Module: Irc::Utils

Defined in:
lib/rbot/core/utils/utils.rb,
lib/rbot/core/utils/httputil.rb

Overview

Miscellaneous useful functions

Defined Under Namespace

Classes: HttpUtil

Constant Summary collapse

SEC_PER_MIN =

Seconds per minute

60
SEC_PER_HR =

Seconds per hour

SEC_PER_MIN * 60
SEC_PER_DAY =

Seconds per day

SEC_PER_HR * 24
SEC_PER_MNTH =

Seconds per (30-day) month

SEC_PER_DAY * 30
SEC_PER_YR =

Second per (30*12 = 360 day) year

SEC_PER_MNTH * 12
@@bot = 
nil
@@safe_save_dir = 
nil

Class Method Summary collapse

Class Method Details

.bot ⇒ Object

The bot instance



164
165
166
 
# File 'lib/rbot/core/utils/utils.rb', line 164

def Utils.bot
  @@bot
end

.bot=(b) ⇒ Object

Set up some Utils routines which depend on the associated bot.



169
170
171
172
173
 
# File 'lib/rbot/core/utils/utils.rb', line 169

def Utils.bot=(b)
  debug "initializing utils"
  @@bot = b
  @@safe_save_dir = "#{@@bot.botclass}/safe_save"
end

.check_location(ds, rx) ⇒ Object

HTML info filters often need to check if the webpage location of a passed DataStream ds matches a given Regexp.



642
643
644
645
646
647
648
649
650
 
# File 'lib/rbot/core/utils/utils.rb', line 642

def Utils.check_location(ds, rx)
  debug ds[:headers]
  if h = ds[:headers]
    loc = [h['x-rbot-location'],h['location']].flatten.grep(rx)
  end
  loc ||= []
  debug loc
  return loc.empty? ? nil : loc
end

.decode_html_entities(str) ⇒ Object

Decode HTML entities in the String str, using HTMLEntities if the package was found, or UNESCAPE_TABLE otherwise.



324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
 
# File 'lib/rbot/core/utils/utils.rb', line 324

def Utils.decode_html_entities(str)
  if defined? ::HTMLEntities
    return HTMLEntities.decode_entities(str)
  else
    str.gsub(/(&(.+?);)/) {
      symbol = $2
      # remove the 0-paddng from unicode integers
      if symbol =~ /^#(\d+)$/
        symbol = $1.to_i.to_s
      end

      # output the symbol's irc-translated character, or a * if it's unknown
      UNESCAPE_TABLE[symbol] || (symbol.match(/^\d+$/) ? [symbol.to_i].pack("U") : '*')
    }
  end
end

.distance_of_time_in_words(minutes) ⇒ Object

Translates a number of minutes into verbal distances. e.g. 0.5 => less than a minute

70 => about one hour



263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
 
# File 'lib/rbot/core/utils/utils.rb', line 263

def Utils.distance_of_time_in_words(minutes)
  case
    when minutes < 1
      _("less than a minute")
    when minutes < 50
      _("%{m} minutes") % {:m => minutes}
    when minutes < 90
      _("about one hour")
    when minutes < 1080
      _("%{m} hours") % {:m => (minutes / 60).round}
    when minutes < 1440
      _("one day")
    when minutes < 2880
      _("about one day")
    else
      _("%{m} days") % {:m => (minutes / 1440).round}
  end
end

.get_first_pars(urls, count, opts = {}) ⇒ Object

Get the first pars of the first count urls. The pages are downloaded using the bot httputil service. Returns an array of the first paragraphs fetched. If (optional) opts :message is specified, those paragraphs are echoed as replies to the IRC message passed as opts :message



688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
 
# File 'lib/rbot/core/utils/utils.rb', line 688

def Utils.get_first_pars(urls, count, opts={})
  idx = 0
  msg = opts[:message]
  retval = Array.new
  while count > 0 and urls.length > 0
    url = urls.shift
    idx += 1

    begin
      info = Utils.get_html_info(URI.parse(url), opts)

      par = info[:content]
      retval.push(par)

      if par
        msg.reply "[#{idx}] #{par}", :overlong => :truncate if msg
        count -=1
      end
    rescue
      debug "Unable to retrieve #{url}: #{$!}"
      next
    end
  end
  return retval
end

.get_html_info(doc, opts = {}) ⇒ Object

This method extracts title, content (first par) and extra information from the given document doc.

doc can be an URI, a Net::HTTPResponse or a String.

