sitediff 0.0.1 → 1.0.0
Sign up to get free protection for your applications and to get access to all the features.
- checksums.yaml +5 -5
- data/bin/sitediff +10 -4
- data/lib/sitediff.rb +179 -91
- data/lib/sitediff/cache.rb +106 -0
- data/lib/sitediff/cli.rb +391 -60
- data/lib/sitediff/config.rb +383 -37
- data/lib/sitediff/config/creator.rb +114 -0
- data/lib/sitediff/config/preset.rb +75 -0
- data/lib/sitediff/crawler.rb +131 -0
- data/lib/sitediff/diff.rb +57 -12
- data/lib/sitediff/exception.rb +5 -0
- data/lib/sitediff/fetch.rb +76 -0
- data/lib/sitediff/files/diff.html.erb +20 -2
- data/lib/sitediff/files/jquery.min.js +2 -0
- data/lib/sitediff/files/normalize.css +349 -0
- data/lib/sitediff/files/report.html.erb +144 -0
- data/lib/sitediff/files/sidebyside.html.erb +16 -0
- data/lib/sitediff/files/sitediff.css +236 -29
- data/lib/sitediff/files/sitediff.js +176 -0
- data/lib/sitediff/report.rb +238 -0
- data/lib/sitediff/result.rb +63 -26
- data/lib/sitediff/sanitize.rb +160 -141
- data/lib/sitediff/sanitize/dom_transform.rb +130 -0
- data/lib/sitediff/sanitize/regexp.rb +82 -0
- data/lib/sitediff/uriwrapper.rb +114 -35
- data/lib/sitediff/webserver.rb +94 -0
- data/lib/sitediff/webserver/resultserver.rb +134 -0
- metadata +103 -43
- data/lib/sitediff/files/html_report.html.erb +0 -47
- data/lib/sitediff/util/cache.rb +0 -32
- data/lib/sitediff/util/webserver.rb +0 -77
@@ -0,0 +1,238 @@
|
|
1
|
+
# frozen_string_literal: true
|
2
|
+
|
3
|
+
require 'fileutils'
|
4
|
+
require 'json'
|
5
|
+
require 'minitar'
|
6
|
+
require 'sitediff'
|
7
|
+
require 'sitediff/config'
|
8
|
+
require 'zlib'
|
9
|
+
|
10
|
+
class SiteDiff
|
11
|
+
##
|
12
|
+
# SiteDiff Report Helper.
|
13
|
+
class Report
|
14
|
+
attr_reader :results, :cache
|
15
|
+
|
16
|
+
##
|
17
|
+
# Directory where diffs will be generated.
|
18
|
+
DIFFS_DIR = 'diffs'
|
19
|
+
|
20
|
+
##
|
21
|
+
# Name of file containing a list of pages with diffs.
|
22
|
+
FAILURES_FILE = 'failures.txt'
|
23
|
+
|
24
|
+
##
|
25
|
+
# Name of file containing HTML report of diffs.
|
26
|
+
REPORT_FILE_HTML = 'report.html'
|
27
|
+
|
28
|
+
##
|
29
|
+
# Name of file containing JSON report of diffs.
|
30
|
+
REPORT_FILE_JSON = 'report.json'
|
31
|
+
|
32
|
+
##
|
33
|
+
# Name of file containing exported file archive.
|
34
|
+
REPORT_FILE_TAR = 'report.tgz'
|
35
|
+
|
36
|
+
##
|
37
|
+
# Name of directory in which to build the portable report.
|
38
|
+
REPORT_BUILD_DIR = '_tmp_report'
|
39
|
+
|
40
|
+
##
|
41
|
+
# Name of the portable report directory.
|
42
|
+
REPORT_DIR = 'report'
|
43
|
+
|
44
|
+
##
|
45
|
+
# Path to settings used for report.
|
46
|
+
SETTINGS_FILE = 'settings.yaml'
