143d570960
Binary encoding ( #1679 )
...
Noms format is now binary encoded.
2016-05-31 16:13:51 -07:00
b79f5b0a04
Implement JS BlobReader.seek ( #1673 )
2016-05-27 18:39:59 -07:00
dbd4c572df
Add ListBuilder ( #1669 )
...
A ListBuilder allows creating large lists efficiently
2016-05-27 14:49:22 -07:00
db9bb15786
Remove es6 support from js sdk ( #1668 )
...
We're not using it, and it just makes compiles slower
2016-05-27 14:08:48 -07:00
18838b7c2a
js sdk: expose type.describe() for convenience ( #1667 )
2016-05-27 14:08:34 -07:00
f52a062f81
Increment version ( #1664 )
2016-05-27 11:24:16 -07:00
5792f901c6
Use native browser fetch if available ( #1655 )
2016-05-27 10:40:20 -07:00
d2880c5d2a
Export escapeStructField ( #1654 )
2016-05-26 17:01:54 -07:00
e271576f06
Forward all query parameters when calling remote server api.
2016-05-25 17:34:49 -07:00
998b7a9df0
Update flow to 0.25 ( #1619 )
2016-05-25 10:30:17 -07:00
0a2dd97d4a
Enable database authorization on JS dataspecs.
...
Accomplished by forwarding query params specified on database spec to
remote server api.
Fixes #1595
2016-05-24 17:47:46 -07:00
68b1b88a69
js/json-convert: encode field names ( #1620 )
2016-05-24 16:47:22 -07:00
61848137ad
increasing the mocha timeout so travis builds fail less ( #1622 )
2016-05-24 14:49:31 -07:00
43d991495c
noms/js: export isSubtype ( #1617 )
2016-05-24 11:55:19 -07:00
e34551cd54
Standardise database/dataset accessors as database() and dataset() ( #1607 )
...
In JS they are either set() or store(). In Go they are either Dataset()
Store(), or Database().
2016-05-24 10:25:16 -07:00
a84ecf41eb
Fix code review issues
2016-05-23 16:51:45 -07:00
77de0b0e54
Introduce js/json-convert.js ( #1601 )
2016-05-23 15:07:35 -07:00
56c7f44aa9
Make JS walk() return value mean "skip" not "recurse" ( #1604 )
...
This is consistent with Go Walk(), and it's arguably less surprising
than returning undefined implies false.
2016-05-23 14:32:45 -07:00
6aa35802e8
Declare NodeJS version dependency for NOMS
...
Fixes #1057
2016-05-23 13:51:53 -07:00
859eca4650
Increment JS version
2016-05-20 19:18:32 -07:00
c97c9ac78b
Export makeUnionType in noms.js ( #1567 )
2016-05-20 16:44:06 -07:00
c6fa8b1029
Rename valueOrPrimite to Value ( #1566 )
...
This renames Value to ValueBase, valueOrPrimite to Value and make Value the default export.
Fixes #1528
2016-05-20 16:39:30 -07:00
01a50e9541
Change the async "newList" and friends to sync "new List" constructors ( #1554 )
...
Friends: List, Set, Map, Blob, and Commit. This lets us remove a lot of
Promise and async/await bloat.
2016-05-20 15:01:55 -07:00
f653a20511
Make Splore Great Again! ( #1541 )
2016-05-18 17:57:57 -07:00
778acf8454
Make OrderedPutCache throw in browsers ( #1539 )
...
This is a stop gap to ensure that the browser version does not depend
on nodejs modules.
Fixes #1531
2016-05-18 15:28:09 -07:00
d67d1fb924
Make Map use Array<[K, V]> instead of Array<K | V> ( #1538 )
2016-05-18 15:16:08 -07:00
4d47b9cf18
jsPath ( #1527 )
2016-05-18 14:35:26 -07:00
643c12172b
Noms data structure names ( #1525 )
...
* Change to default exports
* Rename NomsX to X in (Blob, List, Map, Set)
* Increment version
2016-05-18 09:26:36 -07:00
df1c91843d
JS: No need to compute the union type in decode ( #1523 )
...
* JS: No need to compute the union type in decode
Instead trust that the serialization is correct
Towards #1491
2016-05-17 15:39:18 -07:00
d125446955
Clear dist before publish ( #1480 )
...
This is so that we do not keep old files around.
