Merge branch 'main' into facu/fix-prep-prop-2

docs/architecture/adr-050-sign-mode-textual-annex1.md

@@ -5,6 +5,7 @@
 * Dec 06, 2021: Initial Draft
 * Feb 07, 2022: Draft read and concept-ACKed by the Ledger team.
 * Dec 01, 2022: Remove `Object: ` prefix on Any header screen.
+* Dec 13, 2022: Sign over bytes hash when bytes length > 32.
 
 ## Status
 
@@ -254,7 +255,20 @@ Examples:
 
 ### bytes
 
-* Bytes are rendered in hexadecimal, all capital letters, without the `0x` prefix.
+* Bytes of length shorter or equal to 32 are rendered in hexadecimal, all capital letters, without the `0x` prefix.
+* Bytes of length greater than 32 are hashed using SHA256. The rendered text is `SHA-256=`, followed by the 32-byte hash, in hexadecimal, all capital letters, without the `0x` prefix.
+* The hexadecimal string is finally separated into groups of 4 digits, with a space `' '` as separator. If the bytes length is odd, the 2 remaining hexadecimal characters are at the end.
+
+Note: Data longer than 32 bytes are not rendered in a way that can be inverted. See ADR-050's [section about invertability](./adr-050-sign-mode-textual.md#invertible-rendering) for a discussion.
+
+#### Examples
+
+Inputs are displayed as byte arrays.
+
+* `[0]`: `00`
+* `[0,1,2]`: `0001 02`
+* `[0,1,2,..,31]`: `0001 0203 0405 0607 0809 0A0B 0C0D 0E0F 1011 1213 1415 1617 1819 1A1B 1C1D 1E1F` 
+* `[0,1,2,..,32]`: `SHA-256=5D8F CFEF A9AE EB71 1FB8 ED1E 4B7D 5C8A 9BAF A46E 8E76 E68A A18A DCE5 A10D F6AB`
 
 ### address bytes
 

docs/architecture/adr-050-sign-mode-textual.md

@@ -9,6 +9,7 @@
 * Sep 07, 2022: Add custom `Msg`-renderers.
 * Sep 18, 2022: Structured format instead of lines of text
 * Nov 23, 2022: Specify CBOR encoding.
+* Dec 14, 2022: Mention exceptions for invertability.
 
 ## Status
 
@@ -76,6 +77,14 @@ Note that the existence of an inverse function ensures that the
 rendered text contains the full information of the original transaction,
 not a hash or subset.
 
+We make an exception for invertibility for data which are too large to
+meaningfully display, such as byte strings longer than 32 bytes. We may then
+selectively render them with a cryptographically-strong hash. In these cases,
+it is still computationally infeasible to find a different transaction which
+has the same rendering. However, we must ensure that the hash computation is
+simple enough to be reliably executed independently, so at least the hash is
+itself reasonably verifiable when the raw byte string is not.
+
 ### Chain State
 
 The rendering function (and parsing function) may depend on the current chain state.

tx/textual/internal/testdata/bytes.json

@@ -1,10 +1,12 @@
 [
-    ["", ""],
-    ["00", "AA=="],
-    ["66", "Zg=="],
-    ["666F", "Zm8="],
-    ["666F6F", "Zm9v"],
-    ["666F6F62", "Zm9vYg=="],
-    ["666F6F6261", "Zm9vYmE="],
-    ["666F6F626172", "Zm9vYmFy"]
-  ]
+	["", ""],
+	["AA==", "00"],
+	["Zg==", "66"],
+	["Zm8=", "666F"],
+	["Zm9v", "666F 6F"],
+	["Zm9vYg==", "666F 6F62"],
+	["Zm9vYmE=", "666F 6F62 61"],
+	["Zm9vYmFy", "666F 6F62 6172"],
+	["AAECAwQFBgcICQoLDA0ODxAREhMUFRYXGBkaGxwdHh8=", "0001 0203 0405 0607 0809 0A0B 0C0D 0E0F 1011 1213 1415 1617 1819 1A1B 1C1D 1E1F"],
+	["AAECAwQFBgcICQoLDA0ODxAREhMUFRYXGBkaGxwdHh8g", "SHA-256=5D8F CFEF A9AE EB71 1FB8 ED1E 4B7D 5C8A 9BAF A46E 8E76 E68A A18A DCE5 A10D F6AB"]
+]

tx/textual/valuerenderer/bytes.go

@@ -1,14 +1,21 @@
 package valuerenderer
 
 import (
+	"bytes"
 	"context"
+	"crypto/sha256"
 	"encoding/hex"
 	"fmt"
 	"strings"
 
 	"google.golang.org/protobuf/reflect/protoreflect"
 )
 
+var (
+	hashPrefix = "SHA-256="
+	maxByteLen = 32 // Maximum byte length to be displayed as is. Longer than this, we hash.
+)
+
 // NewBytesValueRenderer returns a ValueRenderer for Protobuf bytes, which are
 // encoded as capital-letter hexadecimal, without the '0x' prefix.
 func NewBytesValueRenderer() ValueRenderer {
@@ -20,20 +27,56 @@ type bytesValueRenderer struct{}
 var _ ValueRenderer = bytesValueRenderer{}
 
 func (vr bytesValueRenderer) Format(ctx context.Context, v protoreflect.Value) ([]Screen, error) {
-	text := strings.ToUpper(hex.EncodeToString(v.Bytes()))
+	bz := v.Bytes()
+
+	if len(bz) <= maxByteLen {
+		text := toHex(bz)
+		return []Screen{{Text: text}}, nil
+	}
+
+	// For long bytes, we display the hash.
+	hasher := sha256.New()
+	_, err := hasher.Write(bz)
+	if err != nil {
+		return nil, err
+	}
+	h := hasher.Sum(nil)
+
+	text := fmt.Sprintf("%s%s", hashPrefix, toHex(h))
 	return []Screen{{Text: text}}, nil
 }
 
 func (vr bytesValueRenderer) Parse(_ context.Context, screens []Screen) (protoreflect.Value, error) {
 	if len(screens) != 1 {
-		return protoreflect.ValueOfBytes([]byte{}), fmt.Errorf("expected single screen: %v", screens)
+		return nilValue, fmt.Errorf("expected single screen: %v", screens)
 	}
 	formatted := screens[0].Text
 
-	data, err := hex.DecodeString(string(formatted))
+	// If the formatted string starts with `SHA-256=`, then we can't actually
+	// invert to get the original bytes. In this case, we error.
+	if strings.HasPrefix(formatted, hashPrefix) {
+		return nilValue, fmt.Errorf("cannot parse bytes hash")
+	}
+
+	// Remove all spaces between every 4th char, then we can decode hex.
+	data, err := hex.DecodeString(strings.ReplaceAll(formatted, " ", ""))
 	if err != nil {
-		return protoreflect.ValueOfBytes([]byte{}), err
+		return nilValue, err
 	}
 
 	return protoreflect.ValueOfBytes(data), nil
 }
+
+// toHex converts bytes to hex, and inserts a space every 4th character.
+func toHex(bz []byte) string {
+	text := strings.ToUpper(hex.EncodeToString(bz))
+
+	var buffer bytes.Buffer
+	for i, r := range text {
+		buffer.WriteRune(r)
+		if i < len(text)-1 && i%4 == 3 {
+			buffer.WriteRune(' ')
+		}
+	}
+	return buffer.String()
+}

tx/textual/valuerenderer/bytes_test.go

@@ -37,17 +37,23 @@ func TestBytesJsonTestCases(t *testing.T) {
 
 		// Round trip
 		val, err := valrend.Parse(context.Background(), screens)
+		if err != nil {
+			// Make sure Parse() only errors because of hashed bytes.
+			require.Equal(t, "cannot parse bytes hash", err.Error())
+			require.Greater(t, len(tc.base64), 32)
+			continue
+		}
 		require.NoError(t, err)
 		require.Equal(t, tc.base64, base64.StdEncoding.EncodeToString(val.Bytes()))
 	}
 }
 
 type bytesTest struct {
-	hex    string
 	base64 string
+	hex    string
 }
 
 func (t *bytesTest) UnmarshalJSON(b []byte) error {
-	a := []interface{}{&t.hex, &t.base64}
+	a := []interface{}{&t.base64, &t.hex}
 	return json.Unmarshal(b, &a)
 }