Packages

floki

0.15.0
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

7 files changed
+38 additions
-16 deletions
  @@ -127,6 +127,7 @@ and a more accurate parser. However `html5ever` is being under active developmen
127 127
128 128 Since it's written in Rust, we need to install Rust and compile the project. Luckily we have have the
129 129 [html5ever Elixir NIF](https://github.com/hansihe/html5ever_elixir) that makes the integration very easy.
130 + For more info, check the article [Rustler - Safe Erlang and Elixir NIFs in Rust](http://hansihe.com/2017/02/05/rustler-safe-erlang-elixir-nifs-in-rust.html).
130 131
131 132 You still need to install Rust in your system. To do that, please
132 133 [follow the instruction](https://www.rust-lang.org/en-US/install.html) presented in the official page.
  @@ -139,7 +140,7 @@ After setup Rust, you need to add `html5ever` NIF to your dependency list:
139 140 defp deps do
140 141 [
141 142 {:floki, "~> 0.14.0"},
142 - {:html5ever, "~> 0.2.0"}
143 + {:html5ever, "~> 0.3.0"}
143 144 ]
144 145 end
145 146 ```
  @@ -156,9 +157,6 @@ config :floki, :html_parser, Floki.HTMLParser.Html5ever
156 157
157 158 After that you are able to use `html5ever` as your HTML parser with Floki.
158 159
159 - To know more about NIFs, check the [Erlang docs](http://erlang.org/doc/man/erl_nif.html).
160 - And to know why Rust NIFs can be better, check the [Rustler](https://github.com/hansihe/rustler) project.
161 -
162 160 ## More about Floki API
163 161
164 162 To parse a HTML document, try:
  @@ -27,4 +27,4 @@
27 27 {<<"name">>,<<"mochiweb">>},
28 28 {<<"optional">>,false},
29 29 {<<"requirement">>,<<"~> 2.15">>}]]}.
30 - {<<"version">>,<<"0.14.0">>}.
30 + {<<"version">>,<<"0.15.0">>}.
  @@ -101,6 +101,8 @@ defmodule Floki.Finder do
101 101 PseudoClass.match_nth_child?(tree, html_node, pseudo_class)
102 102 "first-child" ->
103 103 PseudoClass.match_nth_child?(tree, html_node, %PseudoClass{name: "nth-child", value: 1})
104 + "not" ->
105 + !Selector.match?(html_node, pseudo_class.value)
104 106 unknown_pseudo_class ->
105 107 Logger.warn("Pseudo-class #{inspect unknown_pseudo_class} is not implemented. Ignoring.")
106 108 false
  @@ -110,7 +112,7 @@ defmodule Floki.Finder do
110 112 # The stack serves as accumulator when there is another combinator to traverse.
111 113 # So the scope of one combinator is the stack (or acc) or the parent one.
112 114 defp traverse_with(_, _, []), do: []
113 - defp traverse_with(nil, _, result), do: result
115 + defp traverse_with(nil, _, results), do: results
114 116 defp traverse_with(%Combinator{match_type: :child, selector: s}, tree, stack) do
115 117 results =
116 118 Enum.flat_map(stack, fn(html_node) ->
  @@ -118,7 +120,7 @@ defmodule Floki.Finder do
118 120 |> Enum.reverse
119 121 |> get_nodes(tree)
120 122
121 - Enum.filter(nodes, fn(html_node) -> Selector.match?(html_node, s) end)
123 + Enum.filter(nodes, fn(html_node) -> selector_match?(tree, html_node, s) end)
122 124 end)
123 125
124 126 traverse_with(s.combinator, tree, results)
  @@ -134,7 +136,7 @@ defmodule Floki.Finder do
134 136 nodes = get_nodes(sibling_id, tree)
135 137
136 138 # Finally, try to match those siblings with the selector
137 - Enum.filter(nodes, fn(html_node) -> Selector.match?(html_node, s) end)
139 + Enum.filter(nodes, fn(html_node) -> selector_match?(tree, html_node, s) end)
138 140 end)
139 141
140 142 traverse_with(s.combinator, tree, results)