If doc is a String, only title and content information are retrieved (if possible), using standard methods.

If doc is an URI or a Net::HTTPResponse, additional information is retrieved, and special title/summary extraction routines are used if possible.



557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
 
# File 'lib/rbot/core/utils/utils.rb', line 557

def Utils.get_html_info(doc, opts={})
  case doc
  when String
    Utils.get_string_html_info(doc, opts)
  when Net::HTTPResponse
    Utils.get_resp_html_info(doc, opts)
  when URI
    ret = DataStream.new
    @@bot.httputil.get_response(doc) { |resp|
      ret.replace Utils.get_resp_html_info(resp, opts)
    }
    return ret
  else
    raise
  end
end

.get_resp_html_info(resp, opts = {}) ⇒ Object

This method extracts title, content (first par) and extra information from the given Net::HTTPResponse resp.

Currently, the only accepted options (in opts) are

uri_fragment

the URI fragment of the original request

full_body

get the whole body instead of @@bot.config bytes only

Returns a DataStream with the following keys:

text

the (partial) body

title

the title of the document (if any)

content

the first paragraph of the document (if any)

headers

the headers of the Net::HTTPResponse. The value is a Hash whose keys are lowercase forms of the HTTP header fields, and whose values are Arrays.



594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
 
# File 'lib/rbot/core/utils/utils.rb', line 594

def Utils.get_resp_html_info(resp, opts={})
  case resp
  when Net::HTTPSuccess
    loc = URI.parse(resp['x-rbot-location'] || resp['location']) rescue nil
    if loc and loc.fragment and not loc.fragment.empty?
      opts[:uri_fragment] ||= loc.fragment
    end
    ret = DataStream.new(opts.dup)
    ret[:headers] = resp.to_hash
    ret[:text] = partial = opts[:full_body] ? resp.body : resp.partial_body(@@bot.config['http.info_bytes'])

    filtered = Utils.try_htmlinfo_filters(ret)

    if filtered
      return filtered
    elsif resp['content-type'] =~ /^text\/|(?:x|ht)ml/
      ret.merge!(Utils.get_string_html_info(partial, opts))
    end
    return ret
  else
    raise UrlLinkError, "getting link (#{resp.code} - #{resp.message})"
  end
end

.get_string_html_info(text, opts = {}) ⇒ Object

This method extracts title and content (first par) from the given HTML or XML document text, using standard methods (String#ircify_html_title, Utils.ircify_first_html_par)

Currently, the only accepted option (in opts) is

uri_fragment

the URI fragment of the original request



660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
 
# File 'lib/rbot/core/utils/utils.rb', line 660

def Utils.get_string_html_info(text, opts={})
  debug "getting string html info"
  txt = text.dup
  title = txt.ircify_html_title
  debug opts
  if frag = opts[:uri_fragment] and not frag.empty?
    fragreg = /<a\s+(?:[^>]+\s+)?(?:name|id)=["']?#{frag}["']?[^>]*>/im
    debug fragreg
    debug txt
    if txt.match(fragreg)
      # grab the post-match
      txt = $'
    end
    debug txt
  end
  c_opts = opts.dup
  c_opts[:strip] ||= title
  content = Utils.ircify_first_html_par(txt, c_opts)
  content = nil if content.empty?
  return {:title => title, :content => content}
end

.ircify_first_html_par(xml_org, opts = {}) ⇒ Object

Try to grab and IRCify the first HTML par (<p> tag) in the given string. If possible, grab the one after the first heading

It is possible to pass some options to determine how the stripping occurs. Currently supported options are

strip

Regex or String to strip at the beginning of the obtained text

min_spaces

minimum number of spaces a paragraph should have



350
351
352
353
354
355
356
 
# File 'lib/rbot/core/utils/utils.rb', line 350

def Utils.ircify_first_html_par(xml_org, opts={})
  if defined? ::Hpricot
    Utils.ircify_first_html_par_wh(xml_org, opts)
  else
    Utils.ircify_first_html_par_woh(xml_org, opts)
  end
end

