ollama-ruby 0.8.0 → 0.9.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 7a6a94882e3f76e84fbceadd80ec844f9a93febd18ed294b01f7874867c53e68
-  data.tar.gz: 47cda3eef10438ff1cd136ffcc4469d2275bc2e26cdab985c7a001e2d0178b7f
+  metadata.gz: d0b3ea2bd67e3f4fa2ca4da88a25cb869fc81e35578f1e7fd97ba43e84fc2278
+  data.tar.gz: 626c534bd2fd0580ab0520e8b70587489edbc963e179be0f800ee63235690b33
 SHA512:
-  metadata.gz: 5c42f47be513e41465693c47dee68447ac42a967f7ecaaf88ea51df6ea30eb9beb61b25595fc145909a0c9b8838d22756f47e71bc6acdd33bb47e4876545ccdc
-  data.tar.gz: b4ac6bf59882a23774fbfb3503607c1a22d200037b31613a8f9d565ebf174b99d66259cc62d99f0635e7ac99fd07cc51aec735dd7dbb31259dc767e403eaa5d7
+  metadata.gz: 149986d8144b1bc2c20771e245ead20f6e6bc07b46dbe26f8f355e0bc61eca146cc085a920dcbe3e3e08de65632693a35b6995c9dffef288aa929dd6b034de16
+  data.tar.gz: 4214aa2911db561635011b8ea200074d15c74b93bf7abb3c8e77af9c40ac5d8994e908e2ade8e5f288c2919678e03ea5d84d07684a63ba1f8039de52b6091a86

data/CHANGES.md

@@ -1,5 +1,23 @@
 # Changes
 
+## 2024-10-18 v0.9.0
+
+* Add document policy chooser and modify embedding/importing/summarizing
+  behavior:
+  + Add `/document_policy` command to choose a scan policy for document
+    references
+  + Modify `embed_source`, `import_source`, and `summarize_source` methods to
+    use the chosen document policy
+  + Update `choose_model` method to set `$document_policy` based on
+    configuration or chat command
+* Fix regular expression in `ollama_chat` script:
+  + Updated regular expression for `/info` to `/^/info$`
+* Improve ask prompt to ask about clearing messages and collection.
+* Update specs to use `expect` instead of `allow`
+* Fix library homepage URL in README.md
+* Refactor Markdown handler to remove unnecessary puts statement
+* Reorder chapters in README.md a bit
+
 ## 2024-10-07 v0.8.0
 
 * **Refactor source handling in Ollama chat**:

data/README.md

@@ -26,161 +26,23 @@ gem 'ollama-ruby'
 
 to your Gemfile and run `bundle install` in your terminal.
 
-## Executables
-
-### ollama\_chat
-
-This a chat client, that can be used to connect to an ollama server and enter a
-chat converstation with a LLM. It can be called with the following arguments:
-
-```
-Usage: ollama_chat [OPTIONS]
-
-  -f CONFIG      config file to read
-  -u URL         the ollama base url, OLLAMA_URL
-  -m MODEL       the ollama model to chat with, OLLAMA_CHAT_MODEL
-  -s SYSTEM      the system prompt to use as a file, OLLAMA_CHAT_SYSTEM
-  -c CHAT        a saved chat conversation to load
-  -C COLLECTION  name of the collection used in this conversation
-  -D DOCUMENT    load document and add to embeddings collection (multiple)
-  -M             use (empty) MemoryCache for this chat session
-  -E             disable embeddings for this chat session
-  -V             display the current version number and quit
-  -h             this help
-```
-
-The base URL can be either set by the environment variable `OLLAMA_URL` or it
-is derived from the environment variable `OLLAMA_HOST`. The default model to
-connect can be configured in the environment variable `OLLAMA_MODEL`.
