'raquo' => '»',
'quot' => '"',
'apos' => '\'',
+ 'deg' => '°',
'micro' => 'µ',
'copy' => '©',
'trade' => '™',
'gt' => '>',
'hellip' => '…',
'nbsp' => ' ',
+ 'ndash' => '–',
'Agrave' => 'À',
'Aacute' => 'Á',
'Acirc' => 'Â',
# Some blogging and forum platforms use spans or divs with a 'body' or 'message' or 'text' in their class
# to mark actual text
- AFTER_PAR1_REGEX = /<\w+\s+[^>]*(?:body|message|text)[^>]*>.*?<\/?(?:p|div|html|body|table|td|tr)(?:\s+[^>]*)?>/im
+ AFTER_PAR1_REGEX = /<\w+\s+[^>]*(?:body|message|text|post)[^>]*>.*?<\/?(?:p|div|html|body|table|td|tr)(?:\s+[^>]*)?>/im
# At worst, we can try stuff which is comprised between two <br>
AFTER_PAR2_REGEX = /<br(?:\s+[^>]*)?\/?>.*?<\/?(?:br|p|div|html|body|table|td|tr)(?:\s+[^>]*)?\/?>/im
# Execute an external program, returning a String obtained by redirecting
# the program's standards errors and output
#
+ # TODO: find a way to expose some common errors (e.g. Errno::NOENT)
+ # to the caller
def Utils.safe_exec(command, *args)
- IO.popen("-") { |p|
+ output = IO.popen("-") { |p|
if p
- return p.readlines.join("\n")
+ break p.readlines.join("\n")
else
begin
$stderr.reopen($stdout)
Kernel::exit! 1
end
}
+ raise "safe execution of #{command} returned #{$?}" unless $?.success?
+ return output
end
# Try executing an external program, returning true if the run was successful
# Decode HTML entities in the String _str_, using HTMLEntities if the
# package was found, or UNESCAPE_TABLE otherwise.
#
- def Utils.decode_html_entities(str)
- if defined? ::HTMLEntities
- return HTMLEntities.decode_entities(str)
+
+ if defined? ::HTMLEntities
+ if ::HTMLEntities.respond_to? :decode_entities
+ def Utils.decode_html_entities(str)
+ return HTMLEntities.decode_entities(str)
+ end
else
- str.gsub(/(&(.+?);)/) {
+ @@html_entities = HTMLEntities.new
+ def Utils.decode_html_entities(str)
+ return @@html_entities.decode str
+ end
+ end
+ else
+ def Utils.decode_html_entities(str)
+ return str.gsub(/(&(.+?);)/) {
symbol = $2
# remove the 0-paddng from unicode integers
- if symbol =~ /^#(\d+)$/
+ case symbol
+ when /^#x([0-9a-fA-F]+)$/
+ symbol = $1.to_i(16).to_s
+ when /^#(\d+)$/
symbol = $1.to_i.to_s
end
# HTML first par grabber without hpricot
def Utils.ircify_first_html_par_woh(xml_org, opts={})
- xml = xml_org.gsub(/<!--.*?-->/m, '').gsub(/<script(?:\s+[^>]*)?>.*?<\/script>/im, "").gsub(/<style(?:\s+[^>]*)?>.*?<\/style>/im, "")
+ xml = xml_org.gsub(/<!--.*?-->/m,
+ "").gsub(/<script(?:\s+[^>]*)?>.*?<\/script>/im,
+ "").gsub(/<style(?:\s+[^>]*)?>.*?<\/style>/im,
+ "").gsub(/<select(?:\s+[^>]*)?>.*?<\/select>/im,
+ "")
strip = opts[:strip]
strip = Regexp.new(/^#{Regexp.escape(strip)}/) if strip.kind_of?(String)