Class: Jekyll::Site

Inherits:
Object
  • Object
show all
Defined in:
lib/jekyll/polyglot/patches/jekyll/site.rb

Instance Attribute Summary collapse

Instance Method Summary collapse

Instance Attribute Details

#active_lang ⇒ Object

Returns the value of attribute active_lang.



8
9
10
 
# File 'lib/jekyll/polyglot/patches/jekyll/site.rb', line 8

def active_lang
  @active_lang
end

#default_lang ⇒ Object (readonly)

Returns the value of attribute default_lang.



7
8
9
 
# File 'lib/jekyll/polyglot/patches/jekyll/site.rb', line 7

def default_lang
  @default_lang
end

#exclude_from_localization ⇒ Object (readonly)

Returns the value of attribute exclude_from_localization.



7
8
9
 
# File 'lib/jekyll/polyglot/patches/jekyll/site.rb', line 7

def exclude_from_localization
  @exclude_from_localization
end

#file_langs ⇒ Object

Returns the value of attribute file_langs.



8
9
10
 
# File 'lib/jekyll/polyglot/patches/jekyll/site.rb', line 8

def file_langs
  @file_langs
end

#lang_from_path ⇒ Object (readonly)

Returns the value of attribute lang_from_path.



7
8
9
 
# File 'lib/jekyll/polyglot/patches/jekyll/site.rb', line 7

def lang_from_path
  @lang_from_path
end

#lang_vars ⇒ Object (readonly)

Returns the value of attribute lang_vars.



7
8
9
 
# File 'lib/jekyll/polyglot/patches/jekyll/site.rb', line 7

def lang_vars
  @lang_vars
end

#languages ⇒ Object (readonly)

Returns the value of attribute languages.



7
8
9
 
# File 'lib/jekyll/polyglot/patches/jekyll/site.rb', line 7

def languages
  @languages
end

Instance Method Details

#absolute_url_regex(url, disabled = false) ⇒ Object

a regex that matches absolute urls in a html document matches href=“baseurl/foo/bar-baz” and others like it avoids matching excluded files. prepare makes sure that all @exclude dirs have a trailing slash.



265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
 
# File 'lib/jekyll/polyglot/patches/jekyll/site.rb', line 265

def absolute_url_regex(url, disabled = false)
  regex = ''
  unless disabled
    @exclude.each do |x|
      escaped_x = glob_to_regex(x)
      regex += "(?!#{escaped_x})"
    end
    @languages.each do |x|
      escaped_x = Regexp.escape(x)
      regex += "(?!#{escaped_x}/)"
    end
  end
  start = disabled ? 'ferh' : 'href'
  neglookbehind = disabled ? "" : "(?<!hreflang=\"#{@default_lang}\" |rel=\"canonical\" )"
  %r{#{neglookbehind}#{start}="?#{url}#{@baseurl}/((?:#{regex}[^,'"\s/?.]+\.?)*(?:/[^\]\[)("'\s]*)?)"}
end

#assignPageLanguagePermalinks(doc, docs) ⇒ Object 



196
197
198
199
200
201
202
203
204
205
206
207
208
 
# File 'lib/jekyll/polyglot/patches/jekyll/site.rb', line 196

def assignPageLanguagePermalinks(doc, docs)
  pageId = doc.data['page_id']
  if !pageId.nil? && !pageId.empty?
    unless doc.data['permalink_lang'] then doc.data['permalink_lang'] = {} end
    permalinkDocs = docs.select do |dd|
      dd.data['page_id'] == pageId
    end
    permalinkDocs.each do |dd|
      doclang = dd.data['lang'] || derive_lang_from_path(dd) || @default_lang
      doc.data['permalink_lang'][doclang] = dd.data['permalink']
    end
  end
end

#assignPageRedirects(doc, docs) ⇒ Object 



177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
 
# File 'lib/jekyll/polyglot/patches/jekyll/site.rb', line 177

def assignPageRedirects(doc, docs)
  pageId = doc.data['page_id']
  if !pageId.nil? && !pageId.empty?
    redirects = []

    docs_with_same_id = docs.select { |dd| dd.data['page_id'] == pageId }

    # For each document with the same page_id
    docs_with_same_id.each do |dd|
      # Add redirect if it's a different permalink
      if dd.data['permalink'] != doc.data['permalink']
        redirects << dd.data['permalink']
      end
    end

    doc.data['redirect_from'] = redirects
  end
end

#coordinate_documents(docs) ⇒ Object

assigns natural permalinks to documents and prioritizes documents with active_lang languages over others. If lang is not set in front matter, then this tries to derive from the path, if the lang_from_path is set. otherwise it will assign the document to the default_lang



148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
 
# File 'lib/jekyll/polyglot/patches/jekyll/site.rb', line 148

def coordinate_documents(docs)
  regex = document_url_regex
  approved = {}
  docs.each do |doc|
    lang = doc.data['lang'] || derive_lang_from_path(doc) || @default_lang
    lang_exclusive = doc.data['lang-exclusive'] || []
    url = doc.url.gsub(regex, '/')
    page_id = doc.data['page_id'] || url
    doc.data['permalink'] = url if doc.data['permalink'].to_s.empty? && !doc.data['lang'].to_s.empty?

