Packages

floki

0.19.3
0.38.4 0.38.3 0.38.2 0.38.1 0.38.0 0.37.1 0.37.0 0.36.3 0.36.2 0.36.1 0.36.0 0.35.4 0.35.3 0.35.2 0.35.1 0.35.0 0.34.3 0.34.2 0.34.1 0.34.0 0.33.1 0.33.0 0.32.1 0.32.0 0.31.0 0.30.1 0.30.0 0.29.0 0.28.0 0.27.0 0.26.0 0.25.0 0.24.0 0.23.1 0.23.0 0.22.0 0.21.0 0.20.4 0.20.3 0.20.2 0.20.1 0.20.0 0.19.3 0.19.2 0.19.1 0.19.0 0.18.1 0.18.0 0.17.2 0.17.1 0.17.0 0.16.0 0.15.0 0.14.0 0.13.2 0.13.1 0.13.0 0.12.1 0.12.0 0.11.0 0.10.1 0.10.0 0.9.0 0.8.1 0.8.0 0.7.2 0.7.1 0.7.0 0.6.1 0.6.0 0.5.0 0.4.1 0.4.0 0.3.3 0.3.2 0.3.1 0.3.0 0.2.1 0.2.0 0.1.1 0.1.0 0.0.5 0.0.4 0.0.3 0.0.2 0.0.1

Floki is a simple HTML parser that enables search for nodes using CSS selectors.

HexDocs

Current section

86 Versions

Jump to
Readme
86 Versions
2 Dependencies
Dependants
Files
Activity
Hide whitespace

