fix(neon): Allow extracting Json<Option<T>> when value is undefined

[kjvalencik]

In JavaScript, it's very common to have optional arguments, especially with options object. Consider the following:

function createClient(opts?: Partial<ClientOptions>);

In this case, a user can call createClient() without any options or they can call it with any subset of the keys from ClientOptions. If we want to model this in Neon, we can use the Json extractor.

#[dervie(Default, Deserialize)]
struct ClientOptions {
    host: Option<String>,
    /* ... */
}

#[neon::export]
fn create_client(Json(opts): Json<ClientOptions>) {}

This already works well, except for case where opts isn't provided at all. For that, we want to Option<T> wrap the options.

fn create_client(Json(opts): Json<Option<ClientOptions>>) {
    let opts = opts.unwrap_or_default();
}

This works if the user explicitly passes null for opts (since it is serialized as "null"), but fails if it's undefined (not passed at all) because of the behavior of JSON.stringify. This PR fixes it to match expectations.

[github-actions]

Bencher Report

Branch	kv/json-option
Testbed	ubuntu-latest

Click to view all benchmark results

Benchmark	Latency	Benchmark Result nanoseconds (ns) (Result Δ%)	Lower Boundary nanoseconds (ns) (Limit %)	Upper Boundary nanoseconds (ns) (Limit %)	Throughput	Benchmark Result operations / second (ops/s) (Result Δ%)	Lower Boundary operations / second (ops/s) (Limit %)	Upper Boundary operations / second (ops/s) (Limit %)
JsFunction::bind	📈 view plot 🚷 view threshold	209.48 ns (+3.04%) Baseline: 203.29 ns	182.96 ns (87.34%)	406.59 ns (51.52%)	📈 view plot 🚷 view threshold	4,753,296.93 ops/s (-3.27%) Baseline: 4,913,759.54 ops/s	0.00 ops/s (0.00%)	5,405,135.49 ops/s (87.94%)
JsFunction::call	📈 view plot 🚷 view threshold	219.68 ns (+1.80%) Baseline: 215.80 ns	194.22 ns (88.41%)	431.61 ns (50.90%)	📈 view plot 🚷 view threshold	4,550,537.08 ops/s (-1.51%) Baseline: 4,620,462.24 ops/s	0.00 ops/s (0.00%)	5,082,508.47 ops/s (89.53%)
JsFunction::call_with	📈 view plot 🚷 view threshold	211.44 ns (+4.10%) Baseline: 203.10 ns	182.79 ns (86.45%)	406.21 ns (52.05%)	📈 view plot 🚷 view threshold	4,726,878.67 ops/s (-3.64%) Baseline: 4,905,637.96 ops/s	0.00 ops/s (0.00%)	5,396,201.76 ops/s (87.60%)
auto-exported-noop	📈 view plot 🚷 view threshold	28.59 ns (-3.49%) Baseline: 29.62 ns	26.66 ns (93.25%)	59.25 ns (48.26%)	📈 view plot 🚷 view threshold	34,951,722.76 ops/s (+2.28%) Baseline: 34,170,940.04 ops/s	0.00 ops/s (0.00%)	37,588,034.05 ops/s (92.99%)
hello-world	📈 view plot 🚷 view threshold	58.39 ns (+0.15%) Baseline: 58.30 ns	52.47 ns (89.87%)	116.60 ns (50.07%)	📈 view plot 🚷 view threshold	16,752,208.51 ops/s (-2.44%) Baseline: 17,170,662.37 ops/s	0.00 ops/s (0.00%)	18,887,728.61 ops/s (88.69%)
manually-exported-noop	📈 view plot 🚷 view threshold	27.82 ns (-2.03%) Baseline: 28.39 ns	25.55 ns (91.86%)	56.79 ns (48.99%)	📈 view plot 🚷 view threshold	35,839,475.97 ops/s (+1.79%) Baseline: 35,210,196.62 ops/s	0.00 ops/s (0.00%)	38,731,216.28 ops/s (92.53%)

🐰 View full continuous benchmarking report in Bencher

[codecov]

Codecov Report

❌ Patch coverage is 60.00000% with 2 lines in your changes missing coverage. Please review.
✅ Project coverage is 83.16%. Comparing base (22098e4) to head (c53500b).
⚠️ Report is 2 commits behind head on main.

Files with missing lines	Patch %	Lines
crates/neon/src/types_impl/extract/json.rs	60.00%	0 Missing and 2 partials ⚠️

Additional details and impacted files

@@           Coverage Diff           @@
##             main    #1130   +/-   ##
=======================================
  Coverage   83.16%   83.16%           
=======================================
  Files          80       80           
  Lines        5743     5739    -4     
  Branches     5743     5739    -4     
=======================================
- Hits         4776     4773    -3     
  Misses        848      848           
+ Partials      119      118    -1     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.
• 📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.

[dherman]

This looks nice! It's subtle so I'm still not 100% confident about it, but I definitely know I want it.

I left a couple of questions and one suggestion for an additional test in my review.