sitediff 0.0.3 → 0.0.5
Sign up to get free protection for your applications and to get access to all the features.
- checksums.yaml +4 -4
- data/bin/sitediff +2 -3
- data/lib/sitediff.rb +35 -24
- data/lib/sitediff/cache.rb +53 -47
- data/lib/sitediff/cli.rb +127 -114
- data/lib/sitediff/config.rb +35 -59
- data/lib/sitediff/config/creator.rb +95 -90
- data/lib/sitediff/crawler.rb +83 -72
- data/lib/sitediff/diff.rb +7 -5
- data/lib/sitediff/exception.rb +3 -1
- data/lib/sitediff/fetch.rb +47 -41
- data/lib/sitediff/files/html_report.html.erb +3 -0
- data/lib/sitediff/files/rules/drupal.yaml +36 -6
- data/lib/sitediff/result.rb +13 -11
- data/lib/sitediff/rules.rb +47 -47
- data/lib/sitediff/sanitize.rb +145 -150
- data/lib/sitediff/sanitize/dom_transform.rb +73 -74
- data/lib/sitediff/sanitize/regexp.rb +55 -52
- data/lib/sitediff/uriwrapper.rb +37 -26
- data/lib/sitediff/webserver.rb +80 -77
- data/lib/sitediff/webserver/resultserver.rb +117 -76
- metadata +32 -44
data/lib/sitediff/diff.rb
CHANGED
@@ -1,3 +1,5 @@
|
|
1
|
+
# frozen_string_literal: true
|
2
|
+
|
1
3
|
require 'sitediff'
|
2
4
|
require 'diffy'
|
3
5
|
require 'erb'
|
@@ -9,26 +11,26 @@ class SiteDiff
|
|
9
11
|
|
10
12
|
def html_diffy(before_html, after_html)
|
11
13
|
diff = Diffy::Diff.new(before_html, after_html)
|
12
|
-
diff.first ?
|
14
|
+
diff.first ? # Is it non-empty?
|
13
15
|
diff.to_s(:html) : nil
|
14
16
|
end
|
15
17
|
|
16
18
|
def terminal_diffy(before_html, after_html)
|
17
19
|
args = []
|
18
20
|
args << :color if Rainbow.enabled
|
19
|
-
|
20
|
-
|
21
|
+
Diffy::Diff.new(before_html, after_html, context: 3)
|
22
|
+
.to_s(*args)
|
21
23
|
end
|
22
24
|
|
23
25
|
def generate_html_report(results, before, after, cache)
|
24
26
|
erb_path = File.join(SiteDiff::FILES_DIR, 'html_report.html.erb')
|
25
27
|
report_html = ERB.new(File.read(erb_path)).result(binding)
|
26
|
-
|
28
|
+
report_html
|
27
29
|
end
|
28
30
|
|
29
31
|
def generate_diff_output(result)
|
30
32
|
erb_path = File.join(SiteDiff::FILES_DIR, 'diff.html.erb')
|
31
|
-
|
33
|
+
ERB.new(File.read(erb_path)).result(binding)
|
32
34
|
end
|
33
35
|
|
34
36
|
def css
|
data/lib/sitediff/exception.rb
CHANGED
data/lib/sitediff/fetch.rb
CHANGED
@@ -1,55 +1,61 @@
|
|
1
|
+
# frozen_string_literal: true
|
2
|
+
|
1
3
|
require 'sitediff/uriwrapper'
|
2
4
|
require 'typhoeus'
