ollama-ruby 0.6.0 → 0.8.0

data/bin/ollama_chat

@@ -4,9 +4,6 @@ require 'ollama'
 include Ollama
 require 'term/ansicolor'
 include Term::ANSIColor
-require 'tins'
-require 'tins/xt/full'
-require 'tins/xt/hash_union'
 include Tins::GO
 require 'reline'
 require 'reverse_markdown'
@@ -31,6 +28,11 @@ class OllamaChatConfig
       name: <%= ENV.fetch('OLLAMA_CHAT_MODEL', 'llama3.1') %>
       options:
         num_ctx: 8192
+    location:
+      enabled: false
+      name: Berlin
+      decimal_degrees: [ 52.514127, 13.475211 ]
+      units: SI (International System of Units) # or USCS (United States Customary System)
     prompts:
       embed: "This source was now embedded: %{source}"
       summarize: |
@@ -47,7 +49,7 @@ class OllamaChatConfig
     voice:
       enabled: false
       default: Samantha
-      list: <%= `say -v ?`.lines.map { _1[/^(.+?)\s+[a-z]{2}_[a-zA-Z0-9]{2,}/, 1] }.uniq.sort.to_s.force_encoding('ASCII-8BIT') %>
+      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
     embedding:
@@ -57,6 +59,7 @@ class OllamaChatConfig
         options: {}
         # Retrieval prompt template:
         prompt: 'Represent this sentence for searching relevant passages: %s'
+      batch_size: 10
       collection: <%= ENV['OLLAMA_CHAT_COLLECTION'] %>
       found_texts_size: 4096
       found_texts_count: null
@@ -83,7 +86,7 @@ class OllamaChatConfig
     if @filename == default_path && !retried
       retried = true
       mkdir_p File.dirname(default_path)
-      File.secure_write(default_path.to_s, DEFAULT_CONFIG)
+      File.secure_write(default_path, DEFAULT_CONFIG)
       retry
     else
       raise
@@ -178,7 +181,7 @@ end
 
 class Switch
   def initialize(name, msg:, config: $config)
-    @value = !!config.send("#{name}?")
+    @value = [ false, true ].include?(config) ? config : !!config.send("#{name}?")
     @msg   = msg
   end
 
@@ -259,9 +262,21 @@ def setup_switches
       false => "Embedding is currently not performed.",
     }
   )
+
+  $location = Switch.new(
+    :location,
+    msg: {
+      true  => "Location and localtime enabled.",
+      false => "Location and localtime disabled.",
+    },
+    config: $config.location.enabled
+  )
 end
 
 def search_web(query, n = nil)
+  if l = at_location
+    query += " #{at_location}"
+  end
   n = n.to_i
   n < 1 and n = 1
   query = URI.encode_uri_component(query)
@@ -401,6 +416,32 @@ def parse_atom(source_io)
   end
 end
 
+def pdf_read(io)
+  reader = PDF::Reader.new(io)
+  reader.pages.inject(+'') { |result, page| result << page.text }
+end
+
+def ps_read(io)
+  gs = `which gs`.chomp
+  if gs.present?
+    Tempfile.create do |tmp|
+      IO.popen("#{gs} -q -sDEVICE=pdfwrite -sOutputFile=#{tmp.path} -", 'wb') do |gs_io|
+        until io.eof?
+          buffer = io.read(1 << 17)
+          IO.select(nil, [ gs_io ], nil)
+          gs_io.write buffer
+        end
+        gs_io.close
+        File.open(tmp.path, 'rb') do |pdf|
+          pdf_read(pdf)
+        end
+      end
+    end
+  else
+    STDERR.puts "Cannot convert #{io&.content_type} whith ghostscript, gs not in path."
+  end
+end
+
 def parse_source(source_io)
   case source_io&.content_type
   when 'text/html'
@@ -426,12 +467,11 @@ def parse_source(source_io)
     parse_rss(source_io)
   when 'application/atom+xml'
     parse_atom(source_io)
-  when 'application/json'
-    source_io.read
+  when 'application/postscript'
+    ps_read(source_io)
   when 'application/pdf'
-    reader = PDF::Reader.new(source_io)
-    reader.pages.inject(+'') { |result, page| result << page.text }
-  when %r(\Atext/), nil
+    pdf_read(source_io)
+  when %r(\Aapplication/(json|ld\+json|x-ruby|x-perl|x-gawk|x-python|x-javascript|x-c?sh|x-dosexec|x-shellscript|x-tex|x-latex|x-lyx|x-bibtex)), %r(\Atext/), nil
     source_io.read
   else
     STDERR.puts "Cannot embed #{source_io&.content_type} document."
@@ -439,9 +479,14 @@ def parse_source(source_io)
   end
 end
 
