Crawlib

~~~~~~~~~~~~~~~~~~~
**Features and Improvements**

**Minor Improvements**

**Bugfixes**

**Miscellaneous**

~~~~~~~~~~~~~~~~~~~
**Bugfixes**

- ``fake_useragent`` grabs data from useragentstring.com, and it may fail. so we catch this exception and manually create it.

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
**Features and Improvements**

- add a test mode for selenium downloader.
- move the one-to-one/many job executor to ``crawlib.spider.py``.
- add a implementation validation classmethod for orm object.
- improve logger module.

**Minor Improvements**

- update example project ``crawlib_doc``.

**Bugfixes**

**Miscellaneous**

~~~~~~~~~~~~~~~~~~~
**Features and Improvements**

- add ``crawlib.downloader.selenium_downloader`` module
- move disk cache relative code into ``crawlib.cache.CacheBackedDownloader``
- add a job executor function for mongodb backed, one-to-one, one-to-many architecture.

**Minor Improvements**

- add more methods for ``crawlib.logger.SpiderLogger``

**Miscellaneous**

- remove ``crawlib.spider`` sub package.

~~~~~~~~~~~~~~~~~~~
**Minor Improvements**

- now allows ``OneToManyItem._settings_N_CHILD_1_KEY_optional`` to be None. Then ``{n_child}`` field will not be updated.
- improve code coverage.

~~~~~~~~~~~~~~~~~~~
**Bugfixes**

- add ``\n`` at the head of each line of ``sys.stdout.write(line)``.