feat(expr): enhance expr with JQ functionality

[karatakis]

Summary:

• Enable jq syntax inside the mustache of @expr directives and select operations.
• Report template execution (runtime) errors on response
• Optimized handling of mustache: if it is read-only we use the faster Mustache implementation, otherwise we use Jq

Broken changes:

• On the select argument of @http, ... directives we cannot prepend the .args selector. So the users have to manually add it if they use Jq features only.

From:

@http(..., select: {id: "{{.id | tostring }}")

Becomes

@http(..., select: {id: "{{.args.id | tostring }}")

Example:

schema
  @server(port: 8000) {
  query: Query
}

type Query {
  users: [User] @http(url: "http://jsonplaceholder.typicode.com/users")
}

type User {
  id: Int!
  posts: [Post] @http(url: "http://jsonplaceholder.typicode.com/users/{{.value.id}}/posts")
}

type Post {
  id: Int!
  title: String!
  body: String!
  bodyParts: BodyParts! @expr(body: "{{ .value.body | split(\" \") | {first: .[0], second: .[1]} }}")
}

type BodyParts {
  first: String!
  second: String!
}

Build & Testing:

• I ran cargo test successfully.
• I have run ./lint.sh --mode=fix to fix all linting issues raised by ./lint.sh --mode=check.

Checklist:

• I have added relevant unit & integration tests.
• I have updated the documentation accordingly.
• I have performed a self-review of my code.
• PR follows the naming convention of <type>(<optional scope>): <title>

[codecov]

Codecov Report

Attention: Patch coverage is 56.08365% with 462 lines in your changes missing coverage. Please review.

Project coverage is 85.65%. Comparing base (0b05b83) to head (ed65705).

Files with missing lines	Patch %	Lines
src/core/mustache/jq_value.rs	21.51%	332 Missing ⚠️
src/core/mustache/jq_transform.rs	77.51%	105 Missing ⚠️
src/core/path.rs	29.41%	12 Missing ⚠️
src/core/mustache/render_value.rs	91.93%	5 Missing ⚠️
src/core/mustache/eval.rs	57.14%	3 Missing ⚠️
src/core/http/request_template.rs	0.00%	2 Missing ⚠️
src/core/blueprint/dynamic_value.rs	87.50%	1 Missing ⚠️
src/core/ir/error.rs	50.00%	1 Missing ⚠️
src/core/mustache/parse.rs	97.29%	1 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #3191      +/-   ##
==========================================
- Coverage   86.78%   85.65%   -1.13%     
==========================================
  Files         281      284       +3     
  Lines       28443    29421     +978     
==========================================
+ Hits        24683    25202     +519     
- Misses       3760     4219     +459     

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

[karatakis]

Benchmarks: before/after JQ Template

TC_LOG_LEVEL=error cargo run --release start main.graphql

schema @server(port: 8000, hostname: "0.0.0.0") {
  query: Query
}

type Query {
  args(text: String!): String! @expr(body: "{{.args.text}}")
}

request = function()
    local headers = {}
    headers["Content-Type"] = "application/json"

    local body = [[
    {
        "query": "query { args(text: \"Hello, World!\") }"
    }
    ]]

    return wrk.format("POST", "/graphql", headers, body)
end

Before

$ wrk -t12 -c400 -d30s -s bench.lua http://127.0.0.1:8000/graphql
Running 30s test @ http://127.0.0.1:8000/graphql
  12 threads and 400 connections
  Thread Stats   Avg      Stdev     Max   +/- Stdev
    Latency     0.90ms  648.63us  10.81ms   75.83%
    Req/Sec    35.51k     3.33k   46.54k    67.11%
  12732725 requests in 30.04s, 1.67GB read
Requests/sec: 423842.17
Transfer/sec:     56.99MB

After

$ wrk -t12 -c400 -d30s -s bench.lua http://127.0.0.1:8000/graphql
Running 30s test @ http://127.0.0.1:8000/graphql
  12 threads and 400 connections
  Thread Stats   Avg      Stdev     Max   +/- Stdev
    Latency     0.88ms  633.52us   9.94ms   76.60%
    Req/Sec    36.78k     3.90k   48.32k    67.31%
  13190038 requests in 30.05s, 1.73GB read
Requests/sec: 439007.72
Transfer/sec:     59.03MB

[github-actions]

Bencher Report

Branch	feat/jq-template-hybrid
Testbed	benchmarking-runner

Click to view all benchmark results

Benchmark	Latency	Benchmark Result nanoseconds (ns) (Result Δ%)	Upper Boundary nanoseconds (ns) (Limit %)
from_json_bench	📈 view plot 🚷 view threshold	6,259,800.00 (+0.80%)	6,804,157.68 (92.00%)
group_by	📈 view plot 🚷 view threshold	512.06 (+5.06%)	541.12 (94.63%)
input/args.missing	📈 view plot 🚷 view threshold	22.21 (-8.03%)	29.10 (76.34%)
input/args.nested.existing	📈 view plot 🚷 view threshold	42.97 (-2.43%)	49.83 (86.23%)
input/args.nested.missing	📈 view plot 🚷 view threshold	37.19 (-1.20%)	43.34 (85.82%)
input/args.root	📈 view plot 🚷 view threshold	38.93 (-1.96%)	45.06 (86.41%)
input/headers.existing	📈 view plot 🚷 view threshold	32.92 (+4.07%)	35.18 (93.56%)
input/headers.missing	📈 view plot 🚷 view threshold	31.67 (+2.59%)	34.88 (90.80%)
input/value.missing	📈 view plot 🚷 view threshold	22.39 (-3.55%)	25.84 (86.67%)
input/value.nested.existing	📈 view plot 🚷 view threshold	42.26 (-1.11%)	47.68 (88.62%)
input/value.nested.missing	📈 view plot 🚷 view threshold	36.00 (-0.52%)	40.98 (87.85%)
input/value.root	📈 view plot 🚷 view threshold	38.74 (-1.24%)	44.21 (87.65%)
input/vars.existing	📈 view plot 🚷 view threshold	7.75 (-4.80%)	9.55 (81.24%)
input/vars.missing	📈 view plot 🚷 view threshold	8.38 (-19.60%)	13.46 (62.27%)
synth_nested	📈 view plot 🚷 view threshold	101,080.00 (-17.47%)	158,101.41 (63.93%)
synth_nested_borrow	📈 view plot 🚷 view threshold	37,887.00 (-44.67%)	110,752.73 (34.21%)
test_batched_body	📈 view plot 🚷 view threshold	2,002.90 (+5.79%)	2,069.55 (96.78%)
test_batched_body #2	📈 view plot 🚷 view threshold	1,562,700.00 (+2.75%)	1,663,623.46 (93.93%)
test_data_loader	📈 view plot 🚷 view threshold	395,050.00 (-0.85%)	432,878.85 (91.26%)
test_handle_request	📈 view plot 🚷 view threshold	128,040.00 (-1.56%)	143,351.25 (89.32%)
test_handle_request_jit	📈 view plot 🚷 view threshold	53,005.00 (-43.48%)	156,028.65 (33.97%)
test_http_execute_method	📈 view plot 🚷 view threshold	14,935.00 (+2.94%)	16,055.79 (93.02%)
with_mustache_expressions	📈 view plot 🚷 view threshold	1,029.70 (-6.45%)	1,242.38 (82.88%)
with_mustache_literal	📈 view plot 🚷 view threshold	616.42 (-12.21%)	803.79 (76.69%)

🐰 View full continuous benchmarking report in Bencher