ul-wukong 4.1.0
Sign up to get free protection for your applications and to get access to all the features.
- checksums.yaml +15 -0
- data/.gitignore +60 -0
- data/.gitmodules +6 -0
- data/.rspec +2 -0
- data/.travis.yml +19 -0
- data/.yardopts +6 -0
- data/CHANGELOG.md +7 -0
- data/Gemfile +17 -0
- data/Guardfile +12 -0
- data/LICENSE.md +95 -0
- data/NOTES-travis.md +31 -0
- data/README-old.md +422 -0
- data/README.md +1308 -0
- data/Rakefile +28 -0
- data/TODO.md +99 -0
- data/bin/cutc +30 -0
- data/bin/cuttab +5 -0
- data/bin/greptrue +6 -0
- data/bin/md5sort +20 -0
- data/bin/setcat +11 -0
- data/bin/tabchar +5 -0
- data/bin/uniq-ord +59 -0
- data/bin/uniqc +3 -0
- data/bin/wu +34 -0
- data/bin/wu-clean-encoding +31 -0
- data/bin/wu-date +13 -0
- data/bin/wu-datetime +13 -0
- data/bin/wu-hist +3 -0
- data/bin/wu-lign +186 -0
- data/bin/wu-local +4 -0
- data/bin/wu-plus +9 -0
- data/bin/wu-source +5 -0
- data/bin/wu-sum +31 -0
- data/diagrams/wu_local.dot +39 -0
- data/diagrams/wu_local.dot.png +0 -0
- data/examples/Gemfile +38 -0
- data/examples/README.md +9 -0
- data/examples/basic/string_reverser.rb +23 -0
- data/examples/basic/tiny_count.rb +8 -0
- data/examples/basic/word_count/accumulator.rb +26 -0
- data/examples/basic/word_count/tokenizer.rb +13 -0
- data/examples/basic/word_count/word_count.rb +6 -0
- data/examples/dataflow/scraper_macro_flow.rb +28 -0
- data/examples/deploy_pack/Gemfile +6 -0
- data/examples/deploy_pack/README.md +6 -0
- data/examples/deploy_pack/a/b/c/.gitkeep +0 -0
- data/examples/deploy_pack/app/processors/string_reverser.rb +5 -0
- data/examples/deploy_pack/config/environment.rb +1 -0
- data/examples/dsl/dataflow/fibonacci_series.rb +101 -0
- data/examples/dsl/dataflow/scraper_macro_flow.rb +28 -0
- data/examples/dsl/dataflow/simple.rb +12 -0
- data/examples/dsl/dataflow/telegram.rb +45 -0
- data/examples/dsl/workflow/cherry_pie.dot +97 -0
- data/examples/dsl/workflow/cherry_pie.md +104 -0
- data/examples/dsl/workflow/cherry_pie.png +0 -0
- data/examples/dsl/workflow/cherry_pie.rb +101 -0
- data/examples/empty/.gitkeep +0 -0
- data/examples/examples_helper.rb +9 -0
- data/examples/geo.rb +4 -0
- data/examples/geo/geo_grids.numbers +0 -0
- data/examples/geo/geolocated.rb +331 -0
- data/examples/geo/quadtile.rb +69 -0
- data/examples/geo/spec/geolocated_spec.rb +247 -0
- data/examples/geo/tile_fetcher.rb +77 -0
- data/examples/graph/implied_geolocation/README.md +63 -0
- data/examples/graph/minimum_spanning_tree/airfares_graphviz.rb +73 -0
- data/examples/improver/tweet_summary.rb +73 -0
- data/examples/loadable.rb +2 -0
- data/examples/munging/airline_flights/airline_flights.rake +83 -0
- data/examples/munging/airline_flights/airplane.rb +0 -0
- data/examples/munging/airline_flights/airport_id_unification.rb +129 -0
- data/examples/munging/airline_flights/airport_ok_chars.rb +4 -0
- data/examples/munging/airline_flights/indexable.rb +75 -0
- data/examples/munging/airline_flights/indexable_spec.rb +90 -0
- data/examples/munging/airline_flights/reconcile_airports.rb +142 -0
- data/examples/munging/airline_flights/tasks.rake +83 -0
- data/examples/munging/airline_flights/topcities.rb +167 -0
- data/examples/munging/geo/geo_json.rb +54 -0
- data/examples/munging/geo/geo_models.rb +69 -0
- data/examples/munging/geo/geonames_models.rb +107 -0
- data/examples/munging/geo/iso_codes.rb +172 -0
- data/examples/munging/geo/reconcile_countries.rb +124 -0
- data/examples/munging/geo/tasks.rake +71 -0
- data/examples/munging/wikipedia/articles/extract_articles-parsed.rb +79 -0
- data/examples/munging/wikipedia/articles/extract_articles-templated.rb +136 -0
- data/examples/munging/wikipedia/articles/textualize_articles.rb +54 -0
- data/examples/munging/wikipedia/articles/verify_structure.rb +43 -0
- data/examples/munging/wikipedia/articles/wp2txt-LICENSE.txt +22 -0
- data/examples/munging/wikipedia/articles/wp2txt_article.rb +259 -0
- data/examples/munging/wikipedia/articles/wp2txt_utils.rb +452 -0
- data/examples/munging/wikipedia/dbpedia/dbpedia_common.rb +5 -0
- data/examples/munging/wikipedia/dbpedia/dbpedia_extract_geocoordinates.rb +78 -0
- data/examples/munging/wikipedia/dbpedia/extract_links-cruft.rb +66 -0
- data/examples/munging/wikipedia/dbpedia/extract_links.rb +260 -0
- data/examples/munging/wikipedia/dbpedia/sameas_extractor.rb +20 -0
- data/examples/rake_helper.rb +97 -0
- data/examples/ruby_project/Gemfile +6 -0
- data/examples/ruby_project/README.md +6 -0
- data/examples/ruby_project/a/b/c/.gitkeep +0 -0
- data/examples/server_logs/geo_ip_mapping/munge_geolite.rb +82 -0
- data/examples/server_logs/logline.rb +95 -0
- data/examples/server_logs/models.rb +66 -0
- data/examples/server_logs/page_counts.pig +48 -0
- data/examples/server_logs/server_logs-01-parse-script.rb +13 -0
- data/examples/server_logs/server_logs-02-histograms-full.rb +33 -0
- data/examples/server_logs/server_logs-02-histograms-mapper.rb +14 -0
- data/examples/server_logs/server_logs-03-breadcrumbs-full.rb +71 -0
- data/examples/server_logs/server_logs-04-page_page_edges-full.rb +40 -0