|
3
5
|
|
4
6
|
class SiteDiff
|
5
|
-
class Fetch
|
6
|
-
|
7
|
-
|
8
|
-
|
9
|
-
|
10
|
-
|
11
|
-
|
12
|
-
|
13
|
-
|
7
|
+
class Fetch
|
8
|
+
# Cache is a cache object, see sitediff/cache
|
9
|
+
# Paths is a list of sub-paths
|
10
|
+
# Tags is a hash of tag names => base URLs.
|
11
|
+
def initialize(cache, paths, concurrency = 3, curl_opts = nil, **tags)
|
12
|
+
@cache = cache
|
13
|
+
@paths = paths
|
14
|
+
@tags = tags
|
15
|
+
@curl_opts = curl_opts || UriWrapper::DEFAULT_CURL_OPTS
|
16
|
+
@concurrency = concurrency
|
17
|
+
end
|
14
18
|
|
15
|
-
|
16
|
-
|
17
|
-
|
18
|
-
|
19
|
-
|
20
|
-
|
21
|
-
|
22
|
-
|
23
|
-
|
19
|
+
# Fetch all the paths, once per tag.
|
20
|
+
# When a path has been fetched for every tag, block will be called with the
|
21
|
+
# path, and a hash of tag => UriWrapper::ReadResult objects.
|
22
|
+
def run(&block)
|
23
|
+
@callback = block
|
24
|
+
@hydra = Typhoeus::Hydra.new(max_concurrency: @concurrency)
|
25
|
+
@paths.each { |path| queue_path(path) }
|
26
|
+
@hydra.run
|
27
|
+
end
|
28
|
+
|
29
|
+
private
|
24
30
|
|
25
|
-
|
26
|
-
|
27
|
-
|
28
|
-
|
29
|
-
|
30
|
-
|
31
|
-
if res = @cache.get(tag, path)
|
32
|
-
results[tag] = res
|
33
|
-
process_results(path, results)
|
34
|
-
elsif !base
|
35
|
-
# We only have the cache, but this item isn't cached!
|
36
|
-
results[tag] = UriWrapper::ReadResult.error("Not cached")
|
37
|
-
process_results(path, results)
|
38
|
-
else
|
39
|
-
uri = UriWrapper.new(base + path)
|
40
|
-
uri.queue(@hydra) do |res|
|
41
|
-
@cache.set(tag, path, res)
|
31
|
+
# Queue a path for fetching
|
32
|
+
def queue_path(path)
|
33
|
+
results = {}
|
34
|
+
|
35
|
+
@tags.each do |tag, base|
|
36
|
+
if (res = @cache.get(tag, path))
|
42
37
|
results[tag] = res
|
43
38
|
process_results(path, results)
|
39
|
+
elsif !base
|
40
|
+
# We only have the cache, but this item isn't cached!
|
41
|
+
results[tag] = UriWrapper::ReadResult.error('Not cached')
|
42
|
+
process_results(path, results)
|
43
|
+
else
|
44
|
+
uri = UriWrapper.new(base + path, @curl_opts)
|
45
|
+
uri.queue(@hydra) do |resl|
|
46
|
+
@cache.set(tag, path, resl)
|
47
|
+
results[tag] = resl
|
48
|
+
process_results(path, results)
|
49
|
+
end
|
44
50
|
end
|
45
51
|
end
|
46
52
|
end
|
47
|
-
end
|
48
53
|
|
49
|
-
|
50
|
-
|
51
|
-
|
52
|
-
|
54
|
+
# Process fetch results
|
55
|
+
def process_results(path, results)
|
56
|
+
return unless results.size == @tags.size
|
57
|
+
|
58
|
+
@callback[path, results]
|
59
|
+
end
|
53
60
|
end
|
54
61
|
end
|
55
|
-
end
|
@@ -2,28 +2,28 @@ sanitization:
|
|
2
2
|