|
47
|
+
|
48
|
+
##
|
49
|
+
# Creates a Reporter object.
|
50
|
+
#
|
51
|
+
# @param [Config] config.
|
52
|
+
# @param [Cache] cache.
|
53
|
+
# @param [Array] results.
|
54
|
+
def initialize(config, cache, results)
|
55
|
+
@config = config
|
56
|
+
@cache = cache
|
57
|
+
@results = results
|
58
|
+
end
|
59
|
+
|
60
|
+
##
|
61
|
+
# Generates an HTML report.
|
62
|
+
#
|
63
|
+
# @param [String] dir
|
64
|
+
# The directory in which the report is to be generated.
|
65
|
+
def generate_html(
|
66
|
+
dir,
|
67
|
+
report_before = nil,
|
68
|
+
report_after = nil
|
69
|
+
)
|
70
|
+
report_before ||= @config.before_url
|
71
|
+
report_after ||= @config.after_url
|
72
|
+
|
73
|
+
dir = SiteDiff.ensure_dir dir
|
74
|
+
|
75
|
+
write_diffs dir
|
76
|
+
write_failures dir
|
77
|
+
|
78
|
+
# Prepare report.
|
79
|
+
report = Diff.generate_html(
|
80
|
+
@results,
|
81
|
+
report_before,
|
82
|
+
report_after,
|
83
|
+
@cache,
|
84
|
+
@config.export
|
85
|
+
)
|
86
|
+
|
87
|
+
# Write report.
|
88
|
+
report_file = dir + REPORT_FILE_HTML
|
89
|
+
report_file.unlink if report_file.file?
|
90
|
+
report_file.open('w') { |f| f.write(report) }
|
91
|
+
|
92
|
+
write_settings dir, report_before, report_after
|
93
|
+
|
94
|
+
if @config.export
|
95
|
+
package_report(dir)
|
96
|
+
else
|
97
|
+
SiteDiff.log 'Report generated to ' + report_file.expand_path.to_s
|
98
|
+
end
|
99
|
+
end
|
100
|
+
|
101
|
+
##
|
102
|
+
# Generates a JSON report.
|
103
|
+
#
|
104
|
+
# @param dir
|
105
|
+
# The directory in which the report is to be generated.
|
106
|
+
def generate_json(dir)
|
107
|
+
dir = SiteDiff.ensure_dir dir
|
108
|
+
write_diffs dir
|
109
|
+
write_failures dir
|
110
|
+
|
111
|
+
# Prepare report.
|
112
|
+
report = {
|
113
|
+
paths_compared: @results.length,
|
114
|
+
paths_diffs: 0,
|
115
|
+
paths: {}
|
116
|
+
}
|
117
|
+
@results.each do |item|
|
118
|
+
report[:paths_diffs] += 1 unless item.success?
|
119
|
+
|
120
|
+
item_report = {
|
121
|
+
path: item.path,
|
122
|
+
status: item.status,
|
123
|
+
message: item.error
|
124
|
+
}
|
125
|
+
report[:paths][item.path] = item_report
|
126
|
+
end
|
127
|
+
report = JSON report
|
128
|
+
|
129
|
+
# Write report.
|
130
|
+
report_file = dir + REPORT_FILE_JSON
|
131
|
+
report_file.unlink if report_file.file?
|
132
|
+
report_file.open('w') { |f| f.write(report) }
|
133
|
+
|
134
|
+
write_settings dir
|
135
|
+
|
136
|
+
SiteDiff.log 'Report generated to ' + report_file.expand_path.to_s
|
137
|
+
end
|
138
|
+
|
139
|
+
##
|
140
|
+
# Package report for export.
|
141
|
+
def package_report(dir)
|
142
|
+
# Create temporaryreport directories.
|
143
|
+
temp_path = dir + REPORT_BUILD_DIR
|
144
|
+
temp_path.rmtree if temp_path.directory?
|
145
|
+
temp_path.mkpath
|
146
|
+
report_path = temp_path + REPORT_DIR
|
147
|
+
report_path.mkpath
|
148
|
+
files_path = report_path + 'files'
|
149
|
+
files_path.mkpath
|
150
|
+
diffs_path = dir + DIFFS_DIR
|
151
|
+
|
152
|
+
# Move files to place.
|
153
|
+
FileUtils.move(dir + REPORT_FILE_HTML, report_path)
|
154
|
+
FileUtils.move(diffs_path, files_path) if diffs_path.directory?