-
-The YAML config file in `$XDG_CONFIG_HOME/ollama_chat/config.yml`, that you can
-use for more complex settings, it looks like this:
-
-```
----
-url: <%= ENV['OLLAMA_URL'] || 'http://%s' % ENV.fetch('OLLAMA_HOST') %>
-model:
-  name: <%= ENV.fetch('OLLAMA_CHAT_MODEL', 'llama3.1') %>
-  options:
-    num_ctx: 8192
-system: <%= ENV.fetch('OLLAMA_CHAT_SYSTEM', 'null') %>
-voice: Samantha
-markdown: true
-embedding:
-  enabled: true
-  model:
-    name: mxbai-embed-large
-    options: {}
-  collection: <%= ENV.fetch('OLLAMA_CHAT_COLLECTION', 'ollama_chat') %>
-  found_texts_size: 4096
-  splitter:
-    name: RecursiveCharacter
-    chunk_size: 1024
-cache: Ollama::Documents::RedisCache
-redis:
-  url: <%= ENV.fetch('REDIS_URL', 'null') %>
-debug: <%= ENV['OLLAMA_CHAT_DEBUG'].to_i == 1 ? true : false %>
-```
-
-If you want to store embeddings persistently, set an environment variable
-`REDIS_URL` or update the `redis.url` setting in your `config.yml` file to
-connect to a Redis server. Without this setup, embeddings will only be stored
-in process memory, which is less durable.
-
-Some settings can be passed as arguments as well, e. g. if you want to choose a
-specific system prompt:
-
-```
-$ ollama_chat -s sherlock.txt
-Model with architecture llama found.
-Connecting to llama3.1@http://ollama.local.net:11434 now…
-Configured system prompt is:
-You are Sherlock Holmes and the user is your new client, Dr. Watson is also in
-the room. You will talk and act in the typical manner of Sherlock Holmes do and
-try to solve the user's case using logic and deduction.
-
-Type /help to display the chat help.
-📨 user:
-Good morning.
-📨 assistant:
-Ah, good morning, my dear fellow! It is a pleasure to make your acquaintance. I
-am Sherlock Holmes, the renowned detective, and this is my trusty sidekick, Dr.
-Watson. Please, have a seat and tell us about the nature of your visit. What
-seems to be the problem that has brought you to our humble abode at 221B Baker
-Street?
-
-(Watson nods in encouragement as he takes notes)
-
-Now, pray tell, what is it that puzzles you, my dear client? A missing item,
-perhaps? Or a mysterious occurrence that requires clarification? The game, as
-they say, is afoot!
-```
-
-This example shows how an image like this can be sent to a vision model for
-analysis:
-
-![cat](spec/assets/kitten.jpg)
-
-```
-$ ollama_chat -m llava-llama3
-Model with architecture llama found.
-Connecting to llava-llama3@http://localhost:11434 now…
-Type /help to display the chat help.
-📸 user> What's on this image? ./spec/assets/kitten.jpg
-📨 assistant:
-The image captures a moment of tranquility featuring a young cat. The cat,
-adorned with gray and white fur marked by black stripes on its face and legs,
-is the central figure in this scene. Its eyes, a striking shade of blue, are
-wide open and directed towards the camera, giving an impression of curiosity or
-alertness.
-
-The cat is comfortably nestled on a red blanket, which contrasts vividly with
-its fur. The blanket, soft and inviting, provides a sense of warmth to the
-image. In the background, partially obscured by the cat's head, is another
-blanket of similar red hue. The repetition of the color adds a sense of harmony
-to the composition.
-
-The cat's position on the right side of the photo creates an interesting
-asymmetry with the camera lens, which occupies the left side of the frame. This
-visual balance enhances the overall composition of the image.
+## Usage
 
-There are no discernible texts or other objects in the image. The focus is
-solely on the cat and its immediate surroundings. The image does not provide
-any information about the location or setting beyond what has been described.
-The simplicity of the scene allows the viewer to concentrate on the main
-subject - the young, blue-eyed cat.
-```
+In your own software the library can be used as shown in this example:
 
-The following commands can be given inside the chat, if prefixed by a `/`:
+```ruby
+require "ollama"
+include Ollama
 
-```
-/copy                           to copy last response to clipboard
-/paste                          to paste content
-/markdown                       toggle markdown output
-/stream                         toggle stream output
-/location                       toggle location submission
-/voice( change)                 toggle voice output or change the voice
-/list [n]                       list the last n / all conversation exchanges
-/clear                          clear the whole conversation
-/clobber                        clear the conversation and collection
-/pop [n]                        pop the last n exchanges, defaults to 1
-/model                          change the model
-/system                         change system prompt (clears conversation)
-/regenerate                     the last answer message
-/collection( clear|change)      change (default) collection or clear
-/info                           show information for current session
-/import source                  import the source's content
-/summarize [n] source           summarize the source's content in n words
-/embedding                      toggle embedding paused or not
-/embed source                   embed the source's content
-/web [n] query                  query web search & return n or 1 results
-/save filename                  store conversation messages
-/load filename                  load conversation messages
-/quit                           to quit
-/help                           to view this help
+ollama = Client.new(base_url: 'http://localhost:11434')
+messages = Message.new(role: 'user', content: 'Why is the sky blue?')