Fixes #1380
2016-05-16 19:13:44 -07:00
e0eb7bc297
JS: back BatchStore's put-cache with tingodb ( #1479 )
...
* JS: back BatchStore's put-cache with tingodb
tingodb is a pure-JS, file-backed mongodb-api-compatible database.
Before this patch, BatchStore was caching all pending chunks in memory,
which seems unlikely to scale. Putting them in a tingodb spills them to
disk as needed, though we can still iterate them in put-order when it's
time to send them.
Fixes #1349
2016-05-16 14:38:25 -07:00
5db7fe8285
JS: Update the collection/struct API to compute the type. ( #1502 )
...
When you create or modify a data structure we now compute the type
based on the type of the values in the data structure.
Towards #1491
2016-05-16 10:58:50 -07:00
58464baef9
Total ordering of values ( #1475 )
...
This allows having a set/map of heterogenous elements/keys. To support
this we have to be able to have a total ordering for all noms values.
The ordering is false < true < -999 < 0 < 999 < "" < "a" < "z" < Hash
In other words, Bool < Number < String < * and for comparing non
primitives we compare the hash of the object.
Fixes #1104
Fixes #1312
2016-05-12 10:35:56 -07:00
1c90201f49
Make JS Sequence not be values ( #1470 )
2016-05-10 14:19:39 -07:00
14c377fc7b
Export BuzHash from js/noms. ( #1462 )
...
This is kind of non-ideal, but I need it in order to easily make a
hash performance test rig.
2016-05-09 14:33:01 -07:00
014e330315
Rename DataStore to Database ( #1448 )
2016-05-06 17:37:19 -07:00
c8ce81d6ba
Serialize meta tuple with ref value ( #1443 )
...
When we serialize a meta tuple we should serialize it with the Ref value (which includes the type). This is because in the future the sub tree might be of a different type than the current tree.
Towards #1312
2016-05-06 13:25:23 -07:00
2e7eafed25
Update noms npm version to 24.0.0 ( #1436 )
2016-05-05 17:12:41 -07:00
057f4e0423
Temporarily raise the test timeout 50% ( #1432 )
2016-05-05 15:26:00 -07:00
d159d25583
Introduce walk.js ( #1419 )
...
* Introduce walk.js
* review comments
2016-05-05 09:25:43 -07:00
c9d176cb28
Cleanup serialization of types ( #1410 )
...
We used to serialize types in two different ways, depending on whether
the type was used as a tag or as a value. Now we serialize it the same
way.
Also, remove makeCompoundType completely.
And remove some dead code.
2016-05-04 11:35:33 -07:00
963fbdaad6
Write type with value ( #1406 )
...
Write type for all values
Fixes #1308
2016-05-03 16:52:22 -07:00
aa0b0fd658
JS: Maintain validation cache on DataStore across Commit
...
While this causes the cache to potentially grow infinitely, having a
DataStore keep track of every value it's read or written makes it
simpler to program with. Once RefValues can only come to be by being
read out of a DataStore, we can do away with this altogether.
The big changes here are that DataStore now has a ValueStore, instead
if extending ValueStore. The DataStore constructor now takes a
ValueStore, so that new instances re-use the same ValueStore can be
constructed on Commit. Since callers don't care about that detail,
this change exports NewDataStore(), which still takes a BatchStore.
2016-05-03 13:11:01 -07:00
3032c76a39
Use a map for the struct fields
...
Fixes #1368
2016-05-02 15:35:27 -07:00
9626ad5b42
Fix JS chunk serialization to use network byte order
...
Fixes #1370
2016-05-02 12:32:59 -07:00
e45aa532a3
Manually update new version for JS SDK
2016-05-02 11:52:12 -07:00
4468625ce0
Update flow to 24.1 ( #1385 )
...
Update flow to 24.1
2016-05-02 11:45:49 -07:00
8bce49c838
Fix noms npm package version
2016-05-02 10:22:23 -07:00
8eba180782
Revert "renaming datastore to database ( #1374 ) - fixes bug #1290 "
...
This reverts commit d77902b323
2016-05-02 09:01:44 -07:00
d77902b323
renaming datastore to database ( #1374 ) - fixes bug #1290
...
* datastore -> database
* fixing bug 1290, renaming datastore to database
2016-05-01 22:02:16 -04:00
Rafael Weinstein