|
155
|
+
|
156
|
+
# Make tar file.
|
157
|
+
Dir.chdir(temp_path) do
|
158
|
+
Minitar.pack(
|
159
|
+
REPORT_DIR,
|
160
|
+
Zlib::GzipWriter.new(File.open(REPORT_FILE_TAR, 'wb'))
|
161
|
+
)
|
162
|
+
end
|
163
|
+
FileUtils.move(temp_path + REPORT_FILE_TAR, dir)
|
164
|
+
temp_path.rmtree
|
165
|
+
SiteDiff.log 'Archived report generated to ' + dir.join(REPORT_FILE_TAR).to_s
|
166
|
+
end
|
167
|
+
|
168
|
+
##
|
169
|
+
# Creates diff files in a directory named "diffs".
|
170
|
+
#
|
171
|
+
# If "dir" is /foo/bar, then diffs will be placed in /foo/bar/diffs.
|
172
|
+
#
|
173
|
+
# @param [Pathname] dir
|
174
|
+
# The directory in which a "diffs" directory is to be generated.
|
175
|
+
def write_diffs(dir)
|
176
|
+
raise Exception 'dir must be a Pathname' unless dir.is_a? Pathname
|
177
|
+
|
178
|
+
# Delete existing "diffs" dir, if exists.
|
179
|
+
diff_dir = dir + DIFFS_DIR
|
180
|
+
diff_dir.rmtree if diff_dir.exist?
|
181
|
+
|
182
|
+
# Write diffs to the diff directory.
|
183
|
+
@results.each { |r| r.dump(dir, @config.export) if r.status == Result::STATUS_FAILURE }
|
184
|
+
SiteDiff.log "All diff files written to #{diff_dir.expand_path}" unless @config.export
|
185
|
+
end
|
186
|
+
|
187
|
+
##
|
188
|
+
# Writes paths with diffs into a file.
|
189
|
+
#
|
190
|
+
# @param [Pathname] dir
|
191
|
+
# The directory in which the report is to be generated.
|
192
|
+
def write_failures(dir)
|
193
|
+
raise Exception 'dir must be a Pathname' unless dir.is_a? Pathname
|
194
|
+
|
195
|
+
failures = dir + FAILURES_FILE
|
196
|
+
SiteDiff.log "All failures written to #{failures.expand_path}"
|
197
|
+
failures.open('w') do |f|
|
198
|
+
@results.each { |r| f.puts r.path unless r.success? }
|
199
|
+
end
|
200
|
+
end
|
201
|
+
|
202
|
+
##
|
203
|
+
# Creates report settings.yaml file.
|
204
|
+
#
|
205
|
+
# TODO: Find a way to avoid having to create this file.
|
206
|
+
#
|
207
|
+
# @param [Pathname] dir
|
208
|
+
# The directory in which the report is to be generated.
|
209
|
+
def write_settings(dir, report_before = nil, report_after = nil)
|
210
|
+
raise Exception 'dir must be a Pathname' unless dir.is_a? Pathname
|
211
|
+
|
212
|
+
settings = {
|
213
|
+
'before' => report_before,
|
214
|
+
'after' => report_after,
|
215
|
+
'cached' => %w[before after]
|
216
|
+
}
|
217
|
+
dir.+(SETTINGS_FILE).open('w') { |f| YAML.dump(settings, f) }
|
218
|
+
end
|
219
|
+
|
220
|
+
##
|
221
|
+
# Returns CSS for HTML report.
|
222
|
+
def self.css
|
223
|
+
output = ''
|
224
|
+
output += File.read(File.join(SiteDiff::FILES_DIR, 'normalize.css'))
|
225
|
+
output += File.read(File.join(SiteDiff::FILES_DIR, 'sitediff.css'))