+ollama.chat(model: 'llama3.1', stream: true, messages:, &Print) # or
+print ollama.chat(model: 'llama3.1', stream: true, messages:).lazy.map { |response|
+  response.message.content
+}
 ```
 
-### ollama\_console
+## Try out things in ollama\_console
 
 This is an interactive console, that can be used to try the different commands
 provided by an `Ollama::Client` instance. For example this command generate a
@@ -197,21 +59,6 @@ Commands: chat,copy,create,delete,embeddings,generate,help,ps,pull,push,show,tag
 > In a small village nestled between two great palm trees 🌳, there lived a
 > brave adventurer named Alex 👦. […]
 
-## Usage
-
-In your own software the library can be used as shown in this example:
-
-```ruby
-require "ollama"
-include Ollama
-
-ollama = Client.new(base_url: 'http://localhost:11434')
-messages = Message.new(role: 'user', content: 'Why is the sky blue?')
-ollama.chat(model: 'llama3.1', stream: true, messages:, &Print) # or
-print ollama.chat(model: 'llama3.1', stream: true, messages:).lazy.map { |response|
-	response.message.content
-}
-```
 
 ## API
 
@@ -463,11 +310,166 @@ If `Ollama::Errors::TimeoutError` is raised, it might help to increase the
 
 For more generic errors an `Ollama::Errors::Error` is raised.
 
+## Other executables
+
+### ollama\_chat
+
+This a chat client, that can be used to connect to an ollama server and enter a
+chat converstation with a LLM. It can be called with the following arguments:
+
+```
+Usage: ollama_chat [OPTIONS]
+
+  -f CONFIG      config file to read
+  -u URL         the ollama base url, OLLAMA_URL
+  -m MODEL       the ollama model to chat with, OLLAMA_CHAT_MODEL
+  -s SYSTEM      the system prompt to use as a file, OLLAMA_CHAT_SYSTEM
+  -c CHAT        a saved chat conversation to load
+  -C COLLECTION  name of the collection used in this conversation
+  -D DOCUMENT    load document and add to embeddings collection (multiple)
+  -M             use (empty) MemoryCache for this chat session
+  -E             disable embeddings for this chat session
+  -V             display the current version number and quit
+  -h             this help
+```
+
+The base URL can be either set by the environment variable `OLLAMA_URL` or it
+is derived from the environment variable `OLLAMA_HOST`. The default model to
+connect can be configured in the environment variable `OLLAMA_MODEL`.
+
+The YAML config file in `$XDG_CONFIG_HOME/ollama_chat/config.yml`, that you can
+use for more complex settings, it looks like this:
+
+```
+---
+url: <%= ENV['OLLAMA_URL'] || 'http://%s' % ENV.fetch('OLLAMA_HOST') %>
+model:
+  name: <%= ENV.fetch('OLLAMA_CHAT_MODEL', 'llama3.1') %>
+  options:
+    num_ctx: 8192
+system: <%= ENV.fetch('OLLAMA_CHAT_SYSTEM', 'null') %>
+voice: Samantha
+markdown: true
+embedding:
+  enabled: true
+  model:
+    name: mxbai-embed-large
+    options: {}
+  collection: <%= ENV.fetch('OLLAMA_CHAT_COLLECTION', 'ollama_chat') %>
+  found_texts_size: 4096
+  splitter:
+    name: RecursiveCharacter
+    chunk_size: 1024
+cache: Ollama::Documents::RedisCache
+redis:
+  url: <%= ENV.fetch('REDIS_URL', 'null') %>
+debug: <%= ENV['OLLAMA_CHAT_DEBUG'].to_i == 1 ? true : false %>
+```
+
+If you want to store embeddings persistently, set an environment variable
+`REDIS_URL` or update the `redis.url` setting in your `config.yml` file to
+connect to a Redis server. Without this setup, embeddings will only be stored
+in process memory, which is less durable.
+
+Some settings can be passed as arguments as well, e. g. if you want to choose a
+specific system prompt:
+
+```
+$ ollama_chat -s sherlock.txt
+Model with architecture llama found.
+Connecting to llama3.1@http://ollama.local.net:11434 now…
+Configured system prompt is:
+You are Sherlock Holmes and the user is your new client, Dr. Watson is also in
+the room. You will talk and act in the typical manner of Sherlock Holmes do and
+try to solve the user's case using logic and deduction.