- data/examples/serverlogs/geo_ip_mapping/munge_geolite.rb +82 -0
- data/examples/serverlogs/models/logline.rb +102 -0
- data/examples/serverlogs/parser/apache_parser_widget.rb +46 -0
- data/examples/serverlogs/visit_paths/common.rb +4 -0
- data/examples/serverlogs/visit_paths/page_counts.pig +48 -0
- data/examples/serverlogs/visit_paths/serverlogs-01-parse-script.rb +11 -0
- data/examples/serverlogs/visit_paths/serverlogs-02-histograms-full.rb +31 -0
- data/examples/serverlogs/visit_paths/serverlogs-02-histograms-mapper.rb +12 -0
- data/examples/serverlogs/visit_paths/serverlogs-03-breadcrumbs-full.rb +67 -0
- data/examples/serverlogs/visit_paths/serverlogs-04-page_page_edges-full.rb +38 -0
- data/examples/splitter.rb +94 -0
- data/examples/string_reverser.rb +7 -0
- data/examples/text/pig_latin/pig_latinizer.rb +35 -0
- data/examples/text/pig_latin/pig_latinizer_widget.rb +16 -0
- data/examples/text/regional_flavor/README.md +14 -0
- data/examples/text/regional_flavor/article_wordbags.pig +39 -0
- data/examples/text/regional_flavor/j01-article_wordbags.rb +4 -0
- data/examples/text/regional_flavor/simple_pig_script.pig +27 -0
- data/examples/twitter.rb +5 -0
- data/lib/hanuman.rb +36 -0
- data/lib/hanuman/graph.rb +97 -0
- data/lib/hanuman/graphvizzer.rb +206 -0
- data/lib/hanuman/graphvizzer/gv_models.rb +161 -0
- data/lib/hanuman/graphvizzer/gv_presenter.rb +97 -0
- data/lib/hanuman/link.rb +35 -0
- data/lib/hanuman/registry.rb +46 -0
- data/lib/hanuman/stage.rb +128 -0
- data/lib/hanuman/tree.rb +67 -0
- data/lib/wu/geo.rb +4 -0
- data/lib/wu/geo/geo_grids.numbers +0 -0
- data/lib/wu/geo/geolocated.rb +331 -0
- data/lib/wu/geo/quadtile.rb +69 -0
- data/lib/wu/graph/union_find.rb +62 -0
- data/lib/wu/model/reconcilable.rb +63 -0
- data/lib/wu/munging.rb +71 -0
- data/lib/wu/social/models/twitter.rb +31 -0
- data/lib/wu/wikipedia/models.rb +20 -0
- data/lib/wukong.rb +54 -0
- data/lib/wukong/dataflow.rb +43 -0
- data/lib/wukong/doc_helpers.rb +14 -0
- data/lib/wukong/doc_helpers/dataflow_handler.rb +29 -0
- data/lib/wukong/doc_helpers/field_handler.rb +91 -0
- data/lib/wukong/doc_helpers/processor_handler.rb +29 -0
- data/lib/wukong/driver.rb +214 -0
- data/lib/wukong/driver/event_machine_driver.rb +15 -0
- data/lib/wukong/driver/wiring.rb +68 -0
- data/lib/wukong/local.rb +42 -0
- data/lib/wukong/local/runner.rb +96 -0
- data/lib/wukong/local/stdio_driver.rb +104 -0
- data/lib/wukong/logger.rb +102 -0
- data/lib/wukong/model/faker.rb +136 -0
- data/lib/wukong/model/flatpack_parser/flat.rb +60 -0
- data/lib/wukong/model/flatpack_parser/flatpack.rb +4 -0
- data/lib/wukong/model/flatpack_parser/lang.rb +46 -0
- data/lib/wukong/model/flatpack_parser/parser.rb +55 -0
- data/lib/wukong/model/flatpack_parser/tokens.rb +130 -0
- data/lib/wukong/plugin.rb +48 -0
- data/lib/wukong/processor.rb +110 -0
- data/lib/wukong/rake_helper.rb +6 -0
- data/lib/wukong/runner.rb +169 -0
- data/lib/wukong/runner/boot_sequence.rb +123 -0
- data/lib/wukong/runner/code_loader.rb +52 -0
- data/lib/wukong/runner/command_runner.rb +44 -0
- data/lib/wukong/runner/deploy_pack_loader.rb +75 -0
- data/lib/wukong/runner/help_message.rb +42 -0
- data/lib/wukong/source.rb +33 -0
- data/lib/wukong/source/source_driver.rb +74 -0
- data/lib/wukong/source/source_runner.rb +38 -0
- data/lib/wukong/spec_helpers.rb +74 -0
- data/lib/wukong/spec_helpers/integration_tests.rb +150 -0
- data/lib/wukong/spec_helpers/integration_tests/integration_test_matchers.rb +207 -0
- data/lib/wukong/spec_helpers/integration_tests/integration_test_runner.rb +97 -0
- data/lib/wukong/spec_helpers/shared_examples.rb +22 -0
- data/lib/wukong/spec_helpers/unit_tests.rb +135 -0
- data/lib/wukong/spec_helpers/unit_tests/unit_test_driver.rb +132 -0
- data/lib/wukong/spec_helpers/unit_tests/unit_test_matchers.rb +169 -0
- data/lib/wukong/spec_helpers/unit_tests/unit_test_runner.rb +60 -0
- data/lib/wukong/version.rb +3 -0
- data/lib/wukong/widget/echo.rb +55 -0
- data/lib/wukong/widget/extract.rb +122 -0
- data/lib/wukong/widget/filters.rb +452 -0
- data/lib/wukong/widget/logger.rb +56 -0
- data/lib/wukong/widget/operators.rb +82 -0
- data/lib/wukong/widget/reducers.rb +10 -0
- data/lib/wukong/widget/reducers/accumulator.rb +73 -0
- data/lib/wukong/widget/reducers/bin.rb +368 -0
- data/lib/wukong/widget/reducers/count.rb +73 -0
- data/lib/wukong/widget/reducers/group.rb +128 -0
- data/lib/wukong/widget/reducers/group_concat.rb +98 -0
- data/lib/wukong/widget/reducers/improver.rb +71 -0
- data/lib/wukong/widget/reducers/join_xml.rb +37 -0
- data/lib/wukong/widget/reducers/moments.rb +72 -0
- data/lib/wukong/widget/reducers/sort.rb +180 -0
- data/lib/wukong/widget/reducers/uniq.rb +91 -0
- data/lib/wukong/widget/serializers.rb +317 -0
- data/lib/wukong/widget/utils.rb +46 -0
- data/lib/wukong/widgets.rb +7 -0
- data/spec/examples/dataflow/fibonacci_series_spec.rb +18 -0
- data/spec/examples/dataflow/parse_apache_logs_spec.rb +8 -0
- data/spec/examples/dataflow/parsing_spec.rb +14 -0
- data/spec/examples/dataflow/simple_spec.rb +34 -0
