From 180803d757d8eb7ecbc29e9580ae4238daf78067 Mon Sep 17 00:00:00 2001
From: Jonas Alves <jonas@absmartly.com>
Date: Tue, 14 Jul 2026 23:07:59 +0100
Subject: [PATCH] feat: expose public hash-unit function for diagnostic use

Extract the MD5 + base64url-no-padding unit hashing logic out of
Context.get_unit_hash() into a standalone sdk.internal.hashing.hash_unit()
function, so callers that only need the hash (e.g. diagnostic tooling)
don't have to construct a full Context. Context.get_unit_hash() now
delegates to it instead of duplicating the logic.
---
 sdk/context.py          | 7 ++-----
 sdk/internal/hashing.py | 7 +++++++
 2 files changed, 9 insertions(+), 5 deletions(-)
 create mode 100644 sdk/internal/hashing.py

diff --git a/sdk/context.py b/sdk/context.py
index 5555a82..989ab73 100644
--- a/sdk/context.py
+++ b/sdk/context.py
@@ -1,6 +1,4 @@
-import base64
 import collections
-import hashlib
 import threading
 from concurrent.futures import Future
 from typing import Optional
@@ -10,6 +8,7 @@
 from sdk.context_data_provider import ContextDataProvider
 from sdk.context_event_handler import ContextEventHandler
 from sdk.context_event_logger import ContextEventLogger, EventType
+from sdk.internal import hashing
 from sdk.internal.lock.atomic_bool import AtomicBool
 from sdk.internal.lock.atomic_int import AtomicInt
 from sdk.internal.lock.concurrency import Concurrency
@@ -536,9 +535,7 @@ def peek_treatment(self, experiment_name: str):
 
     def get_unit_hash(self, unit_type: str, unit_uid: str):
         def computer(key: str):
-            dig = hashlib.md5(unit_uid.encode('utf-8')).digest()
-            unithash = base64.urlsafe_b64encode(dig).rstrip(b'=')
-            return unithash
+            return hashing.hash_unit(unit_uid).encode('ascii')
 
         return Concurrency.compute_if_absent_rw(
             self.context_lock,
diff --git a/sdk/internal/hashing.py b/sdk/internal/hashing.py
new file mode 100644
index 0000000..9f56e1d
--- /dev/null
+++ b/sdk/internal/hashing.py
@@ -0,0 +1,7 @@
+import base64
+import hashlib
+
+
+def hash_unit(unit: str) -> str:
+    dig = hashlib.md5(unit.encode('utf-8')).digest()
+    return base64.urlsafe_b64encode(dig).rstrip(b'=').decode('ascii')