+
+Type /help to display the chat help.
+📨 user:
+Good morning.
+📨 assistant:
+Ah, good morning, my dear fellow! It is a pleasure to make your acquaintance. I
+am Sherlock Holmes, the renowned detective, and this is my trusty sidekick, Dr.
+Watson. Please, have a seat and tell us about the nature of your visit. What
+seems to be the problem that has brought you to our humble abode at 221B Baker
+Street?
+
+(Watson nods in encouragement as he takes notes)
+
+Now, pray tell, what is it that puzzles you, my dear client? A missing item,
+perhaps? Or a mysterious occurrence that requires clarification? The game, as
+they say, is afoot!
+```
+
+This example shows how an image like this can be sent to a vision model for
+analysis:
+
+![cat](spec/assets/kitten.jpg)
+
+```
+$ ollama_chat -m llava-llama3
+Model with architecture llama found.
+Connecting to llava-llama3@http://localhost:11434 now…
+Type /help to display the chat help.
+📸 user> What's on this image? ./spec/assets/kitten.jpg
+📨 assistant:
+The image captures a moment of tranquility featuring a young cat. The cat,
+adorned with gray and white fur marked by black stripes on its face and legs,
+is the central figure in this scene. Its eyes, a striking shade of blue, are
+wide open and directed towards the camera, giving an impression of curiosity or
+alertness.
+
+The cat is comfortably nestled on a red blanket, which contrasts vividly with
+its fur. The blanket, soft and inviting, provides a sense of warmth to the
+image. In the background, partially obscured by the cat's head, is another
+blanket of similar red hue. The repetition of the color adds a sense of harmony
+to the composition.
+
+The cat's position on the right side of the photo creates an interesting
+asymmetry with the camera lens, which occupies the left side of the frame. This
+visual balance enhances the overall composition of the image.
+
+There are no discernible texts or other objects in the image. The focus is
+solely on the cat and its immediate surroundings. The image does not provide
+any information about the location or setting beyond what has been described.
+The simplicity of the scene allows the viewer to concentrate on the main
+subject - the young, blue-eyed cat.
+```
+
+The following commands can be given inside the chat, if prefixed by a `/`:
+
+```
+/copy                           to copy last response to clipboard
+/paste                          to paste content
+/markdown                       toggle markdown output
+/stream                         toggle stream output
+/location                       toggle location submission
+/voice( change)                 toggle voice output or change the voice
+/list [n]                       list the last n / all conversation exchanges
+/clear                          clear the whole conversation
+/clobber                        clear the conversation and collection
+/pop [n]                        pop the last n exchanges, defaults to 1
+/model                          change the model
+/system                         change system prompt (clears conversation)
+/regenerate                     the last answer message
+/collection( clear|change)      change (default) collection or clear
+/info                           show information for current session
+/document_policy                pick a scan policy for document references
+/import source                  import the source's content
+/summarize [n] source           summarize the source's content in n words
+/embedding                      toggle embedding paused or not
+/embed source                   embed the source's content
+/web [n] query                  query web search & return n or 1 results
+/save filename                  store conversation messages
+/load filename                  load conversation messages
+/quit                           to quit
+/help                           to view this help
+```
+
 ## Download
 
 The homepage of this library is located at
 
-* https://github.com/flori/ollama
+* https://github.com/flori/ollama-ruby
 
 ## Author
 

data/bin/ollama_chat

@@ -52,6 +52,7 @@ class OllamaChatConfig
       list: <%= `say -v ? 2>/dev/null`.lines.map { _1[/^(.+?)\s+[a-z]{2}_[a-zA-Z0-9]{2,}/, 1] }.uniq.sort.to_s.force_encoding('ASCII-8BIT') %>
     markdown: true
     stream: true
+    document_policy: importing
     embedding:
       enabled: true
       model:
@@ -479,59 +480,6 @@ def parse_source(source_io)
   end
 end
 
-def embed_source(source_io, source, count: nil)
-  $embedding.on? or return parse_source(source_io)
-  m = "Embedding #{italic { source_io&.content_type }} document #{source.to_s.inspect}."
-  if count
-    puts '%u. %s' % [ count, m ]
-  else
-    puts m
-  end
-  text = parse_source(source_io) or return
-  text.downcase!
