deba 0.12.0 → 0.13.0
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- checksums.yaml +4 -4
- data/lib/deba.rb +1 -1
- data/lib/deba/document.rb +3 -1
- data/lib/deba/extractor.rb +29 -2
- metadata +1 -1
checksums.yaml
CHANGED
|
@@ -1,7 +1,7 @@
|
|
|
1
1
|
---
|
|
2
2
|
SHA1:
|
|
3
|
-
metadata.gz:
|
|
4
|
-
data.tar.gz:
|
|
3
|
+
metadata.gz: bc1db7aa56a4e3a035d8cbefe54cd49742faee24
|
|
4
|
+
data.tar.gz: 7d1f0a909f5675ae614792642711fdf94f865469
|
|
5
5
|
SHA512:
|
|
6
|
-
metadata.gz:
|
|
7
|
-
data.tar.gz:
|
|
6
|
+
metadata.gz: 78ca7576ddd7698df8851acf773cf566fbb29ceed3748cc73d0b5844add6a6665e760adecb14207358a2217e59b9d48caa1727cd92e71fc65adaa976119f55af
|
|
7
|
+
data.tar.gz: b2f5628703302d6cad7c5f63a841c53a3e24c973e1d6bf37e3a8216a128c5b2d011ba346d8238b8a933d3cfa551012b8f5a5bbd5da663f207a6cf6583d9b8639
|
data/lib/deba.rb
CHANGED
data/lib/deba/document.rb
CHANGED
|
@@ -1,4 +1,6 @@
|
|
|
1
1
|
class Deba::Document
|
|
2
|
+
BLOCKQUOTE = "> "
|
|
3
|
+
|
|
2
4
|
attr_reader :content
|
|
3
5
|
|
|
4
6
|
def initialize(extractor)
|
|
@@ -22,7 +24,7 @@ class Deba::Document
|
|
|
22
24
|
|
|
23
25
|
@args.unshift(@segments)
|
|
24
26
|
block = @block_type.new(*@args).to_a
|
|
25
|
-
block.unshift(
|
|
27
|
+
block.unshift(BLOCKQUOTE) if @extractor.in_blockquote?
|
|
26
28
|
|
|
27
29
|
@content << Deba::Stringifier.new(block).stringify
|
|
28
30
|
end
|
data/lib/deba/extractor.rb
CHANGED
|
@@ -1,7 +1,34 @@
|
|
|
1
1
|
class Deba::Extractor
|
|
2
2
|
HEADING_TAGS = %w(h1 h2 h3 h4 h5 h6)
|
|
3
|
-
BLOCK_INITIATING_TAGS = %w(
|
|
3
|
+
BLOCK_INITIATING_TAGS = %w(
|
|
4
|
+
address
|
|
5
|
+
article
|
|
6
|
+
aside
|
|
7
|
+
body
|
|
8
|
+
blockquote
|
|
9
|
+
div
|
|
10
|
+
dd
|
|
11
|
+
dl
|
|
12
|
+
dt
|
|
13
|
+
figure
|
|
14
|
+
footer
|
|
15
|
+
header
|
|
16
|
+
li
|
|
17
|
+
main
|
|
18
|
+
nav
|
|
19
|
+
ol
|
|
20
|
+
p
|
|
21
|
+
pre
|
|
22
|
+
section
|
|
23
|
+
td
|
|
24
|
+
th
|
|
25
|
+
ul)
|
|
4
26
|
ENHANCERS = { %w(b strong) => "*", %w(i em) => "_" }
|
|
27
|
+
SKIP_TAGS = %w(
|
|
28
|
+
head
|
|
29
|
+
style
|
|
30
|
+
script
|
|
31
|
+
)
|
|
5
32
|
|
|
6
33
|
attr_reader :blocks
|
|
7
34
|
|
|
@@ -27,7 +54,7 @@ class Deba::Extractor
|
|
|
27
54
|
|
|
28
55
|
node_name = node.name.downcase
|
|
29
56
|
|
|
30
|
-
return if node_name
|
|
57
|
+
return if SKIP_TAGS.include?(node_name)
|
|
31
58
|
|
|
32
59
|
#Handle repeated brs by making a paragraph break
|
|
33
60
|
if node_name == 'br'
|