- title: Strip Drupal.settings
|
3
3
|
selector: script
|
4
4
|
pattern: '^(<script>)?jQuery.extend\(Drupal.settings.*$'
|
5
|
+
- title: Strip IE CSS/JS cache IDs
|
6
|
+
pattern: '("[^"]*ie\d?\.(js|css))\?[a-z0-9]{6}"'
|
7
|
+
substitute: '\1'
|
5
8
|
- title: Strip form build ID
|
6
9
|
selector: input
|
7
|
-
pattern: 'name="form_build_id" value="form-[-\w]{43}"'
|
10
|
+
pattern: 'name="form_build_id" value="form-[-\w]{40,43}"'
|
8
11
|
substitute: 'name="form_build_id" value="form-DRUPAL_FORM_BUILD_ID"'
|
9
12
|
- title: Strip view DOM ID
|
10
13
|
pattern: '(class="view .*) view-dom-id-[a-f0-9]{32}"'
|
11
14
|
substitute: '\1 view-dom-id-DRUPAL_VIEW_DOM_ID"'
|
12
15
|
- title: Strip CSS aggregation filenames
|
13
16
|
selector: link[rel=stylesheet]
|
14
|
-
pattern: '(href="[^"]*/files/css/css_)[-\w]{43}\.css"'
|
17
|
+
pattern: '(href="[^"]*/files/css/css_)[-\w]{40,43}\.css"'
|
15
18
|
substitute: '\1DRUPAL_AGGREGATED_CSS.css"'
|
16
19
|
- title: Strip JS aggregation filenames
|
17
20
|
selector: script
|
18
|
-
pattern: '(src="[^"]*/files/js/js_)[-\w]{43}\.js"'
|
21
|
+
pattern: '(src="[^"]*/files/js/js_)[-\w]{40,43}\.js"'
|
19
22
|
substitute: '\1DRUPAL_AGGREGATED_JS.js"'
|
20
23
|
- title: Strip CSS/JS cache IDs
|
21
24
|
selector: style, script
|
22
25
|
pattern: '("[^"]*\.(js|css))\?[a-z0-9]{6}"'
|
23
26
|
substitute: '\1'
|
24
|
-
- title: Strip IE CSS/JS cache IDs
|
25
|
-
pattern: '("[^"]*ie\d?\.(js|css))\?[a-z0-9]{6}"'
|
26
|
-
substitute: '\1'
|
27
27
|
- title: Strip Drupal JS version tags
|
28
28
|
selector: script
|
29
29
|
pattern: '(src="[^"]*/misc/\w+\.js)?v=\d+\.\d+"'
|
@@ -31,3 +31,33 @@ sanitization:
|
|
31
31
|
- title: Strip domain names from absolute URLs
|
32
32
|
pattern: 'http:\/\/[a-zA-Z0-9.:-]+'
|
33
33
|
substitute: '__domain__'
|
34
|
+
- title: Strip form build ID
|
35
|
+
selector: input
|
36
|
+
pattern: 'autocomplete="off" data-drupal-selector="form-[-\w]{40,43}"'
|
37
|
+
substitute: 'autocomplete="off" data-drupal-selector="form-DRUPAL_FORM_BUILD_ID"'
|
38
|
+
- title: Strip form build ID 2
|
39
|
+
selector: input
|
40
|
+
pattern: 'name="form_build_id" value="form-[-\w]{40,43}"'
|
41
|
+
substitute: 'name="form_build_id" value="form-DRUPAL_FORM_BUILD_ID"'
|
42
|
+
- title: Strip Drupal CSS link queries
|
43
|
+
selector: link
|
44
|
+
pattern: '\.css\?(\w*)'
|
45
|
+
substitute: '\.css'
|
46
|
+
- title: Strip Drupal JS link queries
|
47
|
+
selector: script
|
48
|
+
pattern: '\.js\?(\w*)'
|
49
|
+
substitute: '\.js'
|
50
|
+
- title: Strip Drupal View-DOM ID
|
51
|
+
pattern: 'view-dom-id-\w*'
|
52
|
+
substitute: 'view-dom-id-_ID_'
