c5d5cfdc65
Merge pull request #861 from quantopian/pipeline-isnan
...
ENH: Add isnan, notnan, and isfinite Factor methods.
2015-11-18 22:41:37 -05:00
b43c4f4c0b
ENH: Add isnan, notnan, and isfinite Factor methods.
2015-11-18 21:44:53 -05:00
0cdd88a069
Merge pull request #847 from quantopian/fix-numpy1.10-failures
...
MAINT: Fix type coercion warnings with numpy 1.10.
2015-11-17 11:39:23 -05:00
aef38c4a74
MAINT: Remove unused import.
2015-11-17 11:22:08 -05:00
fed6655204
Merge pull request #817 from quantopian/speedup-retrieve-all
...
Speedup retrieve all
2015-11-16 18:22:59 -05:00
0f349fc3ed
MAINT: Fix type coercion warnings with numpy 1.10.
...
Numpy warns about adding Python integers to uint32s and converting date
objects to datetime64.
2015-11-15 22:42:36 -05:00
4832004c33
TEST: Test fallback to start_date/end_date sorting.
...
Adds tests asserting that we resolve conflicts in accordance with the
following rules when we have multiple assets holding the same symbol at
the same time:
If multiple SIDs exist for symbol S at time T, return the candidate
SID whose start_date is highest. (200 cases)
If multiple SIDs exist for symbol S at time T, the best candidate
SIDs share the highest start_date, return the SID with the highest
end_date. (34 cases)
It is the opinion of the author (ssanderson) that we should consider
this malformed input and fail here. But this is the current indended
behavior of the code, and I accidentally broke it while refactoring.
These will serve as regression tests until the time comes that we
decide to enforce this as an error.
See https://github.com/quantopian/zipline/issues/837 for more
details.
2015-11-13 18:26:54 -05:00
06ab766544
MAINT: De-dupe shared filtering logic.
...
We filter for assets that were alive at some reference date in multiple
places.
2015-11-13 18:26:54 -05:00
a3bafd8dc7
MAINT/STY: Use dict.get instead of in check.
2015-11-13 18:26:54 -05:00
26611b5d60
MAINT: Don't double-coerce to Timestamp.
...
`normalize` would already fail if the input isn't a Timestamp.
2015-11-13 18:26:54 -05:00
b9e4b62374
PERF: Fewer conversions in _convert_row_to_equity.
2015-11-13 18:26:54 -05:00
657a132f1e
ENH: Make retrieve specific type functions public.
...
We rely on these upstream, for better or worse, so add tests and docs.
Also adds distinct `EquitiesNotFound` and `FutureContractsNotFound`
exceptions.
2015-11-13 18:26:54 -05:00
3619a24e4d
TEST: Add support for futures to tmp_asset_finder.
2015-11-13 18:26:54 -05:00
9e463fd5d8
MAINT: make_rotating_asset -> make_rotating_equity.
2015-11-13 18:26:54 -05:00
4109640acb
MAINT: make_simple_asset_info -> make_simple_equity_info.
2015-11-13 18:26:54 -05:00
a684063f95
DOC: Don't lie about caching in _retrieve_asset.
2015-11-13 18:26:54 -05:00
9101636bf6
MAINT: Remove unused method.
2015-11-13 18:26:54 -05:00
60b988a777
MAINT: Remove unused method.
2015-11-13 18:26:54 -05:00
1c52977964
DOC: group_by_type returns a regular dict.
2015-11-13 18:26:54 -05:00
c763d8f4d5
STY: Remove unused import.
2015-11-13 18:26:54 -05:00
654edaa851
BUG: Clear asset caches when mapping DataFrame.
...
Our DataFrame index resolution logic relies on failed lookups **not**
being cached, but not caching failed lookups is a nontrivial performance
hit when repeatedly looking up sids. The "solution" here is to clear
the caches after writing in new assets.
The real fix for this is either:
1. Don't construct an AssetFinder until we have the datasource in hand
in run(), or
2. Don't symbol-map the user's input source if it's a DataFrame.
Instead we should make our data loaders pre-map the data.
