lnmsg.decode_msg: dict values for numbers are int, instead of BE bytes

Will be useful for TLVs where it makes sense to do the conversion in lnmsg,
as it might be more complicated than just int.from_bytes().

electrum/lnmsg.py

@@ -1,7 +1,7 @@
 import os
 import csv
 import io
-from typing import Callable, Tuple, Any, Dict, List, Sequence, Union
+from typing import Callable, Tuple, Any, Dict, List, Sequence, Union, Optional
 
 
 class MalformedMsg(Exception):
@@ -24,8 +24,7 @@ def _assert_can_read_at_least_n_bytes(fd: io.BytesIO, n: int) -> None:
         raise UnexpectedEndOfStream(f"cur_pos={cur_pos}. end_pos={end_pos}. wants to read: {n}")
 
 
-# TODO return int when it makes sense
-def _read_field(*, fd: io.BytesIO, field_type: str, count: int) -> bytes:
+def _read_field(*, fd: io.BytesIO, field_type: str, count: int) -> Union[bytes, int]:
     if not fd: raise Exception()
     assert isinstance(count, int) and count >= 0, f"{count!r} must be non-neg int"
     if count == 0:
@@ -35,10 +34,19 @@ def _read_field(*, fd: io.BytesIO, field_type: str, count: int) -> bytes:
         type_len = 1
     elif field_type == 'u16':
         type_len = 2
+        assert count == 1, count
+        _assert_can_read_at_least_n_bytes(fd, type_len)
+        return int.from_bytes(fd.read(type_len), byteorder="big", signed=False)
     elif field_type == 'u32':
         type_len = 4
+        assert count == 1, count
+        _assert_can_read_at_least_n_bytes(fd, type_len)
+        return int.from_bytes(fd.read(type_len), byteorder="big", signed=False)
     elif field_type == 'u64':
         type_len = 8
+        assert count == 1, count
+        _assert_can_read_at_least_n_bytes(fd, type_len)
+        return int.from_bytes(fd.read(type_len), byteorder="big", signed=False)
     # TODO tu16/tu32/tu64
     elif field_type == 'chain_hash':
         type_len = 32
@@ -203,7 +211,8 @@ class LNSerializer:
                         try:
                             field_count = int(field_count_str)
                         except ValueError:
-                            field_count = int.from_bytes(parsed[field_count_str], byteorder="big")
+                            field_count = parsed[field_count_str]
+                    assert isinstance(field_count, int)
                     #print(f">> count={field_count}. parsed={parsed}")
                     try:
                         parsed[field_name] = _read_field(fd=fd,