.ircify_first_html_par_wh(xml_org, opts = {}) ⇒ Object

HTML first par grabber using hpricot



359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
 
# File 'lib/rbot/core/utils/utils.rb', line 359

def Utils.ircify_first_html_par_wh(xml_org, opts={})
  doc = Hpricot(xml_org)

  # Strip styles and scripts
  (doc/"style|script").remove

  debug doc

  strip = opts[:strip]
  strip = Regexp.new(/^#{Regexp.escape(strip)}/) if strip.kind_of?(String)

  min_spaces = opts[:min_spaces] || 8
  min_spaces = 0 if min_spaces < 0

  txt = String.new

  pre_h = pars = by_span = nil

  while true
    debug "Minimum number of spaces: #{min_spaces}"

    # Initial attempt: <p> that follows <h\d>
    if pre_h.nil?
      pre_h = Hpricot::Elements[]
      found_h = false
      doc.search("*") { |e|
        next if e.bogusetag?
        case e.pathname
        when /^h\d/
          found_h = true
        when 'p'
          pre_h << e if found_h
        end
      }
      debug "Hx: found: #{pre_h.pretty_inspect}"
    end

    pre_h.each { |p|
      debug p
      txt = p.to_html.ircify_html
      txt.sub!(strip, '') if strip
      debug "(Hx attempt) #{txt.inspect} has #{txt.count(" ")} spaces"
      break unless txt.empty? or txt.count(" ") < min_spaces
    }

    return txt unless txt.empty? or txt.count(" ") < min_spaces

    # Second natural attempt: just get any <p>
    pars = doc/"p" if pars.nil?
    debug "par: found: #{pars.pretty_inspect}"
    pars.each { |p|
      debug p
      txt = p.to_html.ircify_html
      txt.sub!(strip, '') if strip
      debug "(par attempt) #{txt.inspect} has #{txt.count(" ")} spaces"
      break unless txt.empty? or txt.count(" ") < min_spaces
    }

    return txt unless txt.empty? or txt.count(" ") < min_spaces

    # Nothing yet ... let's get drastic: we look for non-par elements too,
    # but only for those that match something that we know is likely to
    # contain text

    # Some blogging and forum platforms use spans or divs with a 'body' or
    # 'message' or 'text' in their class to mark actual text. Since we want
    # the class match to be partial and case insensitive, we collect
    # the common elements that may have this class and then filter out those
    # we don't need. If no divs or spans are found, we'll accept additional
    # elements too (td, tr, tbody, table).
    if by_span.nil?
      by_span = Hpricot::Elements[]
      extra = Hpricot::Elements[]
      doc.search("*") { |el|
        next if el.bogusetag?
        case el.pathname
        when AFTER_PAR_PATH
          by_span.push el if el[:class] =~ AFTER_PAR_CLASS or el[:id] =~ AFTER_PAR_CLASS
        when AFTER_PAR_EX
          extra.push el if el[:class] =~ AFTER_PAR_CLASS or el[:id] =~ AFTER_PAR_CLASS
        end
      }
      if by_span.empty? and not extra.empty?
        by_span.concat extra
      end
      debug "other \#1: found: #{by_span.pretty_inspect}"
    end

    by_span.each { |p|
      debug p
      txt = p.to_html.ircify_html
      txt.sub!(strip, '') if strip
      debug "(other attempt \#1) #{txt.inspect} has #{txt.count(" ")} spaces"
      break unless txt.empty? or txt.count(" ") < min_spaces
    }

    return txt unless txt.empty? or txt.count(" ") < min_spaces

    # At worst, we can try stuff which is comprised between two <br>
    # TODO

    debug "Last candidate #{txt.inspect} has #{txt.count(" ")} spaces"
    return txt unless txt.count(" ") < min_spaces
    break if min_spaces == 0
    min_spaces /= 2
  end
end