|
226
|
+
output
|
227
|
+
end
|
228
|
+
|
229
|
+
##
|
230
|
+
# Returns JS for HTML report.
|
231
|
+
def self.js
|
232
|
+
output = ''
|
233
|
+
output += File.read(File.join(SiteDiff::FILES_DIR, 'jquery.min.js'))
|
234
|
+
output += File.read(File.join(SiteDiff::FILES_DIR, 'sitediff.js'))
|
235
|
+
output
|
236
|
+
end
|
237
|
+
end
|
238
|
+
end
|
data/lib/sitediff/result.rb
CHANGED
@@ -1,73 +1,110 @@
|
|
1
|
-
|
1
|
+
# frozen_string_literal: true
|
2
|
+
|
3
|
+
require 'sitediff'
|
4
|
+
require 'sitediff/diff'
|
5
|
+
require 'sitediff/report'
|
2
6
|
require 'digest/sha1'
|
7
|
+
require 'fileutils'
|
3
8
|
|
4
9
|
class SiteDiff
|
5
|
-
|
10
|
+
# SiteDiff Result Object.
|
11
|
+
class Result < Struct.new(
|
12
|
+
:path,
|
13
|
+
:before,
|
14
|
+
:after,
|
15
|
+
:before_encoding,
|
16
|
+
:after_encoding,
|
17
|
+
:error,
|
18
|
+
:verbose
|
19
|
+
)
|
6
20
|
STATUS_SUCCESS = 0 # Identical before and after
|
7
21
|
STATUS_FAILURE = 1 # Different before and after
|
8
22
|
STATUS_ERROR = 2 # Couldn't fetch page
|
9
|
-
STATUS_TEXT = %w[
|
23
|
+
STATUS_TEXT = %w[unchanged changed error].freeze
|
10
24
|
|
11
25
|
attr_reader :status, :diff
|
12
26
|
|
27
|
+
##
|
28
|
+
# Creates a Result.
|
13
29
|
def initialize(*args)
|
14
30
|
super
|
15
31
|
if error
|
16
32
|
@status = STATUS_ERROR
|
17
33
|
else
|
18
|
-
|
34
|
+
if !before_encoding || !after_encoding
|
35
|
+
@diff = Diff.binary_diffy(
|
36
|
+
before,
|
37
|
+
after,
|
38
|
+
before_encoding,
|
39
|
+
after_encoding
|
40
|
+
)
|
41
|
+
else
|
42
|
+
@diff = Diff.html_diffy(before, after)
|
43
|
+
end
|
19
44
|
@status = @diff ? STATUS_FAILURE : STATUS_SUCCESS
|
20
45
|
end
|
21
46
|
end
|
22
47
|
|
48
|
+
##
|
49
|
+
# Whether the result has no diff.
|
50
|
+
#
|
51
|
+
# If there is a diff, it is not a success.
|
52
|
+
#
|
53
|
+
# TODO: Change "Success" to unchanged.
|
23
54
|
def success?
|
24
55
|
status == STATUS_SUCCESS
|
25
56
|
end
|
26
57
|
|
58
|
+
##
|
59
|
+
# Whether the result has an error.
|
60
|
+
def error?
|
61
|
+
status == STATUS_ERROR
|
62
|
+
end
|
63
|
+
|
27
64
|
# Textual representation of the status
|
28
65
|
def status_text
|
29
|
-
|
66
|
+
STATUS_TEXT[status]
|
30
67
|
end
|
31
68
|
|
32
69
|
# Printable URL
|
33
|
-
def url(prefix)
|
34
|
-
|
70
|
+
def url(tag, prefix, cache)
|
71
|
+
base = cache.read_tags.include?(tag) ? "/cache/#{tag}" : prefix
|
72
|
+
base.to_s + path
|
35
73
|
end
|
36
74
|
|
37
75
|
# Filename to store diff
|
38
76
|
def filename
|
39
|
-
File.join(
|
77
|
+
File.join(Report::DIFFS_DIR, Digest::SHA1.hexdigest(path) + '.html')