- data/spec/examples/dataflow/telegram_spec.rb +43 -0
- data/spec/examples/graph/minimum_spanning_tree_spec.rb +34 -0
- data/spec/examples/munging/airline_flights/identifiers_spec.rb +16 -0
- data/spec/examples/munging/airline_flights_spec.rb +202 -0
- data/spec/examples/text/pig_latin_spec.rb +18 -0
- data/spec/examples/workflow/cherry_pie_spec.rb +36 -0
- data/spec/hanuman/graph_spec.rb +119 -0
- data/spec/hanuman/hanuman_spec.rb +10 -0
- data/spec/hanuman/registry_spec.rb +123 -0
- data/spec/hanuman/stage_spec.rb +81 -0
- data/spec/hanuman/tree_spec.rb +119 -0
- data/spec/spec.opts +1 -0
- data/spec/spec_helper.rb +43 -0
- data/spec/support/example_test_helpers.rb +95 -0
- data/spec/support/hanuman_test_helpers.rb +92 -0
- data/spec/support/integration_helper.rb +38 -0
- data/spec/support/model_test_helpers.rb +115 -0
- data/spec/support/shared_context_for_graphs.rb +57 -0
- data/spec/support/shared_context_for_reducers.rb +37 -0
- data/spec/support/shared_examples_for_builders.rb +94 -0
- data/spec/support/shared_examples_for_shortcuts.rb +57 -0
- data/spec/wu/model/reconcilable_spec.rb +152 -0
- data/spec/wukong/dataflow_spec.rb +87 -0
- data/spec/wukong/driver_spec.rb +154 -0
- data/spec/wukong/local/runner_spec.rb +29 -0
- data/spec/wukong/local/stdio_driver_spec.rb +73 -0
- data/spec/wukong/local_spec.rb +6 -0
- data/spec/wukong/logger_spec.rb +49 -0
- data/spec/wukong/model/faker_spec.rb +132 -0
- data/spec/wukong/processor_spec.rb +21 -0
- data/spec/wukong/runner_spec.rb +132 -0
- data/spec/wukong/source_spec.rb +6 -0
- data/spec/wukong/widget/extract_spec.rb +101 -0
- data/spec/wukong/widget/filters_spec.rb +79 -0
- data/spec/wukong/widget/logger_spec.rb +23 -0
- data/spec/wukong/widget/operators_spec.rb +25 -0
- data/spec/wukong/widget/reducers/bin_spec.rb +92 -0
- data/spec/wukong/widget/reducers/count_spec.rb +11 -0
- data/spec/wukong/widget/reducers/group_spec.rb +21 -0
- data/spec/wukong/widget/reducers/join_xml_spec.rb +25 -0
- data/spec/wukong/widget/reducers/moments_spec.rb +36 -0
- data/spec/wukong/widget/reducers/sort_spec.rb +26 -0
- data/spec/wukong/widget/reducers/uniq_spec.rb +14 -0
- data/spec/wukong/widget/serializers_spec.rb +114 -0
- data/spec/wukong/widget/sink_spec.rb +19 -0
- data/spec/wukong/widget/source_spec.rb +65 -0
- data/spec/wukong/wu-local_spec.rb +109 -0
- data/spec/wukong/wu-source_spec.rb +32 -0
- data/spec/wukong/wu_spec.rb +14 -0
- data/spec/wukong/wukong_spec.rb +10 -0
- data/wukong.gemspec +35 -0
- metadata +465 -0
@@ -0,0 +1,71 @@
|
|
1
|
+
require_relative('../../rake_helper')
|
2
|
+
|
3
|
+
Pathname.register_paths(
|
4
|
+
geo_data: [:data, 'geo'],
|
5
|
+
geo_work: [:work, 'geo'],
|
6
|
+
geo_code: File.dirname(__FILE__),
|
7
|
+
#
|
8
|
+
iso_3166: [:geo_data, 'iso_codes', "iso_3166.tsv" ],
|
9
|
+
geonames_countries: [:geo_data, 'geonames', "geonames_countries.json" ],
|
10
|
+
#
|
11
|
+
countries_json: [:geo_work, "countries.json" ],
|
12
|
+
country_name_lookup: [:geo_work, "country_name_lookup.tsv" ],
|
13
|
+
)
|
14
|
+
|
15
|
+
chain :geo do
|
16
|
+
code_files = FileList[Pathname.of(:geo_code, '*.rb').to_s]
|
17
|
+
chain(:countries) do
|
18
|
+
|
19
|
+
task(:load) do
|
20
|
+
require_relative('./geo_models')
|
21
|
+
require_relative('./geo_json')
|
22
|
+
require_relative('./geonames_models')
|
23
|
+
require_relative('./iso_codes')
|
24
|
+
require_relative('./reconcile_countries')
|
25
|
+
CountryReconciler.load_reconciled_countries
|
26
|
+
end
|
27
|
+
|
28
|
+
# desc 'load the ISO 3166 countries'
|
29
|
+
# task(:countries_iso_3166, after: [code_files, :force]) do |dest|
|
30
|
+
# require_relative('./iso_codes')
|
31
|
+
# p Wukong::Data::CountryCode.for_any_name('Bolivia')
|
32
|
+
# end
|
33
|
+
|
34
|
+
# step(:geonames_countries, doc: 'load the Geonames countries',
|
35
|
+
# invoke: 'geo:countries:load',
|
36
|
+
# # , after: [code_files, :force]
|
37
|
+
# ) do |dest|
|
38
|
+
# Wukong::Data::GeonamesGeoJson.load(:geonames_countries)
|
39
|
+
# end
|
40
|
+
|
41
|
+
desc 'Add the iso_codes data to the geonames countries'
|
42
|
+
create_file(:countries_json, invoke: 'geo:countries:load', after: [code_files, :force]) do |dest|
|
43
|
+
Geo::Country.values.each do |country|
|
44
|
+
dest << country.to_json << "\n"
|
45
|
+
end
|
46
|
+
end
|
47
|
+
|
48
|
+
desc 'Add the iso_codes data to the geonames countries'
|
49
|
+
create_file(:country_name_lookup, invoke: 'geo:countries:load', after: [code_files, :force]) do |dest|
|
50
|
+
Geo::Country.values.each do |ct|
|
51
|
+
ct.names.each do |alt_name|
|
52
|
+
dest << [ct.country_id, ct.country_al3id, ct.country_numid,
|
53
|
+
ct.tld_id, ct.geonames_id,
|
54
|
+
ct.name,
|
55
|
+
Geo::Place.slugify_name(alt_name), alt_name
|
56
|
+
].join("\t") << "\n"
|
57
|
+
end
|
58
|
+
end
|
59
|
+
end
|
60
|
+
|
61
|
+
# task(:country_name_lookup => :load) do
|
62
|
+
# Geo::CountryNameLookup.load
|
63
|
+
# end
|
64
|
+
|
65
|
+
end
|
66
|
+
end
|
67
|
+
|
68
|
+
task :default => [
|
69
|
+
# 'geo:countries',
|
70
|
+
'geo:countries:country_name_lookup'
|
71
|
+
]
|
@@ -0,0 +1,79 @@
|
|
1
|
+
#!/usr/bin/env ruby
|
2
|
+
|
3
|
+
# Extracts wikipedia articles from bzipped xml, outputs them in TSV. Article
|
4
|
+
# text is XML encoded, but all newlines and tabs (in fact, all control
|
5
|
+
# characters) are converted to XML entities, making it safe to truck around as
|
6
|
+
# TSV.
|
7
|
+
#
|
8
|
+
# Sample Pig LOAD Statement:
|
9
|
+
#
|
10
|
+
# all_articles = LOAD '$articles' AS
|
11
|
+
# (id:long, namespace:int, title:chararray, revision_timestamp:long, redirect:chararray, text:chararray);
|
12
|
+
#
|
13
|
+
|
14
|
+
# ## Usage
|
15
|
+
#
|
16
|
+
# Flattens the wikipedia 'enwiki-latest-pages-articles.xml.gz' into a
|
17
|
+
# one-line-per-record heap.
|
18
|
+
#
|
19
|
+
# examples/munging/wikipedia/articles/extract_articles-templated.rb --rm --run \
|
20
|
+
# /data/origin/dumps.wikimedia.org/enwiki/20120601/enwiki-20120601-pages-articles.xml
|
21
|
+
# /data/results/wikipedia/full/articles.json.tsv
|
22
|
+
#
|
23
|
+
|
24
|
+
require 'wukong'
|
25
|
+
require 'wukong/streamer/encoding_cleaner'
|
26
|
+
require 'crack/xml'
|
27
|
+
require 'multi_json'
|
28
|
+
require 'oj'
|
29
|
+
require_relative '../utils/munging_utils.rb'
|
30
|
+
|
31
|
+
module ArticlesExtractor
|
32
|
+
class Mapper < Wukong::Streamer::LineStreamer
|
33
|
+
include Wukong::Streamer::EncodingCleaner
|
34
|
+
include MungingUtils
|
35
|
+
|
36
|
+
def lines
|
37
|
+
@lines ||= []
|
38
|
+
end
|
39
|
+
|
40
|
+
def recordize line
|
41
|
+
lines << line
|
42
|
+
if line =~ /<\/page>/
|
43
|
+
result = Crack::XML::parse(lines.join("\n"))
|
44
|
+
@lines = []
|
45
|
+
return [result]
|
46
|
+
else
|
47
|
+
return nil
|
48
|
+
end
|
49
|
+
end
|
50
|
+
|
51
|
+
def process record
|
52
|
+
if record.has_key? 'mediawiki'
|
53
|
+
record = record['mediawiki']
|
54
|
+
end
|
55
|
+
|
56
|
+
redirect = record['page']['redirect'] ? record['page']['redirect']['title'] : ''
|
57
|
+
timestamp = Time.iso8601(record['page']['revision']['timestamp']).to_flat
|
58
|
+
raw_text = record['page']['revision']['text']
|
59
|
+
|
60
|
+
# some few parts per million articles have an empty body -- workaround
|
61
|
+
raw_text = '' if not raw_text.is_a?(String)
|
62
|
+
|
63
|
+
result = [
|
64
|
+
record['page']['id'],
|
65
|
+
record['page']['ns'],
|
66
|
+
scrub_control_chars(record['page']['title']),
|
67
|
+
record['page']['revision']['id'],
|
68
|
+
timestamp,
|
69
|
+
scrub_control_chars(redirect),
|
70
|
+
safe_json_encode(raw_text)
|
71
|
+
]
|
72
|
+
yield result
|
73
|
+
end
|
74
|
+
end
|
75
|
+
end
|
76
|
+
|
77
|
+
# Force it to run in a single map task, to avoid writing a custom input format.
|
78
|
+
# The job runs in 2 hours, once; much less than the time it'd take me to do so.
|
79
|
+
Wukong::Script.new(ArticlesExtractor::Mapper, nil, min_split_size: 1152921504606846976).run
|
@@ -0,0 +1,136 @@
|
|
1
|
+
#!/usr/bin/env ruby
|
2
|
+
|
3
|
+
# Extracts wikipedia articles from bzipped xml, outputs them in TSV. Article
|
4
|
+
# text is XML encoded, but all newlines and tabs (in fact, all control
|
5
|
+
# characters) are converted to XML entities, making it safe to truck around as
|
6
|
+
# TSV.
|
7
|
+
|
8
|
+
# ## Schema
|
9
|
+
#
|
10
|
+
# Sample Pig LOAD Statement:
|
11
|
+
#
|
12
|
+
# all_articles = LOAD '$articles' AS
|
13
|
+
# (id:long, namespace:int, title:chararray, revision_timestamp:long, redirect:chararray, text:chararray);
|
14
|
+
#
|
15
|
+
|
16
|
+
# ## Usage
|
17
|
+
#
|
18
|
+
# Flattens the wikipedia 'enwiki-latest-pages-articles.xml.gz' into a
|
19
|
+
# one-line-per-record heap.
|
20
|
+
#
|
21
|
+
# examples/munging/wikipedia/articles/extract_articles-templated.rb --rm --run \
|
22
|
+
# /data/origin/dumps.wikimedia.org/enwiki/20120601/enwiki-20120601-pages-articles.xml
|
23
|
+
# /data/results/wikipedia/full/articles.tsv
|
24
|
+
#
|
25
|
+
|
26
|
+
require 'wukong'
|
27
|
+
require 'wukong/streamer/encoding_cleaner'
|
28
|
+
require 'crack/xml'
|
29
|
+
require 'multi_json'
|
30
|
+
require_relative '../utils/munging_utils.rb'
|
31
|
+
|
32
|
+
# <page>
|
33
|
+
# <title>Anarchism</title>
|
34
|
+
# <ns>0</ns>
|
35
|
+
# <id>12</id>
|
36
|
+
# <revision>
|
37
|
+
# <id>370845941</id>
|
38
|
+
# <timestamp>2010-06-29T20:14:56Z</timestamp>
|
39
|
+
# <contributor>
|
40
|
+
# <username>Centographer</username>
|
41
|
+
# <id>12640258</id>
|
42
|
+
# </contributor>
|
43
|
+
# <minor />
|
44
|
+
# <comment>clarifying not ordinary anarcho-socialism</comment>
|
45
|
+
# <text xml:space="preserve">
|
46
|
+
# ...snip ...
|
47
|
+
# </text>
|
48
|
+
# <sha1>...</sha1>
|
49
|
+
# </revision>
|
50
|
+
# </page>
|
51
|
+
#
|
52
|
+
module ArticlesExtractor
|
53
|
+
class Mapper < Wukong::Streamer::LineStreamer
|
54
|
+
include Wukong::Streamer::EncodingCleaner
|
55
|
+
include MungingUtils
|
56
|
+
|
57
|
+
def initialize(*)
|
58
|
+
super
|
59
|
+
@lines = []
|
60
|
+
@state = :out_of_article
|
61
|
+
@num_lines = 0
|
62
|
+
end
|
63
|
+
|
64
|
+
# Bolt together all lines between a <page> and a </page> marker.
|
65
|
+
def recordize line
|
66
|
+
@num_lines += 1
|
67
|
+
return if @state == :out_of_article && (ARTICLE_BEG_RE !~ line)
|
68
|
+
@state = :in_article
|
69
|
+
#
|
70
|
+
@lines << line
|
71
|
+
if ARTICLE_END_RE =~ line
|
72
|
+
result = @lines.join("\n")
|
73
|
+
@lines = []
|
74
|
+
@state = :out_of_article
|
75
|
+
return [result]
|
76
|
+
else
|
77
|
+
return nil
|
78
|
+
end
|
79
|
+
end
|
80
|
+
|
81
|
+
def process article
|
82
|
+
info = ARTICLE_RE.match(article)
|
83
|
+
if not info then warn "Bad match line #{@lines}: #{article.to_s[0..2000]}" ; return ; end
|
84
|
+
|
85
|
+
timestamp = [info[:rts_yr], info[:rts_mo], info[:rts_day], info[:rts_hr], info[:rts_min], info[:rts_sec], 'Z'].join
|
86
|
+
text = Crack::XML::parse("<text>#{info[:text]}</text>")['text'] || ''
|
87
|
+
redirect = info[:redirect] || ''
|
88
|
+
|
89
|
+
record = [
|
90
|
+
info[:id],
|
91
|
+
info[:ns],
|
92
|
+
scrub_control_chars(info[:title]),
|
93
|
+
info[:revision_id],
|
94
|
+
timestamp,
|
95
|
+
scrub_control_chars(redirect),
|
96
|
+
safe_json_encode(text)
|
97
|
+
]
|
98
|
+
yield record
|
99
|
+
end
|
100
|
+
end
|
101
|
+
|
102
|
+
ARTICLE_BEG_RE = %r{\A\s*<page>\z}
|
103
|
+
ARTICLE_END_RE = %r{\A\s*</page>\z}
|
104
|
+
ARTICLE_RE = %r{\A
|
105
|
+
\s*<page>
|
106
|
+
\s* <title>(?<title>[^<]*)</title>
|
107
|
+
\s* <ns>(?<ns>\d+)</ns>
|
108
|
+
\s* <id>(?<id>\d+)</id>
|
109
|
+
\s* (?:<redirect\stitle=\"(?<redirect>[^\"]+)\"\s/>)?
|
110
|
+
\s* (?:<restrictions>(?<restrictions>[^<]+)</restrictions>)?
|
111
|
+
\s* <revision>
|
112
|
+
\s* <id>(?<revision_id>\d+)</id>
|
113
|
+
\s* <timestamp>(?<rts_yr>\d\d\d\d)-(?<rts_mo>\d\d)-(?<rts_day>\d\d)T(?<rts_hr>\d\d):(?<rts_min>\d\d):(?<rts_sec>\d\d)Z</timestamp>
|
114
|
+
\s* (?:
|
115
|
+
<contributor>\s*<username>[^<]+</username>\s*<id>\d+</id>\s*</contributor> |
|
116
|
+
<contributor>\s*<ip>[\d\.]+</ip>\s*</contributor> |
|
117
|
+
<contributor\sdeleted="deleted"\s/>
|
118
|
+
)
|
119
|
+
\s* (?:<minor\s/>)?
|
120
|
+
\s* (?:<comment>[^<]*</comment>|<comment\sdeleted="deleted"\s/>)?
|
121
|
+
\s* (?:
|
122
|
+
<text\sxml:space="preserve">
|
123
|
+
(?<text>.*)
|
124
|
+
</text>
|
125
|
+
|
|
126
|
+
<text\sxml:space="preserve"\s/>
|
127
|
+
)
|
128
|
+
\s* (?:<sha1>(?<sha1>[a-z0-9]+)</sha1> | <sha1\s/>)
|
129
|
+
\s* </revision>
|
130
|
+
\s*</page>\s*\z}xmo
|
131
|
+
|
132
|
+
end
|
133
|
+
|
134
|
+
# Force it to run in a single map task, to avoid writing a custom input format.
|
135
|
+
# The job runs in 2 hours, once; much less than the time it'd take me to do so.
|
136
|
+
Wukong::Script.new(ArticlesExtractor::Mapper, nil, min_split_size: 1152921504606846976).run