2015-11-13 18:26:54 -05:00
5bddf32062
BUG: Raise if we fail to symbol map a Frame/Panel.
...
Previously, we just warned here, but then we'd fail immediately trying
to write the found values into an index that's too long.
2015-11-13 18:26:54 -05:00
ff51510d88
PERF: Cache the result of failed lookups.
...
Otherwise we'll keep trying to look them up.
2015-11-13 18:26:54 -05:00
c818463051
MAINT: Simplify conditional.
2015-11-13 18:26:54 -05:00
abbda17239
Merge pull request #841 from quantopian/remember-downloads
...
Remember downloads
2015-11-13 18:25:06 -05:00
182cf9754c
DOC: Add a whatsnew.
2015-11-13 18:12:10 -05:00
43ac9eab5c
ENH: Check getmtime on download locations.
...
Rather than repeatedly try and fail to download data that's not yet
available, only try to download again if we haven't successfully
downloaded in the last hour.
2015-11-13 18:06:04 -05:00
1b7d0c9477
MAINT: Add __future__ print function import.
...
We do print(stock) in this file, which happens to work in py2, but is
confusing.
2015-11-13 18:06:04 -05:00
15c146b4a6
Merge pull request #840 from quantopian/coveralls_please
...
BLD: Add back installation of coveralls on travis
2015-11-13 17:38:51 -05:00
19eb6d0f59
DOC: Updated whatsnew
2015-11-13 17:14:29 -05:00
c6d4da4ca5
BLD: Add back installation of coveralls on travis
...
Was mistakenly removed, so coveralls has not been running.
2015-11-13 16:34:08 -05:00
7f38f9056e
Merge pull request #838 from quantopian/limit_timer
...
TST: Limit the timer report to just the 15 longest tests
2015-11-13 15:38:21 -05:00
f3f5d8a425
DOC: Updated whatsnew
2015-11-13 09:45:05 -05:00
526e9f1719
TST: Limit the timer report to just the 15 longest tests
...
The default of all the tests is pretty spammy.
2015-11-13 09:43:17 -05:00
f73003442a
Merge pull request #815 from quantopian/asset-db-versioning
...
ENH: Adds versions to Asset databases
2015-11-12 15:06:31 -05:00
7a1abc88f8
DOC: Adds whatsnew for asset db versioning
2015-11-12 14:44:20 -05:00
e1e160a128
MAINT: Factors-out asset db table schemas for cleanliness
2015-11-12 14:43:11 -05:00
85dd4b70dd
MAINT: Renames 'version' table to 'version_info' for clarity
2015-11-12 14:43:11 -05:00
6aac54544e
MAINT: Only write asset db version on db creation
2015-11-12 14:43:11 -05:00
48e488d423
ENH: Adds versions to asset databases
2015-11-12 14:43:11 -05:00
1fe4dfe229
Merge pull request #830 from quantopian/equity_caching
...
Equity caching
2015-11-12 14:01:29 -05:00
df492ecf7f
DOC: add whatsnew entry for AssetFinderCachedEquities.
2015-11-12 11:02:30 -05:00
d0cb5bd513
ENH: add extension to AssetFinder that caches all equities in memory and uses that cache in lookup_symbol.
2015-11-12 11:02:26 -05:00
f0dd92772f
MAINT: typos
2015-11-11 19:24:25 -05:00
9c61c77344
TST: cleanup test_history param output
2015-11-11 19:23:20 -05:00
0f5cf78492
MAINT: use subtest to quiet the output from test_events
2015-11-11 19:23:20 -05:00
fb90e71e6b
DOC: whatsnew for subtest
2015-11-11 19:23:20 -05:00
b24c8fc7f4
TST: Adds a subtest decorator.
...
Allows us to run subtests while still seeing which parameter
combination caused the tests to fail. The decorator can be used to
create subtests inside a test or to parameterize an entire test method.
We cannot use something like unittest2.TestCase.subTest because our test
runner does not support that.
2015-11-11 19:22:53 -05:00
b37b7cd89c
DOC: add entry for release-process docs
2015-11-11 19:21:26 -05:00
715baeb269
DOC: document the release process
2015-11-11 19:21:26 -05:00
Scott Sanderson