|
40
78
|
end
|
41
79
|
|
42
|
-
#
|
43
|
-
|
44
|
-
|
45
|
-
|
46
|
-
|
47
|
-
|
48
|
-
end
|
80
|
+
# Returns a URL to the result diff.
|
81
|
+
#
|
82
|
+
# Returns nil if the result has no diffs.
|
83
|
+
def diff_url(relative = false)
|
84
|
+
prefix = relative ? 'files/' : '/files/'
|
85
|
+
return prefix + filename if status == STATUS_FAILURE
|
49
86
|
end
|
50
87
|
|
51
88
|
# Log the result to the terminal
|
52
|
-
def log
|
89
|
+
def log(verbose = true)
|
53
90
|
case status
|
54
|
-
when STATUS_SUCCESS
|
55
|
-
SiteDiff
|
56
|
-
when STATUS_ERROR
|
57
|
-
SiteDiff
|
58
|
-
when STATUS_FAILURE
|
59
|
-
SiteDiff
|
60
|
-
puts Diff
|
91
|
+
when STATUS_SUCCESS
|
92
|
+
SiteDiff.log path, :success, 'UNCHANGED'
|
93
|
+
when STATUS_ERROR
|
94
|
+
SiteDiff.log path + " (#{error})", :warning, 'ERROR'
|
95
|
+
when STATUS_FAILURE
|
96
|
+
SiteDiff.log path, :error, 'CHANGED'
|
97
|
+
puts Diff.terminal_diffy(before, after) if verbose
|
61
98
|
end
|
62
99
|
end
|
63
100
|
|
64
101
|
# Dump the result to a file
|
65
|
-
def dump(dir)
|
102
|
+
def dump(dir, relative = false)
|
66
103
|
dump_path = File.join(dir, filename)
|
67
104
|
base = File.dirname(dump_path)
|
68
|
-
FileUtils
|
105
|
+
FileUtils.mkdir_p(base) unless File.exist?(base)
|
69
106
|
File.open(dump_path, 'w') do |f|
|
70
|
-
f.write(Diff
|
107
|
+
f.write(Diff.generate_diff_output(self, relative))
|
71
108
|
end
|
72
109
|
end
|
73
110
|
end
|
data/lib/sitediff/sanitize.rb
CHANGED
@@ -1,104 +1,152 @@
|
|
1
|
+
# frozen_string_literal: true
|
2
|
+
|
3
|
+
require 'sitediff'
|
4
|
+
require 'sitediff/exception'
|
5
|
+
require 'sitediff/sanitize/dom_transform'
|
6
|
+
require 'sitediff/sanitize/regexp'
|
1
7
|
require 'nokogiri'
|
2
8
|
require 'set'
|
3
9
|
|
4
10
|
class SiteDiff
|
5
|
-
|
6
|
-
|
11
|
+
# SiteDiff Sanitizer.
|
12
|
+
class Sanitizer
|
13
|
+
class InvalidSanitization < SiteDiffException; end
|
7
14
|
|
8
15
|
TOOLS = {
|
9
|
-
:
|
10
|
-
:
|
11
|
-
}
|
12
|
-
DOM_TRANSFORMS = Set.new(%w[remove unwrap_root unwrap remove_class])
|
13
|
-
|
14
|
-
|
15
|
-
|
16
|
-
|
17
|
-
|
18
|
-
|
19
|
-
|
20
|
-
# * { :type => "unwrap_root" }
|
21
|
-
# * { :type => "unwrap", :selector => "div.field-item" }
|
22
|
-
# * { :type => "remove", :selector => "div.extra-stuff" }
|
23
|
-
#
|
24
|
-
# @arg node - Nokogiri document or Node
|
25
|
-
# @arg rules - array of dom_transform rules
|
26
|
-
# @return - transformed Nokogiri document node
|
27
|
-
def perform_dom_transforms(node, rules)
|
28
|
-
rules.each do |rule|
|
29
|
-
type = rule['type'] or
|
30
|
-
raise InvalidSanitization, "DOM transform needs a type"
|
31
|
-
DOM_TRANSFORMS.include?(type) or
|
32
|
-
raise InvalidSanitization, "No DOM transform named #{type}"
|
33
|
-
|
34
|
-
meth = 'transform_' + type
|
35
|
-
|
36
|
-
if sels = rule['selector']
|
37
|
-
sels = [sels].flatten # Either array or scalar is fine
|
38
|
-
# Call method for each node the selectors find
|
39
|
-
sels.each do |sel|
|
40
|
-
node.css(sel).each { |e| send(meth, rule, e) }
|
41
|
-
end
|
42
|
-
else
|
43
|
-
send(meth, rule, node)
|
44
|
-
end
|
45
|
-
end
|
16
|
+
array: %w[dom_transform sanitization],
|
17
|
+
scalar: %w[selector remove_spacing ignore_whitespace]
|
18
|
+
}.freeze
|
19
|
+
DOM_TRANSFORMS = Set.new(%w[remove strip unwrap_root unwrap remove_class])
