core
/
oga
3
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Tests for lexing a basic HTML document. 

This also comes with some changes to the lexer so that it advances column/line
numbers correctly.
This commit is contained in:
Yorick Peterse 2014-03-13 23:55:18 +01:00
parent edf2e4112b
commit 6b2f682c5c
2 changed files with 63 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

11
lib/oga/lexer.rl
View File

@ -264,7 +264,13 @@ module Oga
element_start = '<' element_name; element_start = '<' element_name;
element_text := |* element_text := |*
^'<' => buffer_text; newline => {
emit_text_buffer
t(:T_TEXT)
advance_line
};
^('<' | newline) => buffer_text;
'<' => { '<' => {
emit_text_buffer emit_text_buffer
@ -284,7 +290,8 @@ module Oga
advance_column(2) advance_column(2)
# Advance the column for the closing name. # Advance the column for the closing name.
advance_column(@te - p) advance_column(text.length)
fret; fret;
}; };

54
spec/oga/lexer/documents_spec.rb Normal file
View File

@ -0,0 +1,54 @@
require 'spec_helper'
describe Oga::Parser do
context 'HTML documents' do
example 'lex a basic HTML document' do
html = <<-EOF
<!DOCTYPE html>
<html>
<head>
<title>Title</title>
</head>
<body></body>
</html>
EOF
lex(html).should == [
[:T_DOCTYPE_START, '<!DOCTYPE html', 1, 1],
[:T_DOCTYPE_END, '>', 1, 15],
[:T_TEXT, "\n", 1, 16],
# <html>
[:T_ELEM_OPEN, nil, 2, 1],
[:T_ELEM_NAME, 'html', 2, 2],
[:T_TEXT, "\n", 2, 7],
# <head>
[:T_ELEM_OPEN, nil, 3, 1],
[:T_ELEM_NAME, 'head', 3, 2],
[:T_TEXT, "\n", 3, 7],
# <title>Title</title>
[:T_ELEM_OPEN, nil, 4, 1],
[:T_ELEM_NAME, 'title', 4, 2],
[:T_TEXT, 'Title', 4, 8],
[:T_ELEM_CLOSE, nil, 4, 13],
[:T_TEXT, "\n", 4, 21],
# </head>
[:T_ELEM_CLOSE, nil, 5, 1],
[:T_TEXT, "\n", 5, 8],
# <body></body>
[:T_ELEM_OPEN, nil, 6, 1],
[:T_ELEM_NAME, 'body', 6, 2],
[:T_ELEM_CLOSE, nil, 6, 7],
[:T_TEXT, "\n", 6, 14],
# </html>
[:T_ELEM_CLOSE, nil, 7, 1],
[:T_TEXT, "\n", 7, 8]
]
end
end
end