Module: Metanorma::Utils

Defined in:
lib/utils/image.rb,
lib/utils/cjk.rb,
lib/utils/log.rb,
lib/utils/xml.rb,
lib/utils/main.rb,
lib/utils/version.rb,
lib/utils/namespace.rb,
lib/utils/linestatus.rb,
lib/utils/hash_transform_keys.rb,
lib/utils/hash_transform_keys.rb

Overview

Image methods were moved to the Vectory gem

Defined Under Namespace

Modules: Array, Hash Classes: LineStatus, Log, Namespace

Constant Summary collapse

HAN =

Basic CJK scripts

"\\p{Han}".freeze
BOPOMOFO = 
"\\p{Bopomofo}".freeze
HANGUL = 
"\\p{Hangul}".freeze
HIRAGANA = 
"\\p{Hiragana}".freeze
KATAKANA = 
"\\p{Katakana}".freeze
CJK_SYMBOLS =

CJK Symbols and Punctuation (U+3000–U+303F) Used across all CJK scripts

"[\\u3000-\\u303F]".freeze
CJK_PUNCTUATION =

CJK Punctuation (subset of CJK Symbols commonly used)

"[\\u3001-\\u3003\\u3008-\\u3011\\u3014-\\u301F]".freeze
CJK_HALFWIDTH_FULLWIDTH =

Halfwidth and Fullwidth Forms (U+FF00–U+FFEF) Used in all CJK contexts

"[\\uFF00-\\uFFEF]".freeze
CJK_COMPAT =

CJK Compatibility Forms (U+FE30–U+FE4F) Primarily used with Han but relevant for all CJK

"[\\uFE30-\\uFE4F]".freeze
CJK_VERTICAL =

Vertical Forms (U+FE10–U+FE1F) Used in vertical text layout for all CJK

"[\\uFE10-\\uFE1F]".freeze
CJK_SMALL_FORMS =

Small Form Variants (U+FE50–U+FE6F) Used in all CJK contexts

"[\\uFE50-\\uFE6F]".freeze
HAN_IDC =

Ideographic Description Characters (U+2FF0–U+2FFF) Used with Han script

"[\\u2FF0-\\u2FFF]".freeze
KANBUN =

Kanbun (U+3190–U+319F) Used with Han script for Japanese

"[\\u3190-\\u319F]".freeze
CJK_COMPAT_IDEOGRAPHS =

CJK Compatibility (U+3300–U+33FF) Used with Han script

"[\\u3300-\\u33FF]".freeze
HAN_COMPAT_IDEOGRAPHS =

CJK Compatibility Ideographs (U+F900–U+FAFF)

"[\\uF900-\\uFAFF]".freeze
HAN_EXTENSIONS =

Script extensions by primary script

[
  HAN,
  CJK_SYMBOLS,
  CJK_PUNCTUATION,
  CJK_HALFWIDTH_FULLWIDTH,
  CJK_COMPAT,
  CJK_VERTICAL,
  CJK_SMALL_FORMS,
  HAN_IDC,
  KANBUN,
  CJK_COMPAT_IDEOGRAPHS,
  HAN_COMPAT_IDEOGRAPHS
].join("|").freeze
HANGUL_EXTENSIONS = 
[
  HANGUL,
  CJK_SYMBOLS,
  CJK_PUNCTUATION,
  CJK_HALFWIDTH_FULLWIDTH,
  CJK_VERTICAL,
  CJK_SMALL_FORMS
].join("|").freeze
HIRAGANA_EXTENSIONS = 
[
  HIRAGANA,
  CJK_SYMBOLS,
  CJK_PUNCTUATION,
  CJK_HALFWIDTH_FULLWIDTH,
  CJK_VERTICAL,
  CJK_SMALL_FORMS
].join("|").freeze
KATAKANA_EXTENSIONS = 
[
  KATAKANA,
  CJK_SYMBOLS,
  CJK_PUNCTUATION,
  CJK_HALFWIDTH_FULLWIDTH,
  CJK_VERTICAL,
  CJK_SMALL_FORMS
].join("|").freeze
BOPOMOFO_EXTENSIONS = 
[
  BOPOMOFO,
  CJK_SYMBOLS,
  CJK_PUNCTUATION,
  CJK_HALFWIDTH_FULLWIDTH
].join("|").freeze
CJK =

