JSON encoder: align keys to camelCase and support flat/nested formats by freebacca · Pull Request #6 · wineslab/libe3

freebacca · 2026-03-24T19:14:44Z

Align all JSON key names to camelCase to match the ASN.1 naming convention (e.g., dAppName, responseCode, ranFunctionList)
Align PDU type values to camelCase (e.g., setupRequest instead of SetupRequest)
Support both flat and nested JSON message formats in the decoder: if the incoming message has a data wrapper, payload is read from it; otherwise, payload is read from root
Encoder mirrors the format of the last decoded message (flat or nested), preserving backward compatibility
Existing tests updated and passing (14/14)

Copilot

Pull request overview

This PR updates the JSON encoder/decoder to use camelCase field names and camelCase PDU type strings, and introduces support for decoding both “flat” JSON messages (payload at root) and “nested” messages (payload under a data wrapper), with the encoder attempting to mirror the last decoded format.

Changes:

Renames JSON envelope keys and payload keys to camelCase (e.g., type, id, dAppName, ranFunctionList) and updates PDU type string values to camelCase.
Updates decoding to accept both flat and nested (data-wrapped) formats and makes encoding optionally emit nested payloads based on the last decode.
Adjusts CMake find_package minimum version for nlohmann_json and updates JSON encoder tests for new PDU type strings.

Reviewed changes

Copilot reviewed 4 out of 4 changed files in this pull request and generated 4 comments.

File	Description
`tests/test_json_encoder.cpp`	Updates assertions to match new camelCase PDU type strings in encoded JSON.
`src/encoder/json_encoder.hpp`	Adds `nested_mode_` state used to choose flat vs nested encoding format.
`src/encoder/json_encoder.cpp`	Implements camelCase keys, new PDU type string mapping, and flat/nested decode + stateful encode mirroring.
`cmake/libe3Dependencies.cmake`	Lowers the minimum `nlohmann_json` version requested via `find_package`.

Comments suppressed due to low confidence (1)

cmake/libe3Dependencies.cmake:21

CMake now accepts an installed nlohmann_json as old as 3.10, but the FetchContent fallback still pins v3.11.3. This can lead to different behavior/ABI depending on environment.

Consider aligning the minimum find_package version with the fetched tag (or update the fetched tag to a 3.10.x release if 3.10 support is the goal).

    find_package(nlohmann_json 3.10 QUIET)
    if(NOT nlohmann_json_FOUND)
        include(FetchContent)
        FetchContent_Declare(
            nlohmann_json
            GIT_REPOSITORY https://github.com/nlohmann/json.git
            GIT_TAG v3.11.3
            GIT_SHALLOW TRUE

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

Copilot · 2026-04-14T17:34:08Z

Just as a heads up, I was blocked by some firewall rules while working on your feedback. Expand below for details.

github-actions · 2026-05-15T20:42:46Z

📊 MPMC Queue Benchmark Results (commit `af826b4`)

MpmcQueue End-to-End Latency (SPSC, 30000 items per queue size)

Queue Capacity	P50 (ns)	P95 (ns)	P99 (ns)	P99.9 (ns)
16	647	687	708	24000
64	2441	2561	2671	27207
256	6889	9945	30634	34420
1024	13151	13381	32466	32517
4096	18418	18830	36612	36633

ResponseQueue End-to-End Latency (SPSC, 30000 items per queue size)

Queue Capacity	P50 (ns)	P95 (ns)	P99 (ns)
16	1078	1810	92680
64	15820	22593	37831
256	38963	60162	84528
1024	130515	153958	182912
4096	514372	585956	586518

Throughput (400000 items per configuration)

Configuration	Queue Cap	Throughput (Mops/s)
SPSC (1P×1C)	256	33.81
MPSC (4P×1C)	1024	40.58
MPMC (4P×4C)	4096	54.81
MPMC (8P×4C)	4096	26.86

Stress / Correctness (100000 items each)

Test	Items	Result
SPSC (1P×1C)	100000	✅ PASS
MPSC (4P×1C)	100000	✅ PASS
MPMC (4P×4C)	100000	✅ PASS
MPMC (8P×8C)	100000	✅ PASS

Benchmarked on ubuntu-latest, Release build

Thecave3 · 2026-06-04T22:51:00Z

Once this is ready to be merged i will review it again. one minor note: if possible let's keep json at the 3.11 version. Feel free to set this as ready once we can review it

freebacca · 2026-06-10T18:48:16Z

OSRB has been approved. I have added dvilla in CONTRIBUTORS, and reverted the change of JSON to 3.11. The MR is now ready to be reviewed and merged.

github-actions · 2026-06-10T18:48:19Z

⏱️ Full-loop Latency Benchmark (commit `ac596bb`)

Full-loop latency benchmark (N=1060 after 50 warmup)

All values in microseconds (us). Transport: ZMQ over IPC, encoding: ASN.1 APER.