|
53
|
+
- title: Strip Drupal View-DOM ID 2
|
54
|
+
pattern: '(views?_dom_id"?:"?)\w*'
|
55
|
+
substitute: '\1_ID_'
|
56
|
+
- title: Ignore Drupal CSS file names
|
57
|
+
selector: link
|
58
|
+
pattern: 'css_[-\w]{40,43}(\\|%5C)?\.css'
|
59
|
+
substitute: 'css__ID__.css'
|
60
|
+
- title: Ignore Drupal JS file names
|
61
|
+
selector: script
|
62
|
+
pattern: 'js_[-\w]{40,43}\\?\.js'
|
63
|
+
substitute: 'js__ID__.js'
|
data/lib/sitediff/result.rb
CHANGED
@@ -1,3 +1,5 @@
|
|
1
|
+
# frozen_string_literal: true
|
2
|
+
|
1
3
|
require 'sitediff'
|
2
4
|
require 'sitediff/diff'
|
3
5
|
require 'digest/sha1'
|
@@ -8,7 +10,7 @@ class SiteDiff
|
|
8
10
|
STATUS_SUCCESS = 0 # Identical before and after
|
9
11
|
STATUS_FAILURE = 1 # Different before and after
|
10
12
|
STATUS_ERROR = 2 # Couldn't fetch page
|
11
|
-
STATUS_TEXT = %w[success failure error]
|
13
|
+
STATUS_TEXT = %w[success failure error].freeze
|
12
14
|
|
13
15
|
attr_reader :status, :diff
|
14
16
|
|
@@ -17,7 +19,7 @@ class SiteDiff
|
|
17
19
|
if error
|
18
20
|
@status = STATUS_ERROR
|
19
21
|
else
|
20
|
-
@diff = Diff
|
22
|
+
@diff = Diff.html_diffy(before, after)
|
21
23
|
@status = @diff ? STATUS_FAILURE : STATUS_SUCCESS
|
22
24
|
end
|
23
25
|
end
|
@@ -28,7 +30,7 @@ class SiteDiff
|
|
28
30
|
|
29
31
|
# Textual representation of the status
|
30
32
|
def status_text
|
31
|
-
|
33
|
+
STATUS_TEXT[status]
|
32
34
|
end
|
33
35
|
|
34
36
|
# Printable URL
|
@@ -39,7 +41,7 @@ class SiteDiff
|
|
39
41
|
|
40
42
|
# Filename to store diff
|
41
43
|
def filename
|
42
|
-
File.join(SiteDiff::DIFFS_DIR, Digest::SHA1.hexdigest(
|
44
|
+
File.join(SiteDiff::DIFFS_DIR, Digest::SHA1.hexdigest(path) + '.html')
|
43
45
|
end
|
44
46
|
|
45
47
|
# Text of the link in the HTML report
|
@@ -52,15 +54,15 @@ class SiteDiff
|
|
52
54
|
end
|
53
55
|
|
54
56
|
# Log the result to the terminal
|
55
|
-
def log(verbose=true)
|
57
|
+
def log(verbose = true)
|
56
58
|
case status
|
57
59
|
when STATUS_SUCCESS then
|
58
|
-
SiteDiff
|
60
|
+
SiteDiff.log path, :diff_success, 'UNCHANGED'
|
59
61
|
when STATUS_ERROR then
|
60
|
-
SiteDiff
|
62
|
+
SiteDiff.log path, :warn, "ERROR (#{error})"
|
61
63
|
when STATUS_FAILURE then
|
62
|
-
SiteDiff
|
63
|
-
puts Diff
|
64
|
+
SiteDiff.log path, :diff_failure, 'CHANGED'
|
65
|
+
puts Diff.terminal_diffy(before, after) if verbose
|
64
66
|
end
|
65
67
|
end
|
66
68
|
|
@@ -68,9 +70,9 @@ class SiteDiff
|
|
68
70
|
def dump(dir)
|
69
71
|
dump_path = File.join(dir, filename)
|
70
72
|
base = File.dirname(dump_path)
|
71
|
-
FileUtils
|
73
|
+
FileUtils.mkdir_p(base) unless File.exist?(base)
