wgit 0.10.2 → 0.10.5

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: c712169a7a2cf41bebb38b2c798aa883c9f685e4d3671929ab7a3ead55da0134
-  data.tar.gz: 0fb789510761c01f3d0459415b653f4eb896d194e31c610300ef94141dfab63a
+  metadata.gz: a359a5011cce717a84c0fa353cba658a4abcd11d2e8575da701b39eef35f641c
+  data.tar.gz: 730b2eee3c88d9cd99c1d9754744aa4bae71bc10019113e840f2914b3a7909e5
 SHA512:
-  metadata.gz: e175d2d23877fbb5aefc89aca1f98c31d9fee0ba78ec810bb3cc2260fdd97208c35ba9840b8cdc92ea49bdd8f6fd198910443fd61efdd563767048206f5a9788
-  data.tar.gz: 7dad82e66a10228ba0bce8e7d4cf8d17bdab734f90315813854c0c7e17807fa4c7f500711290a3e5eb28c431bab1c22ba4d25d7b585c1516c8f5470636bf37b4
+  metadata.gz: 4807d488cf03aa3dcf624249bd5169871bc17dd7e2de273cb841801147c040843912edc225a1f7346e427c022ecb2ce5360581323509221560ee99b31ea6a72b
+  data.tar.gz: ea3f1237116d05bbb24b2e85fdba1b6447821377a07ff5c5e6afea859afa5fb36d4e2c3f708c6f5dc201e0802aa61f0eaa56e63a62882330b5225dbbaa08721d

data/CHANGELOG.md

@@ -9,6 +9,33 @@
 - ...
 ---
 
+## v0.10.5
+### Added
+- `Database#last_result` getter method to return the most recent raw mongo result.
+### Changed/Removed
+- ...
+### Fixed
+- ...
+---
+
+## v0.10.4
+### Added
+- `Database#search_text` method which returns a Hash of `url => text_results` instead of `Wgit::Documents` (like `#search`).
+### Changed/Removed
+- ...
+### Fixed
+- ...
+---
+
+## v0.10.3
+### Added
+- ...
+### Changed/Removed
+- Changed `Database#create_collections` and `#create_unique_indexes` by removing `rescue nil` from their database operations. Now any underlying errors with the database client are not masked.
+### Fixed
+- ...
+---
+
 ## v0.10.2
 ### Added
 - `Wgit::Base#setup` and `#teardown` methods (lifecycle hooks) that can be overridden by subclasses.

data/lib/wgit/database/database.rb

@@ -45,6 +45,9 @@ module Wgit
     # A custom setter method is also provided for changing the search logic.
     attr_reader :text_index
 
+    # The raw MongoDB result of the most recent operation.
+    attr_reader :last_result
+
     # Initializes a connected database client using the provided
     # connection_string or ENV['WGIT_CONNECTION_STRING'].
     #
@@ -91,29 +94,27 @@ module Wgit
 
     ### DDL ###
 
-    # Creates the urls and documents collections if they don't already exist.
-    # This method is therefore idempotent.
+    # Creates the 'urls' and 'documents' collections.
     #
     # @return [nil] Always returns nil.
     def create_collections
-      db.client[URLS_COLLECTION].create rescue nil
-      db.client[DOCUMENTS_COLLECTION].create rescue nil
+      @client[URLS_COLLECTION].create
+      @client[DOCUMENTS_COLLECTION].create
 
       nil
     end
 
-    # Creates the urls and documents unique 'url' indexes if they don't already
-    # exist. This method is therefore idempotent.
+    # Creates the urls and documents unique 'url' indexes.
     #
     # @return [nil] Always returns nil.
     def create_unique_indexes
       @client[URLS_COLLECTION].indexes.create_one(
         { url: 1 }, name: UNIQUE_INDEX, unique: true
-      ) rescue nil
+      )
 
       @client[DOCUMENTS_COLLECTION].indexes.create_one(
         { 'url.url' => 1 }, name: UNIQUE_INDEX, unique: true
-      ) rescue nil
+      )
 
       nil
     end
@@ -187,6 +188,8 @@ module Wgit
       result = @client[collection].replace_one(query, data_hash, upsert: true)
 
       result.matched_count.zero?
+    ensure
+      @last_result = result
     end
 
     ### Retrieve Data ###
@@ -296,16 +299,12 @@ module Wgit
       results = retrieve(DOCUMENTS_COLLECTION, query,
                          sort: sort_proj, projection: sort_proj,
                          limit: limit, skip: skip)
-      return [] if results.count < 1 # respond_to? :empty? == false
 
-      # results.respond_to? :map! is false so we use map and overwrite the var.
-      results = results.map do |mongo_doc|
+      results.map do |mongo_doc|
         doc = Wgit::Document.new(mongo_doc)
         yield(doc) if block_given?
         doc
       end
-
-      results
     end
 
     # Searches the database's Documents for the given query and then searches
@@ -350,6 +349,58 @@ module Wgit
       results
     end
 
+    # Searches the database's Documents for the given query and then searches
+    # each result in turn using `doc.search`. Instead of an Array of Documents,
+    # this method returns a Hash of the docs url => search_results creating a
+    # search engine like result set for quick access to text matches.
+    #
+    # @param query [String] The text query to search with.
+    # @param case_sensitive [Boolean] Whether character case must match.
+    # @param whole_sentence [Boolean] Whether multiple words should be searched
+    #   for separately.
+    # @param limit [Integer] The max number of results to return.
+    # @param skip [Integer] The number of results to skip.
+    # @param sentence_limit [Integer] The max length of each search result
+    #   sentence.
+    # @param top_result_only [Boolean] Whether to return all of the documents
+    #   search results or just the top (most relavent) result.
+    # @yield [doc] Given each search result (Wgit::Document) returned from the
+    #   DB.
+    # @return [Hash<String, String | Array<String>>] The search results obtained
+    #   from the DB having mapped the docs url => search_results. The format of
+    #   search_results depends on the value of `top_result_only`.
+    def search_text(
+      query, case_sensitive: false, whole_sentence: true,
+      limit: 10, skip: 0, sentence_limit: 80, top_result_only: false
+    )
+      results = search(
+        query,
+        case_sensitive: case_sensitive,
+        whole_sentence: whole_sentence,
+        limit: limit,
+        skip: skip
+      )
+
+      results
+        .map do |doc|
+          yield(doc) if block_given?
+
+          results = doc.search(
+            query,
+            case_sensitive: case_sensitive,
+            whole_sentence: whole_sentence,
+            sentence_limit: sentence_limit
+          )
+
+          # Only return result if its text has a match - compact is called below.
+          next nil if results.empty?
+
+          [doc.url, (top_result_only ? results.first : results)]
+        end
+        .compact
+        .to_h
+    end
+
     # Returns statistics about the database.
     #
     # @return [BSON::Document#[]#fetch] Similar to a Hash instance.
@@ -456,21 +507,30 @@ module Wgit
     #   0 or 1 because urls are unique.
     def delete(obj)
       collection, query = get_type_info(obj)
-      @client[collection].delete_one(query).n
+      result = @client[collection].delete_one(query)
+      result.n
+    ensure
+      @last_result = result
     end
 
     # Deletes everything in the urls collection.
     #
     # @return [Integer] The number of deleted records.
     def clear_urls
-      @client[URLS_COLLECTION].delete_many({}).n
+      result = @client[URLS_COLLECTION].delete_many({})
+      result.n
+    ensure
+      @last_result = result
     end
 
     # Deletes everything in the documents collection.
     #
     # @return [Integer] The number of deleted records.
     def clear_docs
-      @client[DOCUMENTS_COLLECTION].delete_many({}).n
+      result = @client[DOCUMENTS_COLLECTION].delete_many({})
+      result.n
+    ensure
+      @last_result = result
     end
 
     # Deletes everything in the urls and documents collections. This will nuke
@@ -538,6 +598,8 @@ module Wgit
       else
         raise 'data must be a Hash or an Array of Hashes'
       end
+    ensure
+      @last_result = result
     end
 
     # Return if the write to the DB succeeded or not.
@@ -574,8 +636,8 @@ module Wgit
                  sort: {}, projection: {},
                  limit: 0, skip: 0)
       assert_type(query, Hash)
-      @client[collection.to_sym].find(query).projection(projection)
-                                .skip(skip).limit(limit).sort(sort)
+      @last_result = @client[collection.to_sym].find(query).projection(projection)
+                                               .skip(skip).limit(limit).sort(sort)
     end
 
     # Mutate/update one or more Url or Document records in the DB.
@@ -595,6 +657,8 @@ module Wgit
       raise 'DB write(s) (update) failed' unless write_succeeded?(result)
 
       result.n
+    ensure
+      @last_result = result
     end
 
     alias num_objects num_records

data/lib/wgit/document.rb

@@ -453,7 +453,7 @@ be relative"
 
       if query.is_a?(Regexp)
         regex = query
-      else # respond_to? #to_s == true
+      else # query.respond_to? :to_s == true
         query = query.to_s
         query = query.gsub(' ', '|') unless whole_sentence
         regex = Regexp.new(query, !case_sensitive)

data/lib/wgit/version.rb

@@ -6,7 +6,7 @@
 # @author Michael Telford
 module Wgit
   # The current gem version of Wgit.
-  VERSION = '0.10.2'
+  VERSION = '0.10.5'
 
   # Returns the current gem version of Wgit as a String.
   def self.version

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: wgit
 version: !ruby/object:Gem::Version
-  version: 0.10.2
+  version: 0.10.5
 platform: ruby
 authors:
 - Michael Telford
 autorequire:
 bindir: bin
 cert_chain: []
-date: 2021-11-22 00:00:00.000000000 Z
+date: 2022-07-11 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: addressable