From b16072b66e387ceb9d326bbb233c35c8e344efab Mon Sep 17 00:00:00 2001
From: Titusz Pan <titusz.pan@gmail.com>
Date: Mon, 18 Nov 2024 11:57:47 +0100
Subject: [PATCH] Add multiformats support for codec convenience functions

---
 CHANGELOG.md               |  1 +
 docs/changelog.md          |  1 +
 iscc_core/codec.py         | 57 ++++++++++++++++++++++++--------------
 tests/test_multiformats.py | 23 +++++++++++++++
 4 files changed, 61 insertions(+), 21 deletions(-)
 create mode 100644 tests/test_multiformats.py

diff --git a/CHANGELOG.md b/CHANGELOG.md
index be815a4..a28bfc1 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -4,6 +4,7 @@
 - Set CDC index to size if size is smaller than minimum chunk size
 - Added Python 3.13 support
 - Added .env to .gitignore
+- Added multiformats support for codec convenience functions
 - Updeted locked dependencies
 - Removed EOL Python 3.8/3.9 support
 
diff --git a/docs/changelog.md b/docs/changelog.md
index be815a4..a28bfc1 100644
--- a/docs/changelog.md
+++ b/docs/changelog.md
@@ -4,6 +4,7 @@
 - Set CDC index to size if size is smaller than minimum chunk size
 - Added Python 3.13 support
 - Added .env to .gitignore
+- Added multiformats support for codec convenience functions
 - Updeted locked dependencies
 - Removed EOL Python 3.8/3.9 support
 
diff --git a/iscc_core/codec.py b/iscc_core/codec.py
index 2b373c7..a78214d 100644
--- a/iscc_core/codec.py
+++ b/iscc_core/codec.py
@@ -325,6 +325,33 @@ def decode_base32hex(code):
     return decode_base32(b32)
 
 
+def normalize_multiformat(iscc_code):
+    """
+    Transcode a multiformat encoded ISCC to standard base32 encoding.
+    Returns the input unchanged (but cleaned) if it's not multiformat encoded.
+    """
+    decoders = {
+        MULTIBASE.base16.value: bytes.fromhex,  # f
+        MULTIBASE.base32.value: decode_base32,  # b
+        MULTIBASE.base32hex.value: decode_base32hex,  # v
+        MULTIBASE.base58btc.value: base58.b58decode,  # z
+        MULTIBASE.base64url.value: decode_base64,  # u
+    }
+
+    # Clean the ISCC code first
+    iscc_code = iscc_clean(iscc_code)
+
+    # Check for multibase prefix
+    multibase_prefix = iscc_code[0]
+    if multibase_prefix in decoders.keys():
+        decoder = decoders[multibase_prefix]
+        decoded = decoder(iscc_code[1:])
+        if not decoded.startswith(MC_PREFIX):
+            raise ValueError(f"Malformed multiformat codec: {decoded[:2]}")
+        return encode_base32(decoded[2:])
+    return iscc_code
+
+
 def iscc_decompose(iscc_code):
     # type: (str) -> List[str]
     """
@@ -333,9 +360,10 @@ def iscc_decompose(iscc_code):
     A valid ISCC sequence is a string concatenation of ISCC-UNITS optionally seperated
     by a hyphen.
     """
-    iscc_code = iscc_clean(iscc_code)
-    components = []
+    # Handle multiformat encoding first
+    iscc_code = normalize_multiformat(iscc_code)
 
+    components = []
     raw_code = decode_base32(iscc_code)
     while raw_code:
         mt, st, vs, ln, body = decode_header(raw_code)
@@ -400,26 +428,13 @@ def iscc_normalize(iscc_code):
     """
     from iscc_core.iscc_code import gen_iscc_code_v0
 
-    decoders = {
-        MULTIBASE.base16.value: bytes.fromhex,  # f
-        MULTIBASE.base32.value: decode_base32,  # b
-        MULTIBASE.base32hex.value: decode_base32hex,  # v
-        MULTIBASE.base58btc.value: base58.b58decode,  # z
-        MULTIBASE.base64url.value: decode_base64,  # u
-    }
+    # Handle multiformat encoding first
+    iscc_code = normalize_multiformat(iscc_code)
 
-    # Transcode to base32 if <multibase><multicodec> encoded
-    multibase_prefix = iscc_code[0]
-    if multibase_prefix in decoders.keys():
-        decoder = decoders[multibase_prefix]
-        decoded = decoder(iscc_code[1:])
-        if not decoded.startswith(MC_PREFIX):
-            raise ValueError(f"Malformed multiformat codec: {decoded[:2]}")
-        iscc_code = encode_base32(decoded[2:])
-    else:
-        prefix = iscc_code.upper().replace("ISCC:", "")[:2]
-        if prefix not in PREFIXES:
-            raise ValueError(f"ISCC starts with invalid prefix {prefix}")
+    # Validate prefix
+    prefix = iscc_code.upper()[:2]
+    if prefix not in PREFIXES:
+        raise ValueError(f"ISCC starts with invalid prefix {prefix}")
 
     decomposed = iscc_decompose(iscc_code)
     recomposed = gen_iscc_code_v0(decomposed)["iscc"] if len(decomposed) >= 2 else decomposed[0]
diff --git a/tests/test_multiformats.py b/tests/test_multiformats.py
new file mode 100644
index 0000000..1e53197
--- /dev/null
+++ b/tests/test_multiformats.py
@@ -0,0 +1,23 @@
+# -*- coding: utf-8 -*-
+import iscc_core as ic
+
+CANONICAL = "ISCC:EAAWFH3PX3MCYB6N"
+MF_B32H = "vpg0i00b2jtnrtm1c0v6g"
+MF_B32H_P = "iscc:vpg0i00b2jtnrtm1c0v6g"
+ISCC_OBJ = ic.Code(CANONICAL)
+
+
+def test_iscc_clean():
+    assert ic.iscc_clean(MF_B32H) == MF_B32H
+    assert ic.iscc_clean(MF_B32H_P) == MF_B32H
+
+
+def test_iscc_decode():
+    assert ic.iscc_decode(MF_B32H) == ISCC_OBJ._head + (ISCC_OBJ._body.tobytes(),)
+    assert ic.iscc_decode(MF_B32H_P) == ISCC_OBJ._head + (ISCC_OBJ._body.tobytes(),)
+
+
+def test_iscc_decompose():
+    decomposed = ic.iscc_decompose(ISCC_OBJ.uri)
+    assert ic.iscc_decompose(MF_B32H) == decomposed
+    assert ic.iscc_decompose(MF_B32H_P) == decomposed