Phase	mean	p50	p99	max
1. Collect indication data	0	0	0	7
2. Create & encode indication	0	0	0	8
3. Deliver indication (RAN -> dApp)	83	78	150	518
4. Decode indication	0	0	0	8
5. Process data	0	0	0	0
6. Create & encode control	0	0	0	1
7. Deliver control (dApp -> RAN)	101	103	158	170
8. Decode & handle control	0	0	0	6
Total round-trip	187	174	299	634

Benchmarked on ubuntu-latest, Release build, ZMQ + IPC, ASN.1 APER.

github-actions · 2026-06-10T18:49:35Z

🔄 E2E dApp Integration Results (commit `ac596bb`)

posix/ipc

dApp exit: 0
Indications received: 7

posix/tcp

dApp exit: 0
Indications received: 7

zmq/ipc

dApp exit: 0
Indications received: 7

zmq/tcp

dApp exit: 0
Indications received: 7

example_simple_agent + example_simple_dapp on ubuntu-latest, Release build

github-actions · 2026-06-10T18:49:55Z

🔀 E2E Topologies — multi-dApp / multi-RAN (commit `ac596bb`)

zmq/ipc

✅ 1 RAN - 1 dApp: indications=5
- dapp peer=t11 ran=ran-solo sub=1 indications=5 seq=[0..4] dropped=0 (0%) age_ms(avg=0.4 max=1 @seq=0) hist[<=1:5 2-5:0 6-10:0 >10:0]
✅ 1 RAN - 2 dApps: dApp#1 ind=5 sub=1, dApp#2 ind=6 sub=2, RAN saw 2 dApps
- dapp1 peer=t12 ran=ran-shared sub=1 indications=5 seq=[0..4] dropped=0 (0%) age_ms(avg=0.6 max=1 @seq=2) hist[<=1:5 2-5:0 6-10:0 >10:0]
- dapp2 peer=t12 ran=ran-shared sub=2 indications=6 seq=[0..5] dropped=0 (0%) age_ms(avg=0.5 max=1 @seq=2) hist[<=1:6 2-5:0 6-10:0 >10:0]
✅ 2 RANs - 1 dApp: from ran-a ind=5, from ran-b ind=5
- dapp peer=t2a ran=ran-a sub=1 indications=5 seq=[0..4] dropped=0 (0%) age_ms(avg=0.4 max=1 @seq=2) hist[<=1:5 2-5:0 6-10:0 >10:0]
- dapp peer=t2b ran=ran-b sub=1 indications=5 seq=[0..4] dropped=0 (0%) age_ms(avg=0.4 max=1 @seq=0) hist[<=1:5 2-5:0 6-10:0 >10:0]

zmq/tcp

✅ 1 RAN - 1 dApp: indications=5
- dapp peer=default ran=ran-solo sub=1 indications=5 seq=[0..4] dropped=0 (0%) age_ms(avg=0.2 max=1 @seq=0) hist[<=1:5 2-5:0 6-10:0 >10:0]
✅ 1 RAN - 2 dApps: dApp#1 ind=5 sub=1, dApp#2 ind=6 sub=2, RAN saw 2 dApps
- dapp1 peer=default ran=ran-shared sub=1 indications=5 seq=[0..4] dropped=0 (0%) age_ms(avg=0.6 max=1 @seq=0) hist[<=1:5 2-5:0 6-10:0 >10:0]
- dapp2 peer=default ran=ran-shared sub=2 indications=6 seq=[0..5] dropped=0 (0%) age_ms(avg=0.5 max=1 @seq=0) hist[<=1:6 2-5:0 6-10:0 >10:0]
✅ 2 RANs - 1 dApp: from ran-a ind=5, from ran-b ind=5
- dapp peer=default ran=ran-a sub=1 indications=5 seq=[0..4] dropped=0 (0%) age_ms(avg=0.8 max=1 @seq=0) hist[<=1:5 2-5:0 6-10:0 >10:0]
- dapp peer=off100 ran=ran-b sub=1 indications=5 seq=[0..4] dropped=0 (0%) age_ms(avg=0 max=0 @seq=0) hist[<=1:5 2-5:0 6-10:0 >10:0]

example_simple_agent + example_simple_dapp on ubuntu-latest, Release. Indication age is report-only.

