External/wayback-machine-downloader
1
0
Fork 0
mirror of https://github.com/StrawberryMaster/wayback-machine-downloader.git synced 2025-12-18 10:16:47 +00:00
Code Issues Packages Projects Releases Wiki Activity

Ignore urls with particularly long file name #4

Browse Source
This commit is contained in:
hartator 2015-08-21 20:43:31 -05:00
parent 919210fe64
commit 47ff4bed13
1 changed files with 13 additions and 9 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

22
lib/wayback_machine_downloader.rb
View File

@ -80,17 +80,21 @@ class WaybackMachineDownloader
end end
unless File.exists? file_path unless File.exists? file_path
structure_dir_path dir_path structure_dir_path dir_path
open(file_path, "wb") do |file| begin
begin open(file_path, "wb") do |file|
open("http://web.archive.org/web/#{timestamp}id_/#{file_url}") do |uri| begin
file.write(uri.read) open("http://web.archive.org/web/#{timestamp}id_/#{file_url}") do |uri|
file.write(uri.read)
end
rescue OpenURI::HTTPError => e
puts "#{file_url} # #{e}"
file.write(e.io.read)
rescue StandardError => e
puts "#{file_url} # #{e}"
end end
rescue OpenURI::HTTPError => e
puts "#{file_url} # #{e}"
file.write(e.io.read)
rescue StandardError => e
puts "#{file_url} # #{e}"
end end
rescue Errno::ENAMETOOLONG => e
puts "#{file_url} # #{e}"
end end
puts "#{file_url} -> #{file_path} (#{count}/#{file_list_curated.size})" puts "#{file_url} -> #{file_path} (#{count}/#{file_list_curated.size})"
else else