Compare versions

5 files changed
+98 additions
-84 deletions
  @@ -4,20 +4,23 @@
4 4 <<"Floki is a simple HTML parser that enables search for nodes using CSS selectors.">>}.
5 5 {<<"elixir">>,<<">= 1.3.0">>}.
6 6 {<<"files">>,
7 - [<<"lib/floki.ex">>,<<"lib/floki/deep_text.ex">>,
8 - <<"lib/floki/filter_out.ex">>,<<"lib/floki/finder.ex">>,
9 - <<"lib/floki/flat_text.ex">>,<<"lib/floki/html_parser.ex">>,
7 + [<<"lib">>,<<"lib/floki">>,<<"lib/floki.ex">>,<<"lib/floki/.parser.ex.swp">>,
8 + <<"lib/floki/deep_text.ex">>,<<"lib/floki/filter_out.ex">>,
9 + <<"lib/floki/finder.ex">>,<<"lib/floki/flat_text.ex">>,
10 + <<"lib/floki/html_parser">>,<<"lib/floki/html_parser.ex">>,
10 11 <<"lib/floki/html_parser/html5ever.ex">>,
11 - <<"lib/floki/html_parser/mochiweb.ex">>,<<"lib/floki/html_tree.ex">>,
12 - <<"lib/floki/html_tree/comment.ex">>,<<"lib/floki/html_tree/html_node.ex">>,
12 + <<"lib/floki/html_parser/mochiweb.ex">>,<<"lib/floki/html_tree">>,
13 + <<"lib/floki/html_tree.ex">>,<<"lib/floki/html_tree/comment.ex">>,
14 + <<"lib/floki/html_tree/html_node.ex">>,
13 15 <<"lib/floki/html_tree/id_seeder.ex">>,<<"lib/floki/html_tree/text.ex">>,
16 + <<"lib/floki/raw_html.ex">>,<<"lib/floki/selector">>,
14 17 <<"lib/floki/selector.ex">>,<<"lib/floki/selector/attribute_selector.ex">>,
15 18 <<"lib/floki/selector/combinator.ex">>,
16 19 <<"lib/floki/selector/functional.ex">>,<<"lib/floki/selector/parser.ex">>,
17 20 <<"lib/floki/selector/pseudo_class.ex">>,
18 - <<"lib/floki/selector/tokenizer.ex">>,<<"src/floki_selector_lexer.xrl">>,
19 - <<"mix.exs">>,<<"README.md">>,<<"LICENSE">>,<<"CODE_OF_CONDUCT.md">>,
20 - <<"CONTRIBUTING.md">>]}.
21 + <<"lib/floki/selector/tokenizer.ex">>,<<"lib/floki/xpath_selector">>,
22 + <<"src/floki_selector_lexer.xrl">>,<<"mix.exs">>,<<"README.md">>,
23 + <<"LICENSE">>,<<"CODE_OF_CONDUCT.md">>,<<"CONTRIBUTING.md">>]}.
21 24 {<<"licenses">>,[<<"MIT">>]}.
22 25 {<<"links">>,[{<<"GitHub">>,<<"https://github.com/philss/floki">>}]}.
23 26 {<<"maintainers">>,[<<"Philip Sampaio Silva">>]}.
  @@ -26,9 +29,11 @@
26 29 [[{<<"app">>,<<"html_entities">>},
27 30 {<<"name">>,<<"html_entities">>},
28 31 {<<"optional">>,false},
32 + {<<"repository">>,<<"hexpm">>},
29 33 {<<"requirement">>,<<"~> 0.4.0">>}],
30 34 [{<<"app">>,<<"mochiweb">>},
31 35 {<<"name">>,<<"mochiweb">>},
32 36 {<<"optional">>,false},
37 + {<<"repository">>,<<"hexpm">>},
33 38 {<<"requirement">>,<<"~> 2.15">>}]]}.
34 - {<<"version">>,<<"0.19.2">>}.
39 + {<<"version">>,<<"0.19.3">>}.
  @@ -66,31 +66,12 @@ defmodule Floki do
66 66
67 67 """
68 68
69 - @spec parse(binary) :: html_tree
69 + @spec parse(binary) :: html_tree | String.t()
70 70
71 71 def parse(html) do
72 72 HTMLParser.parse(html)
73 73 end
74 74
75 - @self_closing_tags [
76 - "area",
77 - "base",
78 - "br",
79 - "col",
80 - "command",
81 - "embed",
82 - "hr",
83 - "img",
84 - "input",
85 - "keygen",
86 - "link",
87 - "meta",
88 - "param",
89 - "source",
90 - "track",
91 - "wbr"
92 - ]
93 -
94 75 @doc """
95 76 Converts HTML tree to raw HTML.
96 77 Note that the resultant HTML may be different from the original one.
  @@ -105,59 +86,7 @@ defmodule Floki do
105 86
106 87 @spec raw_html(html_tree | binary) :: binary
107 88
108 - def raw_html(html_tree), do: raw_html(html_tree, "")
109 - defp raw_html([], html), do: html
110 - defp raw_html(string, _html) when is_binary(string), do: HtmlEntities.encode(string)
111 - defp raw_html(tuple, html) when is_tuple(tuple), do: raw_html([tuple], html)
112 -
113 - defp raw_html([string | tail], html) when is_binary(string) do
114 - raw_html(tail, html <> HtmlEntities.encode(string))
115 - end
116 -
117 - defp raw_html([{:comment, comment} | tail], html),
118 - do: raw_html(tail, html <> "<!--#{comment}-->")
119 -
120 - defp raw_html([{:pi, "xml", attrs} | tail], html) do
121 - raw_html(tail, html <> "<?xml " <> tag_attrs(attrs) <> "?>")
122 - end
123 -
124 - defp raw_html([{:doctype, type, public, system} | tail], html) do
125 - attr =
126 - case {public, system} do
127 - {"", ""} -> ""
128 - {"", system} -> " SYSTEM \"#{system}\""
