5 Mon Sep 17 00:00:00 2001
From: Harmen Stoppels <harmenstoppels@gmail.com>
Date: Mon, 14 Aug 2023 15:38:06 +0200
Subject: [PATCH] Fix broken semver regex (#39414)

---
 lib/spack/spack/test/versions.py | 23 +++++++++++++++++++++++
 lib/spack/spack/version.py       | 22 +++++++++++++---------
 2 files changed, 36 insertions(+), 9 deletions(-)

diff --git a/lib/spack/spack/test/versions.py b/lib/spack/spack/test/versions.py
index 7d329336aee..f40f23f7307 100644
--- a/lib/spack/spack/test/versions.py
+++ b/lib/spack/spack/test/versions.py
@@ -17,6 +17,7 @@
 import spack.package_base
 import spack.spec
 from spack.version import (
+    SEMVER_REGEX,
     GitVersion,
     StandardVersion,
     Version,
@@ -976,3 +977,25 @@ def test_unresolvable_git_versions_error(config, mock_packages):
         # The package exists, but does not have a git property set. When dereferencing
         # the version, we should get VersionLookupError, not a generic AttributeError.
         spack.spec.Spec(f"git-test-commit@{'a' * 40}").version.ref_version
+
+
+@pytest.mark.parametrize(
+    "tag,expected",
+    [
+        ("v100.2.3", "100.2.3"),
+        ("v1.2.3", "1.2.3"),
+        ("v1.2.3-pre.release+build.1", "1.2.3-pre.release+build.1"),
+        ("v1.2.3+build.1", "1.2.3+build.1"),
+        ("v1.2.3+build_1", None),
+        ("v1.2.3-pre.release", "1.2.3-pre.release"),
+        ("v1.2.3-pre_release", None),
+        ("1.2.3", "1.2.3"),
+        ("1.2.3.", None),
+    ],
+)
+def test_semver_regex(tag, expected):
+    result = SEMVER_REGEX.search(tag)
+    if expected is None:
+        assert result is None
+    else:
+        assert result.group() == expected
diff --git a/lib/spack/spack/version.py b/lib/spack/spack/version.py
index b8452ab61d7..257aed3cdff 100644
--- a/lib/spack/spack/version.py
+++ b/lib/spack/spack/version.py
@@ -40,11 +40,16 @@
 SEGMENT_REGEX = re.compile(r"(?:(?P<num>[0-9]+)|(?P<str>[a-zA-Z]+))(?P<sep>[_.-]*)")
 
 # regular expression for semantic versioning
-SEMVER_REGEX = re.compile(
-    ".+(?P<semver>([0-9]+)[.]([0-9]+)[.]([0-9]+)"
-    "(?:-([0-9A-Za-z-]+(?:[.][0-9A-Za-z-]+)*))?"
-    "(?:[+][0-9A-Za-z-]+)?)"
-)
+_VERSION_CORE = r"\d+\.\d+\.\d+"
+_IDENT = r"[0-9A-Za-z-]+"
+_SEPARATED_IDENT = rf"{_IDENT}(?:\.{_IDENT})*"
+_PRERELEASE = rf"\-{_SEPARATED_IDENT}"
+_BUILD = rf"\+{_SEPARATED_IDENT}"
+_SEMVER = rf"{_VERSION_CORE}(?:{_PRERELEASE})?(?:{_BUILD})?"
+
+# clamp on the end, so versions like v1.2.3-rc1 will match
+# without the leading 'v'.
+SEMVER_REGEX = re.compile(rf"{_SEMVER}$")
 
 # Infinity-like versions. The order in the list implies the comparison rules
 infinity_versions = ["stable", "trunk", "head", "master", "main", "develop"]
@@ -1319,11 +1324,10 @@ def lookup_ref(self, ref) -> Tuple[Optional[str], int]:
                         commit_to_version[tag_commit] = v
                         break
                 else:
-                    # try to parse tag to copare versions spack does not know
-                    match = SEMVER_REGEX.match(tag)
+                    # try to parse tag to compare versions spack does not know
+                    match = SEMVER_REGEX.search(tag)
                     if match:
-                        semver = match.groupdict()["semver"]
-                        commit_to_version[tag_commit] = semver
+                        commit_to_version[tag_commit] = match.group()
 
             ancestor_commits = []
             for tag_commit in commit_to_version: