mirror of
https://github.com/StrawberryMaster/wayback-machine-downloader.git
synced 2025-12-17 17:56:44 +00:00
44 lines
1.5 KiB
Ruby
44 lines
1.5 KiB
Ruby
require 'json'
|
|
require 'uri'
|
|
|
|
module ArchiveAPI
|
|
|
|
def get_raw_list_from_api(url, page_index, http)
|
|
# Automatically append /* if the URL doesn't contain a path after the domain
|
|
# This is a workaround for an issue with the API and *some* domains.
|
|
# See https://github.com/StrawberryMaster/wayback-machine-downloader/issues/6
|
|
# But don't do this when exact_url flag is set
|
|
if url && !url.match(/^https?:\/\/.*\//i) && !@exact_url
|
|
url = "#{url}/*"
|
|
end
|
|
|
|
request_url = URI("https://web.archive.org/cdx/search/cdx")
|
|
params = [["output", "json"], ["url", url]] + parameters_for_api(page_index)
|
|
request_url.query = URI.encode_www_form(params)
|
|
|
|
begin
|
|
response = http.get(request_url)
|
|
body = response.body.to_s.strip
|
|
return [] if body.empty?
|
|
json = JSON.parse(body)
|
|
|
|
# Check if the response contains the header ["timestamp", "original"]
|
|
json.shift if json.first == ["timestamp", "original"]
|
|
json
|
|
rescue JSON::ParserError, StandardError => e
|
|
warn "Failed to fetch data from API: #{e.message}"
|
|
[]
|
|
end
|
|
end
|
|
|
|
def parameters_for_api(page_index)
|
|
parameters = [["fl", "timestamp,original"], ["collapse", "digest"], ["gzip", "false"]]
|
|
parameters.push(["filter", "statuscode:200"]) unless @all
|
|
parameters.push(["from", @from_timestamp.to_s]) if @from_timestamp && @from_timestamp != 0
|
|
parameters.push(["to", @to_timestamp.to_s]) if @to_timestamp && @to_timestamp != 0
|
|
parameters.push(["page", page_index]) if page_index
|
|
parameters
|
|
end
|
|
|
|
end
|