This is the class that we use to compress JSON string arrays:

https://gist.github.com/anonymous/ee61ff43d3fe76d8d7e0

For example the array

["apple","pear","orange","mango","fig"]

to

["!Jg"]


For that you would need a compression table like this:

0 : apple
1 : pear
2 : orange
3 : mango
4 : fig
5 : ...

where the most frequent words should be at the top, and rarely used
words at the bottom.

Cheers,

Vladimir


On 11.03.2015 16:12, Carsten Bormann wrote:
> Nat Sakimura wrote:
>> We have something similar as well... 
>>
>> 2015-03-10 18:14 GMT+09:00 Vladimir Dzhuvinov <[email protected]
>> <mailto:[email protected]>>:
>>
>>     Arrays would be good. Perhaps even bit fields?
> Do you guys have some examples that we could look at?
>
> (CDDL just got support for bit fields, by the way; the current draft
> only has it for byte strings, the tool already also supports specifying
> them for uints.)
>
> Grüße, Carsten

-- 
Vladimir Dzhuvinov :: [email protected]


_______________________________________________
jose mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/jose

Reply via email to