Packages

floki

0.4.0
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
1 Dependency
Dependants
Files
Activity
Hide whitespace

Compare versions

13 files changed
+1093 additions
-220 deletions
  @@ -91,7 +91,7 @@ You can install Floki by adding a dependency to your mix file (mix.exs):
91 91 ```elixir
92 92 defp deps do
93 93 [
94 - {:floki, "~> 0.3"}
94 + {:floki, "~> 0.4"}
95 95 ]
96 96 end
97 97 ```
  @@ -5,9 +5,13 @@
5 5 <<"A HTML parser and searcher.\n\nYou can search inside HTML documents using CSS like selectors.">>}.
6 6 {<<"elixir">>,<<">= 1.0.0">>}.
7 7 {<<"files">>,
8 - [<<"lib/floki.ex">>,<<"lib/floki/deep_text.ex">>,<<"lib/floki/finder.ex">>,
9 - <<"lib/floki/flat_text.ex">>,<<"lib/floki/matchers.ex">>,
10 - <<"lib/floki/parser.ex">>,<<"mix.exs">>,<<"README.md">>,<<"LICENSE">>]}.
8 + [<<"lib/floki.ex">>,<<"lib/floki/attribute_selector.ex">>,
9 + <<"lib/floki/combinator.ex">>,<<"lib/floki/deep_text.ex">>,
10 + <<"lib/floki/finder.ex">>,<<"lib/floki/flat_text.ex">>,
11 + <<"lib/floki/parser.ex">>,<<"lib/floki/selector.ex">>,
12 + <<"lib/floki/selector_parser.ex">>,<<"lib/floki/selector_tokenizer.ex">>,
13 + <<"src/floki_selector_lexer.erl">>,<<"src/floki_selector_lexer.xrl">>,
14 + <<"mix.exs">>,<<"README.md">>,<<"LICENSE">>]}.
11 15 {<<"licenses">>,[<<"MIT">>]}.
12 16 {<<"links">>,
13 17 [{<<"Docs">>,<<"http://hexdocs.pm/floki">>},
  @@ -18,4 +22,4 @@
18 22 [{<<"app">>,<<"mochiweb">>},
19 23 {<<"optional">>,false},
20 24 {<<"requirement">>,<<"~> 2.12.2">>}]}]}.
21 - {<<"version">>,<<"0.3.3">>}.
25 + {<<"version">>,<<"0.4.0">>}.
  @@ -81,18 +81,13 @@ defmodule Floki do
81 81 Finds elements inside a HTML tree or string.
82 82 You can search by class, tag name or id.
83 83
84 - It is possible to compose searches:
85 -
86 - Floki.find(html_string, ".class")
87 - |> Floki.find(".another-class-inside-small-scope")
88 -
89 84 ## Examples
90 85
91 86 iex> Floki.find("<p><span class=hint>hello</span></p>", ".hint")
92 87 [{"span", [{"class", "hint"}], ["hello"]}]
93 88
94 - iex> "<body><div id=important><div>Content</div></div></body>" |> Floki.find("#important")
95 - {"div", [{"id", "important"}], [{"div", [], ["Content"]}]}
89 + iex> Floki.find("<body><div id=important><div>Content</div></div></body>", "#important")
90 + [{"div", [{"id", "important"}], [{"div", [], ["Content"]}]}]
96 91
97 92 iex> Floki.find("<p><a href='https://google.com'>Google</a></p>", "a")
98 93 [{"a", [{"href", "https://google.com"}], ["Google"]}]
  @@ -101,49 +96,13 @@ defmodule Floki do
101 96
102 97 @spec find(binary | html_tree, binary) :: html_tree
103 98
99 + def find(html, selector) when is_binary(html) do
100 + Floki.Parser.parse(html) |> Finder.find(selector)
101 + end
104 102 def find(html, selector) do
105 103 Finder.find(html, selector)
106 104 end
107 105
108 - @doc """
109 - Returns a list with attribute values for a given selector.
110 -
111 - ## Examples
112 -
113 - iex> Floki.attribute("<a href='https://google.com'>Google</a>", "a", "href")
114 - ["https://google.com"]
115 -
116 - """
117 -
118 - @spec attribute(binary | html_tree, binary, binary) :: list
119 -
120 - def attribute(html, selector, attribute_name) do
121 - html
122 - |> find(selector)
123 - |> Finder.attribute_values(attribute_name)
124 - end
125 -
126 - @doc """
127 - Returns a list with attribute values from elements.
128 -
129 - ## Examples
130 -
131 - iex> Floki.attribute("<a href=https://google.com>Google</a>", "href")
132 - ["https://google.com"]
133 -
134 - """
135 -
136 - @spec attribute(binary | html_tree, binary) :: list
137 -
138 - def attribute(html_tree, attribute_name) when is_binary(html_tree) do
139 - html_tree
140 - |> parse
141 - |> Finder.attribute_values(attribute_name)
142 - end
143 - def attribute(elements, attribute_name) do
