github.com/peermaps/eyros commits | Ecosyste.ms: OpenCollective

fix debug to read branch factor

86e36a8ae190d8e6da319743bb67b292591729f5 authored over 5 years ago by substack <[email protected]>

rust updated how the float elapsed method thing called

587eb8f05ae613a3f6464daa2c24ddf35e92d836 authored over 5 years ago by substack <[email protected]>

clean up params, add list_cache_size arg

2e9cc819fd2590d31fcd86fdf4f3023da1c807bc authored over 5 years ago by substack <[email protected]>

setup field for data list cache size

758586eb5a523d09894c8123e1f23e191f0461dd authored over 5 years ago by substack <[email protected]>

remove max data size < base size restriction with a test

692e9aefc050dc72720fc7908160012e94a8bbf2 authored over 5 years ago by substack <[email protected]>

faster without the cache for data blocks, but possibly only with sync_all() taken out of random-access-disk

71d0394345dad8c696e0317f4ac8fe90a53af40c authored over 5 years ago by substack <[email protected]>

block cache now working with mega_batch test plus more cache tests

7e77921f991c442ea3abb548d5eb188b290fb126 authored over 5 years ago by substack <[email protected]>

failing block cache append test is probably the same bug that the data store runs into

b6411417b7c509aca1685b9cbb4c0f26e6e1c5c8 authored over 5 years ago by substack <[email protected]>

block_cache truncate read/write cache pruning

4c46ba26b174edea66a82f82c1a71d57027d3c3e authored over 5 years ago by substack <[email protected]>

length implementation for block cache. now crashing somewhere in the mega test

e47be564aea441a74681a2796fb7eea3b9545aa5 authored over 5 years ago by substack <[email protected]>

failing block cache test

11b9afdcc9a6f78a5799d0d71c682841941dd415 authored over 5 years ago by substack <[email protected]>

using block cache for trees with passing tests

0f80f87853069c7ddcef7d8104717ef106b3ce30 authored over 5 years ago by substack <[email protected]>

commit implementation and block cache read write

7124c2edc520dfb02e1c638fc447cce8ecc0c496 authored over 5 years ago by substack <[email protected]>

block cache working for full write tests

3158854e691e0277834ada271cc694004eda2849 authored over 5 years ago by substack <[email protected]>

partial block cache implementation

fe88fdd1762add490c90e327c29b30588d281fb4 authored over 5 years ago by substack <[email protected]>

block storage stub, extra block fields, block opts setup for trees and fields setup for the db to keep things more organized

bb03ca7d91d6df09e02b9988cc95e0fd5dd0cc78 authored over 5 years ago by substack <[email protected]>

bbox cache speeds things up slightly

7ec77ac7a376fd50518ad33fe791d91b3f9d55f4 authored over 5 years ago by substack <[email protected]>

extra mega test for a smaller result set to check query perf on unprimed cache

7acd511e01035a02d76f4de2718a2c34c741b72c authored over 5 years ago by substack <[email protected]>

move bbox creation into data to make room for lru cache

5e8c9a6661a06f7ba46f9a904033093e875af27e authored over 5 years ago by substack <[email protected]>

fix stack overflow crash with padding pivots when len==1

f6ae782d275ea29f4d5de5a3b9bde95b3595497b authored over 5 years ago by substack <[email protected]>

grow example for debugging / perf testing

0ef69656ae10b0352b748a84d801b41fc59f9699 authored over 5 years ago by substack <[email protected]>

get rid of data splitter thing that didn't work out but keep batch size assertion checks

a4ce4236c1a5c37f96ae8965958ddc355dd18825 authored over 5 years ago by substack <[email protected]>

take out splitter test case for a perf boost and instead allow branches to be built from smaller number of buckets

44b4cf5bc0b2743789911c8d8112aab8168dd7ae authored over 5 years ago by substack <[email protected]>

another data size assertion

6e18617ac188efdcea92ecd9a9c120f2c2aafdd0 authored over 5 years ago by substack <[email protected]>

implementation for splitter approach. drops a small amount of data and slows things down but queries are faster now that max data size invariants are enforced