|
@@ -0,0 +1,54 @@
|
|
1
|
+
#!/usr/bin/env ruby
|
2
|
+
|
3
|
+
# Generate plain-text versions of articles from the tsv-converted raw article data
|
4
|
+
# (output from extract_articles)
|
5
|
+
#
|
6
|
+
# This strips out template tags, wiki links, and so forth
|
7
|
+
#
|
8
|
+
# Everything that's left is either actual text, or nicely detached punctuation.
|
9
|
+
|
10
|
+
# ## Usage
|
11
|
+
#
|
12
|
+
# Uses the output of extract_articles-templated.rb:
|
13
|
+
#
|
14
|
+
# examples/munging/wikipedia/articles/textualize_articles.rb --rm --run \
|
15
|
+
# /data/results/wikipedia/full/articles.json.tsv \
|
16
|
+
# /data/results/wikipedia/full/article_texts.json.tsv
|
17
|
+
#
|
18
|
+
|
19
|
+
require 'wukong'
|
20
|
+
require 'multi_json'
|
21
|
+
require 'oj'
|
22
|
+
require 'strscan'
|
23
|
+
require 'find'
|
24
|
+
require 'sanitize'
|
25
|
+
#
|
26
|
+
require_relative '../utils/munging_utils.rb'
|
27
|
+
require_relative './wp2txt_article'
|
28
|
+
require_relative './wp2txt_utils'
|
29
|
+
|
30
|
+
module TextualizeArticles
|
31
|
+
|
32
|
+
class Mapper < Wukong::Streamer::RecordStreamer
|
33
|
+
include MungingUtils
|
34
|
+
|
35
|
+
@@errors = 0
|
36
|
+
MAX_ERRORS = 1_000
|
37
|
+
|
38
|
+
def process(id, namespace, title, revision_id, timestamp, redirect, raw_text)
|
39
|
+
|
40
|
+
text = MultiJson.decode(raw_text)
|
41
|
+
article = Wp2txt::Article.new(text, title)
|
42
|
+
jsonized_text = MultiJson.encode(article.polish)
|
43
|
+
|
44
|
+
yield [id, namespace, title, revision_id, timestamp, redirect, jsonized_text]
|
45
|
+
|
46
|
+
rescue StandardError => err
|
47
|
+
Wukong.bad_record("Bad Record", err, raw_text)
|
48
|
+
raise "Too many errors" if (@@errors += 1) > MAX_ERRORS
|
49
|
+
end
|
50
|
+
|
51
|
+
end
|
52
|
+
end
|
53
|
+
|
54
|
+
Wukong::Script.new(TextualizeArticles::Mapper, nil).run
|
@@ -0,0 +1,43 @@
|
|
1
|
+
|
2
|
+
# cat /mnt/data/origin/dumps.wikimedia.org/enwiki/20120601/enwiki-20120601-pages-articles.xml | ruby -ne '$_.chomp!; case $_ when %r{\A(\s*)<redirect title="[^"]+" />\z} then puts %Q{#{$1}<redirect title=\"...\" />} when %r{\A(\s*<[^>]+>)[^<]*(</\w+>)\z} then puts "#{$1}...#{$2}" ; when %r{\A(\s*<[^>]+>)} then puts $1 ; when %r{\A[^<]*(</[\w-]+>)\z} then puts $1 else false end ' > /mnt/data/origin/dumps.wikimedia.org/enwiki/20120601/xml-tags-seen.txt &
|
3
|
+
|
4
|
+
# cat /mnt/data/origin/dumps.wikimedia.org/enwiki/20120601/xml-tags-seen.txt | sort -S1G --temp=/mnt{,2,3,4}/tmp | uniq -c | sort -n | tee xml-tags-census.txt
|
5
|
+
|
6
|
+
# cat /mnt/data/origin/dumps.wikimedia.org/enwiki/20120601/xml-tags-seen.txt | sort -S1800M --temp=/mnt{2,3,4}/tmp | uniq -c | sort -n | tee xml-tags-census.txt &
|
7
|
+
|
8
|
+
|
9
|
+
# 609844 <page>
|
10
|
+
# 609844 <title>...</title>
|
11
|
+
# 609844 <ns>...</ns>
|
12
|
+
# 609844 <id>...</id>
|
13
|
+
# 714 <restrictions>...</restrictions>
|
14
|
+
#
|
15
|
+
# 251254 <redirect title="..." />
|
16
|
+
# 609844 <revision>
|
17
|
+
# 609844 <id>...</id>
|
18
|
+
# 609844 <timestamp>...</timestamp>
|
19
|
+
#
|
20
|
+
# 609843 <contributor>
|
21
|
+
# 548236 <username>...</username>
|
22
|
+
# 61607 <ip>...</ip>
|
23
|
+
# 548236 <id>...</id>
|
24
|
+
# 609843 </contributor>
|
25
|
+
#
|
26
|
+
# 288319 <minor />
|
27
|
+
# 529222 <comment>...</comment>
|
28
|
+
# 108 <comment>
|
29
|
+
# 108 </comment>
|
30
|
+
#
|
31
|
+
# 224818 <text xml:space="preserve">...</text>
|
32
|
+
# 385019 <text xml:space="preserve">
|
33
|
+
# 7 <text xml:space="preserve" />
|
34
|
+
#
|
35
|
+
# 346490 <sha1>...</sha1>
|
36
|
+
# 263354 <sha1 />
|
37
|
+
#
|
38
|
+
# 609843 </revision>
|
39
|
+
# 609843 </page>
|
40
|
+
#
|
41
|
+
# 384998 </text>
|
42
|
+
# 20 </text>
|
43
|
+
# 643676 </page>
|
@@ -0,0 +1,22 @@
|
|
1
|
+
Copyright (c) 2012 Yoichiro Hasebe
|
2
|
+
|
3
|
+
MIT License
|
4
|
+
|
5
|
+
Permission is hereby granted, free of charge, to any person obtaining
|
6
|
+
a copy of this software and associated documentation files (the
|
7
|
+
"Software"), to deal in the Software without restriction, including
|
8
|
+
without limitation the rights to use, copy, modify, merge, publish,
|
9
|
+
distribute, sublicense, and/or sell copies of the Software, and to
|
10
|
+
permit persons to whom the Software is furnished to do so, subject to
|
11
|
+
the following conditions:
|
12
|
+
|
13
|
+
The above copyright notice and this permission notice shall be
|
14
|
+
included in all copies or substantial portions of the Software.
