infoboxer 0.3.0 → 0.4.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
-SHA1:
-  metadata.gz: be65bc91a5370bc24553e500754f413196caed76
-  data.tar.gz: d24dca5a13a64d563ddc473197a732a9f95884fd
+SHA256:
+  metadata.gz: 755e1283e896d2c2b363983a9b04ac92cb14870cfa2ab67fd62777684bac1352
+  data.tar.gz: c5443d788fc06a2310c65b80bc16531b820bb7142f611f5d063af8992cee7525
 SHA512:
-  metadata.gz: 661c06d6703db103035f61f55ebee5f0bc8a5f9ad182fcaf2a22be9f91063a9abd80bac1793fabae436b28bed274fcb7908219ba07453f46271a25d1cba0367a
-  data.tar.gz: 25029633e6516c30a7de21433db1903ab923af2c74082b3a3c9322b50b170cdb104652b95d261b3033a070629ee478a8b5510a7298fbef830d1445dfe56157d0
+  metadata.gz: 1443ecf7dbb485555a275a4d5390417ee48fabb2321a23fb59c208d32c1424259515924e903872b762afebcaaf5574afd713e4eade6c346aeb8e71719c051436
+  data.tar.gz: 71c007554240c40a7e7f9bee81d66d637b8643f1ccf18a5900c24087545d3c430830d8a30d9c4463244b13419d4edb349aee585885b6aecf03dc8d8b37d701a7

data/.github/workflows/ci.yml

@@ -0,0 +1,32 @@
+name: CI
+
+on:
+  push:
+    branches: [ master ]
+  pull_request:
+    branches: [ master ]
+
+jobs:
+  main:
+    name: >-
+      ${{ matrix.ruby }}
+    runs-on: ubuntu-latest
+    strategy:
+      fail-fast: false
+      matrix:
+        ruby: [ 2.6, 2.7, 3.0, head ]
+
+    steps:
+      - name: checkout
+        uses: actions/checkout@v2
+      - name: set up Ruby
+        uses: ruby/setup-ruby@v1
+        with:
+          ruby-version: ${{ matrix.ruby }}
+
+      - name: install dependencies
+        run:  bundle install --jobs 3 --retry 3
+      - name: spec
+        run:  bundle exec rake spec
+      - name: rubocop
+        run:  bundle exec rake rubocop

data/.rubocop_todo.yml

@@ -1,16 +1 @@
-# This configuration was generated by
-# `rubocop --auto-gen-config`
-# on 2017-06-23 13:52:16 +0300 using RuboCop version 0.49.1.
-# The point is for the user to remove these configuration records
-# one by one as the offenses are removed from the code base.
-# Note that changes in the inspected code, or installation of new
-# versions of RuboCop, may require this file to be generated again.
-
-# Offense count: 1
-Metrics/AbcSize:
-  Max: 29
-
-# Offense count: 1
-Metrics/PerceivedComplexity:
-  Max: 10
 

data/CHANGELOG.md

@@ -1,5 +1,48 @@
 # Infoboxer's change log
 
+## 0.4.0 (2021-05-30)
+
+* A cluster of bugs found in #81 fixed:
+  * Empty comment (`<!---->`) now processed properly;
+  * Templates that are implicitly inside tables (put on a separate row) now always create
+    an implicit `<TableCell>`
+  * Heading after non-closed table closes the table implicitly instead of being inserted
+    into the last cell.
+* Drop Ruby < 2.6, and support 3.0 instead.
+
+PS: Yeah, year-and-almost-half is much better than 2 years between releases, I guess.. And let's call
+it non-patch version then.
+
+## 0.3.3 (2020-02-09)
+
+* Fixed table captions handling (thanks @robfors for reporting)
+
+PS: Funny that this small bugfix release is exactly two years after the previous one :(
+
+## 0.3.2 (2018-02-09)
+
+* Updated MediaWiktory to finally turn on gzip encoding of responses;
+* Utility methods to expose some internals (`MediaWiki#api`, `Infoboxer#url_for(:wikipedia)`,
+  `Page#namespaces`, `Template#named_variables` and so on);
+* Fix parsing of lowercase `file:` links in `<gallery>`.
+
+## 0.3.1 (2017-12-04)
+
+* (Experimental) new representation of templates, much more readable;
+* More access to querying process and underlying `MediaWiktory::Wikipedia::Query`;
+* Finally, `limit` parameter for multi-page queries (category, search, prefixsearch).
+
+## 0.3.1.pre (2017-09-16)
+
+* Introduce interwiki links following (and proper handling of interwikis, in general);
+* Add `<gallery>` tag support;
+* Introduce `Navigation::Selector#===`;
+* Much more `Enumerable`'s methods supported by `Nodes`;
+* Lot of small simplifications, cleanups and bugfixes.
+
+TBH, it should be 0.4.0 or more, but it would be a shame to change versions so fast :) So, at least
+until it is `-pre`, let it be 0.3.1.
+
 ## 0.3.0 (2017-07-23)
 
 * Change logic of navigation through templates; now templates contents aren't hidden from global

data/Gemfile.lock

@@ -0,0 +1,172 @@
+GIT
+  remote: https://github.com/zverok/dokaz.git
+  revision: a8a6f0bbeab5589326fe2714cf89842b5f32b850
+  specs:
+    dokaz (0.0.4)
+      ansi
+      rouge
+      slop (~> 3)
+
+PATH
+  remote: .
+  specs:
+    infoboxer (0.4.0)
+      addressable
+      htmlentities
+      mediawiktory (= 0.1.3)
+      terminal-table
+
+GEM
+  remote: https://rubygems.org/
+  specs:
+    addressable (2.7.0)
+      public_suffix (>= 2.0.2, < 5.0)
+    ansi (1.5.0)
+    ast (2.4.2)
+    backports (3.21.0)
+    byebug (11.1.3)
+    coveralls (0.8.23)
+      json (>= 1.8, < 3)
+      simplecov (~> 0.16.1)
+      term-ansicolor (~> 1.3)
+      thor (>= 0.19.4, < 2.0)
+      tins (~> 1.6)
+    crack (0.4.5)
+      rexml
+    diff-lcs (1.4.4)
+    docile (1.4.0)
+    faraday (1.4.2)
+      faraday-em_http (~> 1.0)
+      faraday-em_synchrony (~> 1.0)
+      faraday-excon (~> 1.1)
+      faraday-net_http (~> 1.0)
+      faraday-net_http_persistent (~> 1.1)
+      multipart-post (>= 1.2, < 3)
+      ruby2_keywords (>= 0.0.4)
+    faraday-em_http (1.0.0)
+    faraday-em_synchrony (1.0.0)
+    faraday-excon (1.1.0)
+    faraday-net_http (1.0.1)
+    faraday-net_http_persistent (1.1.0)
+    faraday_middleware (1.0.0)
+      faraday (~> 1.0)
+    hashdiff (1.0.1)
+    hashie (4.1.0)
+    htmlentities (4.3.4)
+    io-console (0.5.9)
+    irb (1.3.5)
+      reline (>= 0.1.5)
+    json (2.5.1)
+    mediawiktory (0.1.3)
+      addressable
+      faraday
+      faraday_middleware
+      hashie
+      naught
+      nokogiri
+    multipart-post (2.1.1)
+    naught (1.1.0)
+    nokogiri (1.11.6-x86_64-linux)
+      racc (~> 1.4)
+    parallel (1.20.1)
+    parser (3.0.1.1)
+      ast (~> 2.4.1)
+    public_suffix (4.0.6)
+    racc (1.5.2)
+    rainbow (3.0.0)
+    rake (13.0.3)
+    redcarpet (3.5.1)
+    regexp_parser (2.1.1)
+    reline (0.2.5)
+      io-console (~> 0.5)
+    rexml (3.2.5)
+    rouge (3.26.0)
+    rspec (3.10.0)
+      rspec-core (~> 3.10.0)
+      rspec-expectations (~> 3.10.0)
+      rspec-mocks (~> 3.10.0)
+    rspec-core (3.10.1)
+      rspec-support (~> 3.10.0)
+    rspec-expectations (3.10.1)
+      diff-lcs (>= 1.2.0, < 2.0)
+      rspec-support (~> 3.10.0)
+    rspec-its (1.3.0)
+      rspec-core (>= 3.0.0)
+      rspec-expectations (>= 3.0.0)
+    rspec-mocks (3.10.2)
+      diff-lcs (>= 1.2.0, < 2.0)
+      rspec-support (~> 3.10.0)
+    rspec-support (3.10.2)
+    rubocop (1.15.0)
+      parallel (~> 1.10)
+      parser (>= 3.0.0.0)
+      rainbow (>= 2.2.2, < 4.0)
+      regexp_parser (>= 1.8, < 3.0)
+      rexml
+      rubocop-ast (>= 1.5.0, < 2.0)
+      ruby-progressbar (~> 1.7)
+      unicode-display_width (>= 1.4.0, < 3.0)
+    rubocop-ast (1.7.0)
+      parser (>= 3.0.1.1)
+    rubocop-rspec (2.3.0)
+      rubocop (~> 1.0)
+      rubocop-ast (>= 1.1.0)
+    ruby-prof (1.4.3)
+    ruby-progressbar (1.11.0)
+    ruby2_keywords (0.0.4)
+    rubygems-tasks (0.2.5)
+      irb (~> 1.0)
+    saharspec (0.0.4)
+    simplecov (0.16.1)
+      docile (~> 1.1)
+      json (>= 1.8, < 3)
+      simplecov-html (~> 0.10.0)
+    simplecov-html (0.10.2)
+    slop (3.6.0)
+    sync (0.5.0)
+    term-ansicolor (1.7.1)
+      tins (~> 1.0)
+    terminal-table (3.0.1)
+      unicode-display_width (>= 1.1.1, < 3)
+    thor (1.1.0)
+    timecop (0.9.4)
+    tins (1.29.1)
+      sync
+    unicode-display_width (2.0.0)
+    vcr (6.0.0)
+    webmock (3.13.0)
+      addressable (>= 2.3.6)
+      crack (>= 0.3.2)
+      hashdiff (>= 0.4.0, < 2.0.0)
+    yard (0.9.26)
+    yard-junk (0.0.9)
+      backports (>= 3.18)
+      rainbow
+      yard
+
+PLATFORMS
+  ruby
+  x86_64-linux
+
+DEPENDENCIES
+  byebug
+  coveralls
+  dokaz!
+  infoboxer!
+  rake
+  redcarpet
+  rspec (~> 3)
+  rspec-its (~> 1)
+  rubocop (~> 1.15.0)
+  rubocop-rspec (~> 2.3.0)
+  ruby-prof
+  rubygems-tasks
+  saharspec (= 0.0.4)
+  timecop
+  vcr
+  webmock
+  yard (~> 0.9)
+  yard-junk (~> 0.0.7)
+
+BUNDLED WITH
+   2.2.0

data/README.md

@@ -1,7 +1,7 @@
 # Infoboxer
 
 [![Gem Version](https://badge.fury.io/rb/infoboxer.svg)](http://badge.fury.io/rb/infoboxer)
-[![Build Status](https://travis-ci.org/molybdenum-99/infoboxer.svg?branch=master)](https://travis-ci.org/molybdenum-99/infoboxer)
+![Build Status](https://github.com/molybdenum-99/infoboxer/workflows/CI/badge.svg?branch=master)
 [![Coverage Status](https://coveralls.io/repos/molybdenum-99/infoboxer/badge.svg?branch=master&service=github)](https://coveralls.io/github/molybdenum-99/infoboxer?branch=master)
 [![Code Climate](https://codeclimate.com/github/molybdenum-99/infoboxer/badges/gpa.svg)](https://codeclimate.com/github/molybdenum-99/infoboxer)
 [![Infoboxer Gitter](https://badges.gitter.im/molybdenum-99/infoboxer.svg)](https://gitter.im/molybdenum-99/infoboxer)

data/infoboxer.gemspec

@@ -32,7 +32,7 @@ Gem::Specification.new do |s|
   s.executables << 'infoboxer'
 
   s.add_dependency 'htmlentities'
-  s.add_dependency 'mediawiktory', '>= 0.1.0'
+  s.add_dependency 'mediawiktory', '= 0.1.3'
   s.add_dependency 'addressable'
   s.add_dependency 'terminal-table'
 end

data/lib/infoboxer.rb

@@ -1,4 +1,4 @@
-# encoding: utf-8
+# frozen_string_literal: true
 
 # Main client module for entire infoboxer functionality. If you're lucky,
 # there's no other classes/modules you need to instantiate or call
@@ -47,9 +47,8 @@
 #
 module Infoboxer
   # @private
-  WIKIA_API_URL = 'http://%s.wikia.com/api.php'.freeze
+  WIKIA_API_URL = 'http://%s.wikia.com/api.php'
 
-  # @private
   WIKIMEDIA_PROJECTS = {
     wikipedia: 'wikipedia.org',
     wikivoyage: 'wikivoyage.org',
@@ -61,7 +60,6 @@ module Infoboxer
     wikisource: 'wikisource.org'
   }.freeze
 
-  # @private
   WIKIMEDIA_COMMONS = {
     commons: 'commons.wikimedia.org',
     species: 'species.wikimedia.org',
@@ -72,11 +70,11 @@ module Infoboxer
   end
 
   # Includeable version of {Infoboxer.wiki}
-  def wiki(api_url, options = {})
-    wikis[api_url] ||= MediaWiki.new(api_url, options || {})
+  def wiki(api_url, **options)
+    wikis[api_url] ||= MediaWiki.new(api_url, **options)
   end
 
-  class << self
+  class << self # rubocop:disable Lint/EmptyClass -- that's for YARD!
     # @!method wiki(api_url, options = {})
     # Default method for creating MediaWiki API client.
     #
@@ -168,21 +166,35 @@ module Infoboxer
   end
 
   WIKIMEDIA_PROJECTS.each do |name, domain|
-    define_method name do |lang = 'en', options = {}|
+    define_method name do |lang = 'en', **options|
       lang, options = 'en', lang if lang.is_a?(Hash)
 
-      wiki("https://#{lang}.#{domain}/w/api.php", options)
+      wiki("https://#{lang}.#{domain}/w/api.php", **options)
     end
   end
 
   alias_method :wp, :wikipedia
 
   WIKIMEDIA_COMMONS.each do |name, domain|
-    define_method name do |options = {}|
+    define_method name do |**options|
       wiki("https://#{domain}/w/api.php", options)
     end
   end
 
+  # Returns URL of API entry-point for a well-known Wiki-project (wikipedia, wikivoyage etc.)
+  # by project's name.
+  #
+  # @param symbol [Symbol] One of {WIKIMEDIA_PROJECTS} or {WIKIMEDIA_COMMONS} keys.
+  # @param lang [String, Symbol] Language of the project, if applicable.
+  # @return [String]
+  def url_for(symbol, lang = 'en')
+    if (domain = WIKIMEDIA_PROJECTS[symbol])
+      "https://#{lang}.#{domain}/w/api.php"
+    elsif (domain = WIKIMEDIA_COMMONS[symbol])
+      "https://#{domain}/w/api.php"
+    end
+  end
+
   # @!method wikipedia(lang = 'en', options = {})
   # Includeable version of {Infoboxer.wikipedia}
 
@@ -216,7 +228,7 @@ module Infoboxer
   # Includeable version of {Infoboxer.wikia}
   def wikia(*domains)
     options = domains.last.is_a?(Hash) ? domains.pop : {}
-    wiki(WIKIA_API_URL % domains.reverse.join('.'), options)
+    wiki(WIKIA_API_URL % domains.reverse.join('.'), **options)
   end
 
   # Sets user agent string globally. Default user agent is

data/lib/infoboxer/core_ext.rb

@@ -1,4 +1,4 @@
-# encoding: utf-8
+# frozen_string_literal: true
 
 # @private
 class Object

data/lib/infoboxer/definitions/en.wikipedia.org.rb

@@ -1,5 +1,6 @@
-# encoding: utf-8
+# frozen_string_literal: true
 
+# rubocop:disable Layout/EmptyLinesAroundArguments
 module Infoboxer
   MediaWiki::Traits.for('en.wikipedia.org') do
     templates do
@@ -372,3 +373,4 @@ module Infoboxer
     end
   end
 end
+# rubocop:enable Layout/EmptyLinesAroundArguments

data/lib/infoboxer/media_wiki.rb

@@ -1,4 +1,4 @@
-# encoding: utf-8
+# frozen_string_literal: true
 
 require 'mediawiktory'
 require 'addressable/uri'
@@ -27,7 +27,7 @@ module Infoboxer
     # You can set yours as an option to {Infoboxer.wiki} and its shortcuts,
     # or to {#initialize}
     UA = "Infoboxer/#{Infoboxer::VERSION} "\
-      '(https://github.com/molybdenum-99/infoboxer; zverok.offline@gmail.com)'.freeze
+      '(https://github.com/molybdenum-99/infoboxer; zverok.offline@gmail.com)'
 
     class << self
       # User agent getter/setter.
@@ -43,43 +43,45 @@ module Infoboxer
     # @private
     attr_reader :api_base_url, :traits
 
+    # @return [MediaWiktory::Wikipedia::Client]
+    attr_reader :api
+
     # Creating new MediaWiki client. {Infoboxer.wiki} provides shortcut
     # for it, as well as shortcuts for some well-known wikis, like
     # {Infoboxer.wikipedia}.
     #
-    # @param api_base_url URL of `api.php` file in your MediaWiki
+    # @param api_base_url [String] URL of `api.php` file in your MediaWiki
     #   installation. Typically, its `<domain>/w/api.php`, but can vary
     #   in different wikis.
-    # @param options Only one option is currently supported:
-    #   * `:user_agent` (also aliased as `:ua`) -- custom User-Agent header.
-    def initialize(api_base_url, options = {})
+    # @param user_agent [String] (also aliased as `:ua`) Custom User-Agent header.
+    def initialize(api_base_url, ua: nil, user_agent: ua)
       @api_base_url = Addressable::URI.parse(api_base_url)
-      @client = MediaWiktory::Wikipedia::Api.new(api_base_url, user_agent: user_agent(options))
-      @traits = Traits.get(@api_base_url.host, namespaces: extract_namespaces)
+      @api = MediaWiktory::Wikipedia::Api.new(api_base_url, user_agent: user_agent(user_agent))
+      @traits = Traits.get(@api_base_url.host, siteinfo)
     end
 
     # Receive "raw" data from Wikipedia (without parsing or wrapping in
     # classes).
     #
     # @param titles [Array<String>] List of page titles to get.
-    # @param prop [Array<Symbol>] List of additional page properties to get, refer to
-    #   [MediaWiktory::Actions::Query#prop](http://www.rubydoc.info/gems/mediawiktory/MediaWiktory/Wikipedia/Actions/Query#prop-instance_method)
-    #   for the list of available properties.
+    # @param processor [Proc] Optional block to preprocess MediaWiktory query. Refer to
+    #   [MediaWiktory::Actions::Query](http://www.rubydoc.info/gems/mediawiktory/MediaWiktory/Wikipedia/Actions/Query)
+    #   for its API. Infoboxer assumes that the block returns new instance of `Query`, so be careful
+    #   while using it.
     #
     # @return [Hash{String => Hash}] Hash of `{requested title => raw MediaWiki object}`. Note that
     #   even missing (does not exist in current Wiki) or invalid (impossible title) still be present
     #   in response, just will have `"missing"` or `"invalid"` key, just like MediaWiki returns them.
-    def raw(*titles, prop: [])
+    def raw(*titles, &processor)
       # could emerge on "automatically" created page lists, should work
       return {} if titles.empty?
 
       titles.each_slice(50).map do |part|
-        response = @client
-                   .query
-                   .titles(*part)
-                   .prop(:revisions, :info, *prop).prop(:content, :timestamp, :url)
-                   .redirects
-                   .response
+        request = prepare_request(@api.query.titles(*part), &processor)
+        response = request.response
+
+        # If additional props are required, there may be additional pages, even despite each_slice(50)
+        response = response.continue while response.continue?
 
         sources = response['pages'].values.map { |page| [page['title'], page] }.to_h
         redirects =
@@ -103,9 +105,11 @@ module Infoboxer
     # `(titles.count / 50.0).ceil` requests)
     #
     # @param titles [Array<String>] List of page titles to get.
-    # @param prop [Array<Symbol>] List of additional page properties to get, refer to
-    #   [MediaWiktory::Actions::Query#prop](http://www.rubydoc.info/gems/mediawiktory/MediaWiktory/Wikipedia/Actions/Query#prop-instance_method)
-    #   for the list of available properties.
+    # @param interwiki [Symbol] Identifier of other wiki, related to current, to fetch pages from.
+    # @param processor [Proc] Optional block to preprocess MediaWiktory query. Refer to
+    #   [MediaWiktory::Actions::Query](http://www.rubydoc.info/gems/mediawiktory/MediaWiktory/Wikipedia/Actions/Query)
+    #   for its API. Infoboxer assumes that the block returns new instance of `Query`, so be careful
+    #   while using it.
     #
     # @return [Page, Tree::Nodes<Page>] array of parsed pages. Notes:
     #   * if you call `get` with only one title, one page will be
@@ -123,8 +127,10 @@ module Infoboxer
     #     and obtain meaningful results instead of `NoMethodError` or
     #     `SomethingNotFound`.
     #
-    def get(*titles, prop: [])
-      pages = get_h(*titles, prop: prop).values.compact
+    def get(*titles, interwiki: nil, &processor)
+      return interwikis(interwiki).get(*titles, &processor) if interwiki
+
+      pages = get_h(*titles, &processor).values.compact
       titles.count == 1 ? pages.first : Tree::Nodes[*pages]
     end
 
@@ -141,14 +147,15 @@ module Infoboxer
     # you've received.
     #
     # @param titles [Array<String>] List of page titles to get.
-    # @param prop [Array<Symbol>] List of additional page properties to get, refer to
-    #   [MediaWiktory::Actions::Query#prop](http://www.rubydoc.info/gems/mediawiktory/MediaWiktory/Wikipedia/Actions/Query#prop-instance_method)
-    #   for the list of available properties.
+    # @param processor [Proc] Optional block to preprocess MediaWiktory query. Refer to
+    #   [MediaWiktory::Actions::Query](http://www.rubydoc.info/gems/mediawiktory/MediaWiktory/Wikipedia/Actions/Query)
+    #   for its API. Infoboxer assumes that the block returns new instance of `Query`, so be careful
+    #   while using it.
     #
     # @return [Hash<String, Page>]
     #
-    def get_h(*titles, prop: [])
-      raw_pages = raw(*titles, prop: prop)
+    def get_h(*titles, &processor)
+      raw_pages = raw(*titles, &processor)
                   .tap { |ps| ps.detect { |_, p| p['invalid'] }.tap { |_, i| i && fail(i['invalidreason']) } }
                   .reject { |_, p| p.key?('missing') }
       titles.map { |title| [title, make_page(raw_pages, title)] }.to_h
@@ -156,59 +163,59 @@ module Infoboxer
 
     # Receive list of parsed MediaWiki pages from specified category.
     #
-    # **NB**: currently, this API **always** fetches all pages from
-    # category, there is no option to "take first 20 pages". Pages are
-    # fetched in 50-page batches, then parsed. So, for large category
-    # it can really take a while to fetch all pages.
-    #
     # @param title [String] Category title. You can use namespaceless title (like
     #     `"Countries in South America"`), title with namespace (like
     #     `"Category:Countries in South America"`) or title with local
     #     namespace (like `"Catégorie:Argentine"` for French Wikipedia)
+    # @param limit [Integer, "max"]
+    # @param processor [Proc] Optional block to preprocess MediaWiktory query. Refer to
+    #   [MediaWiktory::Actions::Query](http://www.rubydoc.info/gems/mediawiktory/MediaWiktory/Wikipedia/Actions/Query)
+    #   for its API. Infoboxer assumes that the block returns new instance of `Query`, so be careful
+    #   while using it.
     #
     # @return [Tree::Nodes<Page>] array of parsed pages.
     #
-    def category(title)
+    def category(title, limit: 'max', &processor)
       title = normalize_category_title(title)
 
-      list(@client.query.generator(:categorymembers).title(title).limit('max'))
+      list(@api.query.generator(:categorymembers).title(title), limit, &processor)
     end
 
     # Receive list of parsed MediaWiki pages for provided search query.
     # See [MediaWiki API docs](https://www.mediawiki.org/w/api.php?action=help&modules=query%2Bsearch)
     # for details.
     #
-    # **NB**: currently, this API **always** fetches all pages from
-    # category, there is no option to "take first 20 pages". Pages are
-    # fetched in 50-page batches, then parsed. So, for large search query
-    # it can really take a while to fetch all pages.
-    #
     # @param query [String] Search query. For old installations, look at
     #     https://www.mediawiki.org/wiki/Help:Searching
     #     for search syntax. For new ones (including Wikipedia), see at
     #     https://www.mediawiki.org/wiki/Help:CirrusSearch.
+    # @param limit [Integer, "max"]
+    # @param processor [Proc] Optional block to preprocess MediaWiktory query. Refer to
+    #   [MediaWiktory::Actions::Query](http://www.rubydoc.info/gems/mediawiktory/MediaWiktory/Wikipedia/Actions/Query)
+    #   for its API. Infoboxer assumes that the block returns new instance of `Query`, so be careful
+    #   while using it.
     #
     # @return [Tree::Nodes<Page>] array of parsed pages.
     #
-    def search(query)
-      list(@client.query.generator(:search).search(query).limit('max'))
+    def search(query, limit: 'max', &processor)
+      list(@api.query.generator(:search).search(query), limit, &processor)
     end
 
     # Receive list of parsed MediaWiki pages with titles startin from prefix.
     # See [MediaWiki API docs](https://www.mediawiki.org/w/api.php?action=help&modules=query%2Bprefixsearch)
     # for details.
     #
-    # **NB**: currently, this API **always** fetches all pages from
-    # category, there is no option to "take first 20 pages". Pages are
-    # fetched in 50-page batches, then parsed. So, for large search query
-    # it can really take a while to fetch all pages.
-    #
     # @param prefix [String] Page title prefix.
+    # @param limit [Integer, "max"]
+    # @param processor [Proc] Optional block to preprocess MediaWiktory query. Refer to
+    #   [MediaWiktory::Actions::Query](http://www.rubydoc.info/gems/mediawiktory/MediaWiktory/Wikipedia/Actions/Query)
+    #   for its API. Infoboxer assumes that the block returns new instance of `Query`, so be careful
+    #   while using it.
     #
     # @return [Tree::Nodes<Page>] array of parsed pages.
     #
-    def prefixsearch(prefix)
-      list(@client.query.generator(:prefixsearch).search(prefix).limit('max'))
+    def prefixsearch(prefix, limit: 'max', &processor)
+      list(@api.query.generator(:prefixsearch).search(prefix), limit, &processor)
     end
 
     # @return [String]
@@ -224,14 +231,11 @@ module Infoboxer
       Page.new(self, Parser.paragraphs(source['revisions'].first['*'], traits), source)
     end
 
-    def list(query)
-      response = query
-                 .prop(:revisions, :info)
-                 .prop(:content, :timestamp, :url)
-                 .redirects
-                 .response
+    def list(query, limit, &processor)
+      request = prepare_request(query.limit(limit), &processor)
+      response = request.response
 
-      response = response.continue while response.continue?
+      response = response.continue while response.continue? && (limit == 'max' || response['pages'].count < limit)
 
       return Tree::Nodes[] if response['pages'].nil?
 
@@ -242,6 +246,11 @@ module Infoboxer
       Tree::Nodes[*pages]
     end
 
+    def prepare_request(request)
+      request = request.prop(:revisions, :info).prop(:content, :timestamp, :url).redirects
+      block_given? ? yield(request) : request
+    end
+
     def normalize_category_title(title)
       # FIXME: shouldn't it go to MediaWiktory?..
       namespace, titl = title.include?(':') ? title.split(':', 2) : [nil, title]
@@ -251,17 +260,26 @@ module Infoboxer
       [namespace, titl].join(':')
     end
 
-    def user_agent(options)
-      options[:user_agent] || options[:ua] || self.class.user_agent || UA
+    def user_agent(custom)
+      custom || self.class.user_agent || UA
+    end
+
+    def siteinfo
+      @siteinfo ||= @api.query.meta(:siteinfo).prop(:namespaces, :namespacealiases, :interwikimap).response.to_h
     end
 
-    def extract_namespaces
-      siteinfo = @client.query.meta(:siteinfo).prop(:namespaces, :namespacealiases).response
-      siteinfo['namespaces'].map do |_, namespace|
-        aliases =
-          siteinfo['namespacealiases'].select { |a| a['id'] == namespace['id'] }.map { |a| a['*'] }
-        namespace.merge('aliases' => aliases)
-      end
+    def interwikis(prefix)
+      @interwikis ||= Hash.new { |h, pre|
+        interwiki = siteinfo['interwikimap'].detect { |iw| iw['prefix'] == prefix } or
+          fail ArgumentError, "Undefined interwiki: #{prefix}"
+
+        # FIXME: fragile, but what can we do?..
+        m = interwiki['url'].match(%r{^(.+)/wiki/\$1$}) or
+          fail ArgumentError, "Interwiki #{interwiki} seems not to be a MediaWiki instance"
+        h[pre] = self.class.new("#{m[1]}/w/api.php") # TODO: copy useragent
+      }
+
+      @interwikis[prefix]
     end
   end
 end