|
20
|
+
|
21
|
+
##
|
22
|
+
# Creates a Sanitizer.
|
23
|
+
def initialize(html, config, opts = {})
|
24
|
+
@html = html
|
25
|
+
@config = config
|
26
|
+
@opts = opts
|
46
27
|
end
|
47
28
|
|
48
|
-
|
49
|
-
|
50
|
-
|
51
|
-
|
52
|
-
|
53
|
-
|
29
|
+
##
|
30
|
+
# Performs sanitization.
|
31
|
+
def sanitize
|
32
|
+
return '' if @html == '' # Quick return on empty input
|
33
|
+
|
34
|
+
@node = Sanitizer.domify(@html)
|
35
|
+
@html = nil
|
36
|
+
|
37
|
+
remove_spacing
|
38
|
+
selector
|
39
|
+
dom_transforms
|
40
|
+
regexps
|
41
|
+
|
42
|
+
@html || Sanitizer.prettify(@node)
|
54
43
|
end
|
55
|
-
|
56
|
-
|
57
|
-
|
58
|
-
|
59
|
-
|
44
|
+
|
45
|
+
# Return whether or not we want to keep a rule
|
46
|
+
def want_rule(rule)
|
47
|
+
return false unless rule
|
48
|
+
return false if rule['disabled']
|
49
|
+
|
50
|
+
# Filter out if path regexp doesn't match
|
51
|
+
if (pathre = rule['path']) && (path = @opts[:path])
|
52
|
+
return ::Regexp.new(pathre).match(path)
|
60
53
|
end
|
61
|
-
|
62
|
-
|
63
|
-
node.children.size == 1 or
|
64
|
-
raise InvalidSanitization, "Multiple root elements in unwrap_root"
|
65
|
-
node.children = node.children[0].children
|
54
|
+
|
55
|
+
true
|
66
56
|
end
|
67
57
|
|
68
|
-
|
69
|
-
|
70
|
-
|
71
|
-
|
58
|
+
# Canonicalize a simple rule, eg: 'remove_spacing' or 'selector'.
|
59
|
+
# It may be a simple value, or a hash, or an array of hashes.
|
60
|
+
# Turn it into an array of hashes.
|
61
|
+
def canonicalize_rule(name)
|
62
|
+
(rules = @config[name]) || (return nil)
|
63
|
+
|
64
|
+
# Already an array? Do nothing.
|
65
|
+
if rules[0]&.respond_to?('each') && rules[0]&.fetch('value')
|
66
|
+
# If it is a hash, put it in an array.
|
67
|
+
elsif rules['value']
|
68
|
+
rules = [rules]
|
69
|
+
# If it is a scalar value, put it in an array.
|
72
70
|
else
|
73
|
-
|
71
|
+
rules = [{ 'value' => rules }]
|
74
72
|
end
|
75
|
-
|
76
|
-
|
77
|
-
|
78
|
-
|
73
|
+
|
74
|
+
want = rules.select { |r| want_rule(r) }
|
75
|
+
return nil if want.empty?
|
76
|
+
raise "Too many matching rules of type #{name}" if want.size > 1
|
77
|
+
|
78
|
+
want.first
|
79
|
+
end
|
80
|
+
|
81
|
+
# Perform 'remove_spacing' action
|
82
|
+
def remove_spacing
|
83
|
+
(rule = canonicalize_rule('remove_spacing')) || return
|
84
|
+
Sanitizer.remove_node_spacing(@node) if rule['value']
|
85
|
+
end
|
86
|
+
|
87
|
+
# Perform 'selector' action, to choose a new root
|
88
|
+
def selector
|
89
|
+
(rule = canonicalize_rule('selector')) || return
|
90
|
+
@node = Sanitizer.select_fragments(@node, rule['value'])
|
91
|
+
end
|
92
|
+
|
93
|
+
# Applies regexps. Also
|
94
|
+
def regexps
|
95
|
+
(rules = @config['sanitization']) || return
|
96
|
+
rules = rules.select { |r| want_rule(r) }
|
97
|
+
|
98
|
+
rules.map! { |r| Regexp.create(r) }
|
99
|
+
selector, global = rules.partition(&:selector?)