|
72
74
|
File.open(dump_path, 'w') do |f|
|
73
|
-
f.write(Diff
|
75
|
+
f.write(Diff.generate_diff_output(self))
|
74
76
|
end
|
75
77
|
end
|
76
78
|
end
|
data/lib/sitediff/rules.rb
CHANGED
@@ -1,65 +1,65 @@
|
|
1
|
+
# frozen_string_literal: true
|
2
|
+
|
1
3
|
require 'sitediff/sanitize/regexp'
|
2
4
|
require 'pathname'
|
3
5
|
require 'set'
|
4
6
|
|
5
7
|
class SiteDiff
|
6
|
-
# Find appropriate rules for a given site
|
7
|
-
class Rules
|
8
|
-
|
9
|
-
|
10
|
-
|
11
|
-
|
12
|
-
|
13
|
-
|
8
|
+
# Find appropriate rules for a given site
|
9
|
+
class Rules
|
10
|
+
def initialize(config, disabled = false)
|
11
|
+
@disabled = disabled
|
12
|
+
@config = config
|
13
|
+
find_sanitization_candidates
|
14
|
+
@rules = Hash.new { |h, k| h[k] = Set.new }
|
15
|
+
end
|
14
16
|
|
15
|
-
|
16
|
-
|
17
|
+
def find_sanitization_candidates
|
18
|
+
@candidates = Set.new
|
17
19
|
|
18
|
-
|
19
|
-
|
20
|
-
|
21
|
-
conf = YAML.load_file(f)
|
22
|
-
@candidates.merge(conf['sanitization'])
|
23
|
-
end
|
24
|
-
end
|
20
|
+
rules_dir = Pathname.new(__FILE__).dirname + 'files' + 'rules'
|
21
|
+
rules_dir.children.each do |f|
|
22
|
+
next unless f.file? && f.extname == '.yaml'
|
25
23
|
|
26
|
-
|
27
|
-
|
28
|
-
|
29
|
-
|
24
|
+
conf = YAML.load_file(f)
|
25
|
+
@candidates.merge(conf['sanitization'])
|
26
|
+
end
|
27
|
+
end
|
30
28
|
|
31
|
-
|
32
|
-
|
33
|
-
|
34
|
-
|
29
|
+
def handle_page(tag, html, doc)
|
30
|
+
found = find_rules(html, doc)
|
31
|
+
@rules[tag].merge(found)
|
32
|
+
end
|
35
33
|
|
36
|
-
|
37
|
-
|
38
|
-
|
34
|
+
# Yield a set of rules that seem reasonable for this HTML
|
35
|
+
# assumption: the YAML file is a list of regexp rules only
|
36
|
+
def find_rules(html, doc)
|
37
|
+
@candidates.select do |rule|
|
38
|
+
re = SiteDiff::Sanitizer::Regexp.create(rule)
|
39
|
+
re.applies?(html, doc)
|
40
|
+
end
|
39
41
|
end
|
40
|
-
end
|
41
42
|
|
42
|
-
|
43
|
-
|
44
|
-
|
43
|
+
# Find all rules from all rulesets that apply for all pages
|
44
|
+
def add_config
|
45
|
+
have_both = @rules.include?(:before)
|
45
46
|
|
46
|
-
|
47
|
-
|
48
|
-
|
49
|
-
|
50
|
-
|
51
|
-
|
52
|
-
|
47
|
+
r1, r2 = *@rules.values_at(:before, :after)
|
48
|
+
if have_both
|
49
|
+
add_section('before', r1 - r2)
|
50
|
+
add_section('after', r2 - r1)
|
51
|
+
add_section(nil, r1 & r2)
|
52
|
+
else
|
53
|
+
add_section(nil, r2)
|
54
|
+
end
|
53
55
|
end
|
54
|
-
end
|
55
56
|
|
56
|
-
|
57
|
-
|
58
|
-
|
59
|
-
|
60
|
-
rules.each { |r| r['disabled'] = true }
|
57
|
+
def add_section(name, rules)
|
58
|
+
return if rules.empty?
