Methods for manipulating comment text
Maps markup formats to classes that can parse them. If the format is unknown, “rdoc” format is used.
Maps an encoding to a Hash of characters properly transcoded for that encoding.
See also encode_fallback.
Transcodes character
to encoding
with a
fallback
character.
# File lib/rdoc/text.rb, line 58
def self.encode_fallback character, encoding, fallback
character.encode(encoding, :fallback => { character => fallback },
:undef => :replace, :replace => fallback)
end
Expands tab characters in text
to eight spaces
# File lib/rdoc/text.rb, line 66
def expand_tabs text
expanded = []
text.each_line do |line|
line.gsub!(%r^(.{8}*?)([^\t\r\n]{0,7})\t/) do
r = "#{$1}#{$2}#{' ' * (8 - $2.size)}"
r.force_encoding text.encoding if Object.const_defined? :Encoding
r
end until line !~ %r\t/
expanded << line
end
expanded.join
end
Flush text
left based on the shortest line
# File lib/rdoc/text.rb, line 85
def flush_left text
indent = 9999
text.each_line do |line|
line_indent = line =~ %r\S/ || 9999
indent = line_indent if indent > line_indent
end
empty = ''
empty.force_encoding text.encoding if Object.const_defined? :Encoding
text.gsub(%r^ {0,#{indent}}/, empty)
end
Convert a string in markup format into HTML.
Requires the including class to implement formatter
# File lib/rdoc/text.rb, line 104
def markup text
parse(text).accept formatter
end
Strips hashes, expands tabs then flushes text
to the left
# File lib/rdoc/text.rb, line 111
def normalize_comment text
return text if text.empty?
text = strip_stars text
text = strip_hashes text
text = expand_tabs text
text = flush_left text
text = strip_newlines text
text
end
Normalizes text
then builds a RDoc::Markup::Document from it
# File lib/rdoc/text.rb, line 125
def parse text, format = 'rdoc'
return text if RDoc::Markup::Document === text
return text.parse if RDoc::Comment === text
text = normalize_comment text # TODO remove, should not be necessary
return RDoc::Markup::Document.new if text =~ %r\A\n*\z/
MARKUP_FORMAT[format].parse text
end
The first limit
characters of text
as HTML
# File lib/rdoc/text.rb, line 139
def snippet text, limit = 100
document = parse text
RDoc::Markup::ToHtmlSnippet.new(limit).convert document
end
Strips leading # characters from text
# File lib/rdoc/text.rb, line 148
def strip_hashes text
return text if text =~ %r^(?>\s*)[^\#]/
empty = ''
empty.force_encoding text.encoding if Object.const_defined? :Encoding
text.gsub(%r^\s*(#+)/) { $1.tr '#', ' ' }.gsub(%r^\s+$/, empty)
end
Strips leading and trailing n characters from text
# File lib/rdoc/text.rb, line 160
def strip_newlines text
text.gsub(%r\A\n*(.*?)\n*\z/) do $1 end # block preserves String encoding
end
Strips style comments
# File lib/rdoc/text.rb, line 167
def strip_stars text
return text unless text =~ %r/\*.*\*/%
encoding = text.encoding if Object.const_defined? :Encoding
text = text.gsub %rDocument-method:\s+[\w:.#=!?]+%, ''
space = ' '
space.force_encoding encoding if encoding
text.sub! %r/\*+% do space * $&.length end
text.sub! %r\*+/% do space * $&.length end
text.gsub! %r^[ \t]*\*% do space * $&.length end
empty = ''
empty.force_encoding encoding if encoding
text.gsub(%r^\s+$/, empty)
end
Converts ampersand, dashes, ellipsis, quotes, copyright and registered
trademark symbols in text
to properly encoded characters.
# File lib/rdoc/text.rb, line 190
def to_html text
if Object.const_defined? :Encoding then
html = ''.encode text.encoding
encoded = RDoc::Text::TO_HTML_CHARACTERS[text.encoding]
else
html = ''
encoded = {
:close_dquote => '”',
:close_squote => '’',
:copyright => '©',
:ellipsis => '…',
:em_dash => '—',
:en_dash => '–',
:open_dquote => '“',
:open_squote => '‘',
:trademark => '®',
}
end
s = StringScanner.new text
insquotes = false
indquotes = false
after_word = nil
until s.eos? do
case
when s.scan(%r<(tt|code)>.*?<\/\11>>/) then # skip contents of tt
html << s.matched.gsub('\\', '\')
when s.scan(%r<(tt|code)>.*?/) then
warn "mismatched <#{s[1]}> tag" # TODO signal file/line
html << s.matched
when s.scan(%r<[^>]+\/?s*>/) then # skip HTML tags
html << s.matched
when s.scan(%r\\(\S)/) then # unhandled suppressed crossref
html << s[1]
after_word = nil
when s.scan(%r\.\.\.(\.?)/) then
html << s[1] << encoded[:ellipsis]
after_word = nil
when s.scan(%r\(c\)/) then
html << encoded[:copyright]
after_word = nil
when s.scan(%r\(r\)/) then
html << encoded[:trademark]
after_word = nil
when s.scan(%r---/) then
html << encoded[:em_dash]
after_word = nil
when s.scan(%r--/) then
html << encoded[:en_dash]
after_word = nil
when s.scan(%r"|"/) then
html << encoded[indquotes ? :close_dquote : :open_dquote]
indquotes = !indquotes
after_word = nil
when s.scan(%r``/) then # backtick double quote
html << encoded[:open_dquote]
after_word = nil
when s.scan(%r''/) then # tick double quote
html << encoded[:close_dquote]
after_word = nil
when s.scan(%r'/) then # single quote
if insquotes
html << encoded[:close_squote]
insquotes = false
elsif after_word
# Mary's dog, my parents' house: do not start paired quotes
html << encoded[:close_squote]
else
html << encoded[:open_squote]
insquotes = true
end
after_word = nil
else # advance to the next potentially significant character
match = s.scan(%r.+?(?=[<\\.("'`&-])/) #"
if match then
html << match
after_word = match =~ %r\w$/
else
html << s.rest
break
end
end
end
html
end
Wraps txt
to line_len
# File lib/rdoc/text.rb, line 284
def wrap(txt, line_len = 76)
res = []
sp = 0
ep = txt.length
while sp < ep
# scan back for a space
p = sp + line_len - 1
if p >= ep
p = ep
else
while p > sp and txt[p] != \s\
p -= 1
end
if p <= sp
p = sp + line_len
while p < ep and txt[p] != \s\
p += 1
end
end
end
res << txt[sp...p] << "\n"
sp = p
sp += 1 while sp < ep and txt[sp] == \s\
end
res.join.strip
end
/ | Search |
---|---|
? | Show this help |