On 7 December 2011 17:06, Ferran Mesas wrote:
> Hello again,
>
> http://agora.panocha.org.mx/snippet/gDqF/
> http://agora.panocha.org.mx/snippet/gDqF/
> I think you can commit this code. I added 5 tests to the file and made a
> more strict input checking (now it checks for binary arrays).
Sweet.
Hello again,
http://agora.panocha.org.mx/snippet/gDqF/
http://agora.panocha.org.mx/snippet/gDqF/
I think you can commit this code. I added 5 tests to the file and made a
more strict input checking (now it checks for binary arrays).
Could you please also change huffmandict?
on line 118:
- cw_li
On 6 December 2011 23:52, Carnë Draug wrote:
> On 6 December 2011 23:23, Ferran Mesas wrote:
>> Hello Carnë,
>>
>> Yes, i think it can be made a subfunction. The thing is, the dictionary from
>> huffmandict() is very efficient for encoding and memory-wise, but the
>> array-tree is good at decodin
On 6 December 2011 23:23, Ferran Mesas wrote:
> Hello Carnë,
>
> Yes, i think it can be made a subfunction. The thing is, the dictionary from
> huffmandict() is very efficient for encoding and memory-wise, but the
> array-tree is good at decoding but not memory-efficient.
> Embedding dict2tree int
Hello Carnë,
Yes, i think it can be made a subfunction. The thing is, the dictionary from
huffmandict() is very efficient for encoding and memory-wise, but the
array-tree is good at decoding but not memory-efficient.
Embedding dict2tree into huffmandeco seems like a good idea.
I copied the licens
On 6 December 2011 18:25, Carnë Draug wrote:
> On 3 December 2011 21:43, Ferran Mesas wrote:
>> Hello everyone,
>>
>> I rewrote huffmandeco() from the communications package. The one in
>> communications 1.1.0 uses a brute-force O(n^3) or O(n^4) solution.
>> I implemented it using a binary tree a
On 3 December 2011 21:43, Ferran Mesas wrote:
> Hello everyone,
>
> I rewrote huffmandeco() from the communications package. The one in
> communications 1.1.0 uses a brute-force O(n^3) or O(n^4) solution.
> I implemented it using a binary tree and now takes O(n) operations
> (please note that I am
Hello everyone,
I rewrote huffmandeco() from the communications package. The one in
communications 1.1.0 uses a brute-force O(n^3) or O(n^4) solution.
I implemented it using a binary tree and now takes O(n) operations
(please note that I am not used to big O notation and numbers may be
off).
I att