0ce6e370ffe6f90758246caeb99022bfe7270c48 authored over 5 years ago by substack <[email protected]>

write cache bug removing overlaps

bd38b2b6bb57e34581a163bd5ece75418b7a41d7 authored over 5 years ago by substack <[email protected]>

print out all branches in debug command

da9f6da4c47bf02f11423f90d6f01f261534071f authored over 5 years ago by substack <[email protected]>

time-query shows that query results are very slow for really large databases

f168d92f6399935ca1a498197ca5607457a9b880 authored over 5 years ago by substack <[email protected]>

load_mixed test that fails when the P::size_of isn't used

95c726067249c7c257199c9d14b090ee6ba3051c authored over 5 years ago by substack <[email protected]>

load test

2cf9688563af8a0e50b0b6b70ae52adf4ff4605e authored over 5 years ago by substack <[email protected]>

fix overlap test for intervals

84b29ab9615563e82e5662bcda13d69a177536fc authored over 5 years ago by substack <[email protected]>

mega batch test

09faab660692a3e88f70e3881c0df71c60ff4b81 authored over 5 years ago by substack <[email protected]>

failing test with mixed types and intervals

566cb251550eff04a170e92a9371f1e949d70a21 authored over 5 years ago by substack <[email protected]>

Point::size_of to match lack of padding in bincode

9e81acf339101b83d43e4539ee0d880174e34d46 authored over 5 years ago by substack <[email protected]>

remove assert_eq (which calls panic) from ensure_eq

30d3a6aa397040542eae748cc4c1b5a18b4ec3b6 authored over 5 years ago by substack <[email protected]>

assertion in staging to catch when serde pads records so size_of doesn't work

63bd9544abbcd4e2b2a815ba9a5a563f442761a1 authored over 5 years ago by substack <[email protected]>

failing test for mixed point types

dfe3235aadf376e0d2676ac61f210b2ca1a49221 authored over 5 years ago by substack <[email protected]>

fixed write cache and staging bug. mega test now passes

cea901cfc0c8a767f351f6dd03e8c3ff5bb151e4 authored over 5 years ago by substack <[email protected]>

write cahce fix for offset bounds

28c58b0b52896dc0e0f35f62897833f8361de709 authored over 5 years ago by substack <[email protected]>

fix issue resetting the counter in pivot comparisons

04e34d8ef9d1cdbd1e7d7795ba3d4714cf7bcddf authored over 5 years ago by substack <[email protected]>

fix vec index bug

53de84e3d94fafe3c988e6eb575c770528d253a5 authored over 5 years ago by substack <[email protected]>

fixed issue with branching, re-introduced some other bugs

8cc91b2bf97a1a8317c26069edc2c85c233a9db7 authored over 5 years ago by substack <[email protected]>

fix bug with non-9 branch_factor. single batch param tests

8fce207127cdcdeb9ca5dd1c1d8e02de4885ae58 authored over 5 years ago by substack <[email protected]>

builder

4412b147e9f70e33eab2694d8d56265dfacf72a1 authored over 5 years ago by substack <[email protected]>

implement reads and autoflush for write cache

b4797ec40380848b0b7f95036d3fd640a290362f authored over 5 years ago by substack <[email protected]>

basic write cache, could use some work

eb84dc75dc745b948e36115b96c3e900d341e089 authored over 5 years ago by substack <[email protected]>

deduplicate pivots before padding

e6fd88e62c1971b29eee832aa40990c64f05ac87 authored over 5 years ago by substack <[email protected]>

pad pivots

cc7c4aa6783f18c15930e6bec69c6638f1506066 authored almost 6 years ago by substack <[email protected]>

refactor so merges take into account the number of records they represent

69907d5d347376c427966a3f7b79d29824ee80b4 authored almost 6 years ago by substack <[email protected]>

intersection branch case

bdacddbbb18e4f07bd871d06b96f072e765a162d authored almost 6 years ago by substack <[email protected]>

use macros to set result types correctly instead of panic! from assert_eq!

