Updated asn1 experiment, now exploring PER and OER

asn1/GajumaruSerialization.asn

@@ -4,8 +4,13 @@
 -- (the layer on top of RLP).
 --
 -- Purpose:
---   * Provide a formal, toolable description of the data.
---   * Enable migration from the legacy RLP-based format to DER.
+--   * Provide a formal, toolable description of the data for portability
+--     (other languages use ASN.1 compilers to get types/parsers).
+--   * Define compact canonical wire format using standard ASN.1 techniques
+--     (primarily Unaligned PER / UPER with constraints for packing).
+--   * Static encoding only (templates from gmserialization).
+--   * Legacy RLP compatibility is achieved via separate translation layer
+--     (deferred in current focus).
 --
 -- Detection of legacy vs new:
 --   Legacy data (produced by gmserialization:serialize / gmser_chain_objects:serialize
@@ -17,15 +22,21 @@
 --   top-level SEQUENCE we define). 0x30 < 0xC0, so a single-byte prefix
 --   check reliably distinguishes the two formats.
 --
--- Usage in Erlang (after compiling with asn1ct):
---   {ok, Mod} = asn1ct:compile(GajumaruSerialization.asn, [ber, der]).
---   {ok, Term} = 'GajumaruSerialization':decode('GajumaruData', DerBinary).
+-- Usage in Erlang (after compiling with asn1ct for compact wire):
+--   {ok, Mod} = asn1ct:compile(GajumaruSerialization.asn, [uper]).
+--   {ok, Bytes} = 'GajumaruSerialization':encode('GajumaruData', Asn1Value).
+--   Bytes is the compact UPER wire format (deterministic with this schema).
+--   For legacy RLP, use the model-to-RLP layer instead (see gmser_asn1_rlp).
 --
--- Notes on type mapping from gmserialization.erl:
---   - int     -> INTEGER          (new data uses canonical DER INTEGER)
+-- Notes on type mapping from gmserialization.erl (static templates):
+--   - int (bignum) -> BigInt ::= INTEGER (0..MAX)
+--     In practice used for amounts/balances in Pucks (up to 1*10^30).
+--     Encoded as non-negative bignum.
+--   - New smaller integer types will be added to templates (uint64, uint32, ...)
+--     for cases where range is known → enables tighter UPER packing.
 --   - binary  -> OCTET STRING
 --   - bool    -> BOOLEAN
---   - id      -> Id (SEQUENCE)    (cleaner than the legacy packed 33-byte form)
+--   - id      -> Id (SEQUENCE)
 --   - [T]     -> SEQUENCE OF T
 --   - {T1,...} -> SEQUENCE        (fixed-size, order matters)
 --   - #{items := [{K,T},...]} -> SEQUENCE with the fields in template order
@@ -33,9 +44,6 @@
 --   - tag + vsn are preserved as the first two fields for compatibility
 --     with existing dispatch logic.
 --
---   Legacy integers had strict unsigned-minimal no-leading-zero (except for 0)
---   and RLP-level rules. New DER data does not need to follow those rules.
---
 -- WARNING:
 --   Any cryptographic hash or signature computed over a serialized object
 --   will change when switching from RLP to DER for that object. Plan a
@@ -53,11 +61,20 @@ EXPORTS ALL;
 -- ============================================================
 
 GajumaruData ::= SEQUENCE {
-    tag     INTEGER,
-    vsn     INTEGER,
+    -- Constrained for better PER packing
+    tag     INTEGER (0..65535),
+    vsn     INTEGER (0..255),
     content Content
 }
 
+-- Preferred top-level for the compact static wire format.
+-- Avoids the extra Content CHOICE tag when the structure is known to be static.
+CompactStatic ::= SEQUENCE {
+    tag    INTEGER (0..65535),
+    vsn    INTEGER (0..255),
+    fields StaticFields
+}
+
 -- Content can be a specific structured type (preferred) or a generic
 -- representation of a template-driven object.
 Content ::= CHOICE {
@@ -65,6 +82,10 @@ Content ::= CHOICE {
     -- without having a pre-defined SEQUENCE for every object.
     templateFields  [0] TemplateFields,
 
+    -- Static-optimized: no field names on wire (matches legacy static behavior exactly)
+    -- Preferred for compact wire format of known templates.
+    staticFields    [10] StaticFields,
+
     -- Examples of concrete versioned types (extend as needed)
     account         [1] Account,
     signedTx        [2] SignedTx,
@@ -87,17 +108,41 @@ TemplateField ::= SEQUENCE {
     value   Value
 }
 
+-- Optimized for static wire format: just the values in order, no names.
+-- This matches the legacy static encoding where maps/records are positional only.
+StaticFields ::= SEQUENCE OF Value
+
+
 Value ::= CHOICE {
-    intValue    [0] INTEGER,
+    -- "int" in static templates is used for bignums in practice
+    -- (e.g. balances and amounts in Pucks, up to 1*10^30).
+    -- We keep it as unbounded non-negative integer.
+    bigIntValue [0] BigInt,
+
     boolValue   [1] BOOLEAN,
     binaryValue [2] OCTET STRING,
     idValue     [3] Id,
     listValue   [4] SEQUENCE OF Value,
     tupleValue  [5] SEQUENCE OF Value,
-    mapValue    [6] SEQUENCE OF KeyValue   -- only needed if you want to
-                                           -- represent dynamic-style maps
+    mapValue    [6] SEQUENCE OF KeyValue,   -- only needed if you want to
+                                            -- represent dynamic-style maps
+
+    -- Additional integer types for smaller ranges (future use in templates
+    -- for better UPER packing when the range is known).
+    uint64Value  [7] Uint64,
+    uint32Value  [8] Uint32,
+    uint128Value [9] Uint128
 }
 
+-- Bignum integer (non-negative). Used for the traditional "int" in static
+-- templates. Max practical value mentioned: 1*10^30 (≈ 2^100 bits).
+BigInt  ::= INTEGER (0..MAX)
+
+-- Convenience sized unsigned integer types.
+Uint64  ::= INTEGER (0..18446744073709551615)
+Uint32  ::= INTEGER (0..4294967295)
+Uint128 ::= INTEGER (0..340282366920938463463374607431768211455)
+
 KeyValue ::= SEQUENCE {
     key     Value,
     val     Value
@@ -203,23 +248,20 @@ TypeInfoV3 ::= SEQUENCE {
 --    always falling back to TemplateFields. This gives you better
 --    validation and generated types.
 --
--- 3. INTEGER in DER is signed and uses a different minimal encoding
---    than the legacy unsigned big-endian no leading zero form. This is
---    fine for new data.
+-- 3. For the compact wire format we use UPER (unaligned PER).
+--    It is stable/deterministic for a given schema (no extensibility
+--    markers on these types, fixed order).
 --
--- 4. If you need to preserve exact legacy integer wire bytes inside
---    the new format (e.g. for some hash preimage reason), you can
---    carry selected integers as OCTET STRING (legacy unsigned minimal bytes).
+-- 4. INTEGER uses ASN.1 PER encoding (canonical for the constraints).
+--    This may differ from legacy RLP minimal unsigned; new format
+--    will have different hashes (expected when introducing new encoding).
 --
--- 5. Dynamic encoding (gmser_dyn) is intentionally not modeled here
---    in full, because it is runtime-schema driven (type codes 246-255,
---    labels, alt/switch, etc.). You can still use the generic
---    TemplateFields + Value for some dynamic cases, or model specific
---    message schemas as additional CHOICE arms.
+-- 5. Dynamic encoding (gmser_dyn) is not in scope here.
 --
--- 6. Compile with DER for canonical output:
---       asn1ct:compile(GajumaruSerialization, [der]).
+-- 6. To generate the compact wire:
+--       asn1ct:compile(GajumaruSerialization, [uper]).
+--       {ok, CompactBytes} = 'GajumaruSerialization':encode('GajumaruData', Value).
 --
---    The ber option is also accepted; der implies the stricter rules.
+--    Use staticFields (not templateFields) for best compactness on static data.
 
 END