-  splitter_config = $config.embedding.splitter
-  inputs = nil
-  case splitter_config.name
-  when 'Character'
-    splitter = Ollama::Documents::Splitters::Character.new(
-      chunk_size: splitter_config.chunk_size,
-    )
-    inputs = splitter.split(text)
-  when 'RecursiveCharacter'
-    splitter = Ollama::Documents::Splitters::RecursiveCharacter.new(
-      chunk_size: splitter_config.chunk_size,
-    )
-    inputs = splitter.split(text)
-  when 'Semantic'
-    splitter = Ollama::Documents::Splitters::Semantic.new(
-      ollama:, model: $config.embedding.model.name,
-      chunk_size: splitter_config.chunk_size,
-    )
-    inputs = splitter.split(
-      text,
-      breakpoint: splitter_config.breakpoint.to_sym,
-      percentage: splitter_config.percentage?,
-      percentile: splitter_config.percentile?,
-    )
-    inputs = splitter.split(text)
-  end
-  inputs or return
-  source = source.to_s
-  if source.start_with?(?!)
-    source = Ollama::Utils::Width.truncate(
-      source[1..-1].gsub(/\W+/, ?_),
-      length: 10
-    )
-  end
-  $documents.add(inputs, source:, batch_size: $config.embedding.batch_size?)
-end
-
-def add_image(images, source_io, source)
-  STDERR.puts "Adding #{source_io&.content_type} image #{source.to_s.inspect}."
-  image = Image.for_io(source_io, path: source.to_s)
-  (images << image).uniq!
-end
-
 def http_options(url)
   options = {}
   if ssl_no_verify = $config.ssl_no_verify?
@@ -573,30 +521,90 @@ rescue => e
   STDERR.puts "Cannot fetch source #{source.to_s.inspect}: #{e}\n#{e.backtrace * ?\n}"
 end
 
+def add_image(images, source_io, source)
+  STDERR.puts "Adding #{source_io&.content_type} image #{source.to_s.inspect}."
+  image = Image.for_io(source_io, path: source.to_s)
+  (images << image).uniq!
+end
+
+def import_source(source_io, source)
+  source = source.to_s
+  puts "Importing #{italic { source_io&.content_type }} document #{source.inspect} now."
+  "Imported #{source.inspect}:\n%s\n\n" % parse_source(source_io)
+end
+
 def import(source)
-  puts "Now importing #{source.to_s.inspect}."
   fetch_source(source) do |source_io|
-    content = parse_source(source_io)
-    content.present? or return
+    content = import_source(source_io, source) or return
     source_io.rewind
     content
   end
 end
 
-def summarize(source, words: nil)
+def summarize_source(source_io, source, words: nil)
+  puts "Summarizing #{italic { source_io&.content_type }} document #{source.inspect} now."
   words = words.to_i
   words < 1 and words = 100
-  puts "Now summarizing #{source.to_s.inspect}."
-  source_content =
-    fetch_source(source) do |source_io|
-      content = parse_source(source_io)
-      content.present? or return
-      source_io.rewind
-      content
-    end
+  source_content = parse_source(source_io)
+  source_content.present? or return
   $config.prompts.summarize % { source_content:, words: }
 end
 
+def summarize(source, words: nil)
+  fetch_source(source) do |source_io|
+    content = summarize_source(source_io, source, words:) or return
+    source_io.rewind
+    content
+  end
+end
+
+def embed_source(source_io, source, count: nil)
+  $embedding.on? or return parse_source(source_io)
+  m = "Embedding #{italic { source_io&.content_type }} document #{source.to_s.inspect}."
+  if count
+    puts '%u. %s' % [ count, m ]
+  else
+    puts m
+  end
+  text = parse_source(source_io) or return
+  text.downcase!
+  splitter_config = $config.embedding.splitter
+  inputs = nil
+  case splitter_config.name
+  when 'Character'
+    splitter = Ollama::Documents::Splitters::Character.new(
+      chunk_size: splitter_config.chunk_size,
+    )
+    inputs = splitter.split(text)
+  when 'RecursiveCharacter'
+    splitter = Ollama::Documents::Splitters::RecursiveCharacter.new(
+      chunk_size: splitter_config.chunk_size,
+    )
+    inputs = splitter.split(text)
+  when 'Semantic'
+    splitter = Ollama::Documents::Splitters::Semantic.new(
+      ollama:, model: $config.embedding.model.name,
+      chunk_size: splitter_config.chunk_size,
+    )
+    inputs = splitter.split(
+      text,
+      breakpoint: splitter_config.breakpoint.to_sym,
+      percentage: splitter_config.percentage?,
+      percentile: splitter_config.percentile?,
+    )
+    inputs = splitter.split(text)
+  end
+  inputs or return
+  source = source.to_s
+  if source.start_with?(?!)
