tailcall feat(2652): use LLM to infer arg and field names

trafficstars

fixes: #2652 /claim #2652

Output

 INFO Suggestions for Address: [Address, Location, Place, Site, Point] - 1/8
 INFO Suggestions for Comment: [Comment, Post, Reply, Message, Feedback] - 2/8
 INFO Suggestions for Company: [Company, Business, Entity, Organization, Profile] - 3/8
 INFO Suggestions for Geo: [Location, Coordinate, Point, Position, Geo] - 4/8
 INFO Suggestions for Photo: [Photo, Image, Picture, Media, Asset] - 5/8
 INFO Suggestions for Post: [Post, Article, Entry, Content, Item] - 6/8
 INFO Suggestions for Todo: [Task, Todo, Item, Entry, Activity] - 7/8
 INFO Suggestions for User: [Contact, User, Profile, Account, Person] - 8/8
 INFO Suggestions for todo's argument p1: [id, index, count, limit, offset] - 1/6
 INFO Suggestions for post's argument p1: [postId, postCount, postLimit, postOffset, postNumber] - 2/6
 INFO Suggestions for user's argument p1: [userId, userCount, id, limit, page] - 3/6
 INFO Suggestions for postComments's argument postId: [postId, commentId, post, comment, id] - 4/6
 INFO Suggestions for comment's argument p1: [commentId, commentCount, postId, replyId, threadId] - 5/6
 INFO Suggestions for photo's argument p1: [photoId, photoIndex, photoNumber, photoCount, photoOrder] - 6/6

type Query {
  comment(commentId: Int!): Reply @http(path: "/comments/{{.args.commentId}}")
  comments: [Reply] @http(path: "/comments")
  photo(photoId: Int!): Image @http(path: "/photos/{{.args.photoId}}")
  photos: [Image] @http(path: "/photos")
  post(postId: Int!): Article @http(path: "/posts/{{.args.postId}}")
  postComments(commentId: Int): [Reply] @http(path: "/comments", query: [{key: "commentId", value: "{{.args.commentId}}"}])
  posts: [Article] @http(path: "/posts")
  todo(id: Int!): Task @http(path: "/todos/{{.args.id}}")
  todos: [Task] @http(path: "/todos")
  user(userId: Int!): Contact @http(path: "/users/{{.args.userId}}")
  users: [Contact] @http(path: "/users")
}

Aug 14 '24 05:08 beelchester

Codecov Report

Attention: Patch coverage is 28.72928% with 258 lines in your changes missing coverage. Please review.

Project coverage is 86.48%. Comparing base (38ae551) to head (72b4d4c). Report is 121 commits behind head on main.

Files with missing lines	Patch %	Lines
src/cli/llm/infer_arg_name.rs	0.00%	106 Missing :warning:
src/cli/llm/infer_field_name.rs	0.00%	103 Missing :warning:
src/cli/generator/generator.rs	0.00%	22 Missing :warning:
src/core/config/transformer/rename_fields.rs	0.00%	20 Missing :warning:
src/core/config/transformer/rename_args.rs	92.94%	6 Missing :warning:
src/cli/llm/infer_type_name.rs	0.00%	1 Missing :warning:

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #2687      +/-   ##
==========================================
- Coverage   87.30%   86.48%   -0.83%     
==========================================
  Files         254      258       +4     
  Lines       25311    25644     +333     
==========================================
+ Hits        22099    22177      +78     
- Misses       3212     3467     +255

:umbrella: View full report in Codecov by Sentry.
:loudspeaker: Have feedback on the report? Share it here.

Aug 16 '24 08:08 codecov[bot]

Bencher Report

Branch	2687/merge
Testbed	benchmarking-runner

🚨 2 ALERTS: Threshold Boundary Limits exceeded!

Benchmark	Measure Units	View	Benchmark Result (Result Δ%)	Lower Boundary (Limit %)	Upper Boundary (Limit %)
from_json_bench	Latency nanoseconds (ns)	📈 plot 🚨 alert 🚷 threshold	6,318,300.00 (+2.24%)		6,312,461.49 (100.09%)
input/headers.missing	Latency nanoseconds (ns)	📈 plot 🚨 alert 🚷 threshold	33.64 (+9.77%)		33.48 (100.50%)

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 alert 🚷 view threshold	6,318,300.00 (+2.24%)	6,312,461.49 (100.09%)
group_by	📈 view plot 🚷 view threshold	497.10 (+1.44%)	513.97 (96.72%)
input/args.missing	📈 view plot 🚷 view threshold	25.64 (+0.00%)	27.74 (92.42%)
input/args.nested.existing	📈 view plot 🚷 view threshold	45.94 (+0.99%)	48.32 (95.07%)
input/args.nested.missing	📈 view plot 🚷 view threshold	39.22 (+1.34%)	40.87 (95.94%)
input/args.root	📈 view plot 🚷 view threshold	41.28 (-0.20%)	45.03 (91.68%)
input/headers.existing	📈 view plot 🚷 view threshold	33.65 (+6.60%)	34.08 (98.73%)
input/headers.missing	📈 view plot 🚨 view alert 🚷 view threshold	33.64 (+9.77%)	33.48 (100.50%)
input/value.missing	📈 view plot 🚷 view threshold	23.60 (+0.90%)	24.68 (95.63%)
input/value.nested.existing	📈 view plot 🚷 view threshold	43.36 (+0.60%)	45.76 (94.76%)
input/value.nested.missing	📈 view plot 🚷 view threshold	37.05 (+1.17%)	38.54 (96.12%)
input/value.root	📈 view plot 🚷 view threshold	41.15 (+6.26%)	41.69 (98.71%)
input/vars.existing	📈 view plot 🚷 view threshold	7.82 (-5.86%)	9.55 (81.87%)
input/vars.missing	📈 view plot 🚷 view threshold	11.71 (+29.55%)	11.88 (98.53%)
synth_nested	📈 view plot 🚷 view threshold	136,190.00 (+21.85%)	173,896.59 (78.32%)
synth_nested_borrow	📈 view plot 🚷 view threshold	80,633.00 (+55.89%)	103,227.60 (78.11%)
test_batched_body	📈 view plot 🚷 view threshold	1,880.90 (-3.10%)	2,077.97 (90.52%)
test_batched_body #2	📈 view plot 🚷 view threshold	1,494,400.00 (-5.50%)	1,728,379.54 (86.46%)
test_data_loader	📈 view plot 🚷 view threshold	399,810.00 (-2.14%)	457,471.29 (87.40%)
test_handle_request	📈 view plot 🚷 view threshold	131,380.00 (-5.62%)	149,962.52 (87.61%)
test_handle_request_jit	📈 view plot 🚷 view threshold	97,663.00 (-12.13%)	149,554.78 (65.30%)
test_http_execute_method	📈 view plot 🚷 view threshold	14,234.00 (-0.88%)	14,885.93 (95.62%)
with_mustache_expressions	📈 view plot 🚷 view threshold	1,138.30 (+2.52%)	1,178.17 (96.62%)
with_mustache_literal	📈 view plot 🚷 view threshold	713.65 (+0.32%)	763.06 (93.52%)

🐰 View full continuous benchmarking report in Bencher