Combined CJK pattern including all script extensions

[
  HAN_EXTENSIONS,
  HANGUL_EXTENSIONS,
  HIRAGANA_EXTENSIONS,
  KATAKANA_EXTENSIONS,
  BOPOMOFO_EXTENSIONS
].join("|").freeze
NAMECHAR = 
"\u0000-\u002c\u002f\u003a-\u0040\\u005b-\u005e" \
"\u0060\u007b-\u00b6\u00b8-\u00bf\u00d7\u00f7\u037e" \
"\u2000-\u200b" \
"\u200e-\u203e\u2041-\u206f\u2190-\u2bff\u2ff0-\u3000".freeze
NAMESTARTCHAR = 
"\\u002d\u002e\u0030-\u0039\u00b7\u0300-\u036f" \
"\u203f-\u2040".freeze
NOKOHEAD = 
<<~HERE.freeze
  <!DOCTYPE html SYSTEM
  "http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd">
  <html xmlns="http://www.w3.org/1999/xhtml">
  <head> <title></title> <meta charset="UTF-8" /> </head>
  <body> </body> </html>
HERE
LONGSTR_THRESHOLD = 
10
LONGSTR_NOPUNCT = 
2
STR_BREAKUP_RE = 
%r{
 (?<=[=_—–\u2009→?+;]) | # break after any of these
 (?<=[,.:])(?!\d) | # break on punct only if not preceding digit
 (?<=[>])(?![>]) | # > not >->
 (?<=[\]])(?![\]]) | # ] not ]-]
 (?<=//) | # //
 (?<=[/])(?![/]) | # / not /-/
 (?<![<])(?=[<]) | # < not <-<
 (?<=\p{L})(?=[(\{\[]\p{L}) # letter and bracket, followed by letter
}x.freeze
CAMEL_CASE_RE = 
%r{
  (?<=\p{Ll}\p{Ll})(?=\p{Lu}\p{Ll}\p{Ll}) # 2 lowerc / upperc, 2 lowerc
}x.freeze
VERSION = 
"1.11.2".freeze

Class Method Summary collapse

Class Method Details

.anchor_attributes(presxml: false) ⇒ Object

all element/attribute pairs that are ID anchors in Metanorma



128
129
130
131
132
133
134
135
136
137
138
139
140
 
# File 'lib/utils/xml.rb', line 128

def anchor_attributes(presxml: false)
  ret = [%w(review from), %w(review to), %w(callout target), %w(xref to),
         %w(eref bibitemid), %w(citation bibitemid), %w(xref target),
         %w(label for), %w(location target), %w(index to),
         %w(termsource bibitemid), %w(admonition target)]
  ret1 = [%w(fn target), %w(semx source), %w(fmt-title source),
          %w(fmt-xref to), %w(fmt-xref target), %w(fmt-eref bibitemid),
          %w(fmt-xref-label container), %w(fmt-fn-body target),
          %w(fmt-review-start source), %w(fmt-review-start end),
          %w(fmt-review-start target), %w(fmt-review-end source),
          %w(fmt-review-end start), %w(fmt-review-end target)]
  presxml ? ret + ret1 : ret
end

.anchor_or_uuid(node = nil) ⇒ Object 



43
44
45
46
 
# File 'lib/utils/xml.rb', line 43

def anchor_or_uuid(node = nil)
  uuid = UUIDTools::UUID.random_create
  node.nil? || node.id.nil? || node.id.empty? ? "_#{uuid}" : node.id
end

.asciidoc_sub(text, flavour = :standoc) ⇒ Object 



22
23
24
25
26
27
28
29
30
31
32
 
# File 'lib/utils/main.rb', line 22

def asciidoc_sub(text, flavour = :standoc)
  return nil if text.nil?
  return "" if text.empty?

  d = Asciidoctor::Document.new(
    text.lines.entries,
    { header_footer: false, backend: flavour },
  )
  b = d.parse.blocks.first
  b.apply_subs(b.source)
end

.attr_code(attributes) ⇒ Object 



24
25
26
27
28
 
# File 'lib/utils/xml.rb', line 24

def attr_code(attributes)
  attributes.compact.transform_values do |v|
    v.is_a?(String) ? HTMLEntities.new.decode(v) : v
  end
end

.break_up_long_str(text, threshold = LONGSTR_THRESHOLD, nopunct = LONGSTR_NOPUNCT) ⇒ Object

break on punct every LONGSTRING_THRESHOLD chars, with zero width space if punct fails, try break on camel case, with soft hyphen break regardless every LONGSTRING_THRESHOLD * LONGSTR_NOPUNCT, with soft hyphen



