documentrix 0.1.1 → 0.2.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: ef2b0d8150c99cc22cc17d7d29d0331729b06d116e593fd32289852eb7739228
-  data.tar.gz: 386b85971a711fd6ba73aad746e98df33c14109c5524b48dd8fa20f3059fc875
+  metadata.gz: c45e75b570207ac77d9a04e95939050c33d8f5e0645c7ebac96c862de4e252f3
+  data.tar.gz: 18b0f526ec16115483de74c027a7d8eaa8c8ba2461ac41d4695f53f1ceff32c7
 SHA512:
-  metadata.gz: 7b1abe342b523199e58724d1b9ec66ce56654389dc010fc59c2847647825876b8c03e1511c5b007dd97ee8f19f4e79caf132760dbb549906d5febdc9ac403bc1
-  data.tar.gz: 5b638e7af884173350e48dcdb83a81d2adeb6c6a4f90dc20243b690d888f083d95edcd34e00fd3bffc9435aa8a29db151b0c1c6311f696d39379e397379ef9df
+  metadata.gz: dddf96ef71ab25c35c6872905cea070c8a27036caf841d97626fadc1db172e496a3b8f13748000cbb9a096b66ae59f459bae145fc2a4a06abffad7de267362a9
+  data.tar.gz: cab52c41f1749fe0ff56538cd01e4525787921e700dca91e44f10d939ec942afc59478c4f8256033fdbf5ef4a38ac649d80eadfbc711042bccb85e38b056a569

data/CHANGES.md

@@ -1,5 +1,28 @@
 # Changes
 
+## 2026-05-12 v0.2.0
+
+### Added
+
+- Implemented source-based document removal by adding the `remove` method to
+  `Documentrix::Documents`.
+- Added `clear_by_source` to `Documentrix::Documents::Cache::Common` as the
+  default cache implementation.
+- Added an optimized `clear_by_source` override in
+  `Documentrix::Documents::Cache::SQLiteCache` utilizing a direct SQL `DELETE`
+  query.
+
+### Changed
+
+- Updated `documentrix.gemspec` to use `rubygems_version` **4.0.10**.
+- Updated `gem_hadar` dependency to **2.17.1**.
+
+### Testing
+
+- Expanded test coverage in `spec/documents_spec.rb`,
+  `spec/documentrix/documents/cache/interface_spec.rb`, and all specific cache
+  specs.
+
 ## 2026-03-31 v0.1.1
 
 - Improved compatibility and reliability by ensuring the gem uses a stable,

data/Rakefile

@@ -33,7 +33,7 @@ GemHadar do
   dependency 'infobar',         '~> 0.9'
   dependency 'json',            '~> 2.0'
   dependency 'tins',            '~> 1.34'
-  dependency 'sqlite-vec',      '>= 0.1.8'
+  dependency 'sqlite-vec',      '>= 0.1.9'
   dependency 'sqlite3',         '~> 2.0', '>= 2.0.1'
   dependency 'kramdown-ansi',   '~> 0.0', '>= 0.0.1'
   dependency 'numo-narray-alt', '~> 0.9'

data/documentrix.gemspec

@@ -1,9 +1,9 @@
 # -*- encoding: utf-8 -*-
-# stub: documentrix 0.1.1 ruby lib
+# stub: documentrix 0.2.0 ruby lib
 
 Gem::Specification.new do |s|
   s.name = "documentrix".freeze
-  s.version = "0.1.1".freeze
+  s.version = "0.2.0".freeze
 
   s.required_rubygems_version = Gem::Requirement.new(">= 0".freeze) if s.respond_to? :required_rubygems_version=
   s.require_paths = ["lib".freeze]
@@ -17,13 +17,13 @@ Gem::Specification.new do |s|
   s.licenses = ["MIT".freeze]
   s.rdoc_options = ["--title".freeze, "Documentrix - Ruby library for embedding vector database".freeze, "--main".freeze, "README.md".freeze]
   s.required_ruby_version = Gem::Requirement.new(">= 3.1".freeze)
-  s.rubygems_version = "4.0.8".freeze
+  s.rubygems_version = "4.0.10".freeze
   s.summary = "Ruby library for embedding vector database".freeze
   s.test_files = ["spec/documentrix/documents/cache/interface_spec.rb".freeze, "spec/documentrix/documents/cache/memory_cache_spec.rb".freeze, "spec/documentrix/documents/cache/redis_cache_spec.rb".freeze, "spec/documentrix/documents/cache/sqlite_cache_spec.rb".freeze, "spec/documentrix/documents/splitters/character_spec.rb".freeze, "spec/documentrix/documents/splitters/semantic_spec.rb".freeze, "spec/documents_spec.rb".freeze, "spec/spec_helper.rb".freeze, "spec/utils/colorize_texts_spec.rb".freeze, "spec/utils/tags_spec.rb".freeze]
 