+    source = Ollama::Utils::Width.truncate(
+      source[1..-1].gsub(/\W+/, ?_),
+      length: 10
+    )
+  end
+  $documents.add(inputs, source:, batch_size: $config.embedding.batch_size?)
+end
+
 def embed(source)
   if $embedding.on?
     puts "Now embedding #{source.to_s.inspect}."
@@ -618,6 +626,7 @@ def parse_content(content, images)
   images.clear
   tags = Utils::Tags.new
 
+  contents = [ content ]
   content.scan(%r((?:\.\.|[.~])?/\S+|https?://\S+|#\S+)).each do |source|
     case source
     when /\A#(\S+)/
@@ -628,8 +637,15 @@ def parse_content(content, images)
         case source_io&.content_type&.media_type
         when 'image'
           add_image(images, source_io, source)
-        when 'text', 'application'
-          embed_source(source_io, source)
+        when 'text', 'application', nil
+          case $document_policy
+          when 'importing'
+            contents << import_source(source_io, source)
+          when 'embedding'
+            embed_source(source_io, source)
+          when 'summarizing'
+            contents << summarize_source(source_io, source)
+          end
         else
           STDERR.puts(
             "Cannot fetch #{source.to_s.inspect} with content type "\
@@ -639,8 +655,8 @@ def parse_content(content, images)
       end
     end
   end
-
-  return content, (tags unless tags.empty?)
+  new_content = contents.select(&:present?).compact * "\n\n"
+  return new_content, (tags unless tags.empty?)
 end
 
 def choose_model(cli_model, current_model)
@@ -674,7 +690,29 @@ def choose_collection(current_collection)
   end
 ensure
   puts "Using collection #{bold{$documents.collection}}."
-  collection_stats
+  info
+end
+
+def choose_document_policy
+  policies = %w[ importing embedding summarizing ].sort
+  current  = if policies.index($document_policy)
+               $document_policy
+             elsif policies.index($config.document_policy)
+               $config.document_policy
+             else
+               policies.first
+             end
+  policies.unshift('[EXIT]')
+  policy = Ollama::Utils::Chooser.choose(policies)
+  case policy
+  when nil, '[EXIT]'
+    puts "Exiting chooser."
+    policy = current
+  end
+  $document_policy = policy
+ensure
+  puts "Using document policy #{bold{$document_policy}}."
+  info
 end
 
 def collection_stats
@@ -756,6 +794,7 @@ def info
   $markdown.show
   $stream.show
   $location.show
+  puts "Document policy for references in user text: #{bold{$document_policy}}"
   if $voice.on?
     puts "Using voice #{bold{$current_voice}} to speak."
   end
@@ -799,6 +838,7 @@ def display_chat_help
     /regenerate                     the last answer message
     /collection( clear|change)      change (default) collection or clear
     /info                           show information for current session
+    /document_policy                pick a scan policy for document references
     /import source                  import the source's content
     /summarize [n] source           summarize the source's content in n words
     /embedding                      toggle embedding paused or not
@@ -853,10 +893,11 @@ $opts[?V] and version
 base_url = $opts[?u] || $config.url
 $ollama      = Client.new(base_url:, debug: $config.debug)
 
-$model       = choose_model($opts[?m], $config.model.name)
-options      = Options[$config.model.options]
-model_system = pull_model_unless_present($model, options)
-messages     = []
+$document_policy = $config.document_policy
+$model           = choose_model($opts[?m], $config.model.name)
+options          = Options[$config.model.options]
+model_system     = pull_model_unless_present($model, options)
+messages         = []
 $embedding_enabled.set($config.embedding.enabled && !$opts[?E])
 
 if $opts[?c]
@@ -969,7 +1010,7 @@ loop do
     puts "Cleared messages."
     next
   when %r(^/clobber$)
-    if ask?(prompt: 'Are you sure? (y/n) ') =~ /\Ay/i
+    if ask?(prompt: 'Are you sure to clear messages and collection? (y/n) ') =~ /\Ay/i
       clear_messages(messages)
       $documents.clear
       puts "Cleared messages and collection #{bold{$documents.collection}}."