    # skip entirely if nothing to check
    next if @file_langs.nil?
    # skip this document if it has already been processed
    next if @file_langs[page_id] == @active_lang
    # skip this document if it has a fallback and it isn't assigned to the active language
    next if @file_langs[page_id] == @default_lang && lang != @active_lang
    # skip this document if it has lang-exclusive defined and the active_lang is not included
    next if !lang_exclusive.empty? && !lang_exclusive.include?(@active_lang)

    approved[page_id] = doc
    @file_langs[page_id] = lang
  end
  approved.each_value do |doc|
    assignPageRedirects(doc, docs)
    assignPageLanguagePermalinks(doc, docs)
  end
  approved.values
end

#correct_nonrelativized_absolute_urls(doc, regex, url) ⇒ Object 



298
299
300
301
302
303
304
 
# File 'lib/jekyll/polyglot/patches/jekyll/site.rb', line 298

def correct_nonrelativized_absolute_urls(doc, regex, url)
  return if doc.output.nil?

  modified_output = doc.output.dup
  modified_output.gsub!(regex, "href=\"#{url}#{@baseurl}/\\1\"")
  doc.output = modified_output
end

#correct_nonrelativized_urls(doc, regex) ⇒ Object 



306
307
308
309
310
311
312
 
# File 'lib/jekyll/polyglot/patches/jekyll/site.rb', line 306

def correct_nonrelativized_urls(doc, regex)
  return if doc.output.nil?

  modified_output = doc.output.dup
  modified_output.gsub!(regex, "href=\"#{@baseurl}/\\1\"")
  doc.output = modified_output
end

#derive_lang_from_path(doc) ⇒ Object 



128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
 
# File 'lib/jekyll/polyglot/patches/jekyll/site.rb', line 128

def derive_lang_from_path(doc)
  unless @lang_from_path
    return nil
  end

  segments = split_on_multiple_delimiters(doc.path)
  # loop through all segments and check if they match the language regex
  segments.each do |segment|
    if @languages.include?(segment)
      return segment
    end
  end

  nil
end

#document_url_regex ⇒ Object

a regex that matches urls or permalinks with i18n prefixes or suffixes matches /en/foo , .en/foo , foo.en/ and other simmilar default urls made by jekyll when parsing documents without explicitly set permalinks



232
233
234
235
236
237
238
239
 
# File 'lib/jekyll/polyglot/patches/jekyll/site.rb', line 232

def document_url_regex
  regex = ''
  (@languages || []).each do |lang|
    regex += "([/.]#{lang}[/.])|"
  end
  regex.chomp! '|'
  /#{regex}/
end

#fetch_languages ⇒ Object 



24
25
26
27
28
29
30
 
# File 'lib/jekyll/polyglot/patches/jekyll/site.rb', line 24

def fetch_languages
  @default_lang = config.fetch('default_lang', 'en')
  @languages = config.fetch('languages', ['en']).uniq
  @keep_files += (@languages - [@default_lang])
  @active_lang = @default_lang
  @lang_vars = config.fetch('lang_vars', [])
end

#glob_to_regex(pattern) ⇒ Object

Convert glob pattern to regex pattern

  • matches any characters except /

? matches any single character except /



121
122
123
124
125
126
 
# File 'lib/jekyll/polyglot/patches/jekyll/site.rb', line 121

def glob_to_regex(pattern)
  # Escape special regex characters first
  escaped = Regexp.escape(pattern)
  # Convert glob patterns to regex patterns
  escaped.gsub("\\*", '.*').gsub("\\?", '.')
end

#prepare ⇒ Object 



10
11
12
13
14
15
16
17
18
19
20
21
22
 
# File 'lib/jekyll/polyglot/patches/jekyll/site.rb', line 10

def prepare
  @file_langs = {}
  fetch_languages
  @parallel_localization = config.fetch('parallel_localization', true)
  @lang_from_path = config.fetch('lang_from_path', false)
  @exclude_from_localization = config.fetch('exclude_from_localization', []).map do |e|
    if File.directory?(e) && e[-1] != '/'
      "#{e}/"
    else
      e
    end
  end
end

#process ⇒ Object 



33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
 
# File 'lib/jekyll/polyglot/patches/jekyll/site.rb', line 33

def process
  prepare
  all_langs = ([@default_lang] + @languages).uniq
  if @parallel_localization
    nproc = Etc.nprocessors
    pids = {}
    begin
      all_langs.each do |lang|
        pids[lang] = fork do
          process_language lang
        end
        while pids.length >= (lang == all_langs[-1] ? 1 : nproc)
          sleep 0.1
          pids.map do |pid_lang, pid|
            next unless waitpid pid, Process::WNOHANG

            pids.delete pid_lang
            raise "Polyglot subprocess #{pid} (#{lang}) failed (#{$CHILD_STATUS.exitstatus})" unless $CHILD_STATUS.success?
          end
        end
      end
    rescue Interrupt
      all_langs.each do |lang|
        next unless pids.key? lang

        puts "Killing #{pids[lang]} : #{lang}"
        kill('INT', pids[lang])
      end
    end
  else
    all_langs.each do |lang|
      process_language lang
    end
  end
  Jekyll::Hooks.trigger :polyglot, :post_write, self
end

