sep-006.trac

= SEP-006: Rename of Selectors to Extractors =

[[PageOutline(2-5, Contents)]]

||'''SEP'''||6||
||'''Title'''||Extractors||
||'''Author'''||Ismael Carnales and a bunch of rabid mice||
||'''Created'''||2009-07-28||
||'''Status'''||Obsolete (discarded)||

== Abstract ==

This SEP proposes a more meaningful naming of XPathSelectors or "Selectors" and their `x` method.

== Motivation ==

When you use Selectors in Scrapy, your final goal is to "extract" the data that you've selected, as the [http://doc.scrapy.org/topics/selectors.html XPath Selectors documentation] says (bolding by me):

   "When you’re scraping web pages, the most common task you need to perform is to '''extract''' data from the HTML source."

   ...

   "Scrapy comes with its own mechanism for '''extracting''' data. They’re called XPath selectors (or just “selectors”, for short) because they “select” certain parts of the HTML document specified by XPath expressions."

   ...

   "To actually '''extract''' the textual data you must call the selector extract() method, as follows"

   ...

   "Selectors also have a re() method for '''extracting''' data using regular expressions."

   ...

   "For example, suppose you want to '''extract''' all <p> elements inside <div> elements. First you get would get all <div> elements"

== Rationale ==

As and there is no Extractor object in Scrapy and what you want to finally perform with Selectors is extracting data, we propose the renaming of Selectors to Extractors. (In Scrapy for extracting you use selectors is really weird :) )

=== Additional changes ===

As the name of the method for performing selection (the `x` method) is not descriptive nor mnemotechnic enough and clearly clashes with extract method (x sounds like a short for extract in english), we propose to rename it to `select`, `sel` (is shortness if required), or `xpath` after [http://codespeak.net/lxml/xpathxslt.html lxml's xpath method]

== Bonus (!ItemBuilder) ==

After this renaming we propose also renaming !ItemBuilder to !ItemExtractor, because the !ItemBuilder/Extractor will act as a bridge between a set of Extractors and an Item and because it will literally "extract" an item from a webpage or set of pages.

== References ==

 1. XPath Selectors (http://doc.scrapy.org/topics/selectors.html)
 2. XPath and XSLT with lxml (http://codespeak.net/lxml/xpathxslt.html)