  @@ -147,7 +149,7 @@ defmodule Floki.Finder do
147 149 nodes = get_nodes(sibling_ids, tree)
148 150
149 151 # Finally, try to match those siblings with the selector
150 - Enum.filter(nodes, fn(html_node) -> Selector.match?(html_node, s) end)
152 + Enum.filter(nodes, fn(html_node) -> selector_match?(tree, html_node, s) end)
151 153 end)
152 154
153 155 traverse_with(s.combinator, tree, results)
  @@ -158,7 +160,7 @@ defmodule Floki.Finder do
158 160 ids_to_match = get_descendant_ids(html_node.node_id, tree)
159 161 nodes = get_nodes(ids_to_match, tree)
160 162
161 - Enum.filter(nodes, fn(html_node) -> Selector.match?(html_node, s) end)
163 + Enum.filter(nodes, fn(html_node) -> selector_match?(tree, html_node, s) end)
162 164 end)
163 165
164 166 traverse_with(s.combinator, tree, results)
  @@ -196,11 +198,12 @@ defmodule Floki.Finder do
196 198 end)
197 199 end
198 200
199 - # finds all descendant node ids recursively through the tree
201 + # finds all descendant node ids recursively through the tree preserving the order
200 202 defp get_descendant_ids(node_id, tree) do
201 203 case get_node(node_id, tree) do
202 204 %{children_nodes_ids: node_ids} ->
203 - Enum.reverse(node_ids) ++ Enum.flat_map(node_ids, &(get_descendant_ids(&1, tree)))
205 + reversed_ids = Enum.reverse(node_ids)
206 + reversed_ids ++ Enum.flat_map(reversed_ids, &(get_descendant_ids(&1, tree)))
204 207 _ ->
205 208 []
206 209 end
  @@ -4,7 +4,7 @@ defmodule Floki.SelectorParser do
4 4 @moduledoc false
5 5 # Parses a list of tokens returned from `SelectorTokenizer` and transfor into a `Selector`.
6 6
7 - alias Floki.{Selector, AttributeSelector, Combinator}
7 + alias Floki.{Selector, SelectorTokenizer, AttributeSelector, Combinator}
8 8 alias Floki.Selector.PseudoClass
9 9
10 10 @attr_match_types [:equal, :dash_match, :includes, :prefix_match, :sufix_match, :substring_match]
  @@ -12,6 +12,10 @@ defmodule Floki.SelectorParser do
12 12 # Returns a `Selector` struct with the parsed selector.
13 13 # Note that this parser does not deal with groups of selectors.
14 14
15 + def parse(selector) when is_binary(selector) do
16 + token_list = SelectorTokenizer.tokenize(selector)
17 + parse(token_list)
18 + end
15 19 def parse(tokens) do
16 20 do_parse(tokens, %Selector{})
17 21 end
  @@ -56,6 +60,23 @@ defmodule Floki.SelectorParser do
56 60 pseudo_class = selector.pseudo_class
57 61 do_parse(t, %{selector | pseudo_class: %{pseudo_class | value: to_string(pattern)}})
58 62 end
63 + defp do_parse([{:pseudo_class_generic_value, _, value} | t], selector) do
64 + s = case selector.pseudo_class do
65 + %PseudoClass{name: "not"} ->
66 + not_selector = parse(to_string(value))
67 +
68 + if not_selector.combinator do
69 + Logger.warn("Only simple selectors are allowed in :not() pseudo-class. Ignoring.")
70 + %{selector | pseudo_class: nil}
71 + else
72 + %{selector | pseudo_class: %{selector.pseudo_class | value: not_selector}}
73 + end
74 + _ ->
75 + selector
76 + end
77 +
78 + do_parse(t, s)
79 + end
59 80 defp do_parse([{:space, _} | t], selector) do
60 81 {t, combinator} = consume_combinator(t, :descendant)
  @@ -3,8 +3,6 @@ defmodule Floki.SelectorTokenizer do
3 3
4 4 # It decodes a given selector and returns the tokens that represents it.
5 5 # Check the rules in "src/floki_selector_lexer.xrl"
6 -
7 - def tokenize(""), do: []
8 6 def tokenize(selector) do
9 7 char_list = selector |> String.strip |> String.to_char_list
Loading more files…