-def embed_source(source_io, source)
+def embed_source(source_io, source, count: nil)
   $embedding.on? or return parse_source(source_io)
-  puts "Embedding #{italic { source_io&.content_type }} document #{source.to_s.inspect}."
+  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
@@ -478,7 +523,7 @@ def embed_source(source_io, source)
       length: 10
     )
   end
-  $documents.add(inputs, source: source)
+  $documents.add(inputs, source:, batch_size: $config.embedding.batch_size?)
 end
 
 def add_image(images, source_io, source)
@@ -515,7 +560,7 @@ def fetch_source(source, &block)
     ) do |tmp|
       block.(tmp)
     end
-  when %r(\Afile://(?:(?:[.-]|[[:alnum:]])*)(/\S*)|([~.]?/\S*))
+  when %r(\Afile://(/\S*)|\A((?:\.\.|[~.]?)/\S*))
     filename = $~.captures.compact.first
     filename = File.expand_path(filename)
     Utils::Fetcher.read(filename) do |tmp|
@@ -573,10 +618,10 @@ def parse_content(content, images)
   images.clear
   tags = Utils::Tags.new
 
-  content.scan(%r([.~]?/\S+|https?://\S+|#\S+)).each do |source|
+  content.scan(%r((?:\.\.|[.~])?/\S+|https?://\S+|#\S+)).each do |source|
     case source
     when /\A#(\S+)/
-      tags << $1
+      tags.add($1, source:)
     else
       source = source.sub(/(["')]|\*+)\z/, '')
       fetch_source(source) do |source_io|
@@ -598,39 +643,53 @@ def parse_content(content, images)
   return content, (tags unless tags.empty?)
 end
 
-def choose_model(cli_model, default_model)
+def choose_model(cli_model, current_model)
   models = ollama.tags.models.map(&:name).sort
   model = if cli_model == ''
-            Ollama::Utils::Chooser.choose(models) || default_model
+            Ollama::Utils::Chooser.choose(models) || current_model
           else
-            cli_model || default_model
+            cli_model || current_model
           end
 ensure
   puts green { "Connecting to #{model}@#{ollama.base_url} now…" }
 end
 
-def choose_collection(default_collection)
-  collections = [ default_collection ] + $documents.collections
+def ask?(prompt:)
+  print prompt
+  STDIN.gets.chomp
+end
+
+def choose_collection(current_collection)
+  collections = [ current_collection ] + $documents.collections
   collections = collections.compact.map(&:to_s).uniq.sort
-  collections.unshift('[NEW]')
-  collection = Ollama::Utils::Chooser.choose(collections) || default_collection
-  if collection == '[NEW]'
-      print "Enter name of the new collection: "
-      collection = STDIN.gets.chomp
+  collections.unshift('[EXIT]').unshift('[NEW]')
+  collection = Ollama::Utils::Chooser.choose(collections) || current_collection
+  case collection
+  when '[NEW]'
+    $documents.collection = ask?(prompt: "Enter name of the new collection: ")
+  when nil, '[EXIT]'
+    puts "Exiting chooser."
+  when /./
+    $documents.collection = collection
   end
-  $documents.collection = collection
 ensure
-  puts "Changing to collection #{bold{collection}}."
+  puts "Using collection #{bold{$documents.collection}}."
   collection_stats
 end
 
 def collection_stats
+  list = $documents.collections.sort.map { |c|
+    '  ' + ($documents.collection == c ? bold { c } : c).to_s
+  }.join(?\n)
   puts <<~EOT
-    Collection
+    Current Collection
       Name: #{bold{$documents.collection}}
       Embedding model: #{bold{$embedding_model}}
       #Embeddings: #{$documents.size}
+      #Tags: #{$documents.tags.size}
       Tags: #{$documents.tags}
+    List:
+    #{list}
   EOT
 end
 