|
100
|
+
|
101
|
+
selector.each { |r| r.apply(@node) }
|
102
|
+
@html = Sanitizer.prettify(@node)
|
103
|
+
@node = nil
|
104
|
+
# Prevent potential UTF-8 encoding errors by removing bytes
|
105
|
+
# Not the only solution. An alternative is to return the
|
106
|
+
# string unmodified.
|
107
|
+
@html = @html.encode(
|
108
|
+
'UTF-8',
|
109
|
+
'binary',
|
110
|
+
invalid: :replace,
|
111
|
+
undef: :replace,
|
112
|
+
replace: ''
|
113
|
+
)
|
114
|
+
global.each { |r| r.apply(@html) }
|
115
|
+
end
|
116
|
+
|
117
|
+
# Perform DOM transforms
|
118
|
+
def dom_transforms
|
119
|
+
(rules = @config['dom_transform']) || return
|
120
|
+
rules = rules.select { |r| want_rule(r) }
|
121
|
+
|
122
|
+
rules.each do |rule|
|
123
|
+
transform = DomTransform.create(rule)
|
124
|
+
transform.apply(@node)
|
79
125
|
end
|
80
|
-
doc
|
81
126
|
end
|
82
127
|
|
83
|
-
|
84
|
-
|
85
|
-
|
86
|
-
|
87
|
-
|
88
|
-
|
89
|
-
|
90
|
-
# This ought to work, and would be faster,
|
91
|
-
# but seems to segfault Nokogiri
|
92
|
-
# doc = Nokogiri::HTML('<html><body>')
|
93
|
-
# doc.at('body').children = obj.children
|
94
|
-
# return doc
|
95
|
-
else
|
96
|
-
return to_document(parse(obj))
|
128
|
+
##### Implementations of actions #####
|
129
|
+
|
130
|
+
# Remove double-spacing inside text nodes
|
131
|
+
def self.remove_node_spacing(node)
|
132
|
+
# remove double spacing, but only inside text nodes (eg not attributes)
|
133
|
+
node.xpath('//text()').each do |el|
|
134
|
+
el.content = el.content.gsub(/ +/, ' ')
|
97
135
|
end
|
98
136
|
end
|
99
137
|
|
100
|
-
#
|
101
|
-
def
|
138
|
+
# Get a fragment consisting of the elements matching the selector(s)
|
139
|
+
def self.select_fragments(node, sel)
|
140
|
+
# When we choose a new root, we always become a DocumentFragment,
|
141
|
+
# and lose any DOCTYPE and such.
|
142
|
+
ns = node.css(sel)
|
143
|
+
node = Nokogiri::HTML.fragment('') unless node.fragment?
|
144
|
+
node.children = ns
|
145
|
+
node
|
146
|
+
end
|
147
|
+
|
148
|
+
# Pretty-print some HTML
|
149
|
+
def self.prettify(obj)
|
102
150
|
@stylesheet ||= begin
|
103
151
|
stylesheet_path = File.join(SiteDiff::FILES_DIR, 'pretty_print.xsl')
|
104
152
|
Nokogiri::XSLT(File.read(stylesheet_path))
|
@@ -109,10 +157,22 @@ class SiteDiff
|
|
109
157
|
# but that tends to segfault Nokogiri
|
110
158
|
str = @stylesheet.apply_to(to_document(obj))
