Counter Batches and Reverse table scans

[philip-sterne]

• Ability to batch counter increments and decrements. This functions in exactly the same form as table.batch(), except one now calls:

cb = table.counter_batch()
cb.counter_inc('row-key', 'cf1:counter1')
cb.counter_inc('row-key', 'cf1:counter2')
cb.send()

Or alternatively, one can use a context manager:

with table.counter_batch(batch_size=1000) as cb:
    cb.counter_inc('row-key', 'cf1:counter1')
    cb.counter_inc('row-key', 'cf1:counter2')

• Added reverse table scans:

for key, data in table.scan(reverse=True):
    print key, data

which will return the data in reversed order

[landscape-bot]

Repository health increased by 0.31% when pulling c9048bf on TAKEALOT:master into 9cbd718 on wbolster:master.

• 3 new problems were found (including 0 errors and 3 code smells).
• No problems were fixed.

[landscape-bot]

Repository health increased by 0.35% when pulling 6663a88 on TAKEALOT:master into 9cbd718 on wbolster:master.

• 2 new problems were found (including 0 errors and 2 code smells).
• No problems were fixed.

[landscape-bot]

Repository health increased by 0.33% when pulling 2d4ed1f on TAKEALOT:master into 9cbd718 on wbolster:master.

• 2 new problems were found (including 0 errors and 2 code smells).
• No problems were fixed.

[wbolster]

i think this is incorrect. if the (row, column) tuple was already present this count will be wrong. i guess len(self.batch) is a better count ;)

[gabrii]

I've solved the merge conflicts and the patch_counter problem, out of the need to use this in pyhton3. Third PR about this feature. Time to merge! :)
#192