@@ -652,13 +711,25 @@ def show_system_prompt
   EOT
 end
 
+def at_location
+  if $location.on?
+    location_name            = $config.location.name
+    location_decimal_degrees = $config.location.decimal_degrees * ', '
+    localtime                = Time.now.iso8601
+    units                    = $config.location.units
+    $config.prompts.location % {
+      location_name:, location_decimal_degrees:, localtime:, units:,
+    }
+  end.to_s
+end
+
 def set_system_prompt(messages, system)
   $system = system
   messages.clear
   messages << Message.new(role: 'system', content: system)
 end
 
-def change_system_prompt(messages)
+def change_system_prompt(messages, default)
   prompts = $config.system_prompts.attribute_names.compact
   chosen  = Ollama::Utils::Chooser.choose(prompts)
   system  = if chosen
@@ -684,6 +755,7 @@ def info
   puts "Documents database cache is #{$documents.nil? ? 'n/a' : bold{$documents.cache.class}}"
   $markdown.show
   $stream.show
+  $location.show
   if $voice.on?
     puts "Using voice #{bold{$current_voice}} to speak."
   end
@@ -716,6 +788,7 @@ def display_chat_help
     /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
@@ -724,7 +797,8 @@ def display_chat_help
     /model                          change the model
     /system                         change system prompt (clears conversation)
     /regenerate                     the last answer message
-    /collection clear [tag]|change  clear or show stats of current collection
+    /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
@@ -739,7 +813,7 @@ end
 
 def usage
   puts <<~EOT
-    #{File.basename($0)} [OPTIONS]
+    Usage: #{File.basename($0)} [OPTIONS]
 
       -f CONFIG      config file to read
       -u URL         the ollama base url, OLLAMA_URL
@@ -790,7 +864,7 @@ if $opts[?c]
 else
   default = $config.system_prompts.default? || model_system
   if $opts[?s] == ??
-    change_system_prompt(messages)
+    change_system_prompt(messages, default)
   else
     system = Ollama::Utils::FileArgument.get_file_argument($opts[?s], default:)
     system.present? and set_system_prompt(messages, system)
@@ -827,11 +901,13 @@ if $embedding.on?
       end
     end
     puts "Collection #{bold{collection}}: Adding #{document_list.size} documents…"
+    count = 1
     document_list.each_slice(25) do |docs|
       docs.each do |doc|
         fetch_source(doc) do |doc_io|
-          embed_source(doc_io, doc)
+          embed_source(doc_io, doc, count:)
         end
+        count += 1
       end
     end
   end
@@ -860,18 +936,21 @@ loop do
   content = Reline.readline(input_prompt, true)&.chomp
 