@@ -1034,9 +1075,12 @@ loop do
       choose_collection($documents.collection)
     end
     next
-  when %r(/info)
+  when %r(^/info$)
     info
     next
+  when %r(^/document_policy$)
+    choose_document_policy
+    next
   when %r(^/import\s+(.+))
     parse_content = false
     content       = import($1) or next

data/lib/ollama/handlers/markdown.rb

@@ -7,7 +7,7 @@ class Ollama::Handlers::Markdown
   def initialize(output: $stdout)
     super
     @output.sync = true
-    @content = ''
+    @content     = ''
   end
 
   def call(response)
@@ -16,7 +16,6 @@ class Ollama::Handlers::Markdown
       markdown_content = Ollama::Utils::ANSIMarkdown.parse(@content)
       @output.print clear_screen, move_home, markdown_content
     end
-    response.done and @output.puts
     self
   end
 end

data/lib/ollama/version.rb

@@ -1,6 +1,6 @@
 module Ollama
   # Ollama version
-  VERSION         = '0.8.0'
+  VERSION         = '0.9.0'
   VERSION_ARRAY   = VERSION.split('.').map(&:to_i) # :nodoc:
   VERSION_MAJOR   = VERSION_ARRAY[0] # :nodoc:
   VERSION_MINOR   = VERSION_ARRAY[1] # :nodoc:

data/ollama-ruby.gemspec

@@ -1,14 +1,14 @@
 # -*- encoding: utf-8 -*-
-# stub: ollama-ruby 0.8.0 ruby lib
+# stub: ollama-ruby 0.9.0 ruby lib
 
 Gem::Specification.new do |s|
   s.name = "ollama-ruby".freeze
-  s.version = "0.8.0".freeze
+  s.version = "0.9.0".freeze
 
   s.required_rubygems_version = Gem::Requirement.new(">= 0".freeze) if s.respond_to? :required_rubygems_version=
   s.require_paths = ["lib".freeze]
   s.authors = ["Florian Frank".freeze]
-  s.date = "2024-10-06"
+  s.date = "2024-10-18"
   s.description = "Library that allows interacting with the Ollama API".freeze
   s.email = "flori@ping.de".freeze
   s.executables = ["ollama_console".freeze, "ollama_chat".freeze, "ollama_update".freeze, "ollama_cli".freeze]
@@ -24,7 +24,7 @@ Gem::Specification.new do |s|
 
   s.specification_version = 4
 
-  s.add_development_dependency(%q<gem_hadar>.freeze, ["~> 1.18.0".freeze])
+  s.add_development_dependency(%q<gem_hadar>.freeze, ["~> 1.19".freeze])
   s.add_development_dependency(%q<all_images>.freeze, ["~> 0.4".freeze])
   s.add_development_dependency(%q<rspec>.freeze, ["~> 3.2".freeze])
   s.add_development_dependency(%q<webmock>.freeze, [">= 0".freeze])

data/spec/ollama/client_spec.rb

@@ -54,21 +54,21 @@ RSpec.describe Ollama::Client do
   end
 
   it 'can raise error on connection error' do
-    allow(excon).to receive(:post).and_raise Excon::Error::Socket
+    expect(excon).to receive(:post).and_raise Excon::Error::Socket
     expect {
       ollama.generate(model: 'llama3.1', prompt: 'Hello World')
     }.to raise_error(Ollama::Errors::SocketError)
   end
 
   it 'can raise error on timeout' do
-    allow(excon).to receive(:post).and_raise Excon::Errors::Timeout
+    expect(excon).to receive(:post).and_raise Excon::Errors::Timeout
     expect {
       ollama.generate(model: 'llama3.1', prompt: 'Hello World')
     }.to raise_error(Ollama::Errors::TimeoutError)
   end
 
   it 'can raise a generic error' do
-    allow(excon).to receive(:post).and_raise Excon::Errors::Error
+    expect(excon).to receive(:post).and_raise Excon::Errors::Error
     expect {
       ollama.generate(model: 'llama3.1', prompt: 'Hello World')
     }.to raise_error(Ollama::Errors::Error)

data/spec/ollama/documents/redis_backed_memory_cache_spec.rb

@@ -78,7 +78,7 @@ RSpec.describe Ollama::Documents::RedisBackedMemoryCache do
     end
 
     it 'returns size' do
-      allow(cache).to receive(:count).and_return 3
+      expect(cache).to receive(:count).and_return 3
       expect(cache.size).to eq 3
     end
 

data/spec/ollama/documents/redis_cache_spec.rb

@@ -57,7 +57,7 @@ RSpec.describe Ollama::Documents::RedisCache do
       key, value = 'foo', { test: true }
       expect(redis).to receive(:set).with('test-' + key, JSON(value), ex: 3_600)
       cache[key] = value
-      allow(redis).to receive(:ttl).with('test-' + key).and_return 3_600
+      expect(redis).to receive(:ttl).with('test-' + key).and_return 3_600
       expect(cache.ttl(key)).to eq 3_600
     end
 

data/spec/ollama/documents_spec.rb

@@ -42,7 +42,7 @@ RSpec.describe Ollama::Documents do
   end
 
   it 'can find strings' do
-    allow(ollama).to receive(:embed).