.ircify_first_html_par_woh(xml_org, opts = {}) ⇒ Object

HTML first par grabber without hpricot



468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
 
# File 'lib/rbot/core/utils/utils.rb', line 468

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, "")

  strip = opts[:strip]
  strip = Regexp.new(/^#{Regexp.escape(strip)}/) if strip.kind_of?(String)

  min_spaces = opts[:min_spaces] || 8
  min_spaces = 0 if min_spaces < 0

  txt = String.new

  while true
    debug "Minimum number of spaces: #{min_spaces}"
    header_found = xml.match(HX_REGEX)
    if header_found
      header_found = $'
      while txt.empty? or txt.count(" ") < min_spaces
        candidate = header_found[PAR_REGEX]
        break unless candidate
        txt = candidate.ircify_html
        header_found = $'
        txt.sub!(strip, '') if strip
        debug "(Hx attempt) #{txt.inspect} has #{txt.count(" ")} spaces"
      end
    end

    return txt unless txt.empty? or txt.count(" ") < min_spaces

    # If we haven't found a first par yet, try to get it from the whole
    # document
    header_found = xml
    while txt.empty? or txt.count(" ") < min_spaces
      candidate = header_found[PAR_REGEX]
      break unless candidate
      txt = candidate.ircify_html
      header_found = $'
      txt.sub!(strip, '') if strip
      debug "(par attempt) #{txt.inspect} has #{txt.count(" ")} spaces"
    end

    return txt unless txt.empty? or txt.count(" ") < min_spaces

    # Nothing yet ... let's get drastic: we look for non-par elements too,
    # but only for those that match something that we know is likely to
    # contain text

    # Attempt #1
    header_found = xml
    while txt.empty? or txt.count(" ") < min_spaces
      candidate = header_found[AFTER_PAR1_REGEX]
      break unless candidate
      txt = candidate.ircify_html
      header_found = $'
      txt.sub!(strip, '') if strip
      debug "(other attempt \#1) #{txt.inspect} has #{txt.count(" ")} spaces"
    end

    return txt unless txt.empty? or txt.count(" ") < min_spaces

    # Attempt #2
    header_found = xml
    while txt.empty? or txt.count(" ") < min_spaces
      candidate = header_found[AFTER_PAR2_REGEX]
      break unless candidate
      txt = candidate.ircify_html
      header_found = $'
      txt.sub!(strip, '') if strip
      debug "(other attempt \#2) #{txt.inspect} has #{txt.count(" ")} spaces"
    end

    debug "Last candidate #{txt.inspect} has #{txt.count(" ")} spaces"
    return txt unless txt.count(" ") < min_spaces
    break if min_spaces == 0
    min_spaces /= 2
  end
end

.safe_exec(command, *args) ⇒ Object

Execute an external program, returning a String obtained by redirecting the program’s standards errors and output



286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
 
# File 'lib/rbot/core/utils/utils.rb', line 286

def Utils.safe_exec(command, *args)
  IO.popen("-") { |p|
    if p
      return p.readlines.join("\n")
    else
      begin
        $stderr.reopen($stdout)
        exec(command, *args)
      rescue Exception => e
        puts "exec of #{command} led to exception: #{e.pretty_inspect}"
        Kernel::exit! 0
      end
      puts "exec of #{command} failed"
      Kernel::exit! 0
    end
  }
end

.safe_save(file) {|temp| ... } ⇒ Object

Safely (atomically) save to file, by passing a tempfile to the block and then moving the tempfile to its final location when done.

call-seq: Utils.safe_save(file, &block)

Yields:

  • (temp) 


310
311
312
313
314
315
316
317
318
 
# File 'lib/rbot/core/utils/utils.rb', line 310

def Utils.safe_save(file)
  raise 'No safe save directory defined!' if @@safe_save_dir.nil?
  basename = File.basename(file)
  temp = Tempfile.new(basename,@@safe_save_dir)
  temp.binmode
  yield temp if block_given?
  temp.close
  File.rename(temp.path, file)
end

.secs_to_short(seconds) ⇒ Object

Turn a number of seconds into a hours:minutes:seconds e.g. 3:18:10 or 5’12“ or 7s