   case content
-  when %r(^/paste$)
-    puts bold { "Paste your content and then press C-d!" }
-    content = STDIN.read
   when %r(^/copy$)
     copy_to_clipboard(messages)
     next
+  when %r(^/paste$)
+    puts bold { "Paste your content and then press C-d!" }
+    content = STDIN.read
   when %r(^/markdown$)
     $markdown.toggle
     next
   when %r(^/stream$)
     $stream.toggle
     next
+  when %r(^/location$)
+    $location.toggle
+    next
   when %r(^/voice(?:\s+(change))?$)
     if $1 == 'change'
       change_voice
@@ -890,33 +969,14 @@ loop do
     puts "Cleared messages."
     next
   when %r(^/clobber$)
-    clear_messages(messages)
-    $documents.clear
-    puts "Cleared messages and collection."
-    next
-  when %r(^/collection\s+(clear|change)(?:\s+(.+))?$)
-    command, arg = $1, $2
-    case command
-    when 'clear'
-      tags = arg.present? ? arg.sub(/\A#*/, '') : nil
-      if tags
-        $documents.clear(tags:)
-        puts "Cleared tag ##{tags} from collection #{bold{collection}}."
-      else
-        $documents.clear
-        puts "Cleared collection #{bold{collection}}."
-      end
-    when 'change'
-      choose_collection(collection)
+    if ask?(prompt: 'Are you sure? (y/n) ') =~ /\Ay/i
+      clear_messages(messages)
+      $documents.clear
+      puts "Cleared messages and collection #{bold{$documents.collection}}."
+    else
+      puts 'Cancelled.'
     end
     next
-  when %r(^/system$)
-    change_system_prompt(messages)
-    info
-    next
-  when %r(/info)
-    info
-    next
   when %r(^/pop(?:\s+(\d*))?$)
     if messages.size > 1
       n = $1.to_i.clamp(1, Float::INFINITY)
@@ -931,6 +991,10 @@ loop do
   when %r(^/model$)
     $model = choose_model('', $model)
     next
+  when %r(^/system$)
+    change_system_prompt(messages, $system)
+    info
+    next
   when %r(^/regenerate$)
     if content = messages[-2]&.content
       content.gsub!(/\nConsider these chunks for your answer.*\z/, '')
@@ -941,6 +1005,38 @@ loop do
     end
     parse_content = false
     content
+  when %r(^/collection(?:\s+(clear|change))?$)
+    case $1 || 'change'
+    when 'clear'
+      loop do
+        tags = $documents.tags.add('[EXIT]').add('[ALL]')
+        tag = Ollama::Utils::Chooser.choose(tags, prompt: 'Clear? %s')
+        case tag
+        when nil, '[EXIT]'
+          puts "Exiting chooser."
+          break
+        when '[ALL]'
+          if ask?(prompt: 'Are you sure? (y/n) ') =~ /\Ay/i
+            $documents.clear
+            puts "Cleared collection #{bold{$documents.collection}}."
+            break
+          else
+            puts 'Cancelled.'
+            sleep 3
+          end
+        when /./
+          $documents.clear(tags: [ tag ])
+          puts "Cleared tag #{tag} from collection #{bold{$documents.collection}}."
+          sleep 3
+        end
+      end
+    when 'change'
+      choose_collection($documents.collection)
+    end
+    next
+  when %r(/info)
+    info
+    next
   when %r(^/import\s+(.+))
     parse_content = false
     content       = import($1) or next
@@ -1007,6 +1103,10 @@ loop do
     end
   end
 
+  if location = at_location.full?
+    content += " [#{location} – do not comment on this information, just consider it for eventual queries]"
+  end
+
   messages << Message.new(role: 'user', content:, images: images.dup)
   images.clear
   handler = FollowChat.new(messages:, markdown: $markdown.on?, voice: ($current_voice if $voice.on?))

data/bin/ollama_cli

@@ -8,8 +8,8 @@ include Tins::GO
 require 'json'
 
 def usage
-  puts <<~end
-    #{File.basename($0)} [OPTIONS]
+  puts <<~EOT
+    Usage: #{File.basename($0)} [OPTIONS]
 
       -u URL         the ollama base url, OLLAMA_URL
       -m MODEL       the ollama model to chat with, OLLAMA_MODEL
@@ -22,7 +22,7 @@ def usage
       -S             use streaming for generation
       -h             this help
 
-  end
+  EOT
   exit 0
 end
 

data/bin/ollama_update

@@ -14,4 +14,6 @@ ollama.tags.models.each do |model|
     "Updating model #{bold {name}} (last modified at #{modified_at.iso8601}):"
   )
   ollama.pull(name:)
+rescue Ollama::Errors::Error => e
+  infobar.puts "Caught #{e.class} for model #{bold { model.name }}: #{e} => Continuing."
 end

data/docker-compose.yml

@@ -1,6 +1,6 @@
 services:
   redis:
-    image: redis:7.2.5-alpine
+    image: valkey/valkey:7.2.7-alpine
     restart: unless-stopped
     ports: [ "127.0.0.1:9736:6379" ]
     volumes:

data/lib/ollama/client.rb

@@ -1,6 +1,3 @@
-require 'tins/xt/string_camelize'
-require 'tins/annotate'
-
 class Ollama::Client
 end
 require 'ollama/client/doc'

data/lib/ollama/documents/cache/redis_cache.rb

@@ -23,7 +23,17 @@ class Ollama::Documents::RedisCache
   end
 
   def []=(key, value)
-    redis.set(pre(key), JSON.generate(value), ex: @ex)
+    set(key, value)
+  end
+
+  def set(key, value, ex: nil)
+    ex ||= @ex
+    if !ex.nil? && ex < 1
+      redis.del(pre(key))
+    else
+      redis.set(pre(key), JSON.generate(value), ex:)
+    end
+    value
   end
 
   def ttl(key)

data/lib/ollama/documents.rb

@@ -25,7 +25,7 @@ class Ollama::Documents
     end
 
     def tags_set
-      Ollama::Utils::Tags.new(tags)
+      Ollama::Utils::Tags.new(tags, source:)
     end
 
     def ==(other)
@@ -55,10 +55,13 @@ class Ollama::Documents
     @cache.prefix = prefix
   end
 