   s.specification_version = 4
 
-  s.add_development_dependency(%q<gem_hadar>.freeze, [">= 2.17.0".freeze])
+  s.add_development_dependency(%q<gem_hadar>.freeze, [">= 2.17.1".freeze])
   s.add_development_dependency(%q<all_images>.freeze, ["~> 0.12".freeze])
   s.add_development_dependency(%q<rspec>.freeze, ["~> 3.2".freeze])
   s.add_development_dependency(%q<kramdown>.freeze, ["~> 2.0".freeze])
@@ -32,7 +32,7 @@ Gem::Specification.new do |s|
   s.add_runtime_dependency(%q<infobar>.freeze, ["~> 0.9".freeze])
   s.add_runtime_dependency(%q<json>.freeze, ["~> 2.0".freeze])
   s.add_runtime_dependency(%q<tins>.freeze, ["~> 1.34".freeze])
-  s.add_runtime_dependency(%q<sqlite-vec>.freeze, [">= 0.1.8".freeze])
+  s.add_runtime_dependency(%q<sqlite-vec>.freeze, [">= 0.1.9".freeze])
   s.add_runtime_dependency(%q<sqlite3>.freeze, ["~> 2.0".freeze, ">= 2.0.1".freeze])
   s.add_runtime_dependency(%q<kramdown-ansi>.freeze, ["~> 0.0".freeze, ">= 0.0.1".freeze])
   s.add_runtime_dependency(%q<numo-narray-alt>.freeze, ["~> 0.9".freeze])

data/lib/documentrix/documents/cache/common.rb

@@ -116,6 +116,19 @@ module Documentrix::Documents::Cache::Common
     self
   end
 
+  # The clear_by_source method removes all records from the cache that
+  # have a source matching the given source.
+  #
+  # @param source [String] the source to filter records by
+  #
+  # @return [self] self
+  def clear_by_source(source)
+    each do |key, record|
+      delete(unpre(key)) if record.source == source
+    end
+    self
+  end
+
   # The clear method removes cached records based on the provided tags or
   # clears all records with the current prefix.
   #

data/lib/documentrix/documents/cache/sqlite_cache.rb

@@ -157,6 +157,17 @@ class Documentrix::Documents::Cache::SQLiteCache
     self
   end
 
+  # The clear_by_source method removes all records from the cache that
+  # have a source matching the given source.
+  #
+  # @param source [String] the source to filter records by
+  #
+  # @return [Documentrix::Documents::Cache::SQLiteCache] self
+  def clear_by_source(source)
+    execute(%{DELETE FROM records WHERE source = ?}, [ source ])
+    self
+  end
+
   # Move a key prefix in the cache.
   #
   # This operation updates every record whose key starts with +old_prefix+,

data/lib/documentrix/documents.rb

@@ -219,6 +219,16 @@ class Documentrix::Documents
     self
   end
 
+  # The remove method removes all documents associated with the given source.
+  #
+  # @param source [String] the source of the documents to remove
+  #
+  # @return [Documentrix::Documents] self
+  def remove(source)
+    @cache.clear_by_source(source)
+    self
+  end
+
   # The find method searches for strings within the cache by computing their
   # similarity scores.
   #

data/lib/documentrix/version.rb

@@ -1,6 +1,6 @@
 module Documentrix
   # Documentrix version
-  VERSION         = '0.1.1'
+  VERSION         = '0.2.0'
   VERSION_ARRAY   = VERSION.split('.').map(&:to_i) # :nodoc:
   VERSION_MAJOR   = VERSION_ARRAY[0] # :nodoc:
   VERSION_MINOR   = VERSION_ARRAY[1] # :nodoc:

data/spec/documentrix/documents/cache/interface_spec.rb

@@ -52,6 +52,9 @@ describe 'Documentrix::Documents::Cache Interface' do
       expect(cache).to respond_to(:clear_for_tags)
       expect(cache.method(:clear_for_tags).owner).to eq Documentrix::Documents::Cache::Common
 
+      expect(cache).to respond_to(:clear_by_source)
+      expect(cache.method(:clear_by_source).owner).to eq Documentrix::Documents::Cache::Common
+
       expect(cache).to respond_to(:clear)
       expect(cache.method(:clear).owner).to eq Documentrix::Documents::Cache::Common
     end
@@ -108,6 +111,9 @@ describe 'Documentrix::Documents::Cache Interface' do
       expect(cache).to respond_to(:clear_for_tags)
       expect(cache.method(:clear_for_tags).owner).to eq Documentrix::Documents::Cache::Common
 
+      expect(cache).to respond_to(:clear_by_source)
+      expect(cache.method(:clear_by_source).owner).to eq Documentrix::Documents::Cache::Common
+
       expect(cache).to respond_to(:clear)
       expect(cache.method(:clear).owner).to eq Documentrix::Documents::Cache::Common
 
@@ -168,6 +174,9 @@ describe 'Documentrix::Documents::Cache Interface' do
       expect(cache).to respond_to(:clear_for_tags)
       expect(cache.method(:clear_for_tags).owner).to eq Documentrix::Documents::Cache::SQLiteCache
 
+      expect(cache).to respond_to(:clear_by_source)
+      expect(cache.method(:clear_by_source).owner).to eq Documentrix::Documents::Cache::SQLiteCache
+
       expect(cache).to respond_to(:clear)
       expect(cache.method(:clear).owner).to eq Documentrix::Documents::Cache::Common
 

data/spec/documentrix/documents/cache/memory_cache_spec.rb

@@ -120,6 +120,17 @@ describe Documentrix::Documents::MemoryCache do
     }.from(1).to(0)
   end
 
+  it 'can clear by source' do
+    cache['foo'] = Documentrix::Documents::Record[text: 'foo', source: 's1', embedding: [0.1]]
+    cache['bar'] = Documentrix::Documents::Record[text: 'bar', source: 's1', embedding: [0.1]]
+    cache['baz'] = Documentrix::Documents::Record[text: 'baz', source: 's2', embedding: [0.1]]
+    expect {
+      cache.clear_by_source('s1')
+    }.to change { cache.size }.from(3).to(1)
+    expect(cache.key?('baz')).to be true
+    expect(cache.key?('foo')).to be false
+  end
+
   it 'can iterate over keys under a prefix' do
     cache['foo'] = 'bar'
     expect(cache.to_a).to eq [ %W[ #{prefix}foo bar ] ]

data/spec/documentrix/documents/cache/redis_cache_spec.rb

@@ -109,6 +109,20 @@ describe Documentrix::Documents::RedisCache do
       expect(cache.clear).to eq cache
     end
 
+    it 'can clear by source' do
+      object_class = Class.new(JSON::GenericObject)
+      cache = described_class.new(prefix:, url: 'something', object_class:)
+      expect(redis).to receive(:scan_each).with(match: 'test-*').and_yield(
+        'test-foo'
+      ).and_yield(
+        'test-bar'
+      )
+      expect(redis).to receive(:get).with('test-foo').and_return(JSON(source: 's1'))
+      expect(redis).to receive(:get).with('test-bar').and_return(JSON(source: 's2'))
+      expect(redis).to receive(:del).with('test-foo')
+      expect(cache.clear_by_source('s1')).to eq cache
+    end
+
     it 'can iterate over keys under a prefix' do
       expect(redis).to receive(:scan_each).with(match: 'test-*')
       cache.to_a

data/spec/documentrix/documents/cache/sqlite_cache_spec.rb

@@ -145,6 +145,20 @@ describe Documentrix::Documents::SQLiteCache do
     expect(cache).to be_key 'bar'
   end
 
+  it 'can clear by source' do
+    val1 = test_value.merge(source: 's1')
+    val2 = test_value.merge(source: 's1')
+    val3 = test_value.merge(source: 's2')
+    cache['foo'] = val1
+    cache['bar'] = val2
+    cache['baz'] = val3
+    expect {
+      cache.clear_by_source('s1')
+    }.to change { cache.size }.from(3).to(1)
+    expect(cache.key?('baz')).to be true
+    expect(cache.key?('foo')).to be false
+  end
+
   it 'can return tags' do
     key, value = 'foo', { tags: %w[ foo ], embedding: [ 0.5 ] * 1_024 }
     cache[key] = value

data/spec/documents_spec.rb

@@ -158,6 +158,24 @@ describe Documentrix::Documents do
       }.to change { documents.size }.from(1).to(0)
     end
 
+    it 'can remove sources' do
+      allow(ollama).to receive(:embed).at_least(:once).
+        and_return(double(embeddings: [ [ 0.1 ] ]))
+
+      documents.add('foo', source: 'source1')
+      documents.add('bar', source: 'source1')
+      documents.add('baz', source: 'source2')
+
+      expect(documents.size).to eq 3
+
+      documents.remove('source1')
+
+      expect(documents.size).to eq 1
+      expect(documents.exist?('baz')).to be true
+      expect(documents.exist?('foo')).to be false
+      expect(documents.exist?('bar')).to be false
+    end
+
     it 'returns collections' do
       expect(documents.collections).to eq [ :default ]
     end

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: documentrix
 version: !ruby/object:Gem::Version
-  version: 0.1.1
+  version: 0.2.0
 platform: ruby
 authors:
 - Florian Frank
@@ -15,14 +15,14 @@ dependencies:
     requirements:
     - - ">="
       - !ruby/object:Gem::Version
-        version: 2.17.0
+        version: 2.17.1
   type: :development
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
     - - ">="
       - !ruby/object:Gem::Version
-        version: 2.17.0
+        version: 2.17.1
 - !ruby/object:Gem::Dependency
   name: all_images
   requirement: !ruby/object:Gem::Requirement
@@ -141,14 +141,14 @@ dependencies:
     requirements:
     - - ">="
       - !ruby/object:Gem::Version
-        version: 0.1.8
+        version: 0.1.9
   type: :runtime
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
     - - ">="
       - !ruby/object:Gem::Version
-        version: 0.1.8
+        version: 0.1.9
 - !ruby/object:Gem::Dependency
   name: sqlite3
   requirement: !ruby/object:Gem::Requirement
@@ -314,7 +314,7 @@ required_rubygems_version: !ruby/object:Gem::Requirement
     - !ruby/object:Gem::Version
       version: '0'
 requirements: []
-rubygems_version: 4.0.8
+rubygems_version: 4.0.10
 specification_version: 4
 summary: Ruby library for embedding vector database
 test_files: