tools: Normalize aggregate-address command when doing frr-reload

[ton31337]

No description provided.

[greptile-apps]

Greptile Summary

This PR adds a get_normalized_aggregate_address_line function to frr-reload.py that canonicalizes the keyword ordering of aggregate-address commands. Because FRR accepts keywords in any order but emits them in a fixed order, a config file with keywords in a non-canonical order caused frr-reload to always see a diff and re-apply the aggregate — briefly withdrawing the prefix on every reload.

• Parses and reorders the six supported keywords (as-set, summary-only, route-map, origin, matching-MED-only, suppress-map) into the exact order FRR's bgp_config_write_network emits them, confirmed in bgpd/bgp_route.c:18926–18948.
• Handles both CIDR (A.B.C.D/M) and old-style (A.B.C.D A.B.C.D) prefix notation; any line containing an unrecognized token is returned untouched to avoid silent drops.

Confidence Score: 4/5

Safe to merge; the normalization is additive and falls back to returning the original line for any unrecognized input.

The canonical keyword order produced by the new function matches exactly what bgpd's bgp_config_write_network outputs (confirmed in bgpd/bgp_route.c). The regex correctly handles both CIDR and legacy A.B.C.D A.B.C.D prefix notation, all six valid keywords are covered, and the else-return-line guard means any future keyword addition degrades gracefully.

tools/frr-reload.py — specifically the prefix-capture regex in get_normalized_aggregate_address_line; verified correct but slightly subtle.

Important Files Changed

Filename	Overview
tools/frr-reload.py	Adds aggregate-address keyword normalization; canonical order confirmed against bgpd config-write output; safe fall-through for unknown tokens.

Flowchart

%%{init: {'theme': 'neutral'}}%%
flowchart TD
    A[line starts with 'aggregate-address '] --> B[regex match prefix + rest]
    B -- no match --> C[return line unchanged]
    B -- match --> D[tokenize rest into list]
    D --> E{for each token}
    E -- as-set --> F[as_set = True]
    E -- summary-only --> G[summary_only = True]
    E -- matching-MED-only --> H[match_med = True]
    E -- route-map --> I[consume next token as route_map]
    E -- origin --> J[consume next token as origin]
    E -- suppress-map --> K[consume next token as suppress_map]
    E -- unknown token --> C
    F & G & H & I & J & K --> E
    E -- done --> L[build canonical string]
    L --> M[return normalized line]

Loading

_{Reviews (1): Last reviewed commit: "tools: Normalize aggregate-address comma..." | Re-trigger Greptile}

[riw777]

looks good

[ton31337]

@Mergifyio backport stable/10.7

[mergify]

backport stable/10.7

🟠 Waiting for conditions to match

Details

• merged [📌 backport requirement]