slaw 10.5.0 → 10.6.0
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/README.md +4 -0
- data/lib/slaw/extract/html_to_akn_text.xsl +29 -23
- data/lib/slaw/grammars/za/act_text.xsl +1 -1
- data/lib/slaw/version.rb +1 -1
- metadata +2 -2
checksums.yaml
CHANGED
|
@@ -1,7 +1,7 @@
|
|
|
1
1
|
---
|
|
2
2
|
SHA256:
|
|
3
|
-
metadata.gz:
|
|
4
|
-
data.tar.gz:
|
|
3
|
+
metadata.gz: 2feb9ea5e726f4300a0920de90b88f14b3a86ce4ae3dddd9a59b3bf8c855e575
|
|
4
|
+
data.tar.gz: 263ef515bc1a81b8cccc1788a3b107b573f3840492913cedf17ae729745d1c4b
|
|
5
5
|
SHA512:
|
|
6
|
-
metadata.gz:
|
|
7
|
-
data.tar.gz:
|
|
6
|
+
metadata.gz: 573bab26b9e880856e74404058208e9fae5f400f0ee64413d771fea872e4b3b7c626ebefb5c020321dc4b4f35919b25b94801c6622e82d2f2a8dfe3507c35f61
|
|
7
|
+
data.tar.gz: 01d73bb0903dbee8213eae364316835361b4d34bbe0ed4a12f9dcaeb114bf1e5ec9ac46133e0570bea1998af0daaac610591da905463d64e63f1449ffd529db7
|
data/README.md
CHANGED
|
@@ -86,6 +86,10 @@ You can create your own grammar by creating a gem that provides these files and
|
|
|
86
86
|
|
|
87
87
|
## Changelog
|
|
88
88
|
|
|
89
|
+
### 10.6.0 (10 May 2021)
|
|
90
|
+
|
|
91
|
+
* Handle sup and sub when extracting from HTML.
|
|
92
|
+
|
|
89
93
|
### 10.5.0 (20 April 2021)
|
|
90
94
|
|
|
91
95
|
* Handle escaping inlines when unparsing.
|
|
@@ -11,9 +11,10 @@
|
|
|
11
11
|
|
|
12
12
|
<xsl:template match="head|style|script|link" />
|
|
13
13
|
|
|
14
|
-
|
|
14
|
+
<!-- block containers that end with newlines -->
|
|
15
|
+
<xsl:template match="ul|ol|section|article|h1|h2|h3|h4|h5">
|
|
15
16
|
<xsl:apply-templates />
|
|
16
|
-
<xsl:text> </xsl:text>
|
|
17
|
+
<xsl:text> </xsl:text>
|
|
17
18
|
</xsl:template>
|
|
18
19
|
|
|
19
20
|
<xsl:template match="ul/li">
|
|
@@ -23,20 +24,23 @@
|
|
|
23
24
|
<xsl:text> </xsl:text>
|
|
24
25
|
</xsl:template>
|
|
25
26
|
|
|
27
|
+
<!-- numbered lists should include a number -->
|
|
26
28
|
<xsl:template match="ol/li">
|
|
27
|
-
<!-- 1. foo -->
|
|
29
|
+
<!-- \1. foo -->
|
|
28
30
|
<xsl:text>\</xsl:text>
|
|
29
|
-
<xsl:
|
|
31
|
+
<xsl:choose>
|
|
32
|
+
<xsl:when test="@value">
|
|
33
|
+
<xsl:value-of select="@value" />
|
|
34
|
+
</xsl:when>
|
|
35
|
+
<xsl:otherwise>
|
|
36
|
+
<xsl:value-of select="position()" />
|
|
37
|
+
</xsl:otherwise>
|
|
38
|
+
</xsl:choose>
|
|
30
39
|
<xsl:text>. </xsl:text>
|
|
31
40
|
<xsl:apply-templates />
|
|
32
41
|
<xsl:text> </xsl:text>
|
|
33
42
|
</xsl:template>
|
|
34
43
|
|
|
35
|
-
<xsl:template match="h1|h2|h3|h4|h5">