129 - {public, system} -> " PUBLIC \"#{public}\" \"#{system}\""
130 - end
131 -
132 - raw_html(tail, html <> "<!DOCTYPE #{type}#{attr}>")
133 - end
134 -
135 - defp raw_html([{type, attrs, children} | tail], html) do
136 - raw_html(tail, html <> tag_for(type, tag_attrs(attrs), children))
137 - end
138 -
139 - defp tag_attrs(attr_list) do
140 - attr_list
141 - |> Enum.reduce("", &build_attrs/2)
142 - |> String.trim()
143 - end
144 -
145 - defp build_attrs({attr, value}, attrs), do: ~s(#{attrs} #{attr}="#{value}")
146 - defp build_attrs(attr, attrs), do: "#{attrs} #{attr}"
147 -
148 - defp tag_for(type, attrs, []) when type in @self_closing_tags do
149 - case attrs do
150 - "" -> "<#{type}/>"
151 - _ -> "<#{type} #{attrs}/>"
152 - end
153 - end
154 -
155 - defp tag_for(type, attrs, children) do
156 - case attrs do
157 - "" -> "<#{type}>#{raw_html(children)}</#{type}>"
158 - _ -> "<#{type} #{attrs}>#{raw_html(children)}</#{type}>"
159 - end
160 - end
89 + defdelegate raw_html(html_tree), to: Floki.RawHTML
161 90
162 91 @doc """
163 92 Find elements inside a HTML tree or string.
  @@ -0,0 +1,80 @@
1 + defmodule Floki.RawHTML do
2 + @self_closing_tags [
3 + "area",
4 + "base",
5 + "br",
6 + "col",
7 + "command",
8 + "embed",
9 + "hr",
10 + "img",
11 + "input",
12 + "keygen",
13 + "link",
14 + "meta",
15 + "param",
16 + "source",
17 + "track",
18 + "wbr"
19 + ]
20 +
21 + def raw_html(html_tree), do: raw_html(html_tree, "")
22 + defp raw_html([], html), do: html
23 + defp raw_html(string, _html) when is_binary(string), do: HtmlEntities.encode(string)
24 + defp raw_html(tuple, html) when is_tuple(tuple), do: raw_html([tuple], html)
25 +
26 + defp raw_html([string | tail], html) when is_binary(string) do
27 + raw_html(tail, html <> HtmlEntities.encode(string))
28 + end
29 +
30 + defp raw_html([{:comment, comment} | tail], html),
31 + do: raw_html(tail, html <> "<!--#{comment}-->")
32 +
33 + defp raw_html([{:pi, "xml", attrs} | tail], html) do
34 + raw_html(tail, html <> "<?xml " <> tag_attrs(attrs) <> "?>")
35 + end
36 +
37 + defp raw_html([{:doctype, type, public, system} | tail], html) do
38 + attr =
39 + case {public, system} do
40 + {"", ""} -> ""
41 + {"", system} -> " SYSTEM \"#{system}\""
42 + {public, system} -> " PUBLIC \"#{public}\" \"#{system}\""
43 + end
44 +
45 + raw_html(tail, html <> "<!DOCTYPE #{type}#{attr}>")
46 + end
47 +
48 + defp raw_html([{type, attrs, children} | tail], html) do
49 + raw_html(tail, html <> tag_for(type, attrs, children))
50 + end
51 +
52 + defp tag_attrs(attr_list) do
53 + attr_list
54 + |> Enum.reduce("", &build_attrs/2)
55 + |> String.trim()
56 + end
57 +
58 + defp tag_with_attrs(type, [], children), do: "<#{type}" <> close_open_tag(type, children)
59 +
60 + defp tag_with_attrs(type, attrs, children) do
61 + "<#{type} #{tag_attrs(attrs)}" <> close_open_tag(type, children)
62 + end
63 +
64 + defp close_open_tag(type, []) when type in @self_closing_tags, do: "/>"
65 + defp close_open_tag(_type, _), do: ">"
66 +
67 + defp close_end_tag(type, []) when type in @self_closing_tags, do: ""
68 + defp close_end_tag(type, _), do: "</#{type}>"
69 +
70 + defp build_attrs({attr, value}, attrs), do: ~s(#{attrs} #{attr}="#{value}")
71 + defp build_attrs(attr, attrs), do: "#{attrs} #{attr}"
72 +
73 + defp tag_for(type, attrs, content) when type in ["script", "style"] do
74 + tag_with_attrs(type, attrs, []) <> Enum.join(content) <> "</#{type}>"
75 + end
76 +
77 + defp tag_for(type, attrs, children) do
78 + tag_with_attrs(type, attrs, children) <> raw_html(children) <> close_end_tag(type, children)
79 + end
80 + end
  @@ -2,7 +2,7 @@ defmodule Floki.Mixfile do
2 2 use Mix.Project
3 3
4 4 @description "Floki is a simple HTML parser that enables search for nodes using CSS selectors."
5 - @version "0.19.2"
5 + @version "0.19.3"
6 6
7 7 def project do
8 8 [
  @@ -19,7 +19,7 @@ defmodule Floki.Mixfile do
19 19 end
20 20
21 21 def application do
22 - [applications: [:logger, :mochiweb]]
22 + [applications: [:logger, :mochiweb, :html_entities]]
23 23 end
24 24
25 25 defp deps do