Update utils and changelog

NEWS.md

@@ -1,3 +1,14 @@
+# ollamar 1.2.0
+
+- All functions calling API endpoints have `endpoint` parameter.
+- All functions calling API endpoints have `...` parameter to pass additional model options to the API.
+- All functions calling API endpoints have `host` parameter to specify the host URL. Default is `NULL`, which uses the default Ollama URL.
+- Add `req` as an output format for `generate()` and `chat()`.
+- Add new functions for calling APIs: `create()`, `show()`, `copy()`, `delete()`, `push()`, `embed()` (supercedes `embeddings()`), `ps()`.
+- Add helper functions to manipulate chat/conversation history for `chat()` function (or other APIs like OpenAI): `create_message()`, `append_message()`, `prepend_message()`, `delete_message()`, `insert_message()`.
+- Add `ohelp()` function to chat with models in real-time.
+- Add helper functions: `model_avail()`, `image_encode_base64()`, `check_option_valid()`, `check_options()`, `search_options()`, `validate_options()`
+
 # ollamar 1.1.1
 
 ## Bug fixes
@@ -11,13 +22,13 @@
 
 ## New features
 
-- Integrated R with Ollama to run language models locally on your own machine. 
-- Included `test_connection()` function to test connection to Ollama server.
-- Included `list_models()` function to list available models.
-- Included `pull()` function to pull a model from Ollama server.
-- Included `delete()` function to delete a model from Ollama server.
-- Included `chat()` function to chat with a model.
-- Included `generate()` function to generate text from a model.
-- Included `embeddings()` function to get embeddings from a model.
-- Included `resp_process()` function to process `httr2_response` objects.
+- Integrate R with Ollama to run language models locally on your own machine. 
+- Include `test_connection()` function to test connection to Ollama server.
+- Include `list_models()` function to list available models.
+- Include `pull()` function to pull a model from Ollama server.
+- Include `delete()` function to delete a model from Ollama server.
+- Include `chat()` function to chat with a model.
+- Include `generate()` function to generate text from a model.
+- Include `embeddings()` function to get embeddings from a model.
+- Include `resp_process()` function to process `httr2_response` objects.
 

R/utils.R

@@ -86,9 +86,13 @@ stream_handler <- function(x, env, endpoint) {
 #' resp_process(resp, "resp") # return input response object
 #' resp_process(resp, "text") # return text/character vector
 resp_process <- function(resp, output = c("df", "jsonlist", "raw", "resp", "text")) {
+
+    if (!inherits(resp, "httr2_response")) {
+        stop("Input must be a httr2 response object")
+    }
+
     if (is.null(resp) || resp$status_code != 200) {
-        warning("Cannot process response")
-        return(NULL)
+        stop("Cannot process response")
     }
 
     endpoints_to_skip <- c("api/delete", "api/embed", "api/embeddings", "api/create")

README.Rmd

@@ -289,7 +289,22 @@ messages <- delete_message(messages, 2)
 
 ### Parallel requests
 
-For the `generate()` and `chat()` endpoints/functions, you can make parallel requests with the `req_perform_parallel` function from the `httr2` library. You need to specify `output = 'req'` in the function so the functions return `httr2_request` objects instead of `httr2_response` objects.
+For the `generate()` and `chat()` endpoints/functions, you can specify `output = 'req'` in the function so the functions return `httr2_request` objects instead of `httr2_response` objects.
+
+```{r eval=FALSE}
+prompt <- "Tell me a 10-word story"
+req <- generate("llama3.1", prompt, output = "req")  # returns a httr2_request object
+# <httr2_request>
+# POST http://127.0.0.1:11434/api/generate
+# Headers:
+# • content_type: 'application/json'
+# • accept: 'application/json'
+# • user_agent: 'ollama-r/1.1.1 (aarch64-apple-darwin20) R/4.4.0'
+# Body: json encoded data
+
+```
+
+When you have multiple `httr2_request` objects in a list, you can make parallel requests with the `req_perform_parallel` function from the `httr2` library. See [`httr2` documentation](https://httr2.r-lib.org/reference/req_perform_parallel.html) for details.
 
 ```{r eval=FALSE}
 library(httr2)

README.md

@@ -328,11 +328,27 @@ messages <- delete_message(messages, 2)
 
 ### Parallel requests
 
-For the `generate()` and `chat()` endpoints/functions, you can make
+For the `generate()` and `chat()` endpoints/functions, you can specify
+`output = 'req'` in the function so the functions return `httr2_request`
+objects instead of `httr2_response` objects.
+
+``` r
+prompt <- "Tell me a 10-word story"
+req <- generate("llama3.1", prompt, output = "req")  # returns a httr2_request object
+# <httr2_request>
+# POST http://127.0.0.1:11434/api/generate
+# Headers:
+# • content_type: 'application/json'
+# • accept: 'application/json'
+# • user_agent: 'ollama-r/1.1.1 (aarch64-apple-darwin20) R/4.4.0'
+# Body: json encoded data
+```
+
+When you have multiple `httr2_request` objects in a list, you can make
 parallel requests with the `req_perform_parallel` function from the
-`httr2` library. You need to specify `output = 'req'` in the function so
-the functions return `httr2_request` objects instead of `httr2_response`
-objects.
+`httr2` library. See [`httr2`
+documentation](https://httr2.r-lib.org/reference/req_perform_parallel.html)
+for details.
 
 ``` r
 library(httr2)