- Add ``RequestDownload.alert_when_cache_missing`` option, allow it print alert when cache is missing. Of course you need to turn on ``RequestDownload.read_cache_first = True``.
**Bugfixes**
- Fix a fetal type bug in creating proxy for session in ``RequestDownload.__init__`.
0.0.21
~~~~~~~~~~~~~~~~~~~ **Bugfixes**
- Fix a fetal bug in creating proxy for session in ``RequestsDownloader.__init__``.
0.0.20
~~~~~~~~~~~~~~~~~~~ **Features and Improvements**
- add ``SpiderLogger`` class.
**Minor Improvements**
- add some ``set_status`` methods for ``ParseResult``. - improve tet coverage.
**Bugfixes**
- it should call ``requests.Session`` instead of ``requests.session`` in ``RequestsDownloader``.
0.0.19
~~~~~~~~~~~~~~~~~~~ **Features and Improvements**
- allow user to customize the cache behavior. ``RequestsDownloader.read_cache_first``, ``RequestsDownloader.always_update_cache`` these two option variable can be used to adjust cache behavior.
0.0.18
~~~~~~~~~~~~~~~~~~~ **Features and Improvements**
- add a downloader middleware, integrate with auto rotate headers, disk cache, tor network support.
0.0.17
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ **Features and Improvements**
- breaks lots of backward compatibility. - rewrite ``crawlib.data_class.ParseResult`` and ``crawlib.data_class.ExtendedItem``. - add a ``crawlib.pipeline`` module, add integration with mongodb and relational database.