From e8deb94befc2a67ad8d2e7703233ca1392fb2ec7 Mon Sep 17 00:00:00 2001
From: Richard Liang <rliang@cfenet.ubc.ca>
Date: Thu, 28 Aug 2025 16:07:10 -0700
Subject: [PATCH 1/2] Fixed a mis-identified precondition that doesn't always
 hold.

Fortunately this didn't require much of a code change.  Added tests
to cover the previously-missed cases.
---
 src/hla_algorithm/models.py |  6 +-----
 tests/models_test.py        | 32 ++++++++++++++++++++++++++++++--
 2 files changed, 31 insertions(+), 7 deletions(-)

diff --git a/src/hla_algorithm/models.py b/src/hla_algorithm/models.py
index 10f4035..69b78a7 100644
--- a/src/hla_algorithm/models.py
+++ b/src/hla_algorithm/models.py
@@ -429,15 +429,11 @@ def _identify_longest_prefix(allele_prefixes: list[GeneCoord]) -> GeneCoord:
 
         Precondition: that the input must all share at least the same first
         coordinate.  The algorithm may not return cogent values if not.
-
-        Precondition: the specified allele prefixes do not all perfectly match,
-        so we lose nothing by trimming one coordinate off the end of all of
-        them.
         """
         longest_prefix: GeneCoord = ()
         if len(allele_prefixes) > 0:
             max_length: int = max([len(allele) for allele in allele_prefixes])
-            for i in range(max_length - 1, 0, -1):
+            for i in range(max_length, 0, -1):
                 curr_prefixes: set[GeneCoord] = {allele[0:i] for allele in allele_prefixes}
                 if len(curr_prefixes) == 1:
                     longest_prefix = curr_prefixes.pop()
diff --git a/tests/models_test.py b/tests/models_test.py
index d3e020e..cb21e59 100644
--- a/tests/models_test.py
+++ b/tests/models_test.py
@@ -1206,8 +1206,31 @@ def test_identify_clean_prefix_in_pairs(
                 (),
                 id="trivial_case",
             ),
-            # Note: we have no single allele tests because that contradicts one
-            # of our preconditions.
+            pytest.param(
+                [("C*01", "02", "03", "04G")],
+                ("C*01", "02", "03", "04G"),
+                id="single_input_length_4",
+            ),
+            pytest.param(
+                [("C*01", "02", "03")],
+                ("C*01", "02", "03"),
+                id="single_input_length_3",
+            ),
+            pytest.param(
+                [("C*01", "02")],
+                ("C*01", "02"),
+                id="single_input_length_2",
+            ),
+            pytest.param(
+                [("C*01",)],
+                ("C*01",),
+                id="single_input_length_1",
+            ),
+            pytest.param(
+                [("C*01", "02", "03", "04G"), ("C*01", "02", "03", "04G")],
+                ("C*01", "02", "03", "04G"),
+                id="best_match_length_4",
+            ),
             pytest.param(
                 [("C*01", "02", "03", "04G"), ("C*01", "02", "03", "110N")],
                 ("C*01", "02", "03"),
@@ -1248,6 +1271,11 @@ def test_identify_clean_prefix_in_pairs(
                 ("C*01",),
                 id="best_match_length_1_different_lengths_one_with_no_excess",
             ),
+            pytest.param(
+                [("C*01", "07", "88"), ("C*01", "07", "01"), ("C*01", "07", "01", "110N")],
+                ("C*01", "07"),
+                id="typical_case",
+            ),
         ],
     )
     def test_identify_longest_prefix(

From fa4e4b6ea0f10917fd66e27eecfc420fad966d73 Mon Sep 17 00:00:00 2001
From: Richard Liang <rliang@cfenet.ubc.ca>
Date: Thu, 28 Aug 2025 16:09:20 -0700
Subject: [PATCH 2/2] Fixed up a docstring's formatting.

---
 src/hla_algorithm/models.py | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/src/hla_algorithm/models.py b/src/hla_algorithm/models.py
index 69b78a7..86ea8b6 100644
--- a/src/hla_algorithm/models.py
+++ b/src/hla_algorithm/models.py
@@ -427,8 +427,9 @@ def _identify_longest_prefix(allele_prefixes: list[GeneCoord]) -> GeneCoord:
         """
         Identify the longest gene coordinate "prefix" in the given allele prefixes.
 
-        Precondition: that the input must all share at least the same first
-        coordinate.  The algorithm may not return cogent values if not.
+        Precondition: all allele prefixes in the input must all share at least
+        the same first coordinate.  The algorithm may not return cogent values
+        if not.
         """
         longest_prefix: GeneCoord = ()
         if len(allele_prefixes) > 0: