infoboxer 0.3.2 → 0.3.3

checksums.yaml

@@ -1,7 +1,7 @@
 ---
-SHA1:
-  metadata.gz: d1eca6a2e6e025b77b1eeed915629b24338380a8
-  data.tar.gz: a852e3d6cefa55b04b2a8ffe530478e8199a884f
+SHA256:
+  metadata.gz: 36fa882706af5f4b483c3723a179934eb906842f4de8c5118a9cf5b0dd91fec1
+  data.tar.gz: 0e11e6635fab15e01a2b4cf53eb8fa209938121f582bf1e925cbeeb841dd7478
 SHA512:
-  metadata.gz: 0cb20f539dcb4fecaf1f3c57a842d407481e37a2763ee35018bd436b058e704382493fd0a0a6a008101a70e9d9c283cc9fa3b7dbd8f48deef894e8d857ea7c42
-  data.tar.gz: b4f9ab8d39bc3c5fca6f7f247fd8c7f54bf76de8b5ed878a9fb935c6ccf3d99f7ed768c16cd1b1529ba56742a0bb1b6a09dafd0fc2900c12f2c0a39264d938fb
+  metadata.gz: 35825da3980f77f9fbaaa91e3cde9d3c2e40c02249577a6a2f337dd5077c247cfbeef97c6e714d1c529cd109f91bf7555e4d8e67f7c4d32a226508d004a2e86b
+  data.tar.gz: df6abe08b07cf597bc0c196900ddffbeff023cbdfa8ec5b185c343c6fa4dd2e9cd924c92ed1deb793e96f285b287360a6c27c32489d3a970e046dbedc481529b

data/CHANGELOG.md

@@ -1,5 +1,11 @@
 # Infoboxer's change log
 
+## 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;

data/Gemfile.lock

@@ -0,0 +1,150 @@
+GIT
+  remote: https://github.com/zverok/dokaz.git
+  revision: a8a6f0bbeab5589326fe2714cf89842b5f32b850
+  specs:
+    dokaz (0.0.4)
+      ansi
+      rouge
+      slop (~> 3)
+
+PATH
+  remote: .
+  specs:
+    infoboxer (0.3.3)
+      addressable
+      htmlentities
+      mediawiktory (= 0.1.3)
+      terminal-table
+
+GEM
+  remote: https://rubygems.org/
+  specs:
+    addressable (2.5.1)
+      public_suffix (~> 2.0, >= 2.0.2)
+    ansi (1.5.0)
+    ast (2.4.0)
+    backports (3.10.3)
+    byebug (9.0.6)
+    coveralls (0.8.21)
+      json (>= 1.8, < 3)
+      simplecov (~> 0.14.1)
+      term-ansicolor (~> 1.3)
+      thor (~> 0.19.4)
+      tins (~> 1.6)
+    crack (0.4.3)
+      safe_yaml (~> 1.0.0)
+    diff-lcs (1.3)
+    docile (1.1.5)
+    faraday (0.15.4)
+      multipart-post (>= 1.2, < 3)
+    faraday_middleware (0.13.0)
+      faraday (>= 0.7.4, < 1.0)
+    hashdiff (0.3.4)
+    hashie (3.6.0)
+    htmlentities (4.3.4)
+    json (2.1.0)
+    json (2.1.0-java)
+    mediawiktory (0.1.3)
+      addressable
+      faraday
+      faraday_middleware
+      hashie
+      naught
+      nokogiri
+    mini_portile2 (2.4.0)
+    multipart-post (2.1.1)
+    naught (1.1.0)
+    nokogiri (1.10.4)
+      mini_portile2 (~> 2.4.0)
+    parallel (1.19.1)
+    parser (2.7.0.2)
+      ast (~> 2.4.0)
+    powerpack (0.1.2)
+    public_suffix (2.0.5)
+    rainbow (3.0.0)
+    rake (12.3.0)
+    redcarpet (3.3.4)
+    rouge (2.1.1)
+    rspec (3.6.0)
+      rspec-core (~> 3.6.0)
+      rspec-expectations (~> 3.6.0)
+      rspec-mocks (~> 3.6.0)
+    rspec-core (3.6.0)
+      rspec-support (~> 3.6.0)
+    rspec-expectations (3.6.0)
+      diff-lcs (>= 1.2.0, < 2.0)
+      rspec-support (~> 3.6.0)
+    rspec-its (1.2.0)
+      rspec-core (>= 3.0.0)
+      rspec-expectations (>= 3.0.0)
+    rspec-mocks (3.6.0)
+      diff-lcs (>= 1.2.0, < 2.0)
+      rspec-support (~> 3.6.0)
+    rspec-support (3.6.0)
+    rubocop (0.52.1)
+      parallel (~> 1.10)
+      parser (>= 2.4.0.2, < 3.0)
+      powerpack (~> 0.1)
+      rainbow (>= 2.2.2, < 4.0)
+      ruby-progressbar (~> 1.7)
+      unicode-display_width (~> 1.0, >= 1.0.1)
+    rubocop-rspec (1.20.1)
+      rubocop (>= 0.51.0)
+    ruby-prof (0.16.2)
+    ruby-progressbar (1.10.1)
+    rubygems-tasks (0.2.4)
+    safe_yaml (1.0.4)
+    saharspec (0.0.4)
+    simplecov (0.14.1)
+      docile (~> 1.1.0)
+      json (>= 1.8, < 3)
+      simplecov-html (~> 0.10.0)
+    simplecov-html (0.10.1)
+    slop (3.6.0)
+    term-ansicolor (1.6.0)
+      tins (~> 1.0)
+    terminal-table (1.8.0)
+      unicode-display_width (~> 1.1, >= 1.1.1)
+    thor (0.19.4)
+    timecop (0.9.0)
+    tins (1.14.0)
+    tty-color (0.4.2)
+    unicode-display_width (1.6.1)
+    vcr (3.0.3)
+    webmock (3.0.1)
+      addressable (>= 2.3.6)
+      crack (>= 0.3.2)
+      hashdiff
+    yard (0.9.12)
+    yard-junk (0.0.7)
+      backports
+      rainbow
+      tty-color
+      yard
+
+PLATFORMS
+  java
+  ruby
+
+DEPENDENCIES
+  byebug
+  coveralls
+  dokaz!
+  infoboxer!
+  rake
+  redcarpet
+  rspec (~> 3)
+  rspec-its (~> 1)
+  rubocop (~> 0.52.1)
+  rubocop-rspec (~> 1.20)
+  ruby-prof
+  rubygems-tasks
+  saharspec (= 0.0.4)
+  timecop
+  vcr
+  webmock
+  yard (~> 0.9)
+  yard-junk (~> 0.0.7)
+
+BUNDLED WITH
+   1.17.2