b8cd1912bd836947037b5dc05c1def8c8a300bf3 authored almost 6 years ago by substack <[email protected]>

guard max_data_size value, cleanup

1d37cb61582afb5213e428d120b9b54f1c2c0354 authored almost 6 years ago by substack <[email protected]>

fix for pivots not being in ascending order and sometimes being equal

77ada88b94a80562eece215ec6d14ed342e8c1dd authored almost 6 years ago by substack <[email protected]>

more understandable construction loop

0b4633aa5e674efaf65031cf0fd347c12b227e36 authored almost 6 years ago by substack <[email protected]>

show staging info in debug command

2be3c031d74a295dd8db199141c1ef5a1fd73054 authored almost 6 years ago by substack <[email protected]>

fix so that the last bucket gets records

44e0085d39757602f95334a483af829f8cc2e570 authored almost 6 years ago by substack <[email protected]>

pass through max data size

90fba8b50e4b22ca5b8f3c7786041f665975d689 authored almost 6 years ago by substack <[email protected]>

fix for duplicate/missing records

ba77b507b81e64592ca6427c3a83d0dbf7146c47 authored almost 6 years ago by substack <[email protected]>

bug setting the correct bucket in branch routine

580f584793f314ca35ed5e7ba01578308d232c5b authored almost 6 years ago by substack <[email protected]>

better info command, rename "size" and "len" to "bytes" to be less ambiguous

f4347e346b3903eccef232f2df3800cd3047459b authored almost 6 years ago by substack <[email protected]>

command for debugging data

2d8cf5525e82574d25aed61e0ebdb58c8f30e0a4 authored almost 6 years ago by substack <[email protected]>

clear not needed

a2b1672f412ed15dbf7c266aeb5cff96e2ad24af authored almost 6 years ago by substack <[email protected]>

fix to get the correct number of results

92c1f366bc0693af4ed5c20f366466c0af1d0335 authored almost 6 years ago by substack <[email protected]>

now there are too many results in the merge output

68c525db6a698ca1230fc4ba091b9f77e10121b1 authored almost 6 years ago by substack <[email protected]>

figured out types so that merge over data ranges can reuse existing branch builder implementation

c6aafd623b86e1ab2c24da739dcdbf6988b4b578 authored almost 6 years ago by substack <[email protected]>

refactor to use a trait for data batching

0b33641eaf742fc59fc403803bb32701d4d6a211 authored almost 6 years ago by substack <[email protected]>

rearrange the builder to reuse the implementation for merges

c42b695073d81c2df3698555545f00e53cf3d425 authored almost 6 years ago by substack <[email protected]>

calculate bounds for the data blocks in a tree during a merge

5945b87c3e3d01a90dca094d031338828a797af6 authored almost 6 years ago by substack <[email protected]>

refactor batch planner handling to satisfy invariants

65af823aa7e1b8a6fdb9868a1dba879bdcd94cd3 authored almost 6 years ago by substack <[email protected]>

failing multi batch test

b3adc8dfcb3c871a1ac1d653efef207c9f20fcd2 authored almost 6 years ago by substack <[email protected]>

fix single batch test and add partial test

ecdddd06e36d4e193562ed4079d4a9ff67c89032 authored almost 6 years ago by substack <[email protected]>

extra constrait for branch factor

d4481cf6a5199a575f7d416c953417fe37824ebb authored almost 6 years ago by substack <[email protected]>

fix order tests for correct branch factor consideration

33f7136bdb9d1e21aa2875cf26e7bf0e16a84313 authored almost 6 years ago by substack <[email protected]>

clean up warnings

10c363d77d4a9e7f35064cc5d20ea942d4ee3f0d authored almost 6 years ago by substack <[email protected]>

