Adithya Kumar
86e8accfb0
Implement UTF-16 encoding/decoding routines
...
- Add identity tests
- Add benchmarks
- Add Word16 to Word8 helpers
2024-02-22 15:19:48 +05:30
Harendra Kumar
b2e7d1dabc
Fix old prelude rate benchmarks
2024-02-18 03:11:09 +05:30
Harendra Kumar
0990626371
Add comparison benchmarks for rate/without rate
2024-02-18 03:10:58 +05:30
Harendra Kumar
14932904bd
Add rate benchmarks for new concurrent streams
2024-02-18 03:10:46 +05:30
Harendra Kumar
dd12c4f7cb
Add maxThreads 1 benchmarks for parConcat
2024-02-14 11:51:39 +05:30
Harendra Kumar
db1cee542a
Add a scan for folds using "Scan" type
2024-02-13 04:21:08 +05:30
Harendra Kumar
fe2dba682c
Add a Scan module as simplified pipe
2024-02-13 04:21:08 +05:30
Harendra Kumar
e73aa914bb
Merge the Scan module into the Pipe module
2024-02-13 04:21:08 +05:30
Harendra Kumar
11845277c7
Cleanup and make consistent the export lists of Array modules
2024-01-04 09:25:36 +05:30
Harendra Kumar
191abebe92
Fix serialize test taking too long, too much mem
2024-01-04 07:41:04 +05:30
Harendra Kumar
a067d53632
Fix hlint issues in benchmarks
2023-12-21 17:17:52 +05:30
Julian Ospald
6d2798cdd9
Relax filepath upper bound
2023-12-15 23:53:51 +05:30
Harendra Kumar
c0d3178fa4
Make SerializeConfig opaque
2023-11-28 03:36:49 +05:30
Harendra Kumar
fd93674cec
Remove the intermediate Concurrent/Time/Exception modules
2023-11-27 15:50:24 +05:30
Harendra Kumar
f432cc36da
Collect all internal Unfold modules in one
2023-11-27 02:11:00 +05:30
Harendra Kumar
13581bfec1
Rename serialize/deserialize/size of Serialize class
2023-11-26 01:57:10 +05:30
Harendra Kumar
1efeed403b
Rename peekByteIndex/pokeByteIndex to peekAt/pokeAt
2023-11-26 01:57:10 +05:30
Harendra Kumar
7cb856b7e6
Move the Array serialization APIs to the Array module
2023-11-26 01:57:10 +05:30
Harendra Kumar
56fd008519
Rename newByteArray to new
...
To keep it consistent with naming in other Array modules.
2023-11-26 01:57:10 +05:30
Harendra Kumar
6705ec00b9
Rename Data.Serialize to Data.MutByteArray
2023-11-26 01:57:10 +05:30
Harendra Kumar
627c72ae65
Rename newBytes and pinnedNewBytes
2023-11-24 13:38:10 +05:30
Harendra Kumar
47b10b3914
Rename MutableByteArray to MutByteArray
2023-11-24 13:38:10 +05:30
Harendra Kumar
dfa2079542
Move Stream.Chunked to Array.Stream
2023-11-24 12:37:42 +05:30
Harendra Kumar
41baea74b8
Move Stream.MutChunked to MutArray.Stream
2023-11-24 12:37:42 +05:30
Harendra Kumar
33bffd614b
Change deriveUnbox to use instance splice
2023-11-22 21:39:39 +05:30
Harendra Kumar
1640a686e7
Rename Config/defaultConfig to S/serializeConfig
2023-11-22 21:39:39 +05:30
Harendra Kumar
7d50e5e03e
Implement deriveSerialize using deriveSerializeWith
2023-11-22 21:39:39 +05:30
Harendra Kumar
e7f6e7e9eb
Add Config setter functions for Serialize derivation
2023-11-22 21:39:39 +05:30
Harendra Kumar
97e0d20ffb
Re-export internal serialize modules and hide them
2023-11-21 02:37:28 +05:30
Harendra Kumar
32390c592e
Add a finalizer action in Fold type
...
This was needed especially for concurrent fold combinators. A fold
combinator that uses concurrent folds needs to wait for the concurrent
folds to finish before it can finish. The finalizing action in folds can
deallocate any resources allocated by the "initial" action and also wait
for folds that it has initialized.
This complicates fold combinators in general. We can potentially
introduce a type for non-failing parsers and support finalization only
in those. The current use cases can be covered by that. Parsers do not
support scanning, which is not required in the use cases where we need
finalization (there is no known use case).
2023-11-18 23:51:09 +05:30
Harendra Kumar
633bb185a8
Remove use-gauge build flag
2023-11-03 01:44:25 +05:30
Harendra Kumar
a11fe00965
Fix warnings for GHC 9.6 and 9.8
2023-11-03 01:44:25 +05:30
Harendra Kumar
35bc11a0fc
Use more memory for running Parser bench with --long
2023-11-02 15:38:07 +05:30
Harendra Kumar
79df417138
Fix the ParserK benchmark module names
2023-11-02 15:38:07 +05:30
Harendra Kumar
74e3cdfcb4
Move some benchmarks from o-1 to o-n category
2023-11-02 15:38:07 +05:30
Harendra Kumar
efecb265d2
Change benchmark mem requirements for GHC 9.6
2023-11-02 15:38:07 +05:30
Harendra Kumar
4b03673010
Disable some inspection tests for GHC-9.6 onwards
2023-10-31 13:37:28 +05:30
Harendra Kumar
104bfc3a3f
Update dependency versions
2023-10-31 13:32:39 +05:30
Adithya Kumar
c6fa580a28
Add benchmarks for the recordSyntaxWithHeader option in Serialize.TH
2023-09-27 08:18:14 +05:30
Adithya Kumar
5fef84310e
Remove Serailize.Size and change the signature of Serialize.size
2023-08-30 14:07:33 +05:30
Harendra Kumar
9e8c35d49e
Add fusion checking code in Data.Fold benchmark
2023-08-24 19:43:51 +05:30
Harendra Kumar
3e93e22228
Disable some redundant serialize benchmarks
2023-08-23 16:20:10 +05:30
Harendra Kumar
71346fd9a3
Add encode benchmark
...
encode = poke + getSize + array allocation
2023-08-23 16:20:10 +05:30
Harendra Kumar
282d79cd9d
Use deepseq instead of equality check
2023-08-23 16:20:10 +05:30
Harendra Kumar
7bc450d725
Pass the array size to deserialize for bound check
2023-08-21 16:28:02 +05:30
Harendra Kumar
43e25a173f
Enable the comparison after deserialization
2023-08-21 16:28:02 +05:30
Ranjeet Ranjan
209e990b06
Re-export entire submodules from Parser ( #2525 )
2023-08-21 14:09:09 +05:30
Adithya Kumar
bfe0169024
Simplify Serialize benchmarking code
2023-08-21 01:45:06 +05:30
Harendra Kumar
4dbdf477a8
Remove the alloc cost from poke op
...
For better comparison with peek cost.
Also now there are two versions of poke benchmarks, one including the
getSize cost and the other without to figure out how much overhead
getSize adds, because the cost may be nothing if it is dominated by
cache miss costs.
2023-08-21 01:45:06 +05:30
Harendra Kumar
4fdc8360ff
Include the size/alloc op costs in poke/roundtrip
2023-08-21 01:45:06 +05:30