goldcollection
/
muwire
mirror of https://github.com/zlatinb/muwire
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity
2,715 Commits
9 Branches
129 Tags
8.5 MiB
Commit Graph

2715 Commits (d61e4245bdb5e2d4074be917f93e6b5b9a85a8bc)

Author SHA1 Message Date
Zlatin Balevsky 1e7e30dee2
Switch to LAX Json parser and parallelize the stream after filtering 2021-06-09 12:50:00 +01:00
Zlatin Balevsky aa2b80cc0d
remove unnecessary canonicalization 2021-06-09 12:48:47 +01:00
Zlatin Balevsky 3c2d6e3952
more efficient (de)serialization of String[] 2021-06-09 00:57:27 +01:00
Zlatin Balevsky 96e359643a
double page size 2021-06-09 00:55:31 +01:00
Zlatin Balevsky 12a052ca55
small optimizations 2021-06-08 23:21:54 +01:00
Zlatin Balevsky fd06fdf7ce
spltt search index into Java and Groovy parts for faster initial loading. GitHub issue #59 2021-06-08 22:30:33 +01:00
Zlatin Balevsky e09b4c6621
do not reload keys, only values 2021-06-08 21:26:18 +01:00
Zlatin Balevsky 5b3416f7bb
perform a periodic gc() just in case, GitHub issue #59 2021-06-08 16:17:04 +01:00
Zlatin Balevsky 94a78bfa4e
GitHub issue #59 - try to accommodate larger libraries without limiting heap size 2021-06-08 15:54:51 +01:00
Zlatin Balevsky bcc94abadd
switch Set -> Array in FileTree 2021-06-05 23:54:04 +01:00
Zlatin Balevsky d00d6d7e18
flatten the nameToFile map as well 2021-06-05 23:39:25 +01:00
Zlatin Balevsky 71d20d075b
reduce memory usage of the rootToFiles structure by switching to array values 2021-06-05 23:18:39 +01:00
Zlatin Balevsky 30abc5af8d
get rid of unused cached value 2021-06-05 22:16:49 +01:00
Zlatin Balevsky fe85887b39
use two SkipLists, thanks zzz for the suggestion 2021-06-05 15:44:53 +01:00
Zlatin Balevsky 72c4998954
increment counters appropriately after increasing value size to integer 2021-06-05 05:29:54 +01:00
Zlatin Balevsky ea1f431323
more efficient splitting, reduces blockfile access 2021-06-05 03:18:18 +01:00
Zlatin Balevsky cb2ab77316
increase max value size to 2^24 2021-06-05 03:11:50 +01:00
Zlatin Balevsky a9c4f612fd
fix re-use of SkipSpan when it goes empty 2021-06-05 03:04:52 +01:00
Zlatin Balevsky 1168a871eb
get rid of the limit for byte[] i/o 2021-06-04 22:27:22 +01:00
Zlatin Balevsky 85043abb82 Merge branch 'blockfile-index' into 'master' 
Off-heap Blockfile storage for search index

See merge request zlatinb/muwire!52
 2021-06-04 17:49:52 +00:00
Zlatin Balevsky 0916e67939
rename package so that it won't conflict in the plugin 2021-06-04 18:22:45 +01:00
Zlatin Balevsky 88f1f381b9
multi-part backing storage 2021-06-04 18:14:07 +01:00
Zlatin Balevsky 4a0275914d
clean or suppress some warnings 2021-06-04 16:05:14 +01:00
Zlatin Balevsky 38bfecbc4f
increase span size 2021-06-04 16:01:14 +01:00
Zlatin Balevsky 40492ac5d2
re-add nulling of values 2021-06-04 16:00:31 +01:00
Zlatin Balevsky 008d82ad1d
side-port the blockfile db changes from i2p.i2p 2021-06-04 14:54:16 +01:00
Zlatin Balevsky 93ac56b353
fix or suppress various warnings 2021-06-04 12:41:01 +01:00
Zlatin Balevsky 84387bef8e
clear out vals storage after retrieving value 2021-06-04 12:35:21 +01:00
Zlatin Balevsky 3d30f49776
do not store values in the heap, just keys 2021-06-04 12:23:22 +01:00
Zlatin Balevsky 9ecb73149c
Split processing of files from directories. Throttle emission of new events based on threadpool queue length 2021-06-04 05:57:50 +01:00
Zlatin Balevsky dfd7be9618
reduce the size of the memmapped file 2021-06-04 03:48:15 +01:00
Zlatin Balevsky 8892dd73a8
switch to a memory-mapped file instead of a RAF 2021-06-04 03:21:43 +01:00
Zlatin Balevsky 76eaca996f
experimental Blockfile db instead of in-memory index 2021-06-04 00:09:33 +01:00
Zlatin Balevsky 1c2972ecfc
add some default file types to ignore 2021-06-03 17:22:38 +01:00
Zlatin Balevsky 34b7496036
prevent corrupt messages from breaking messenger loading 2021-06-03 12:54:00 +01:00
Zlatin Balevsky 6f3a3357ea
more efficient traversal of directories 2021-06-03 12:15:14 +01:00
Zlatin Balevsky 3ac6d085f6
configurable list of file types to not share 2021-06-03 05:27:42 +01:00
Zlatin Balevsky 95fcd1635f
more safeguards against zero-length or unreadable files 2021-06-03 01:59:33 +01:00
Zlatin Balevsky 4b11aca16e
show the number of files being hashed 2021-06-02 20:10:26 +01:00
Zlatin Balevsky 2af996266f
check if file is readable before trying to hash 2021-06-02 20:09:55 +01:00
Zlatin Balevsky c849fa078a
No need to restart when chaning the number of hashing threads , GitHub issue #57 2021-06-02 19:17:17 +01:00
Zlatin Balevsky b1af2e557b
Make the number of hashing threads configurable, GitHub issue #57 2021-06-01 23:35:59 +01:00
Zlatin Balevsky 228a92a58d
filter the hosts that are about to be verified as well 2021-06-01 20:24:38 +01:00
Zlatin Balevsky 0fc5ca7141
fix copy-full-id from message table 2021-06-01 19:25:29 +01:00
Zlatin Balevsky 47e2c32877
update TODO with some items and ideas 2021-06-01 19:09:19 +01:00
Zlatin Balevsky 173fd55afb
update version, get rid of cli completely 2021-05-26 22:34:36 +01:00
Zlatin Balevsky 7a40cdd0c2
return 3 hosts from hostcache 2021-05-26 09:42:41 +01:00
Zlatin Balevsky 837b8d8213
connect more aggressively, send more pongs, fix destination parsing and header check 2021-05-26 08:57:20 +01:00
Zlatin Balevsky 4e09b4ad71
forgot 0x 2021-05-26 08:01:48 +01:00
Zlatin Balevsky cfe1fe98b0
wip on switching to binary pongs 2021-05-26 07:46:25 +01:00