-  def add(inputs, batch_size: 10, source: nil, tags: [])
+  def add(inputs, batch_size: nil, source: nil, tags: [])
     inputs = Array(inputs)
-    tags   = Ollama::Utils::Tags.new(tags)
-    source and tags.add File.basename(source).gsub(/\?.*/, '')
+    batch_size ||= 10
+    tags = Ollama::Utils::Tags.new(tags, source:)
+    if source
+      tags.add(File.basename(source).gsub(/\?.*/, ''), source:)
+    end
     inputs.map! { |i|
       text = i.respond_to?(:read) ? i.read : i.to_s
       text
@@ -70,7 +73,7 @@ class Ollama::Documents
     end
     batches = inputs.each_slice(batch_size).
       with_infobar(
-        label: "Add #{truncate(tags.to_s, percentage: 25)}",
+        label: "Add #{truncate(tags.to_s(link: false), percentage: 25)}",
         total: inputs.size
       )
     batches.each do |batch|
@@ -159,7 +162,11 @@ class Ollama::Documents
   end
 
   def tags
-    @cache.inject(Ollama::Utils::Tags.new) { |t, (_, record)| t.merge(record.tags) }
+    @cache.each_with_object(Ollama::Utils::Tags.new) do |(_, record), t|
+      record.tags.each do |tag|
+        t.add(tag, source: record.source)
+      end
+    end
   end
 
   private

data/lib/ollama/image.rb

@@ -7,6 +7,8 @@ class Ollama::Image
 
   attr_accessor :path
 
+  attr_reader :data
+
   class << self
     def for_base64(data, path: nil)
       obj = new(data)
@@ -31,7 +33,7 @@ class Ollama::Image
   end
 
   def ==(other)
-    @data == other..data
+    @data == other.data
   end
 
   def to_s

data/lib/ollama/utils/cache_fetcher.rb

@@ -13,7 +13,7 @@ class Ollama::Utils::CacheFetcher
     if body && content_type
       io = StringIO.new(body)
       io.rewind
-      io.extend(Ollama::Utils::Fetcher::ContentType)
+      io.extend(Ollama::Utils::Fetcher::HeaderExtension)
       io.content_type = content_type
       block.(io)
     end
@@ -25,8 +25,8 @@ class Ollama::Utils::CacheFetcher
     body.empty? and return
     content_type = io.content_type
     content_type.nil? and return
-    @cache[key(:body, url)]          = body
-    @cache[key(:content_type,  url)] = content_type.to_s
+    @cache.set(key(:body, url), body, ex: io.ex)
+    @cache.set(key(:content_type,  url), content_type.to_s, ex: io.ex)
     self
   end
 

data/lib/ollama/utils/chooser.rb

@@ -8,8 +8,9 @@ module Ollama::Utils::Chooser
 
   module_function
 
-  def choose(entries)
+  def choose(entries, prompt: 'Search? %s')
     entry = Search.new(
+      prompt:,
       match: -> answer {
         matcher = Amatch::PairDistance.new(answer.downcase)
         matches = entries.map { |n| [ n, -matcher.similar(n.to_s.downcase) ] }.
@@ -19,7 +20,7 @@ module Ollama::Utils::Chooser
       },
       query: -> _answer, matches, selector {
         matches.each_with_index.map { |m, i|
-          i == selector ? "#{blue{?⮕}} #{on_blue{m}}" : "  #{m}"
+          i == selector ? "#{blue{?⮕}} #{on_blue{m}}" : "  #{m.to_s}"
         } * ?\n
       },
       found: -> _answer, matches, selector {
@@ -27,6 +28,11 @@ module Ollama::Utils::Chooser
       },
       output: STDOUT
     ).start
-    return entry if entry
+    if entry
+      entry
+    else
+      print clear_screen, move_home
+      nil
+    end
   end
 end