seriously evil bug. correct number of results now

58792867bce44ead63d9a60728c9eb342a351b5d authored almost 6 years ago by substack <[email protected]>

expanded test to compare elements

1d13962d2610b34367982f3c9b61e763cc2db6d6 authored almost 6 years ago by substack <[email protected]>

refactor lib branch

d6d69e6bef3143810817068d3219432c3a1cd465 authored almost 6 years ago by substack <[email protected]>

failing batch test

6b5c519dbd4640fd41992027472851e148943950 authored almost 6 years ago by substack <[email protected]>

mostly working, but drops a few records

59aa7afa0261d100cfdf524023c4c253aed4da89 authored almost 6 years ago by substack <[email protected]>

fix issue in data bitfield calculation

fedbf6dfb7ffab529119cec3ab42b716906670cc authored almost 6 years ago by substack <[email protected]>

walk the buckets, but issues in the construction somewhere

cb28ea928f249101726e293375c8fe091f07cfed authored almost 6 years ago by substack <[email protected]>

using factored out block reader in tree branch reads

73b3a6ae957d4ed4fe90df2fcc3faad419a4cd18 authored almost 6 years ago by substack <[email protected]>

more detailed data format doc, reading blocks intersecting pivots, split out block reading code

872e626aa932847e625815386bbae002d68aba8a authored almost 6 years ago by substack <[email protected]>

refactoring and partial implementation for queries

6ba9e66b75ece1c2ca3ca90597444855cd7c07fd authored almost 6 years ago by substack <[email protected]>

write all branch fields to the tree

a5c4632e59029d7a5bee4ec687c9efa8066764e0 authored almost 6 years ago by substack <[email protected]>

refactor branch construction to write out to separate data store

307c31563220d7a0d70bba66266fc72e1d2b13d1 authored almost 6 years ago by substack <[email protected]>

more detail about branch blocks

29dd176a425673ed33dfd59913901120b8d546c7 authored almost 6 years ago by substack <[email protected]>

factor out pivot_order and add a test

52d22ab925ca3f8fd52394cad2cf3c5294462862 authored almost 6 years ago by substack <[email protected]>

some work on cmp_buf but going to refactor next

0f144292ed397a67816c5af59a9cd7f98738dd43 authored almost 6 years ago by substack <[email protected]>

data format

f5ddbdca050fc94820b0f615a4ab8b4293a2ea98 authored almost 6 years ago by substack <[email protected]>

start of the query traversal implementation. added cmp_buf point method

a290b8f03bca451f3d5e637da8da1ff3e139f294 authored almost 6 years ago by substack <[email protected]>

planner fix for carry on the last bit

5988472f218ac54943d94d455482479589da1a45 authored almost 6 years ago by substack <[email protected]>

load mask from storage

3de7148f5b513e47241dedbedae7e33f7a68f92c authored almost 6 years ago by substack <[email protected]>

calculations to divide batches into staging remainder and trees for merging

6dc5d1e7adb41a3e2c93951b79658c7c4474a851 authored almost 6 years ago by substack <[email protected]>

new planner algorithm with staging and tree index output

dacdbc13dac2893d8e45598d40865c9b89a2484c authored almost 6 years ago by substack <[email protected]>

tree planner with tests

2c78b9d2aa31f6ba4a5154d757d2d38e3d926bbc authored almost 6 years ago by substack <[email protected]>

add batched rows to staging

7416baf50c31ac67eff79951b19b208058d102aa authored almost 6 years ago by substack <[email protected]>

staging implementation

3b56b3e956d5e234d5986c361666e656b85ccf5f authored almost 6 years ago by substack <[email protected]>

round-robin for each tree for query results

40ad3190db93c770dd602dd2becc99a1ed3de0c8 authored almost 6 years ago by substack <[email protected]>

replace interval/point combinations with combination macro up to 6 dimensions

197d35fc317cb103ab2fd407ac7e1f4d1d772ec4 authored almost 6 years ago by substack <[email protected]>