Replace hardcoded `_get_tag()` with data-driven `TAG_TABLE` generated from CSV

[sumanjeet0012]

The _get_tag() function in code.py is ~260 lines of hardcoded if/elif pattern matching. It should be replaced with a simple dict lookup using a TAG_TABLE that is generated from the upstream CSV's tag column, matching how go-multicodec generates its Tag() method.

Problem

Current implementation in multicodec/code.py (lines 178–438):

def _get_tag(code: int) -> str:
    name = CODE_TABLE.get(code, "")

    # CID
    if name in ("cidv1", "cidv2", "cidv3"):
        return "cid"

    # Encryption
    if name in ("aes-gcm-256",):
        return "encryption"

    # ... 200+ more lines of if/elif pattern matching ...

    # Multihash — substring matching (fragile!)
    if any(h in name for h in ("sha2-", "sha3-", "blake2b-", ...)):
        return "multihash"

    return "<unknown>"

Problems with this approach:

• Drift risk: When new codecs are added to the CSV, they may not match existing patterns and silently return "<unknown>"
• Already missing 6 tags: multikey, multisig, nonce, shelter, softhash, vlad (see related issue)
• 260 lines of code that should be ~5 lines with a generated lookup table
• No connection to the authoritative CSV tag column

In contrast, go-multicodec generates its Tag() method directly from the CSV:

// Generated from table.csv — always in sync
func (c Code) Tag() string {
    switch c {
    case Identity, Sha1, Sha2_256, ...:
        return "multihash"
    case Ip4, Tcp, Udp, ...:
        return "multiaddr"
    // ...
    }
}

Proposed Solution

1. Generate a TAG_TABLE dict in constants.py (or a new tags.py) that maps every codec code to its tag, sourced from the CSV:

   TAG_TABLE = {
       0x12: "multihash",     # sha2-256
       0x04: "multiaddr",     # ip4
       0x70: "ipld",          # dag-pb
       # ... all 603 entries ...
   }

2. Replace _get_tag() in code.py with a one-line lookup:

   def _get_tag(code: int) -> str:
       return TAG_TABLE.get(code, "<unknown>")

3. Update tools/update-table.py (or tools/gen_code_table.py) to generate TAG_TABLE from the CSV data.

4. Remove the ~260 lines of hardcoded pattern matching from code.py.

5. Add tests that verify every codec's tag matches the CSV.

Related

• Issue: _get_tag() missing 6 tags
• Go implementation: go-multicodec gen.go generates Tag() from CSV