+    expect(ollama).to receive(:embed).
       with(model:, input: [ 'foo' ], options: nil).
       and_return(double(embeddings: [ [ 0.1 ] ]))
     expect(documents << 'foo').to eq documents
@@ -57,7 +57,7 @@ RSpec.describe Ollama::Documents do
   end
 
   it 'can find only tagged strings' do
-    allow(ollama).to receive(:embed).
+    expect(ollama).to receive(:embed).
       with(model:, input: [ 'foo' ], options: nil).
       and_return(double(embeddings: [ [ 0.1 ] ]))
     expect(documents.add('foo', tags: %i[ test ])).to eq documents
@@ -77,10 +77,10 @@ RSpec.describe Ollama::Documents do
   end
 
   it 'can find strings conditionally' do
-    allow(ollama).to receive(:embed).
+    expect(ollama).to receive(:embed).
       with(model:, input: [ 'foobar' ], options: nil).
       and_return(double(embeddings: [ [ 0.01 ] ]))
-    allow(ollama).to receive(:embed).
+    expect(ollama).to receive(:embed).
       with(model:, input: [ 'foo' ], options: nil).
       and_return(double(embeddings: [ [ 0.1 ] ]))
     expect(documents << 'foobar').to eq documents
@@ -132,7 +132,7 @@ RSpec.describe Ollama::Documents do
     end
 
     it 'can clear texts with tags' do
-      allow(ollama).to receive(:embed).
+      expect(ollama).to receive(:embed).
         with(model:, input: %w[ bar ], options: nil).
         and_return(double(embeddings: [ [ 0.1 ] ]))
       expect(documents.add('foo', tags: %i[ test ])).to eq documents

data/spec/ollama/handlers/markdown_spec.rb

@@ -25,7 +25,6 @@ RSpec.describe Ollama::Handlers::Markdown do
   it 'can markdown response as markdown' do
     output = double('output', :sync= => true)
     expect(output).to receive(:print).with("\e[2J", "\e[1;1H", ansi)
-    expect(output).to receive(:puts)
     markdown = described_class.new(output:)
     response = double('response', response: md, done: false)
     markdown.call(response)
@@ -36,7 +35,6 @@ RSpec.describe Ollama::Handlers::Markdown do
   it 'can markdown message content as markdown' do
     output = double('output', :sync= => true)
     expect(output).to receive(:print).with("\e[2J", "\e[1;1H", ansi)
-    expect(output).to receive(:puts)
     markdown = described_class.new(output:)
     response = double('response', response: nil, message: double(content: md), done: false)
     markdown.call(response)

data/spec/ollama/utils/fetcher_spec.rb

@@ -105,7 +105,7 @@ RSpec.describe Ollama::Utils::Fetcher do
   end
 
   it 'can .execute and fail' do
-    allow(IO).to receive(:popen).and_raise StandardError
+    expect(IO).to receive(:popen).and_raise StandardError
     described_class.execute('foobar') do |file|
       expect(file).to be_a StringIO
       expect(file.read).to be_empty

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: ollama-ruby
 version: !ruby/object:Gem::Version
-  version: 0.8.0
+  version: 0.9.0
 platform: ruby
 authors:
 - Florian Frank
 autorequire:
 bindir: bin
 cert_chain: []
-date: 2024-10-06 00:00:00.000000000 Z
+date: 2024-10-18 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: gem_hadar
@@ -16,14 +16,14 @@ dependencies:
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: 1.18.0
+        version: '1.19'
   type: :development
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: 1.18.0
+        version: '1.19'
 - !ruby/object:Gem::Dependency
   name: all_images
   requirement: !ruby/object:Gem::Requirement