See http://www.sphinx-doc.org/en/stable/changes.html#release-1-4-released-mar-28-2016
sphinx_rtd_theme has become optional, needs to be added to reqs
https://github.com/sphinx-doc/sphinx/pull/2320 changes node entries tuples
to 5 values instead of 4
`sh` syntax highlighting added very locally in selectors.rst
because of this warning/error with Sphinx 1.4:
```
Warning, treated as error:
/home/paul/src/scrapy/docs/topics/selectors.rst:743:
WARNING: Could not lex literal_block as "python". Highlighting skipped.
```
>>> response.xpath('//div/[id="not-exists"]/text()').extract_first() is None
Traceback (most recent call last):
File "<console>", line 1, in <module>
File "/home/vagrant/.virtualenvs/scrapy/lib/python2.7/site-packages/scrapy/http/response/text.py", line 109, in xpath
return self.selector.xpath(query)
File "/home/vagrant/.virtualenvs/scrapy/lib/python2.7/site-packages/scrapy/selector/unified.py", line 100, in xpath
raise ValueError(msg if six.PY3 else msg.encode("unicode_escape"))
ValueError: Invalid XPath: //div/[id="not-exists"]/text()
Ported .get() from _Element and .text_content() from HTMLMixin
Add CSS selectors to scrapy shell
Documenting CSS Selectors: Constructing selectors
Documenting CSS Selectors: Using Selectors
Make CSS Selectors a default feature.
Adds XPath powers to CSS Selectors and some syntactic sugar.
Removes methods copied over from lxml.html.HtmlMixin.
Updating docs to use new CSS Selector super powers.
Documenting CSS Selectors: Regular Expressions
Moving section after Nesting section, since it mentions it.
Documenting CSS Selectors: Nesting Selectors
Fix XPath specificity in lxml.selector.CSSSelectorMixin.text
Cleaning up unused stuff from cssel.py
Changing the behavior of lxml.selector.CSSSelectorMixin.text.
Concatenating all of the descendant text nodes is more useful
than returning it in pieces (there's xpath() if you need that).
Documenting CSS Selectors: CSS Selector objects
Documenting CSS Selectors: CSSSelectorList objects
Documenting CSS Selectors: HtmlCSSSelector objects
Documenting CSS Selectors: XmlCSSSelector objects
Fixing some documentations typos and errors
Enforcing the 80-char width lines
Tidying up CSS selectors and CSSSelectorMixin objects
Adding some missing references in documentation.
Fixing lxml.selector.CSSSelectorList.text
* Added new setting (SELECTORS_BACKEND) to choose which backend to use
* Deprecated the extract_unquoted() function from selectors
* Made libxml2 optional by adding a dummy selector backend. Closes#260
--HG--
rename : scrapy/tests/test_selector.py => scrapy/tests/test_selector_libxml2.py
