File tree Expand file tree Collapse file tree 2 files changed +16
-11
lines changed Expand file tree Collapse file tree 2 files changed +16
-11
lines changed Original file line number Diff line number Diff line change @@ -53,9 +53,15 @@ def each(&block)
5353 end
5454
5555 private
56+ def base_name
57+ "#{ @language } wiki-latest-#{ type_in_path } .xml.bz2"
58+ end
59+
60+ def data_path
61+ cache_dir_path + base_name
62+ end
63+
5664 def open_data ( &block )
57- base_name = "#{ @language } wiki-latest-#{ type_in_path } .xml.bz2"
58- data_path = cache_dir_path + base_name
5965 data_url = "https://dumps.wikimedia.org/#{ @language } wiki/latest/#{ base_name } "
6066 bz2 = Enumerator . new do |yielder |
6167 download ( data_path , data_url ) do |bz2_chunk |
Original file line number Diff line number Diff line change @@ -21,10 +21,10 @@ def teardown
2121 end
2222
2323 test ( "#each" ) do
24- def @dataset . download ( output_path , url )
25- xml_path = output_path . sub_ext ( "" )
26- xml_path . open ( "w" ) do |xml_file |
27- xml_file . puts ( <<-XML )
24+ data_path = @dataset . __send__ ( :data_path )
25+ xml_path = data_path . sub_ext ( "" )
26+ xml_path . open ( "w" ) do |xml_file |
27+ xml_file . puts ( <<-XML )
2828<mediawiki
2929 xmlns="http://www.mediawiki.org/xml/export-0.10/"
3030 xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
@@ -55,11 +55,10 @@ def @dataset.download(output_path, url)
5555 </revision>
5656 </page>
5757</mediawiki>
58- XML
59- end
60- unless system ( "bzip2" , xml_path . to_s )
61- raise "failed to run bzip2"
62- end
58+ XML
59+ end
60+ unless system ( "bzip2" , xml_path . to_s )
61+ raise "failed to run bzip2"
6362 end
6463
6564 contributor = Datasets ::Wikipedia ::Contributor . new ( "user" , 10 )
You can’t perform that action at this time.
0 commit comments