140
141
142
143
144
145
146
147
148
149
150
151
 
# File 'lib/utils/main.rb', line 140

def break_up_long_str(text, threshold = LONGSTR_THRESHOLD,
nopunct = LONGSTR_NOPUNCT)
  /^\s*$/.match?(text) and return text
  text.split(/(?=(?:\s|-))/).map do |w|
    if /^\s*$/.match(w) || (w.size < threshold) then w
    else
      w.scan(/.{,#{threshold}}/o).map.with_index do |w1, i|
        w1.size < threshold ? w1 : break_up_long_str1(w1, i + 1, nopunct)
      end.join
    end
  end.join
end

.break_up_long_str1(text, iteration, nopunct) ⇒ Object 



168
169
170
171
172
173
174
175
176
177
178
 
# File 'lib/utils/main.rb', line 168

def break_up_long_str1(text, iteration, nopunct)
  s, separator = break_up_long_str2(text)
  if s.size == 1 # could not break up
    (iteration % nopunct).zero? and
      text += "\u00ad" # force soft hyphen
    text
  else
    s[-1] = "#{separator}#{s[-1]}"
    s.join
  end
end

.break_up_long_str2(text) ⇒ Object 



180
181
182
183
184
185
186
187
188
 
# File 'lib/utils/main.rb', line 180

def break_up_long_str2(text)
  s = text.split(STR_BREAKUP_RE, -1)
  separator = "\u200b"
  if s.size == 1
    s = text.split(CAMEL_CASE_RE)
    separator = "\u00ad"
  end
  [s, separator]
end

.case_transform_xml(xml, kase) ⇒ Object 



167
168
169
170
171
172
173
174
 
# File 'lib/utils/xml.rb', line 167

def case_transform_xml(xml, kase)
  x = Nokogiri::XML("<root>#{xml}</root>")
  x.traverse do |e|
    e.text? or next
    e.replace(e.text.send(kase))
  end
  x.root.children.to_xml
end

.create_namespace(xmldoc) ⇒ Object 



21
22
23
 
# File 'lib/utils/namespace.rb', line 21

def create_namespace(xmldoc)
  Namespace.new(xmldoc)
end

.csv_split(text, delim = ";") ⇒ Object

, “ => ,” : CSV definition does not deal with space followed by quote at start of field



15
16
17
18
19
20
 
# File 'lib/utils/main.rb', line 15

def csv_split(text, delim = ";")
  text.nil? || text.empty? and return []
  CSV.parse_line(text.gsub(/#{delim} "(?!")/, "#{delim}\""),
                 liberal_parsing: true,
                 col_sep: delim)&.compact&.map(&:strip)
end

.default_script(lang) ⇒ Object 



113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
 
# File 'lib/utils/main.rb', line 113

def default_script(lang)
  case lang
  when "ar", "fa" then "Arab"
  when "ur" then "Aran"
  when "ru", "bg" then "Cyrl"
  when "hi" then "Deva"
  when "el" then "Grek"
  when "zh" then "Hans"
  when "ko" then "Kore"
  when "he" then "Hebr"
  when "ja" then "Jpan"
  else
    "Latn"
  end
end

.dl_to_attrs(elem, dlist, name) ⇒ Object

convert definition list term/value pair into Nokogiri XML attribute



143
144
145
146
147
 
# File 'lib/utils/xml.rb', line 143

def dl_to_attrs(elem, dlist, name)
  e = dlist.at("./dt[text()='#{name}']") or return
  val = e.at("./following::dd/p") || e.at("./following::dd") or return
  elem[name] = val.text
end

.dl_to_elems(ins, elem, dlist, name) ⇒ Object

convert definition list term/value pairs into Nokogiri XML elements



150
151
152
153
154
155
156
157
 
# File 'lib/utils/xml.rb', line 150

def dl_to_elems(ins, elem, dlist, name)
  a = elem.at("./#{name}[last()]")
  ins = a if a
  dlist.xpath("./dt[text()='#{name}']").each do |e|
    ins = dl_to_elems1(e, name, ins)
  end
  ins
end

.dl_to_elems1(term, name, ins) ⇒ Object 



159
160
161
162
163
164
165
 
# File 'lib/utils/xml.rb', line 159

def dl_to_elems1(term, name, ins)
  v = term.at("./following::dd")
  e = v.elements and e.size == 1 && e.first.name == "p" and v = e.first
  v.name = name
  ins.next = v
  ins.next
end

.endash_date(elem) ⇒ Object 



53
54
55
56
57
58
 
# File 'lib/utils/main.rb', line 53

def endash_date(elem)
  elem.traverse do |n|
    n.text? or next
    n.replace(n.text.gsub(/\s+--?\s+/, "&#8211;").gsub("--", "&#8211;"))
  end
end