|
15
|
+
|
16
|
+
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
|
17
|
+
EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
|
18
|
+
MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
|
19
|
+
NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE
|
20
|
+
LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION
|
21
|
+
OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION
|
22
|
+
WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
|
@@ -0,0 +1,259 @@
|
|
1
|
+
#!/usr/bin/env ruby
|
2
|
+
# -*- coding: utf-8 -*-
|
3
|
+
|
4
|
+
# Taken from Yoichiro Yohasebe's [`wp2txt` project](https://github.com/yohasebe/wp2txt)
|
5
|
+
# with liberal modifications for our purposes.
|
6
|
+
#
|
7
|
+
# This software is distributed under the MIT License. Please see the `./wp2txt-LICENSE.txt` file.
|
8
|
+
|
9
|
+
require 'strscan'
|
10
|
+
require_relative 'wp2txt_utils'
|
11
|
+
|
12
|
+
module Wp2txt
|
13
|
+
|
14
|
+
# possible element type, which could be later chosen to print or not to print
|
15
|
+
# :mw_heading
|
16
|
+
# :mw_htable
|
17
|
+
# :mw_quote
|
18
|
+
# :mw_unordered
|
19
|
+
# :mw_ordered
|
20
|
+
# :mw_definition
|
21
|
+
# :mw_pre
|
22
|
+
# :mw_paragraph
|
23
|
+
# :mw_comment
|
24
|
+
# :mw_math
|
25
|
+
# :mw_source
|
26
|
+
# :mw_inputbox
|
27
|
+
# :mw_template
|
28
|
+
# :mw_link
|
29
|
+
# :mw_summary
|
30
|
+
# :mw_blank
|
31
|
+
# :mw_redirect
|
32
|
+
|
33
|
+
# an article contains elements, each of which is [TYPE, string]
|
34
|
+
class Article
|
35
|
+
|
36
|
+
include Wp2txt
|
37
|
+
attr_accessor :elements, :title
|
38
|
+
|
39
|
+
# class varialbes to save resource for generating regexps
|
40
|
+
# those with a trailing number 1 represent opening tag/markup
|
41
|
+
# those with a trailing number 2 represent closing tag/markup
|
42
|
+
# those without a trailing number contain both opening/closing tags/markups
|
43
|
+
|
44
|
+
@@in_template_regex = Regexp.new('^\s*\{\{[^\}]+\}\}\s*$')
|
45
|
+
@@in_link_regex = Regexp.new('^\s*\[.*\]\s*$')
|
46
|
+
|
47
|
+
@@in_inputbox_regex = Regexp.new('<inputbox>.*?<\/inputbox>')
|
48
|
+
@@in_inputbox_regex1 = Regexp.new('<inputbox>')
|
49
|
+
@@in_inputbox_regex2 = Regexp.new('<\/inputbox>')
|
50
|
+
|
51
|
+
@@in_source_regex = Regexp.new('<source.*?>.*?<\/source>')
|
52
|
+
@@in_source_regex1 = Regexp.new('<source.*?>')
|
53
|
+
@@in_source_regex2 = Regexp.new('<\/source>')
|
54
|
+
|
55
|
+
@@in_math_regex = Regexp.new('<math.*?>.*?<\/math>')
|
56
|
+
@@in_math_regex1 = Regexp.new('<math.*?>')
|
57
|
+
@@in_math_regex2 = Regexp.new('<\/math>')
|
58
|
+
|
59
|
+
@@in_heading_regex = Regexp.new('^=+.*?=+$')
|
60
|
+
|
61
|
+
@@in_html_table_regex = Regexp.new('<table.*?><\/table>')
|
62
|
+
@@in_html_table_regex1 = Regexp.new('<table\b')
|
63
|
+
@@in_html_table_regex2 = Regexp.new('<\/\s*table>')
|
64
|
+
|
65
|
+
@@in_table_regex1 = Regexp.new('^\s*\{\|')
|
66
|
+
@@in_table_regex2 = Regexp.new('^\|\}.*?$')
|
67
|
+
|
68
|
+
@@in_unordered_regex = Regexp.new('^\*')
|
69
|
+
@@in_ordered_regex = Regexp.new('^\#')
|
70
|
+
@@in_pre_regex = Regexp.new('^ ')
|
71
|
+
@@in_definition_regex = Regexp.new('^[\;\:]')
|
72
|
+
|
73
|
+
@@blank_line_regex = Regexp.new('^\s*$')
|
74
|
+
|
75
|
+
@@redirect_regex = Regexp.new('#(?:REDIRECT|転送)\s+\[\[(.+)\]\]', Regexp::IGNORECASE)
|
76
|
+
|
77
|
+
def initialize(text, title = "", strip_tmarker = false)
|
78
|
+
@title = title.strip
|
79
|
+
@strip_tmarker = strip_tmarker
|
80
|
+
parse text
|
81
|
+
end
|
82
|
+
|
83
|
+
def create_element(tp, text)
|
84
|
+
[tp, text]
|
85
|
+
end
|
86
|
+
|
87
|
+
def parse(source)
|
88
|
+
self.class.remove_comments(source)
|
89
|
+
@elements = []
|
90
|
+
mode = nil
|
91
|
+
open_stack = []
|
92
|
+
close_stack = []
|
93
|
+
source.each_line do |line|
|
94
|
+
|
95
|
+
case mode
|
96
|
+
when :mw_table
|
97
|
+
if @@in_table_regex2 =~ line
|
98
|
+
mode = nil
|
99
|
+
end
|
100
|
+
@elements.last.last << line
|
101
|
+
next
|
102
|
+
when :mw_inputbox
|
103
|
+
if @@in_inputbox_regex2 =~ line
|
104
|
+
mode = nil
|
105
|
+
end
|
106
|
+
@elements.last.last << line
|
107
|
+
next
|
108
|
+
when :mw_source
|
109
|
+
if @@in_source_regex2 =~ line
|
110
|
+
mode = nil
|
111
|
+
end
|
112
|
+
@elements.last.last << line
|
113
|
+
next
|
114
|
+
when :mw_math
|
115
|
+
if @@in_math_regex2 =~ line
|
116
|
+
mode = nil
|
117
|
+
end
|
118
|
+
@elements.last.last << line
|
119
|
+
next
|
120
|
+
when :mw_htable
|
121
|
+
if @@in_html_table_regex2 =~ line
|
122
|
+
mode = nil
|
123
|
+
end
|
124
|
+
@elements.last.last << line
|
125
|
+
next
|
126
|
+
end
|
127
|
+
|
128
|
+
case line
|
129
|
+
when @@blank_line_regex
|
130
|
+
@elements << create_element(:mw_blank, "\n")
|
131
|
+
when @@redirect_regex
|
132
|
+
@elements << create_element(:mw_redirect, line)
|
133
|
+
when @@in_template_regex
|
134
|
+
@elements << create_element(:mw_template, line)
|
135
|
+
when @@in_heading_regex