data/lib/infoboxer/parser/context.rb

@@ -2,7 +2,7 @@ require 'strscan'
 
 module Infoboxer
   class Parser
-    class Context
+    class Context # rubocop:disable Metrics/ClassLength
       attr_reader :lineno
       attr_reader :traits
 
@@ -128,7 +128,13 @@ module Infoboxer
 
       # state inspection
       def matched_inline?(re)
-        re.nil? ? (matched.empty? && eol?) : matched =~ re
+        if re.nil?
+          matched.empty? && eol?
+        elsif re.inspect.start_with?('/^') # was it REALLY at the beginning of the line?..
+          @scanner.pos == matched.length && matched =~ re
+        else
+          matched =~ re
+        end
       end
 
       def matched?(re)
@@ -144,6 +150,12 @@ module Infoboxer
         fail(ParsingError, "#{text} at line #{@lineno}:\n\t#{current}")
       end
 
+      def unscan_matched!
+        return unless @matched
+        @scanner.pos -= @matched.size
+        @rest = nil
+      end
+
       private
 
       # we do hard use of #matched and #rest, its wiser to memoize them

data/lib/infoboxer/parser/table.rb

@@ -16,8 +16,9 @@ module Infoboxer
 
         @context.next!
 
-        loop do
+        guarded_loop do
           table_next_line(table) or break
+          log 'Next table row'
           @context.next!
         end
 
@@ -66,9 +67,18 @@ module Infoboxer
         log 'Table caption found'
         @context.skip(/^\s*\|\+\s*/)
 
+        params = if @context.check(/[^|{|\[]+\|([^\|]|$)/)
+                   parse_params(@context.scan_until(/\|/))
+                 else
+                   {}
+                 end
+
         children = inline(/^\s*([|!]|{\|)/)
-        @context.prev! if @context.eol? # compensate next! which will be done in table()
-        table.push_children(TableCaption.new(children.strip))
+        if @context.matched
+          @context.unscan_matched!
+          @context.prev! # compensate next! which will be done in table()
+        end
+        table.push_children(TableCaption.new(children.strip, params))
       end
 
       def table_cells(table, cell_class = TableCell)

data/lib/infoboxer/version.rb

@@ -1,7 +1,7 @@
 module Infoboxer
   MAJOR = 0
   MINOR = 3
-  PATCH = 2
+  PATCH = 3
   PRE = nil
   VERSION = [MAJOR, MINOR, PATCH, PRE].compact.join('.')
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: infoboxer
 version: !ruby/object:Gem::Version
-  version: 0.3.2
+  version: 0.3.3
 platform: ruby
 authors:
 - Victor Shepelev
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2018-02-09 00:00:00.000000000 Z
+date: 2020-02-09 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: htmlentities
@@ -81,6 +81,7 @@ files:
 - ".yardopts"
 - CHANGELOG.md
 - CONTRIBUTING.md
+- Gemfile.lock
 - LICENSE.txt
 - Parsing.md
 - README.md
@@ -170,8 +171,7 @@ required_rubygems_version: !ruby/object:Gem::Requirement
     - !ruby/object:Gem::Version
       version: '0'
 requirements: []
-rubyforge_project: 
-rubygems_version: 2.6.14
+rubygems_version: 3.0.3
 signing_key: 
 specification_version: 4
 summary: MediaWiki client and parser, targeting information extraction.