.external_path(path) ⇒ Object 



102
103
104
105
106
107
108
109
110
111
 
# File 'lib/utils/main.rb', line 102

def external_path(path)
  win = !!((RUBY_PLATFORM =~ /(win|w)(32|64)$/) ||
           (RUBY_PLATFORM =~ /mswin|mingw/))
  if win
    path.gsub!(%{/}, "\\")
    path[/\s/] ? "\"#{path}\"" : path
  else
    path
  end
end

.firstchar_xml(line) ⇒ Object

need to deal with both <em> and its reverse string, >me<



80
81
82
83
 
# File 'lib/utils/xml.rb', line 80

def firstchar_xml(line)
  m = /^([<>][^<>]+[<>])*(.)/.match(line) or return ""
  m[2]
end

.guid_anchor?(id) ⇒ Boolean

Returns:

  • (Boolean) 


176
177
178
179
 
# File 'lib/utils/xml.rb', line 176

def guid_anchor?(id)
  /^_[0-9A-F]{8}-[0-9A-F]{4}-[0-9A-F]{4}-[0-9A-F]{4}-[0-9A-F]{12}$/i
    .match?(id)
end

.line_sanitise(ret) ⇒ Object

By default, carriage return in source translates to whitespace; but in CJK, it does not. (Non-CJK text n CJK)



63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
 
# File 'lib/utils/xml.rb', line 63

def line_sanitise(ret)
  ret.size == 1 and return ret
  (0...(ret.size - 1)).each do |i|
    last = firstchar_xml(ret[i].reverse)
    nextfirst = firstchar_xml(ret[i + 1])
    cjk1 = /#{CJK}/o.match?(last)
    cjk2 = /#{CJK}/o.match?(nextfirst)
    text1 = /[^\p{Z}\p{C}]/.match?(last)
    text2 = /[^\p{Z}\p{C}]/.match?(nextfirst)
    cjk1 && (cjk2 || !text2) and next
    !text1 && cjk2 and next
    ret[i] += " "
  end
  ret
end

.localdir(node) ⇒ Object 



34
35
36
37
 
# File 'lib/utils/main.rb', line 34

def localdir(node)
  docfile = node.attr("docfile")
  docfile.nil? ? "./" : "#{Pathname.new(docfile).parent}/"
end

.noko(_script = "Latn", &block) ⇒ Object

block for processing XML document fragments as XHTML, to allow for HTMLentities Unescape special chars used in Asciidoctor substitution processing



51
52
53
54
55
56
57
58
59
 
# File 'lib/utils/xml.rb', line 51

def noko(_script = "Latn", &block)
  fragment = ::Nokogiri::XML.parse(NOKOHEAD).fragment("")
  ::Nokogiri::XML::Builder.with fragment, &block
  fragment
    .to_xml(encoding: "UTF-8", indent: 0,
            save_with: Nokogiri::XML::Node::SaveOptions::AS_XML)
    .gsub("&#150;", "\u0096").gsub("&#151;", "\u0097")
    .gsub("&#x96;", "\u0096").gsub("&#x97;", "\u0097")
end

.noko_html(&block) ⇒ Object 



85
86
87
88
89
90
91
92
93
94
 
# File 'lib/utils/xml.rb', line 85

def noko_html(&block)
  doc = ::Nokogiri::XML.parse(NOKOHEAD)
  fragment = doc.fragment("")
  ::Nokogiri::XML::Builder.with fragment, &block
  fragment.to_xml(encoding: "UTF-8", indent: 0,
                  save_with: Nokogiri::XML::Node::SaveOptions::AS_XML)
    .lines.map do |l|
    l.gsub(/\s*\n/, "")
  end
end

.ns(xpath) ⇒ Object 



101
102
103
104
105
106
 
# File 'lib/utils/xml.rb', line 101

def ns(xpath)
  xpath.gsub(%r{/([a-zA-Z])}, "/xmlns:\\1")
    .gsub(%r{::([a-zA-Z])}, "::xmlns:\\1")
    .gsub(%r{\[([a-zA-Z][a-z0-9A-Z@/-]* ?=)}, "[xmlns:\\1")
    .gsub(%r{\[([a-zA-Z][a-z0-9A-Z@/-]*[/\[\]])}, "[xmlns:\\1")
end

.numeric_escapes(xml) ⇒ Object 



108
109
110
111
112
113
114
115
116
 
# File 'lib/utils/xml.rb', line 108

def numeric_escapes(xml)
  c = HTMLEntities.new
  xml.split(/(&[^ \r\n\t#&;]+;)/).map do |t|
    if /^(&[^ \t\r\n#;]+;)/.match?(t)
      c.encode(c.decode(t), :hexadecimal)
    else t
    end
  end.join
end