Thecave3

Minor comments, if possible can you run and paste the test for the json one or can you update the template include the json in the tests done by the CI?

This would ensure that changes are preserved over the updates

Thecave3 · 2026-06-11T14:23:35Z

Tests for json included in #27 , please update this branch to the latest version once #27 is merged.

Thecave3 · 2026-06-11T22:47:51Z

Encoder threads are all resolved and CI is green. Two items left before merge:

Blocking

Control IDs are dropped end-to-end (subscription_manager.hpp). SubscriptionRequest.control_identifier_list already exists on the wire (since First iteration for the library #4) but is never passed at the add_subscription() call site in e3_interface.cpp, so a dApp's requested control ids are silently discarded. Now that we honor per-dApp telemetry + periodicity, control should be symmetric: add control_ids to SubscriptionDetails, thread it through add_subscription() and the call site, and add a matching e3_agent_get_subscription_control_ids C-API getter.

Cleanup

Redundant fields in SubscriptionDetails — (dapp_id, ran_function_id) is already the composite map key (make_sub_key), so storing them again inside the value duplicates the key and can drift. Drop them; keep only the per-subscription metadata.

Resolved by author's answers (no code change)

e3_interface.cpp subscription-delete negative case: confirmed — remove_subscription_by_id fails when the id is absent or belongs to another dApp.
e3_agent.cpp null get_subscription_details: ternary returns 0 = use SM default periodicity, safe.
e3_agent.cpp empty-vector fallback: required by the ternary's common type; an empty std::vector does no heap allocation.

Thecave3 · 2026-06-12T21:10:55Z

Great work, thanks for this contribution. One final thing, please rebase and bump the version to 0.0.6. Afterwards, I will approve the PR and include it with squash and merge.

…ct unknown PDU types - Derive camelCase PDU type strings from pdu_type_to_string() instead of separate pdu_type_to_json_string() switch - Remove backward-compat fallbacks for legacy key names (pdu_type, message_id) - Return std::optional from string_to_pdu_type; fail decode on unrecognized types - Add Doxygen documentation on JsonE3Encoder class and nested_mode_ member - Add tests: nested format decode, encode mirrors nested format, PascalCase rejection

Drop nested_mode_ member. Encoder always merges payload fields at root. Decoder rejects messages containing a "data" wrapper with DECODE_FAILED. Update indication-message tests to use valid JSON in protocol_data so they exercise the json::parse() path the encoder now requires.

Add per-subscription control_ids alongside telemetry_ids, threaded through add_subscription() and exposed via get_subscription_control_ids() and the matching C API. SubscriptionRequest already carries control_identifier_list on the wire, so no encoder change. Also drop dapp_id/ran_function_id from SubscriptionDetails since they duplicate the composite map key and can drift.

freebacca · 2026-06-12T21:37:26Z

Great work, thanks for this contribution. One final thing, please rebase and bump the version to 0.0.6. Afterwards, I will approve the PR and include it with squash and merge.

Done and tests run with no issues.

Thecave3

Great work, thanks for this contribution.

freebacca requested review from Thecave3 and Copilot March 24, 2026 19:14

Copilot started reviewing on behalf of freebacca March 24, 2026 19:15 View session

Copilot AI reviewed Mar 24, 2026

View reviewed changes

Comment thread src/encoder/json_encoder.cpp

Comment thread src/encoder/json_encoder.cpp Outdated

Comment thread src/encoder/json_encoder.cpp Outdated

Comment thread src/encoder/json_encoder.cpp Outdated

Thecave3 closed this Mar 24, 2026

mychele reopened this Mar 27, 2026

Thecave3 requested changes Mar 29, 2026

View reviewed changes

Thecave3 reviewed Apr 14, 2026

View reviewed changes

Comment thread src/encoder/json_encoder.cpp

Copilot started work on behalf of Thecave3 April 14, 2026 17:28 View session

Copilot finished work on behalf of Thecave3 April 14, 2026 17:34

Copilot AI requested a review from Thecave3 April 14, 2026 17:34

Thecave3 removed their request for review April 14, 2026 17:37

Thecave3 force-pushed the json-format branch 2 times, most recently from b73fb51 to 4ec3cb0 Compare May 15, 2026 20:42

Thecave3 marked this pull request as draft June 4, 2026 22:51

freebacca marked this pull request as ready for review June 10, 2026 19:24