~jameinel/+junk/pybloom : changes

~jameinel/+junk/pybloom » Changes from revision 42

From Revision 42 to 23

Rev	Summary	Authors	Date
42	Remove an !=, which also caught an accidental access to the ... Remove an !=, which also caught an accidental access to the original python object instead of the character buffer.	John Arbash Meinel	2009-04-24
41	play around with some tweaks. when benchmarking murmur, use ... play around with some tweaks. when benchmarking murmur, use 20x the number of keys.	John Arbash Meinel	2009-04-17
40	Lots of little tweaks. Lots of little tweaks. Include a larger ancestry set. Now using 24k bzr revs instead of 8k. Spend some time optimizing the insert side of blooms. Improves insert time from 200+ms down to 50ms. Add BloomMurmur to the benchmark suite, so it is actually being benchmarked. Overall BloomMurmur is about 200 => 30ms for check times (not quite 10x faster). Insert times are approx 250 => 50ms (about 5x). Using the multi-way & aligned path shaves about 10% off. 56ms => 49ms. We really need a longer running test, ideally ~1s rather than .05s.	John Arbash Meinel	2009-04-17
39	Implement the parallel form of multi hash. Implement the parallel form of multi hash.	John Arbash Meinel	2009-02-03
38	Add the framework for a multi-hash version. Add the framework for a multi-hash version. The main win here is that we can factor out some of the common work rather than making multiple passes over the data. Also, we allow for using the fully optimized code when on a little endian device with aligned input.	John Arbash Meinel	2009-02-03
37	Minor optimization of the python implementation of murmur ha... Minor optimization of the python implementation of murmur hash. The big change is to just use struct.unpack() across everything in one pass, rather than lots of ord() calls and bit shifts.	John Arbash Meinel	2009-02-02
36	A bit more info about paging thoughts. A bit more info about paging thoughts.	John Arbash Meinel	2008-07-16
35	Push the __contains__ check down into pyrex. Push the __contains__ check down into pyrex. Gives us another ~10% on indexbench. For some reason get_components_positions still loses.	John Arbash Meinel	2008-07-16
34	Add a filter_by_presence function, to allow filtering multip... Add a filter_by_presence function, to allow filtering multiple nodes in one go. Customize the BloomMurmur.__contains__ so that it doesn't have to compute all hashes when it can know right away that something doesn't hit.	John Arbash Meinel	2008-07-16
33	If you are going to write a compiled form... use it If you are going to write a compiled form... use it	John Arbash Meinel	2008-07-16
32	Switch the _murmur_hash implementations to return 0=>2^31 ra... Switch the _murmur_hash implementations to return 0=>2^31 rather. This allows us to put it into a signed int, and still always have a positive offset. It is unlikely that we would ever need that last bit anyway.	John Arbash Meinel	2008-07-16
31	Strip the 'sign' bit so that we can use regular PyInt rather... Strip the 'sign' bit so that we can use regular PyInt rather than PyLong. Still some failing tests.	John Arbash Meinel	2008-07-16
30	Switching over from AP => Murmur since the py implementation... Switching over from AP => Murmur since the py implementation of AP was wrong anyway	John Arbash Meinel	2008-07-16
29	Found a small discrepancy when the string is exactly a multi... Found a small discrepancy when the string is exactly a multiple of 4 long.	John Arbash Meinel	2008-07-15
28	Export the murmur hash using a pyrex function, and write a p... Export the murmur hash using a pyrex function, and write a python function which matches it. Of course, it won't match the speed, but we only care about the values.	John Arbash Meinel	2008-07-15
27	Add the MurmurHash functions, and start playing with a pytho... Add the MurmurHash functions, and start playing with a python implementation of hashlittle.	John Arbash Meinel	2008-07-15
26	Add the Bob Jenkins' lookup3.hashlittle hash along with some... Add the Bob Jenkins' lookup3.hashlittle hash along with some notes. It looks like the best hash for now, and should be possible to port to python.	John Arbash Meinel	2008-07-15
25	allow using the compiled hash function, not perfect yet, but allow using the compiled hash function, not perfect yet, but interesting	John Arbash Meinel	2008-07-15
24	Start working on a bloom that just uses the AP hash. Start working on a bloom that just uses the AP hash.	John Arbash Meinel	2008-07-15
23	Start playing around with adding a compiled bloom object. Start playing around with adding a compiled bloom object.	John Arbash Meinel	2008-07-15

Older »