|
136
|
+
@elements << create_element(:mw_heading, "\n" + line + "\n")
|
137
|
+
when @@in_inputbox_regex
|
138
|
+
@elements << create_element(:mw_inputbox, line)
|
139
|
+
when @@in_inputbox_regex1
|
140
|
+
mode = :mw_inputbox
|
141
|
+
@elements << create_element(:mw_inputbox, line)
|
142
|
+
when @@in_source_regex
|
143
|
+
@elements << create_element(:mw_source, line)
|
144
|
+
when @@in_source_regex1
|
145
|
+
mode = :mw_source
|
146
|
+
@elements << create_element(:mw_source, line)
|
147
|
+
when @@in_math_regex
|
148
|
+
@elements << create_element(:mw_math, line)
|
149
|
+
when @@in_math_regex1
|
150
|
+
mode = :mw_math
|
151
|
+
@elements << create_element(:mw_math, line)
|
152
|
+
when @@in_html_table_regex
|
153
|
+
@elements << create_element(:mw_htable, line)
|
154
|
+
when @@in_html_table_regex1
|
155
|
+
mode = :mw_htable
|
156
|
+
@elements << create_element(:mw_htable, line)
|
157
|
+
when @@in_table_regex1
|
158
|
+
mode = :mw_table
|
159
|
+
@elements << create_element(:mw_table, line)
|
160
|
+
when @@in_unordered_regex
|
161
|
+
line = line.sub(/\A[\*\#\;\:\ ]+/, "") if @strip_tmarker
|
162
|
+
@elements << create_element(:mw_unordered, line)
|
163
|
+
when @@in_ordered_regex
|
164
|
+
line = line.sub(/\A[\*\#\;\:\ ]+/, "") if @strip_tmarker
|
165
|
+
@elements << create_element(:mw_ordered, line)
|
166
|
+
when @@in_pre_regex
|
167
|
+
line = line.sub(/\A\^\ /, "") if @strip_tmarker
|
168
|
+
@elements << create_element(:mw_pre, line)
|
169
|
+
when @@in_definition_regex
|
170
|
+
line = line.sub(/\A[\;\:\ ]+/, "") if @strip_tmarker
|
171
|
+
@elements << create_element(:mw_definition, line)
|
172
|
+
when @@in_link_regex
|
173
|
+
@elements << create_element(:mw_link, line)
|
174
|
+
else
|
175
|
+
@elements << create_element(:mw_paragraph, line)
|
176
|
+
end
|
177
|
+
end
|
178
|
+
@elements
|
179
|
+
end
|
180
|
+
|
181
|
+
def self.remove_comments(text)
|
182
|
+
# remove all comment texts
|
183
|
+
# and insert as many number of new line chars included in
|
184
|
+
# each comment instead
|
185
|
+
text.gsub!(/\<\!\-\-(.*?)\-\-\>/m) do |content|
|
186
|
+
num_of_newlines = content.count("\n")
|
187
|
+
(num_of_newlines == 0) ? "" : ("\n" * num_of_newlines)
|
188
|
+
end
|
189
|
+
end
|
190
|
+
|
191
|
+
EXCLUDE_SECTIONS = {
|
192
|
+
mw_title: false,
|
193
|
+
mw_heading: false,
|
194
|
+
mw_paragraph: false,
|
195
|
+
mw_link: false,
|
196
|
+
mw_redirect: false,
|
197
|
+
#
|
198
|
+
mw_pre: false,
|
199
|
+
mw_quote: false,
|
200
|
+
mw_unordered: false,
|
201
|
+
mw_ordered: false,
|
202
|
+
mw_definition: false,
|
203
|
+
#
|
204
|
+
mw_table: true,
|
205
|
+
mw_htable: true,
|
206
|
+
mw_blank: true,
|
207
|
+
mw_math: true,
|
208
|
+
mw_source: true,
|
209
|
+
mw_template: true,
|
210
|
+
}
|
211
|
+
|
212
|
+
def polish
|
213
|
+
contents = []
|
214
|
+
elements.each do |el_type, element|
|
215
|
+
contents << "+#{el_type.to_s.upcase}+\t" if $DEBUG_MODE
|
216
|
+
next if EXCLUDE_SECTIONS[el_type]
|
217
|
+
#
|
218
|
+
case el_type
|
219
|
+
when :mw_heading then contents << format_wiki(element)
|
220
|
+
when :mw_paragraph then contents << format_wiki(element)
|
221
|
+
when :mw_link then contents << format_wiki(element)
|
222
|
+
when :mw_redirect then contents << format_wiki(element) << "\n\n"
|
223
|
+
when :mw_pre then contents << element
|
224
|
+
when :mw_quote then contents << format_wiki(element)
|
225
|
+
when :mw_unordered then contents << format_wiki(element)
|
226
|
+
when :mw_ordered then contents << format_wiki(element)
|
227
|
+
when :mw_definition then contents << format_wiki(element)
|
228
|
+
when :mw_table, :mw_htable then contents << format_wiki(element)
|
229
|
+
when :mw_math, :mw_source then contents << format_wiki(element)
|
230
|
+
when :mw_blank then contents << format_wiki(element)
|
231
|
+
else
|
232
|
+
warn "Unknown section #{el_type}, content '#{element.to_s.gsub(/[\r\n]+/m,'')[0..200]}'"
|
233
|
+
contents << format_wiki(element)
|
234
|
+
end
|
235
|
+
end
|
236
|
+
text = contents.join
|
237
|
+
|
238
|
+
# Extract text from <b>..</b> and so forth; remove contents of <ref>...</ref> completely
|
239
|
+
text = clean_html(text)
|
240
|
+
# translate some recognizable special characters
|
241
|
+
text = special_chr(text)
|
242
|
+
# re-hang the no-wiki segments
|
243
|
+
unescape_nowiki(text)
|
244
|
+
# strip out templates. Several parts per million of these will fail for
|
245
|
+
# bad structure; I assume that means some parts per thousand will be
|
246
|
+
# mis-estimated. C'est la UGC.
|
247
|
+
text = remove_templates(text) if EXCLUDE_SECTIONS[:mw_template]
|
248
|
+
|
249
|
+
return '' if /\A\s*\z/m =~ text
|
250
|
+
#
|
251
|
+
result = EXCLUDE_SECTIONS[:mw_title] ? "" : "# #{format_wiki(title)}\n\n"
|
252
|
+
result << text
|
253
|
+
result.gsub!(/\n\n\n+/m){"\n\n"}
|
254
|
+
result << "\n"
|
255
|
+
|
256
|
+
result
|
257
|
+
end
|
258
|
+
end
|
259
|
+
end
|