.rtl_script?(script) ⇒ Boolean

Returns:

  • (Boolean) 


129
130
131
 
# File 'lib/utils/main.rb', line 129

def rtl_script?(script)
  %w(Arab Aran Hebr).include? script
end

.set_nested_value(hash, keys, new_val) ⇒ Object

Set hash value using keys path mod from stackoverflow.com/a/42425884



62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
 
# File 'lib/utils/main.rb', line 62

def set_nested_value(hash, keys, new_val)
  key = keys[0]
  if keys.length == 1
    hash[key] = if hash[key].is_a?(::Array) then (hash[key] << new_val)
                else hash[key].nil? ? new_val : [hash[key], new_val]
                end
  elsif hash[key].is_a?(::Array)
    hash[key][-1] = {} if !hash[key].empty? && hash[key][-1].nil?
    hash[key] << {} if hash[key].empty? || !hash[key][-1].is_a?(::Hash)
    set_nested_value(hash[key][-1], keys[1..-1], new_val)
  elsif hash[key].nil? || hash[key].empty?
    hash[key] = {}
    set_nested_value(hash[key], keys[1..-1], new_val)
  elsif hash[key].is_a?(::Hash) && !hash[key][keys[1]]
    set_nested_value(hash[key], keys[1..-1], new_val)
  elsif !hash[key][keys[1]]
    hash[key] = [hash[key], {}]
    set_nested_value(hash[key][-1], keys[1..-1], new_val)
  else
    set_nested_value(hash[key], keys[1..-1], new_val)
  end
  hash
end

.smartformat(text) ⇒ Object

TODO needs internationalisation of quote



40
41
42
43
44
45
46
47
48
49
50
51
 
# File 'lib/utils/main.rb', line 40

def smartformat(text)
  ret = HTMLEntities.new.decode(
    text.gsub(/ --? /, "&#8201;&#8212;&#8201;")
    .gsub("--", "&#8212;"),
  )
  ret = ret.gsub(%r{(#{CJK})(["'])}o, "\\1\u200a\\2")
    .gsub(%r{(["'])(#{CJK})}o, "\\1\u200a\\2")
  ret = ret.smart_format
  ret = ret.gsub(%r{(#{CJK})\u200a}o, "\\1")
    .gsub(%r{\u200a(#{CJK})}o, "\\1")
  HTMLEntities.new.encode(ret, :basic)
end

.strict_capitalize_first(str) ⇒ Object 



94
95
96
97
98
99
100
 
# File 'lib/utils/main.rb', line 94

def strict_capitalize_first(str)
  str.split(/ /).each_with_index.map do |w, i|
    letters = w.chars
    letters.first.upcase! if i.zero?
    letters.join
  end.join(" ")
end

.strict_capitalize_phrase(str) ⇒ Object 



86
87
88
89
90
91
92
 
# File 'lib/utils/main.rb', line 86

def strict_capitalize_phrase(str)
  str.split(/ /).map do |w|
    letters = w.chars
    letters.first.upcase!
    letters.join
  end.join(" ")
end

.to_ncname(tag, asciionly: true) ⇒ Object 



30
31
32
33
34
35
36
37
38
39
40
41
 
# File 'lib/utils/xml.rb', line 30

def to_ncname(tag, asciionly: true)
  asciionly and tag = HTMLEntities.new.encode(tag, :basic,
                                              :hexadecimal)
  start = tag[0]
  ret1 = if %r([#{NAMECHAR}#])o.match?(start)
           "_"
         else
           (%r([#{NAMESTARTCHAR}#])o.match?(start) ? "_#{start}" : start)
         end
  ret2 = tag[1..-1] || ""
  (ret1 || "") + ret2.gsub(%r([#{NAMECHAR}#])o, "_")
end

.to_xhtml_fragment(xml) ⇒ Object 



96
97
98
99
 
# File 'lib/utils/xml.rb', line 96

def to_xhtml_fragment(xml)
  doc = ::Nokogiri::XML.parse(NOKOHEAD)
  doc.fragment(xml)
end

.wrap_in_para(node, out) ⇒ Object

if the contents of node are blocks, output them to out; else, wrap them in <p>



120
121
122
123
124
125
 
# File 'lib/utils/xml.rb', line 120

def wrap_in_para(node, out)
  if node.blocks? then out << node.content
  else
    out.p { |p| p << node.content }
  end
end