#process_active_language ⇒ Object 



101
102
103
104
105
106
107
108
109
110
 
# File 'lib/jekyll/polyglot/patches/jekyll/site.rb', line 101

def process_active_language
  old_dest = @dest
  old_exclude = @exclude
  @file_langs = {}
  @dest = "#{@dest}/#{@active_lang}"
  @exclude += @exclude_from_localization
  process_orig
  @dest = old_dest
  @exclude = old_exclude
end

#process_default_language ⇒ Object 



95
96
97
98
99
 
# File 'lib/jekyll/polyglot/patches/jekyll/site.rb', line 95

def process_default_language
  old_include = @include
  process_orig
  @include = old_include
end

#process_documents(docs) ⇒ Object

performs any necessary operations on the documents before rendering them



211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
 
# File 'lib/jekyll/polyglot/patches/jekyll/site.rb', line 211

def process_documents(docs)
  # return if @active_lang == @default_lang

  url = config.fetch('url', false)
  rel_regex = relative_url_regex(false)
  abs_regex = absolute_url_regex(url, false)
  non_rel_regex = relative_url_regex(true)
  non_abs_regex = absolute_url_regex(url, true)
  docs.each do |doc|
    unless @active_lang == @default_lang then relativize_urls(doc, rel_regex) end
    correct_nonrelativized_urls(doc, non_rel_regex)
    if url
      unless @active_lang == @default_lang then relativize_absolute_urls(doc, abs_regex, url) end
      correct_nonrelativized_absolute_urls(doc, non_abs_regex, url)
    end
  end
end

#process_language(lang) ⇒ Object 



82
83
84
85
86
87
88
89
90
91
92
93
 
# File 'lib/jekyll/polyglot/patches/jekyll/site.rb', line 82

def process_language(lang)
  @active_lang = lang
  config['active_lang'] = @active_lang
  lang_vars.each do |v|
    config[v] = @active_lang
  end
  if @active_lang == @default_lang
  then process_default_language
  else
    process_active_language
  end
end

#process_orig ⇒ Object 



32
 
# File 'lib/jekyll/polyglot/patches/jekyll/site.rb', line 32

alias process_orig process

#relative_url_regex(disabled = false) ⇒ Object

a regex that matches relative urls in a html document matches href=“baseurl/foo/bar-baz” href=“/foo/bar-baz” and others like it avoids matching excluded files. prepare makes sure that all @exclude dirs have a trailing slash.



245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
 
# File 'lib/jekyll/polyglot/patches/jekyll/site.rb', line 245

def relative_url_regex(disabled = false)
  regex = ''
  unless disabled
    @exclude.each do |x|
      escaped_x = glob_to_regex(x)
      regex += "(?!#{escaped_x})"
    end
    @languages.each do |x|
      escaped_x = Regexp.escape(x)
      regex += "(?!#{escaped_x}/)"
    end
  end
  start = disabled ? 'ferh' : 'href'
  %r{#{start}="?#{@baseurl}/((?:#{regex}[^,'"\s/?.]+\.?)*(?:/[^\]\[)("'\s]*)?)"}
end

#relativize_absolute_urls(doc, regex, url) ⇒ Object 



290
291
292
293
294
295
296
 
# File 'lib/jekyll/polyglot/patches/jekyll/site.rb', line 290

def relativize_absolute_urls(doc, regex, url)
  return if doc.output.nil?

  modified_output = doc.output.dup
  modified_output.gsub!(regex, "href=\"#{url}#{@baseurl}/#{@active_lang}/\\1\"")
  doc.output = modified_output
end

#relativize_urls(doc, regex) ⇒ Object 



282
283
284
285
286
287
288
 
# File 'lib/jekyll/polyglot/patches/jekyll/site.rb', line 282

def relativize_urls(doc, regex)
  return if doc.output.nil?

  modified_output = doc.output.dup
  modified_output.gsub!(regex, "href=\"#{@baseurl}/#{@active_lang}/\\1\"")
  doc.output = modified_output
end

#site_payload ⇒ Object 



71
72
73
74
75
76
77
78
79
80
 
# File 'lib/jekyll/polyglot/patches/jekyll/site.rb', line 71

def site_payload
  payload = site_payload_orig
  payload['site']['default_lang'] = default_lang
  payload['site']['languages'] = languages
  payload['site']['active_lang'] = active_lang
  lang_vars.each do |v|
    payload['site'][v] = active_lang
  end
  payload
end

#site_payload_orig ⇒ Object 



70
 
# File 'lib/jekyll/polyglot/patches/jekyll/site.rb', line 70

alias site_payload_orig site_payload

#split_on_multiple_delimiters(string) ⇒ Object 



112
113
114
115
116
 
# File 'lib/jekyll/polyglot/patches/jekyll/site.rb', line 112

def split_on_multiple_delimiters(string)
  delimiters = ['.', '/']
  regex = Regexp.union(delimiters)
  string.split(regex)
end