|
59
|
+
|
60
|
+
conf = name ? @config[name] : @config
|
61
|
+
rules.each { |r| r['disabled'] = true } if @disabled
|
62
|
+
conf['sanitization'] = rules.to_a.sort_by { |r| r['title'] }
|
61
63
|
end
|
62
|
-
conf['sanitization'] = rules.to_a.sort_by { |r| r['title'] }
|
63
64
|
end
|
64
65
|
end
|
65
|
-
end
|
data/lib/sitediff/sanitize.rb
CHANGED
@@ -1,3 +1,5 @@
|
|
1
|
+
# frozen_string_literal: true
|
2
|
+
|
1
3
|
require 'sitediff'
|
2
4
|
require 'sitediff/exception'
|
3
5
|
require 'sitediff/sanitize/dom_transform'
|
@@ -6,183 +8,176 @@ require 'nokogiri'
|
|
6
8
|
require 'set'
|
7
9
|
|
8
10
|
class SiteDiff
|
9
|
-
class Sanitizer
|
10
|
-
class InvalidSanitization < SiteDiffException; end
|
11
|
-
|
12
|
-
TOOLS = {
|
13
|
-
|
14
|
-
|
15
|
-
}
|
16
|
-
DOM_TRANSFORMS = Set.new(%w[remove unwrap_root unwrap remove_class])
|
17
|
-
|
18
|
-
def initialize(html, config, opts = {})
|
19
|
-
|
20
|
-
|
21
|
-
|
22
|
-
end
|
23
|
-
|
24
|
-
def sanitize
|
25
|
-
return '' if @html == '' # Quick return on empty input
|
11
|
+
class Sanitizer
|
12
|
+
class InvalidSanitization < SiteDiffException; end
|
13
|
+
|
14
|
+
TOOLS = {
|
15
|
+
array: %w[dom_transform sanitization],
|
16
|
+
scalar: %w[selector remove_spacing]
|
17
|
+
}.freeze
|
18
|
+
DOM_TRANSFORMS = Set.new(%w[remove unwrap_root unwrap remove_class])
|
19
|
+
|
20
|
+
def initialize(html, config, opts = {})
|
21
|
+
@html = html
|
22
|
+
@config = config
|
23
|
+
@opts = opts
|
24
|
+
end
|
26
25
|
|
27
|
-
|
26
|
+
def sanitize
|
27
|
+
return '' if @html == '' # Quick return on empty input
|
28
28
|
|
29
|
-
|
30
|
-
|
31
|
-
dom_transforms
|
32
|
-
regexps
|
29
|
+
@node = Sanitizer.domify(@html)
|
30
|
+
@html = nil
|
33
31
|
|
34
|
-
|
35
|
-
|
32
|
+
remove_spacing
|
33
|
+
selector
|
34
|
+
dom_transforms
|
35
|
+
regexps
|
36
36
|
|
37
|
-
|
38
|
-
|
39
|
-
return false unless rule
|
40
|
-
return false if rule['disabled']
|
37
|
+
@html || Sanitizer.prettify(@node)
|
38
|
+
end
|
41
39
|
|
42
|
-
|
43
|
-
|
44
|
-
|
45
|
-
|
40
|
+
# Return whether or not we want to keep a rule
|
41
|
+
def want_rule(rule)
|
42
|
+
return false unless rule
|
43
|
+
return false if rule['disabled']
|
46
44
|
|
47
|
-
|
48
|
-
|
45
|
+
# Filter out if path regexp doesn't match
|
46
|
+
if (pathre = rule['path']) && (path = @opts[:path])
|
47
|
+
return ::Regexp.new(pathre).match(path)
|
48
|
+
end
|
49
49
|
|
50
|
-
|
51
|
-
|
52
|
-
# Turn it into an array of hashes.
|
53
|
-
def canonicalize_rule(name)
|
54
|
-
rules = @config[name] or return nil
|
55
|
-
|
56
|
-
if rules[0] && rules[0].respond_to?(:[]) && rules[0]['value']
|
57
|
-
# Already an array
|
58
|
-
elsif rules['value']
|
59
|
-
# Hash, put it in an array
|
60
|
-
rules = [rules]
|
61
|
-
else
|
62
|
-
# Scalar, put it in a hash
|
63
|
-
rules = [{ 'value' => rules }]
|
64
|
-
end
|
50
|
+
true
|
51
|
+
end
|
65
52
|
|
66
|
-
|
67
|
-
|
68
|
-
|
69
|
-
|
70
|
-
|
53
|
+
# Canonicalize a simple rule, eg: 'remove_spacing' or 'selector'.
|
54
|
+
# It may be a simple value, or a hash, or an array of hashes.
|
55
|
+
# Turn it into an array of hashes.
|
56
|
+
def canonicalize_rule(name)
|
57
|
+
(rules = @config[name]) || (return nil)
|
58
|
+
|
59
|
+
if rules[0]&.respond_to?(:[]) && rules[0]['value']
|
60
|
+
# Already an array
|
61
|
+
elsif rules['value']
|
62
|
+
# Hash, put it in an array
|
63
|
+
rules = [rules]
|
64
|
+
else
|
65
|
+
# Scalar, put it in a hash
|
66
|
+
rules = [{ 'value' => rules }]
|
67
|
+
end
|
68
|
+
|
69
|
+
want = rules.select { |r| want_rule(r) }
|
70
|
+
return nil if want.empty?
