cli(arp): the Interface column spells the name out, as the hardware does #2352

Merged
jaylouisw merged 1 commit from rev/1121-0212 into main 2026-11-21 02:39:24 +00:00
Owner

The question

One narrow, observable thing: does the Interface column of show arp / show ip arp print the interface's whole name, or the abbreviation?

Every primary capture of this table spells it out - Ethernet0/0 (Cisco TAC command reference), GigabitEthernet0/0/6 (IOS XE 16.7 ARP guide), GigabitEthernet0/0/2 (the ASR1001-X FAQ this repo already cites), FastEthernet0/3/0/6 (CRS command reference), and Vlan1 on switches, where an abbreviation would be Vl1. Nothing on cisco.com abbreviates in this table. The engine printed the stored faceplate name (Gi0/0), so one box said GigabitEthernet0/0 in show ip interface brief and Gi0/0 one command later.

Evidence

research/show-arp-interface-spelling.md carries the captures, the one contradicting sample (a blog whose rows are retyped - it also prints Age(min) without the space), and the reason the abbreviation lives in other tables: the same vendor-collector page that prints show ip arp → ARPA GigabitEthernet0/1 prints show mac address-table → Gi0/1.

What changed

  • engine/cli.js: the Interface cell, the box's own local rows and a VRRP group's row print ifaceName.displayName(name) - the same function the brief table uses (now in the small ifaceName alias). The conflict paragraph after the table names the port the same way, so one output cannot spell one interface two ways. Stored names are untouched, so interface GigabitEthernet0/0 typed from a row still resolves.
  • Tests: the five ARP suites (132 checks) compare the printed cell against displayName(recorded) - still proving it is the port the engine learned the binding on - and three checks pin the literal spelling (GigabitEthernet0, GigabitEthernet0.10, GigabitEthernet9/9 for a stored name that no longer resolves).
  • docs/CLI.md sample rows + the new Interface bullet; CHANGELOG.md Unreleased/Fixed.

Gates

node scripts/syntax-check.js OK (510 files). npm test: 430 suites, 14369 assertions, 0 failed.

Still unresolved

Whether any platform abbreviates (no cisco.com capture does; NX-OS has a different five-column table), and how a name wider than the terminal is handled - neither is witnessed.

## The question One narrow, observable thing: **does the Interface column of `show arp` / `show ip arp` print the interface's whole name, or the abbreviation?** Every primary capture of this table spells it out - `Ethernet0/0` (Cisco TAC command reference), `GigabitEthernet0/0/6` (IOS XE 16.7 ARP guide), `GigabitEthernet0/0/2` (the ASR1001-X FAQ this repo already cites), `FastEthernet0/3/0/6` (CRS command reference), and `Vlan1` on switches, where an abbreviation would be `Vl1`. Nothing on cisco.com abbreviates in this table. The engine printed the stored faceplate name (`Gi0/0`), so one box said `GigabitEthernet0/0` in `show ip interface brief` and `Gi0/0` one command later. ## Evidence `research/show-arp-interface-spelling.md` carries the captures, the one contradicting sample (a blog whose rows are retyped - it also prints `Age(min)` without the space), and the reason the abbreviation lives in *other* tables: the same vendor-collector page that prints `show ip arp` → `ARPA GigabitEthernet0/1` prints `show mac address-table` → `Gi0/1`. ## What changed - `engine/cli.js`: the Interface cell, the box's own local rows and a VRRP group's row print `ifaceName.displayName(name)` - the same function the brief table uses (now in the small `ifaceName` alias). The conflict paragraph after the table names the port the same way, so one output cannot spell one interface two ways. Stored names are untouched, so `interface GigabitEthernet0/0` typed from a row still resolves. - Tests: the five ARP suites (132 checks) compare the printed cell against `displayName(recorded)` - still proving it is the port the engine learned the binding on - and three checks pin the literal spelling (`GigabitEthernet0`, `GigabitEthernet0.10`, `GigabitEthernet9/9` for a stored name that no longer resolves). - `docs/CLI.md` sample rows + the new Interface bullet; `CHANGELOG.md` Unreleased/Fixed. ## Gates `node scripts/syntax-check.js` OK (510 files). `npm test`: **430 suites, 14369 assertions, 0 failed**. ## Still unresolved Whether any platform abbreviates (no cisco.com capture does; NX-OS has a different five-column table), and how a name wider than the terminal is handled - neither is witnessed.
cli(arp): the Interface column spells the name out, as the hardware does
All checks were successful
CI / Classify the change (pull_request) Successful in 14s
PR rebase / Rebase onto main and arm auto-merge (pull_request) Successful in 16s
CI / Server smoke and Docker build (pull_request) Successful in 28s
CI / Test (Node 22) (pull_request) Successful in 1m2s
CI / Test (Node 20) (pull_request) Successful in 1m6s
CI / Test (Node 18) (pull_request) Successful in 1m10s
CI / Arm auto-merge (pull_request) Successful in 4s
2b76dd4d63
Every primary capture of show arp / show ip arp prints the interface's whole
name - Ethernet0/0 (TAC command reference), GigabitEthernet0/0/6 (IOS XE ARP
guide), GigabitEthernet0/0/2 (the ASR FAQ this repo already quotes),
FastEthernet0/3/0/6 (CRS), Vlan1 on switches - and none prints Gi0/0/6. The
console wrote the stored faceplate name straight into the cell, so one box said
GigabitEthernet0/0 in show ip interface brief and Gi0/0 one command later.

The cell - neighbour rows, the box's own local rows and a VRRP group's - now
calls the same ifaceName.displayName() the brief table uses, and the conflict
paragraph names the port the same way. The stored name is untouched, so
'interface GigabitEthernet0/0' typed from a row still resolves.

The five ARP suites compare the printed cell against displayName(recorded) and
three checks pin the literal spelling. research/show-arp-interface-spelling.md
has the captures, the one retyped blog counter-sample, and what is still
unwitnessed.

Tests: npm test - 430 suites, 14369 assertions, 0 failed.
forgejo-actions scheduled this pull request to auto merge when all checks succeed 2026-11-21 02:33:32 +00:00
jaylouisw deleted branch rev/1121-0212 2026-11-21 02:39:26 +00:00
Sign in to join this conversation.
No reviewers
No milestone
No project
No assignees
1 participant
Notifications
Due date
The due date is invalid or out of range. Please use the format "yyyy-mm-dd".

No due date set.

Dependencies

No dependencies set.

Reference
jaylouisw/packetforge!2352
No description provided.