|
|
36
|
-
<xsl:apply-templates />
|
|
37
|
-
<xsl:text> </xsl:text>
|
|
38
|
-
</xsl:template>
|
|
39
|
-
|
|
40
44
|
<xsl:template match="p|div">
|
|
41
45
|
<xsl:choose>
|
|
42
46
|
<xsl:when test="starts-with(., '[') and substring(., string-length(.)) = ']'">
|
|
@@ -51,32 +55,27 @@
|
|
|
51
55
|
<xsl:text> </xsl:text>
|
|
52
56
|
</xsl:template>
|
|
53
57
|
|
|
58
|
+
<!-- START tables -->
|
|
59
|
+
|
|
54
60
|
<xsl:template match="table">
|
|
55
61
|
<xsl:text>{| </xsl:text>
|
|
56
|
-
<xsl:text>
|
|
57
|
-
|-</xsl:text>
|
|
62
|
+
<xsl:text> |-</xsl:text>
|
|
58
63
|
<xsl:apply-templates />
|
|
59
|
-
<xsl:text>
|
|
60
|
-
|}
|
|
61
|
-
|
|
62
|
-
</xsl:text>
|
|
64
|
+
<xsl:text> |} </xsl:text>
|
|
63
65
|
</xsl:template>
|
|
64
66
|
|
|
65
67
|
<xsl:template match="tr">
|
|
66
68
|
<xsl:apply-templates />
|
|
67
|
-
<xsl:text>
|
|
68
|
-
|-</xsl:text>
|
|
69
|
+
<xsl:text> |-</xsl:text>
|
|
69
70
|
</xsl:template>
|
|
70
71
|
|
|
71
72
|
<xsl:template match="th|td">
|
|
72
73
|
<xsl:choose>
|
|
73
74
|
<xsl:when test="local-name(.) = 'th'">
|
|
74
|
-
<xsl:text>
|
|
75
|
-
! </xsl:text>
|
|
75
|
+
<xsl:text> ! </xsl:text>
|
|
76
76
|
</xsl:when>
|
|
77
77
|
<xsl:when test="local-name(.) = 'td'">
|
|
78
|
-
<xsl:text>
|
|
79
|
-
| </xsl:text>
|
|
78
|
+
<xsl:text> | </xsl:text>
|
|
80
79
|
</xsl:when>
|
|
81
80
|
</xsl:choose>
|
|
82
81
|
|
|
@@ -118,8 +117,15 @@
|
|
|
118
117
|
</xsl:template>
|
|
119
118
|
|
|
120
119
|
<xsl:template match="br">
|
|
121
|
-
<xsl:text>
|
|
122
|
-
</xsl:
|
|
120
|
+
<xsl:text> </xsl:text>
|
|
121
|
+
</xsl:template>
|
|
122
|
+
|
|
123
|
+
<xsl:template match="sup">
|
|
124
|
+
<xsl:text>^^</xsl:text><xsl:apply-templates /><xsl:text>^^</xsl:text>
|
|
125
|
+
</xsl:template>
|
|
126
|
+
|
|
127
|
+
<xsl:template match="sub">
|
|
128
|
+
<xsl:text>_^</xsl:text><xsl:apply-templates /><xsl:text>^_</xsl:text>
|
|
123
129
|
</xsl:template>
|
|
124
130
|
|
|
125
131
|
|
data/lib/slaw/version.rb
CHANGED
metadata
CHANGED
|
@@ -1,14 +1,14 @@
|
|
|
1
1
|
--- !ruby/object:Gem::Specification
|
|
2
2
|
name: slaw
|
|
3
3
|
version: !ruby/object:Gem::Version
|
|
4
|
-
version: 10.
|
|
4
|
+
version: 10.6.0
|
|
5
5
|
platform: ruby
|
|
6
6
|
authors:
|
|
7
7
|
- Greg Kempe
|
|
8
8
|
autorequire:
|
|
9
9
|
bindir: bin
|
|
10
10
|
cert_chain: []
|
|
11
|
-
date: 2021-
|
|
11
|
+
date: 2021-05-10 00:00:00.000000000 Z
|
|
12
12
|
dependencies:
|
|
13
13
|
- !ruby/object:Gem::Dependency
|
|
14
14
|
name: rake
|