|
71
|
+
raise "Too many matching rules of type #{name}" if want.size > 1
|
72
|
+
|
73
|
+
want.first
|
74
|
+
end
|
71
75
|
|
72
|
-
# Perform 'remove_spacing' action
|
73
|
-
def remove_spacing
|
74
|
-
|
75
|
-
|
76
|
-
end
|
76
|
+
# Perform 'remove_spacing' action
|
77
|
+
def remove_spacing
|
78
|
+
(rule = canonicalize_rule('remove_spacing')) || return
|
79
|
+
Sanitizer.remove_node_spacing(@node) if rule['value']
|
80
|
+
end
|
77
81
|
|
78
|
-
# Perform 'selector' action, to choose a new root
|
79
|
-
def selector
|
80
|
-
|
81
|
-
|
82
|
-
end
|
82
|
+
# Perform 'selector' action, to choose a new root
|
83
|
+
def selector
|
84
|
+
(rule = canonicalize_rule('selector')) || return
|
85
|
+
@node = Sanitizer.select_fragments(@node, rule['value'])
|
86
|
+
end
|
83
87
|
|
84
|
-
# Applies regexps. Also
|
85
|
-
def regexps
|
86
|
-
|
87
|
-
|
88
|
+
# Applies regexps. Also
|
89
|
+
def regexps
|
90
|
+
(rules = @config['sanitization']) || return
|
91
|
+
rules = rules.select { |r| want_rule(r) }
|
88
92
|
|
89
|
-
|
90
|
-
|
93
|
+
rules.map! { |r| Regexp.create(r) }
|
94
|
+
selector, global = rules.partition(&:selector?)
|
91
95
|
|
92
|
-
|
93
|
-
|
94
|
-
|
95
|
-
|
96
|
+
selector.each { |r| r.apply(@node) }
|
97
|
+
@html = Sanitizer.prettify(@node)
|
98
|
+
@node = nil
|
99
|
+
global.each { |r| r.apply(@html) }
|
100
|
+
end
|
96
101
|
|
97
|
-
# Perform DOM transforms
|
98
|
-
def dom_transforms
|
99
|
-
|
100
|
-
|
102
|
+
# Perform DOM transforms
|
103
|
+
def dom_transforms
|
104
|
+
(rules = @config['dom_transform']) || return
|
105
|
+
rules = rules.select { |r| want_rule(r) }
|
101
106
|
|
102
|
-
|
103
|
-
|
104
|
-
|
105
|
-
|
106
|
-
end
|
107
|
+
rules.each do |rule|
|
108
|
+
transform = DomTransform.create(rule)
|
109
|
+
transform.apply(@node)
|
110
|
+
end
|
111
|
+
end
|
107
112
|
|
108
|
-
##### Implementations of actions #####
|
113
|
+
##### Implementations of actions #####
|
109
114
|
|
110
|
-
# Remove double-spacing inside text nodes
|
111
|
-
def self.remove_node_spacing(node)
|
112
|
-
|
113
|
-
|
114
|
-
|
115
|
-
|
116
|
-
end
|
115
|
+
# Remove double-spacing inside text nodes
|
116
|
+
def self.remove_node_spacing(node)
|
117
|
+
# remove double spacing, but only inside text nodes (eg not attributes)
|
118
|
+
node.xpath('//text()').each do |el|
|
119
|
+
el.content = el.content.gsub(/ +/, ' ')
|
120
|
+
end
|
121
|
+
end
|
117
122
|
|
118
|
-
# Get a fragment consisting of the elements matching the selector(s)
|
119
|
-
def self.select_fragments(node, sel)
|
120
|
-
|
121
|
-
|
122
|
-
|
123
|
-
|
124
|
-
|
125
|
-
|
126
|
-
|
127
|
-
return node
|
128
|
-
end
|
123
|
+
# Get a fragment consisting of the elements matching the selector(s)
|
124
|
+
def self.select_fragments(node, sel)
|
125
|
+
# When we choose a new root, we always become a DocumentFragment,
|
126
|
+
# and lose any DOCTYPE and such.
|
127
|
+
ns = node.css(sel)
|
128
|
+
node = Nokogiri::HTML.fragment('') unless node.fragment?
