Method: Addressable::URI.heuristic_parse

Defined in:
lib/addressable/uri.rb

.heuristic_parse(uri, hints = {}) ⇒ Addressable::URI

Converts an input to a URI. The input does not have to be a valid URI — the method will use heuristics to guess what URI was intended. This is not standards-compliant, merely user-friendly.

Parameters:

  • uri (String, Addressable::URI, #to_str) —

    The URI string to parse. No parsing is performed if the object is already an Addressable::URI.

  • hints (Hash) (defaults to: {}) —

    A Hash of hints to the heuristic parser. Defaults to {:scheme => "http"}.

Returns:



191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
 
# File 'lib/addressable/uri.rb', line 191

def self.heuristic_parse(uri, hints={})
  # If we were given nil, return nil.
  return nil unless uri
  # If a URI object is passed, just return itself.
  return uri.dup if uri.kind_of?(self)

  # If a URI object of the Ruby standard library variety is passed,
  # convert it to a string, then parse the string.
  # We do the check this way because we don't want to accidentally
  # cause a missing constant exception to be thrown.
  if uri.class.name =~ /^URI\b/
    uri = uri.to_s
  end

  unless uri.respond_to?(:to_str)
    raise TypeError, "Can't convert #{uri.class} into String."
  end
  # Otherwise, convert to a String
  uri = uri.to_str.dup.strip
  hints = {
    :scheme => "http"
  }.merge(hints)
  case uri
  when /^http:\//i
    uri.sub!(/^http:\/+/i, "http://")
  when /^https:\//i
    uri.sub!(/^https:\/+/i, "https://")
  when /^feed:\/+http:\//i
    uri.sub!(/^feed:\/+http:\/+/i, "feed:http://")
  when /^feed:\//i
    uri.sub!(/^feed:\/+/i, "feed://")
  when %r[^file:/{4}]i
    uri.sub!(%r[^file:/+]i, "file:////")
  when %r[^file://localhost/]i
    uri.sub!(%r[^file://localhost/+]i, "file:///")
  when %r[^file:/+]i
    uri.sub!(%r[^file:/+]i, "file:///")
  when /^\d{1,3}\.\d{1,3}\.\d{1,3}\.\d{1,3}/
    uri.sub!(/^/, hints[:scheme] + "://")
  when /\A\d+\..*:\d+\z/
    uri = "#{hints[:scheme]}://#{uri}"
  end
  match = uri.match(URIREGEX)
  fragments = match.captures
  authority = fragments[3]
  if authority && authority.length > 0
    new_authority = authority.tr("\\", "/").gsub(" ", "%20")
    # NOTE: We want offset 4, not 3!
    offset = match.offset(4)
    uri = uri.dup
    uri[offset[0]...offset[1]] = new_authority
  end
  parsed = self.parse(uri)
  if parsed.scheme =~ /^[^\/?#\.]+\.[^\/?#]+$/
    parsed = self.parse(hints[:scheme] + "://" + uri)
  end
  if parsed.path.include?(".")
    if parsed.path[/\b@\b/]
      parsed.scheme = "mailto" unless parsed.scheme
    elsif new_host = parsed.path[/^([^\/]+\.[^\/]*)/, 1]
      parsed.defer_validation do
        new_path = parsed.path.sub(
          Regexp.new("^" + Regexp.escape(new_host)), EMPTY_STR)
        parsed.host = new_host
        parsed.path = new_path
        parsed.scheme = hints[:scheme] unless parsed.scheme
      end
    end
  end
  return parsed
end