226
227
228
229
230
231
232
233
234
235
236
237
 
# File 'lib/rbot/core/utils/utils.rb', line 226

def Utils.secs_to_short(seconds)
  secs = seconds.to_i # make sure it's an integer
  mins, secs = secs.divmod 60
  hours, mins = mins.divmod 60
  if hours > 0
    return ("%s:%s:%s" % [hours, mins, secs])
  elsif mins > 0
    return ("%s'%s\"" % [mins, secs])
  else
    return ("%ss" % [secs])
  end
end

.secs_to_string(secs) ⇒ Object

Turn a number of seconds into a human readable string, e.g 2 days, 3 hours, 18 minutes and 10 seconds



199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
 
# File 'lib/rbot/core/utils/utils.rb', line 199

def Utils.secs_to_string(secs)
  ret = []
  years, secs = secs.divmod SEC_PER_YR
  secs_to_string_case(ret, years, _("year"), _("years")) if years > 0
  months, secs = secs.divmod SEC_PER_MNTH
  secs_to_string_case(ret, months, _("month"), _("months")) if months > 0
  days, secs = secs.divmod SEC_PER_DAY
  secs_to_string_case(ret, days, _("day"), _("days")) if days > 0
  hours, secs = secs.divmod SEC_PER_HR
  secs_to_string_case(ret, hours, _("hour"), _("hours")) if hours > 0
  mins, secs = secs.divmod SEC_PER_MIN
  secs_to_string_case(ret, mins, _("minute"), _("minutes")) if mins > 0
  secs = secs.to_i
  secs_to_string_case(ret, secs, _("second"), _("seconds")) if secs > 0 or ret.empty?
  case ret.length
  when 0
    raise "Empty ret array!"
  when 1
    return ret.to_s
  else
    return [ret[0, ret.length-1].join(", ") , ret[-1]].join(_(" and "))
  end
end

.secs_to_string_case(array, var, string, plural) ⇒ Object

Auxiliary method needed by Utils.secs_to_string



188
189
190
191
192
193
194
195
 
# File 'lib/rbot/core/utils/utils.rb', line 188

def Utils.secs_to_string_case(array, var, string, plural)
  case var
  when 1
    array << "1 #{string}"
  else
    array << "#{var} #{plural}"
  end
end

.timeago(time, options = {}) ⇒ Object

Returns human readable time. Like: 5 days ago

about one hour ago

options :start_date, sets the time to measure against, defaults to now :date_format, used with <tt>to_formatted_s<tt>, default to :default



245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
 
# File 'lib/rbot/core/utils/utils.rb', line 245

def Utils.timeago(time, options = {})
  start_date = options.delete(:start_date) || Time.new
  date_format = options.delete(:date_format) || :default
  delta_minutes = (start_date.to_i - time.to_i).floor / 60
  if delta_minutes.abs <= (8724*60) # eight weeks? I'm lazy to count days for longer than that
    distance = Utils.distance_of_time_in_words(delta_minutes);
    if delta_minutes < 0
      _("%{d} from now") % {:d => distance}
    else
      _("%{d} ago") % {:d => distance}
    end
  else
    return _("on %{date}") % {:date => system_date.to_formatted_s(date_format)}
  end
end

.try_htmlinfo_filters(ds) ⇒ Object

This method runs an appropriately-crafted DataStream ds through the filters in the :htmlinfo filter group, in order. If one of the filters returns non-nil, its results are merged in ds and returned. Otherwise nil is returned.

The input DataStream shuold have the downloaded HTML as primary key (:text) and possibly a :headers key holding the resonse headers.



626
627
628
629
630
631
632
633
634
635
636
637
638
 
# File 'lib/rbot/core/utils/utils.rb', line 626

def Utils.try_htmlinfo_filters(ds)
  filters = @@bot.filter_names(:htmlinfo)
  return nil if filters.empty?
  cur = nil
  # TODO filter priority
  filters.each { |n|
    debug "testing filter #{n}"
    cur = @@bot.filter(@@bot.global_filter_name(n, :htmlinfo), ds)
    debug "returned #{cur.pretty_inspect}"
    break if cur
  }
  return ds.merge(cur) if cur
end