search

m1guelpf / auto-commit

A CLI tool that automatically writes commit messages for you.
MIT License
1.2k stars 54 forks source link

`invalid_request_error` #7

Closed 0x4007 closed 1 year ago

0x4007 commented 2 years ago 
m1:ubiquity-dollar nv$ auto-commit
Loading Data...
🙈  Analyzing Codebase...thread 'main' panicked at 'Couldn't complete prompt.: Api(ErrorMessage { message: "That model does not exist", error_type: "invalid_request_error" })', src/main.rs:137:10
note: run with `RUST_BACKTRACE=1` environment variable to display a backtrace
🙉  Analyzing Codebase...m1:ubiquity-dollar nv$
m1:ubiquity-dollar nv$ auto-commit --verbose
There are no staged files to commit.
Try running `git add` to stage some files.
Loading Data...
🙈  Analyzing Codebase...[2022-11-01T08:55:12Z DEBUG openai_api] Request: Request { method: Post, url: Url { scheme: "https", cannot_be_a_base: false, username: "", password: None, host: Some(Domain("api.openai.com")), port: None, path: "/v1/engines/code-davinci-002/completions", query: None, fragment: None }, headers: {"content-type": "application/json"}, version: None, body: Body { reader: "<hidden>", length: Some(2686), bytes_read: 0 }, local_addr: None, peer_addr: None, ext: Extensions, trailers_sender: Some(Sender { .. }), trailers_receiver: Some(Receiver { .. }), has_trailers: false }
[2022-11-01T08:55:12Z DEBUG hyper::client::connect::dns] resolving host="api.openai.com"
[2022-11-01T08:55:12Z DEBUG hyper::client::connect::http] connecting to 52.152.96.252:443
[2022-11-01T08:55:12Z DEBUG hyper::client::connect::http] connected to 52.152.96.252:443
🙉  Analyzing Codebase...[2022-11-01T08:55:13Z DEBUG hyper::proto::h1::io] flushed 209 bytes
[2022-11-01T08:55:13Z DEBUG hyper::proto::h1::io] flushed 2686 bytes
[2022-11-01T08:55:13Z DEBUG hyper::proto::h1::io] read 439 bytes
[2022-11-01T08:55:13Z DEBUG hyper::proto::h1::io] parsed 7 headers
[2022-11-01T08:55:13Z DEBUG hyper::proto::h1::conn] incoming body is content-length (158 bytes)
[2022-11-01T08:55:13Z DEBUG hyper::proto::h1::conn] incoming body completed
[2022-11-01T08:55:13Z DEBUG hyper::client::pool] pooling idle connection for ("https", api.openai.com)
[2022-11-01T08:55:13Z DEBUG openai_api] Response: Response { response: Response { status: NotFound, headers: {"connection": "keep-alive", "strict-transport-security": "max-age=15724800; includeSubDomains", "content-length": "158", "date": "Tue, 01 Nov 2022 08:55:13 GMT", "content-type": "application/json; charset=utf-8", "vary": "Origin", "x-request-id": "7dd1ef5d977c0dae864d89fbcbeeaa37"}, version: Some(Http1_1), has_trailers: false, trailers_sender: Some(Sender { .. }), trailers_receiver: Some(Receiver { .. }), upgrade_sender: Some(Sender { .. }), upgrade_receiver: Some(Receiver { .. }), has_upgrade: false, body: Body { reader: "<hidden>", length: Some(158), bytes_read: 0 }, ext: Extensions, local_addr: None, peer_addr: None } }
thread 'main' panicked at 'Couldn't complete prompt.: Api(ErrorMessage { message: "That model does not exist", error_type: "invalid_request_error" })', src/main.rs:137:10
note: run with `RUST_BACKTRACE=1` environment variable to display a backtrace
🙊  Analyzing Codebase...m1:ubiquity-dollar nv$

With RUST_BACKTRACE=full

m1:ubiquity-dollar nv$ auto-commit
Loading Data...
🌎  Analyzing Codebase...thread 'main' panicked at 'Couldn't complete prompt.: Api(ErrorMessage { message: "That model does not exist", error_type: "invalid_request_error" })', src/main.rs:137:10
stack backtrace:
   0:        0x100803cb8 - <std::sys_common::backtrace::_print::DisplayBacktrace as core::fmt::Display>::fmt::h1543c132bc4e188c
   1:        0x10082077c - core::fmt::write::hda8e8eb84b49cbfc
   2:        0x1007fe498 - std::io::Write::write_fmt::hb84c8996aec7120c
   3:        0x1008054c4 - std::panicking::default_hook::{{closure}}::hdf06011cb093de6a
   4:        0x100805228 - std::panicking::default_hook::hd7ceb942fff7b170
   5:        0x10080595c - std::panicking::rust_panic_with_hook::h053d4067a63a6fcb
   6:        0x100805890 - std::panicking::begin_panic_handler::{{closure}}::hea9e6c546a23e8ff
   7:        0x100804194 - std::sys_common::backtrace::__rust_end_short_backtrace::hd64e012cf32134c6
   8:        0x1008055e8 - _rust_begin_unwind
   9:        0x10083422c - core::panicking::panic_fmt::hbfde5533e1c0592e
  10:        0x100834318 - core::result::unwrap_failed::h68832e989a8867c1
  11:        0x1005f5544 - auto_commit::main::{{closure}}::hde9ffac744f15d7c
  12:        0x1005d8a78 - std::thread::local::LocalKey<T>::with::h8299edffc48b47fb
  13:        0x1005e1498 - tokio::runtime::enter::Enter::block_on::h8cd42799fe53fdaa
  14:        0x1005eb9d4 - tokio::runtime::context::enter::hd622a04884cced71
  15:        0x1005dc83c - tokio::runtime::handle::Handle::enter::hb597e6521843e9f1
  16:        0x1005e26a4 - auto_commit::main::h54076be9b6549311
  17:        0x1005fd8c8 - std::sys_common::backtrace::__rust_begin_short_backtrace::h27a8d6ce065a0fc5
  18:        0x1005e84cc - std::rt::lang_start::{{closure}}::h901a2890ae649abe
  19:        0x1007f976c - std::rt::lang_start_internal::hef2161f9571a51d7
  20:        0x1005e2768 - _main
🌏  Analyzing Codebase...m1:ubiquity-dollar nv$
0x4007 commented 2 years ago

Looks like only gpt-3 is public and the others require beta invitations. I think this could also be considered to add to the README.

I've been trying to access the engine named code-davinci-002 which is a private beta version engine. So without access it's not possible to access the engine. It seems only the GPT-3 models are of public usage. We need to need to join the OpenAI Codex Private Beta Waitlist in order to access Codex models through API.

https://stackoverflow.com/questions/71597547/openai-error-invalidrequesterror-engine-not-found/72107883#comment127573953_72107883

m1guelpf commented 2 years ago

This is already on the README! See https://github.com/m1guelpf/auto-commit/blob/64cc176f0d1b6676c7f88f06636259624475cf53/README.md?plain=1#L25

m1guelpf commented 2 years ago

Tho I agree a better error message could be presented

0x4007 commented 2 years ago

This is already on the README! See

https://github.com/m1guelpf/auto-commit/blob/64cc176f0d1b6676c7f88f06636259624475cf53/README.md?plain=1#L25

I didn't realize that they had one public model and other private beta models. I assumed that I signed up for the beta some time ago, which I assumed gave me access at all to any OpenAI APIs 😰

m1guelpf commented 1 year ago

Should be fixed in the latest release! 😄