|
129
|
+
node.children = ns
|
130
|
+
node
|
131
|
+
end
|
129
132
|
|
130
|
-
# Pretty-print some HTML
|
131
|
-
def self.prettify(obj)
|
132
|
-
|
133
|
-
|
134
|
-
|
135
|
-
|
133
|
+
# Pretty-print some HTML
|
134
|
+
def self.prettify(obj)
|
135
|
+
@stylesheet ||= begin
|
136
|
+
stylesheet_path = File.join(SiteDiff::FILES_DIR, 'pretty_print.xsl')
|
137
|
+
Nokogiri::XSLT(File.read(stylesheet_path))
|
138
|
+
end
|
136
139
|
|
137
|
-
|
138
|
-
|
139
|
-
|
140
|
-
|
140
|
+
# Pull out the html element's children
|
141
|
+
# The obvious way to do this is to iterate over pretty.css('html'),
|
142
|
+
# but that tends to segfault Nokogiri
|
143
|
+
str = @stylesheet.apply_to(to_document(obj))
|
141
144
|
|
142
|
-
|
145
|
+
# There's a lot of cruft left over,that we don't want
|
143
146
|
|
144
|
-
|
145
|
-
|
146
|
-
|
147
|
-
|
147
|
+
# Remove xml declaration and <html> tags
|
148
|
+
str.sub!(/\A<\?xml.*$\n/, '')
|
149
|
+
str.sub!(/\A^<html>$\n/, '')
|
150
|
+
str.sub!(%r{</html>\n\Z}, '')
|
148
151
|
|
149
|
-
|
150
|
-
|
151
|
-
|
152
|
+
# Remove top-level indentation
|
153
|
+
indent = /\A(\s*)/.match(str)[1].size
|
154
|
+
str.gsub!(/^\s{,#{indent}}/, '')
|
152
155
|
|
153
|
-
|
154
|
-
|
156
|
+
# Remove blank lines
|
157
|
+
str.gsub!(/^\s*$\n/, '')
|
155
158
|
|
156
|
-
|
157
|
-
end
|
159
|
+
str
|
160
|
+
end
|
158
161
|
|
159
|
-
# Parse HTML into a node
|
160
|
-
def self.domify(str, force_doc = false)
|
161
|
-
|
162
|
-
|
163
|
-
|
164
|
-
|
165
|
-
|
166
|
-
end
|
162
|
+
# Parse HTML into a node
|
163
|
+
def self.domify(str, force_doc = false)
|
164
|
+
if force_doc || /<!DOCTYPE/.match(str[0, 512])
|
165
|
+
Nokogiri::HTML(str)
|
166
|
+
else
|
167
|
+
Nokogiri::HTML.fragment(str)
|
168
|
+
end
|
169
|
+
end
|
167
170
|
|
168
|
-
# Force this object to be a document, so we can apply a stylesheet
|
169
|
-
def self.to_document(obj)
|
170
|
-
|
171
|
-
|
172
|
-
|
173
|
-
|
174
|
-
|
175
|
-
|
176
|
-
|
177
|
-
|
178
|
-
doc = Nokogiri::HTML('<html><body>')
|
179
|
-
doc.at('body').children = obj.children
|
180
|
-
return doc
|
171
|
+
# Force this object to be a document, so we can apply a stylesheet
|
172
|
+
def self.to_document(obj)
|
173
|
+
if Nokogiri::XML::Document == obj.class || Nokogiri::HTML::Document == obj.class
|
174
|
+
obj
|
175
|
+
# node or fragment
|
176
|
+
elsif Nokogiri::XML::Node == obj.class || Nokogiri::HTML::DocumentFragment == obj.class
|
177
|
+
domify(obj.to_s, true)
|
178
|
+
else
|
179
|
+
to_document(domify(obj, false))
|
180
|
+
end
|
181
181
|
end
|
182
|
-
else
|
183
|
-
return to_document(domify(obj))
|
184
182
|
end
|
185
183
|
end
|
186
|
-
|
187
|
-
end
|
188
|
-
end
|