Module: DownloadFiles

Defined in:
lib/download_files.rb,
lib/download_files/version.rb

Constant Summary collapse

VERSION =
"1.0.1"

Class Method Summary collapse

Class Method Details

.download_files(page_address, pattern, dir, ssl_verify, is_verbose, logger) ⇒ Object



21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
# File 'lib/download_files.rb', line 21

def self.download_files(
  page_address,
  pattern,
  dir,
  ssl_verify,
  is_verbose,
  logger
)
  logger.debug(
    "DownloadFiles.download_files: " +
    {
      page_address: page_address,
      pattern: pattern,
      dir: dir,
      ssl_verify: ssl_verify,
      is_verbose: is_verbose,
      logger: logger
    }.inspect
  )

  abs_dir = File.absolute_path(dir)

  logger.info(
    "Downloading files from \"#{page_address}\" to \"#{abs_dir}\" that match /#{pattern}/ " +
    (if ssl_verify then 'requiring' else 'without requiring' end + ' SSL verification') +
    '...'
  )

  Mechanize.go(page_address, ssl_verify) do |agent, page|
    logger.debug page.inspect

    if is_verbose
      logger.info "On #{page.uri}"
    end

    agent.pluggable_parser.default = Mechanize::Download

    page.links.each do |link|
      if is_verbose
        logger.info "Checking link #{link.href}"
      end

      next unless link.href && link.href.match(pattern)

      full_file_name = File.expand_path(File.basename(link.href), abs_dir)

      if is_verbose
        logger.info "Downloading #{link.click.uri} to #{full_file_name}"
      end

      agent.get(link.click.uri).save(full_file_name)
    end
  end
end