|
111
159
|
|
160
|
+
# There's a lot of cruft left over,that we don't want
|
161
|
+
|
162
|
+
# Prevent potential UTF-8 encoding errors by removing invalid bytes.
|
163
|
+
# Not the only solution.
|
164
|
+
# An alternative is to return the string unmodified.
|
165
|
+
str = str.encode(
|
166
|
+
'UTF-8',
|
167
|
+
'binary',
|
168
|
+
invalid: :replace,
|
169
|
+
undef: :replace,
|
170
|
+
replace: ''
|
171
|
+
)
|
112
172
|
# Remove xml declaration and <html> tags
|
113
173
|
str.sub!(/\A<\?xml.*$\n/, '')
|
114
174
|
str.sub!(/\A^<html>$\n/, '')
|
115
|
-
str.sub!(%r
|
175
|
+
str.sub!(%r{</html>\n\Z}, '')
|
116
176
|
|
117
177
|
# Remove top-level indentation
|
118
178
|
indent = /\A(\s*)/.match(str)[1].size
|
@@ -121,73 +181,32 @@ class SiteDiff
|
|
121
181
|
# Remove blank lines
|
122
182
|
str.gsub!(/^\s*$\n/, '')
|
123
183
|
|
124
|
-
|
125
|
-
|
184
|
+
# Remove DOS newlines
|
185
|
+
str.gsub!(/\x0D$/, '')
|
186
|
+
str.gsub!(/ $/, '')
|
126
187
|
|
127
|
-
def remove_spacing(doc)
|
128
|
-
# remove double spacing, but only inside text nodes (eg not attributes)
|
129
|
-
doc.xpath('//text()').each do |node|
|
130
|
-
node.content = node.content.gsub(/ +/, ' ')
|
131
|
-
end
|
132
|
-
end
|
133
|
-
|
134
|
-
# Do one regexp transformation on a string
|
135
|
-
def substitute(str, rule)
|
136
|
-
#FIXME escape forward slashes, right now we are escaping them in YAML!
|
137
|
-
str.gsub!(/#{rule['pattern']}/, rule['substitute'] || '' )
|
138
188
|
str
|
139
189
|
end
|
140
190
|
|
141
|
-
#
|
142
|
-
def
|
143
|
-
|
144
|
-
|
145
|
-
|
146
|
-
|
147
|
-
if sel = rule['selector']
|
148
|
-
node.css(sel).each do |e|
|
149
|
-
e.replace(substitute(e.to_html, rule))
|
150
|
-
end
|
151
|
-
end
|
152
|
-
end
|
153
|
-
|
154
|
-
# If needed, do rules without a selector. We'd rather not convert to
|
155
|
-
# a string unless necessary.
|
156
|
-
global_rules = rules.reject { |r| r['selector'] }
|
157
|
-
return node if global_rules.empty?
|
158
|
-
|
159
|
-
str = node.to_html # Convert to string
|
160
|
-
global_rules.each { |r| substitute(str, r) }
|
161
|
-
return str
|
162
|
-
end
|
163
|
-
|
164
|
-
def select_root(node, sel)
|
165
|
-
return node unless sel
|
166
|
-
|
167
|
-
# When we choose a new root, we always become a DocumentFragment,
|
168
|
-
# and lose any DOCTYPE and such.
|
169
|
-
ns = node.css(sel)
|
170
|
-
unless node.fragment?
|
171
|
-
node = Nokogiri::HTML.fragment('')
|
191
|
+
# Parse HTML into a node
|
192
|
+
def self.domify(str, force_doc = false)
|
193
|
+
if force_doc || /<!DOCTYPE/.match(str[0, 512])
|
194
|
+
Nokogiri::HTML(str)
|
195
|
+
else
|
196
|
+
Nokogiri::HTML.fragment(str)
|
172
197
|
end
|
173
|
-
node.children = ns
|
174
|
-
return node
|
175
198
|
end
|
176
199
|
|
177
|
-
|
178
|
-
|
179
|
-
|
180
|
-
|
181
|
-
|
182
|
-
|
183
|
-
|
184
|
-
|
185
|
-
|
200
|
+
# Force this object to be a document, so we can apply a stylesheet
|
201
|
+
def self.to_document(obj)
|
202
|
+
if Nokogiri::XML::Document == obj.class || Nokogiri::HTML::Document == obj.class
|
203
|
+
obj
|
204
|
+
# node or fragment
|
205
|
+
elsif Nokogiri::XML::Node == obj.class || Nokogiri::HTML::DocumentFragment == obj.class
|
206
|
+
domify(obj.to_s, true)
|
207
|
+
else
|
208
|
+
to_document(domify(obj, false))
|
186
209
|
end
|
187
|
-
|
188
|
-
obj = perform_regexps(node, config['sanitization'])
|
189
|
-
|
190
|
-
return prettify(obj)
|
191
210
|
end
|
192
211
|
end
|
193
212
|
end
|