qrly - a query module for erlang

qrly attempts to be a query module for erlang that provides an API to extract information from structured data using a jquery-like API.

the implementation provides a standard api that allows to extend it to support other languages

supported selectors

for details on selectors check the jquery documentation: http://api.jquery.com/category/selectors/

• .class
• #id
• tagname
• Attribute Contains Selector [name*="value"]
• Attribute Contains Word Selector [name~="value"]
• Attribute Ends With Selector [name$="value"]
• Attribute Equals Selector [name="value"]
• Attribute Not Equal Selector [name!="value"]
• Multiple Attribute Selector [name="value"][name2="value2"]
• Attribute Starts With Selector [name^="value"]
• :first
• :last
• :odd
• :even
• :nth-child(N)
• :eq(N)
• :gt(N)
• :lt(N)
• Multiple Selector (“selector1, selector2, selectorN”)

other selectors are not implemented because of lack of time in spawnfest, they are really easy to implement (already parsed, but not implemented in qrly:filter) and should come after the spawnfest ends and I have some time to add them.

supported data formats

XML

the qrly_xml provides an api to manipulate and query xml using xmerl as a backend

HTML

the qrly_html provides an api to manipulate and query html using mochiweb_html as a backend

usage

use qrly_html:parse or qrly_html:parse_string to load from file or string an html file and convert it to a qrly data structure and then use qrly_html:filter to apply the filtering.

if you want your string back call qrly_html:to_string. if you want to store it on a file call qrly_html:to_file.

the same instructions work for qrly_xml and any other future implementation of the api (imagine querying json structures with this).

example

1> {ok, Qrly} = qrly_html:parse("../extra/test/test.html").
...
2> Q1 = qrly_html:filter(Qrly, "h1").
[{<<"h1">>,
  [{<<"class">>,<<"first-title asdwordclass">>}],
  [<<"personal">>]},
 {<<"h1">>,
  [{<<"class">>,<<"second-title wordclass">>}],
  [<<"projects">>]},
 {<<"h1">>,
  [{<<"title">>,<<"others">>},
   {<<"class">>,<<"third-title wordclass">>}],
  [<<"others">>]}]
3> Q2 = qrly_html:filter(Qrly, "h1.first-title").
[{<<"h1">>,
  [{<<"class">>,<<"first-title asdwordclass">>}],
  [<<"personal">>]}]
4> io:format("~s~n", [qrly_html:to_string(Q1)]).
<div><h1 class="first-title asdwordclass">personal</h1><h1 class="second-title wordclass">projects</h1><h1 title="others" class="third-title wordclass">others</h1></div>
ok
5> io:format("~s~n", [qrly_html:to_string(Q2)]).
<div><h1 class="first-title asdwordclass">personal</h1></div>
ok

testing

qrly uses eunit, just do:

cd src
./test.sh

it will compile and run the tests

logo

let me introduce you: Qrly the bird

https://github.com/spawnfest/qrly/raw/master/extra/qrly.jpg

license

BSD + optional beer invitation to the creator

see LICENSE for details