144 - Finder.attribute_values(elements, attribute_name)
145 - end
146 -
147 106 @doc """
148 107 Returns the text nodes from a HTML tree.
149 108 By default, it will perform a deep search through the HTML tree.
  @@ -176,4 +135,65 @@ defmodule Floki do
176 135
177 136 search_strategy.get(html_tree)
178 137 end
138 +
139 + @doc """
140 + Returns a list with attribute values for a given selector.
141 +
142 + ## Examples
143 +
144 + iex> Floki.attribute("<a href='https://google.com'>Google</a>", "a", "href")
145 + ["https://google.com"]
146 +
147 + """
148 +
149 + @spec attribute(binary | html_tree, binary, binary) :: list
150 +
151 + def attribute(html, selector, attribute_name) do
152 + html
153 + |> find(selector)
154 + |> attribute_values(attribute_name)
155 + end
156 +
157 + @doc """
158 + Returns a list with attribute values from elements.
159 +
160 + ## Examples
161 +
162 + iex> Floki.attribute("<a href=https://google.com>Google</a>", "href")
163 + ["https://google.com"]
164 +
165 + """
166 +
167 + @spec attribute(binary | html_tree, binary) :: list
168 +
169 + def attribute(html_tree, attribute_name) when is_binary(html_tree) do
170 + html_tree
171 + |> parse
172 + |> attribute_values(attribute_name)
173 + end
174 + def attribute(elements, attribute_name) do
175 + attribute_values(elements, attribute_name)
176 + end
177 +
178 + defp attribute_values(element, attr_name) when is_tuple(element) do
179 + attribute_values([element], attr_name)
180 + end
181 + defp attribute_values(elements, attr_name) do
182 + values = Enum.reduce elements, [], fn({_, attributes, _}, acc) ->
183 + case attribute_match?(attributes, attr_name) do
184 + {_attr_name, value} ->
185 + [value|acc]
186 + _ ->
187 + acc
188 + end
189 + end
190 +
191 + Enum.reverse(values)
192 + end
193 +
194 + defp attribute_match?(attributes, attribute_name) do
195 + Enum.find attributes, fn({attr_name, _}) ->
196 + attr_name == attribute_name
197 + end
198 + end
179 199 end
  @@ -0,0 +1,45 @@
1 + defmodule Floki.AttributeSelector do
2 + alias Floki.AttributeSelector
3 +
4 + defstruct match_type: nil, attribute: nil, value: nil
5 +
6 + def match?(attributes, s = %AttributeSelector{match_type: nil, value: nil}) do
7 + attribute_present?(s.attribute, attributes)
8 + end
9 + def match?(attributes, s = %AttributeSelector{match_type: :equal}) do
10 + get_value(s.attribute, attributes) == s.value
11 + end
12 + def match?(attributes, s = %AttributeSelector{match_type: :includes}) do
13 + value = get_value(s.attribute, attributes)
14 +
15 + whitespace_values = String.split(value, " ")
16 +
17 + Enum.any? whitespace_values, fn(v) -> v == s.value end
18 + end
19 + def match?(attributes, s = %AttributeSelector{match_type: :dash_match}) do
20 + value = get_value(s.attribute, attributes)
21 +
22 + value == s.value || String.starts_with?(value, "#{s.value}-")
23 + end
24 + def match?(attributes, s = %AttributeSelector{match_type: :prefix_match}) do
25 + get_value(s.attribute, attributes) |> String.starts_with?(s.value)
26 + end
27 + def match?(attributes, s = %AttributeSelector{match_type: :sufix_match}) do
28 + get_value(s.attribute, attributes) |> String.ends_with?(s.value)
29 + end
30 + def match?(attributes, s = %AttributeSelector{match_type: :substring_match}) do
31 + get_value(s.attribute, attributes) |> String.contains?(s.value)
32 + end
33 +
34 + defp get_value(attr_name, attributes) do
35 + {_attr_name, value} = Enum.find attributes, {attr_name, ""}, fn({k, _v}) ->
36 + k == attr_name
37 + end
38 +
39 + value
40 + end
41 +
42 + defp attribute_present?(name, attributes) do
43 + Enum.any? attributes, fn({k, _v}) -> k == name end
44 + end
45 + end
  @@ -0,0 +1,3 @@
1 + defmodule Floki.Combinator do
2 + defstruct match_type: nil, selector: nil
3 + end
Loading more files…