Merge #4846

4846: Add OpenAI tests r=dureuill a=dureuill

# Pull Request

## Related issue
Part of fixing #4757 

## What does this PR do?
- OpenAI embedder: don't pass apiKey when it is empty (slightly improves error messages)
- rest embedder and rest-based embedders: specialize the authorization denied error message depending on the configuration source
- fix existing tests
- Adds assets containing prerecorded texts to embed and the embeddings obtained from OpenAI
- Adds an asset containing a tokenized long document and the embedding obtained from OpenAI for this token
- Uses the wiremock crate to mock the OpenAI API: parse the openai request, lookup the response in assets, craft an openai response


Co-authored-by: Louis Dureuil <louis@meilisearch.com>

milli/src/vector/error.rs

@@ -62,8 +62,18 @@ pub enum EmbedErrorKind {
     RestResponseDeserialization(std::io::Error),
     #[error("expected a response containing {0} embeddings, got only {1}")]
     RestResponseEmbeddingCount(usize, usize),
-    #[error("could not authenticate against embedding server{}", option_info(.0.as_deref(), "server replied with "))]
-    RestUnauthorized(Option<String>),
+    #[error("could not authenticate against {embedding} server{server_reply}{hint}", embedding=match *.1 {
+        ConfigurationSource::User => "embedding",
+        ConfigurationSource::OpenAi => "OpenAI",
+        ConfigurationSource::Ollama => "ollama"
+    },
+    server_reply=option_info(.0.as_deref(), "server replied with "),
+    hint=match *.1 {
+        ConfigurationSource::User => "\n  - Hint: Check the `apiKey` parameter in the embedder configuration",
+        ConfigurationSource::OpenAi => "\n  - Hint: Check the `apiKey` parameter in the embedder configuration, and the `MEILI_OPENAI_API_KEY` and `OPENAI_API_KEY` environment variables",
+        ConfigurationSource::Ollama => "\n  - Hint: Check the `apiKey` parameter in the embedder configuration"
+    })]
+    RestUnauthorized(Option<String>, ConfigurationSource),
     #[error("sent too many requests to embedding server{}", option_info(.0.as_deref(), "server replied with "))]
     RestTooManyRequests(Option<String>),
     #[error("sent a bad request to embedding server{}{}",
@@ -136,8 +146,14 @@ impl EmbedError {
         }
     }
 
-    pub(crate) fn rest_unauthorized(error_response: Option<String>) -> EmbedError {
-        Self { kind: EmbedErrorKind::RestUnauthorized(error_response), fault: FaultSource::User }
+    pub(crate) fn rest_unauthorized(
+        error_response: Option<String>,
+        configuration_source: ConfigurationSource,
+    ) -> EmbedError {
+        Self {
+            kind: EmbedErrorKind::RestUnauthorized(error_response, configuration_source),
+            fault: FaultSource::User,
+        }
     }
 
     pub(crate) fn rest_too_many_requests(error_response: Option<String>) -> EmbedError {

milli/src/vector/openai.rs

@@ -183,7 +183,7 @@ impl Embedder {
 
         let rest_embedder = RestEmbedder::new(
             RestEmbedderOptions {
-                api_key: Some(api_key.clone()),
+                api_key: (!api_key.is_empty()).then(|| api_key.clone()),
                 distribution: None,
                 dimensions: Some(options.dimensions()),
                 url,

milli/src/vector/rest.rs

@@ -275,7 +275,10 @@ fn check_response(
         Err(ureq::Error::Status(code, response)) => {
             let error_response: Option<String> = response.into_string().ok();
             Err(match code {
-                401 => Retry::give_up(EmbedError::rest_unauthorized(error_response)),
+                401 => Retry::give_up(EmbedError::rest_unauthorized(
+                    error_response,
+                    configuration_source,
+                )),
                 429 => Retry::rate_limited(EmbedError::rest_too_many_requests(error_response)),
                 400 => Retry::give_up(EmbedError::rest_bad_request(
                     error_response,