Commit graph

72 commits

Author SHA1 Message Date
Karl Seguin 692cd618b2 guard access to item.promotions in LayeredCache, which was applied to Cache in 557d56ec6f 2018-12-27 22:54:50 +07:00
Karl Seguin 142396791e
Merge pull request #22 from alexejk/gcOnDelete
Calling OnDelete from gc()
2018-11-26 20:29:43 +07:00
Alexej Kubarev 243f5c8219
Fixes #21. Callong OnDelete during gc() 2018-11-25 15:31:09 -08:00
Alexej Kubarev 7e55af0a9c
Small routine lock file upgrade 2018-11-25 15:29:47 -08:00
Karl Seguin a317416755
Merge pull request #16 from alexejk/onremove
Support for OnDelete() callback
2018-07-22 11:17:02 +07:00
Alexej Kubarev 7421e2d7b4
Adding support for OnDelete callback function
OnDelete will receive an item that is being processed for deletion to support calling cleanup function specific to the item stored
2018-07-16 18:20:17 +02:00
Alexej Kubarev 72059a01e9
Adding missing ignore file to ensure vendor folder is not checked in 2018-07-16 18:18:51 +02:00
Alexej Kubarev 00324cb2d2
Adding dep dependency management files to utilize vendoring 2018-07-16 18:18:02 +02:00
Karl Seguin b425c9ca00 Merge pull request #11 from EdwardBetts/spelling
correct spelling mistake
2017-09-04 13:47:32 +07:00
Edward Betts 0d05fa8278 correct spelling mistake 2017-09-01 11:40:44 +01:00
Karl Seguin 3ba9789cfd Merge pull request #10 from heyitsanthony/test-races
fix data races in tests
2017-02-17 13:08:20 +07:00
Anthony Romano b3c864ded7 cache: make Stop() synchronous and races in tests
worker goroutine running concurrently with tests would cause data race errors
when running tests with -race enabled.
2017-02-13 15:39:24 -08:00
Anthony Romano c69270ce08 layeredcache: add Stop() and fix races in tests
worker goroutine running concurrently with tests would cause data race errors
when running tests with -race enabled.
2017-02-13 15:39:24 -08:00
Karl Seguin 12c7ffdc19 Merge pull request #9 from spicydog/patch-1
Fix an error in example "itemsToPrune"
2016-12-22 18:15:50 +07:00
spicydog 77679ba342 Fix an error in example "itemsToPrune"
itemsToPrune -> ItemsToPrune
2016-12-21 20:07:16 +10:00
Karl Seguin a2d6215577 Merge pull request #8 from jdeppe-pivotal/master
Add a SecondaryCache which exposes the secondary part of a LayeredCache
2016-11-03 22:19:53 +07:00
Jens Deppe a451d7262c Integrate feedback and upstream fixes
- Ensure correct locking in GetOrCreateSecondaryCache
- Fetch now returns a *Item
2016-11-01 23:53:22 -07:00
Jens Deppe d2c2442186 Merge remote-tracking branch 'seguin/master' 2016-11-01 20:33:44 -07:00
Karl Seguin 8adbb5637b return *Item from layered cache fetch instead of interface{} 2016-11-02 09:34:09 +07:00
Jens Deppe c1634a4d00 Add concept of a SecondaryCache which exposes the secondary part of a LayeredCache 2016-11-01 09:01:39 -07:00
Karl Seguin 2f6b517f7b Merge pull request #6 from HasMatthew/nanosecond_ttl
Use nanosecond-resolution TTL instead of second-resolution.
2016-07-07 20:03:45 -07:00
Matthew Dale 162d4e27ca Use nanosecond-resolution TTL instead of second-resolution. 2016-07-07 15:32:49 -07:00
Karl Seguin ddcff8e624 Merge pull request #5 from dvdplm/master
Fetch does not return stale items
2016-02-05 22:16:17 +08:00
David Palm 3665b16e83 Better test 2016-02-05 14:34:58 +01:00
David Palm d5307b40af Fetch does not return stale items 2016-02-03 16:07:59 +01:00
Karl Seguin 74754c77cc Partially fixing #3.
On close, drain the deletables channel (unblocking an waiting goroutines) and
close deletables. Like Gets and Sets against a now-closed promotables,
this means any subsequent to Deletes from deletables will panic.

I'm still not sure that this is ccache's responsibility. If a client closes a DB
connection, we'd expect subsequent operations against the now-closed connection
to fail. My main problems with defer'ing a recover are:

1 - the performance overhead on every single get / set / delete
2 - not communicating with the caller that the requested operatin is no longer
    valid.
2015-07-26 11:05:48 +08:00
Karl Seguin bfa769c6b6 add Stop method to stop the background worker and make it possible for the GC to reap the object 2015-07-23 22:24:50 +08:00
Karl Seguin 41f1a3cfcb gonna be one of those days... 2015-01-07 08:12:17 +07:00
Karl Seguin f9c7f14b7b Fetch's API wasn't usable. It returned different values types based on whether
the fetch was needed or not. It now behaves consistently (with itself and with
Get), returning an *Item.
2015-01-07 08:09:39 +07:00
Karl Seguin 6df1e24ae3 2 changes:
1 -
Previously, we determined if an item should be promoted in the main getter
thread. This required that we protect the item.promotions variable, as both
the getter and the worker were concurrently accessing it. This change pushes
the conditional promotion to the worker (from the getter's point of view, items
are always promoted). Since only the worker ever accesses .promotions, we no
longer must protect access to it.

2 -
The total size of the cache was being maintained by both the worker thread
and the calling code. This required that we protect access to cache.size. Now,
only the worker ever changes the size. While this simplifies much of the code,
it means that we can't easily replace an item (replacement either via Set or
Replace). A replcement now involves creating a new object and deleting the old
one (using the existing deletables and promotable infrastructure). The only
noticeable impact frmo this change is that, despite previous documentation,
Replace WILL cause the item to be promoted (but it still only does so if it
exists and it still doesn't extend the original TTL).
2014-12-28 11:11:32 +07:00
Karl Seguin 557d56ec6f guard all access to item.promotions 2014-12-28 10:35:20 +07:00
Karl Seguin c75dcd4c12 link to rcache 2014-12-06 17:19:23 +07:00
Karl Seguin 78e597cdae replace is size-aware 2014-11-21 15:45:11 +07:00
Karl Seguin 41ccfbb39a renamed MaxItems to MaxSize, updated readme 2014-11-21 15:06:27 +07:00
Karl Seguin c810d4feb3 test + fix for actual size function 2014-11-21 14:59:04 +07:00
Karl Seguin ff8727e847 initial work on tracking cache by item size 2014-11-21 14:39:25 +07:00
Karl Seguin 44cdb043d1 Move size tracking to a variable, away from simply using the length of the list.
This paves the way for more complex size tracking.
2014-11-20 07:03:59 +07:00
Karl Seguin df2f8eb082 Added documentation.
Bucket and LayeredBucket are no longer exported.
2014-11-14 07:56:24 +07:00
Karl Seguin 3e4d668990 blank identifier for tests 2014-11-14 07:41:22 +07:00
Karl Seguin 2131ac5052 better tests 2014-11-13 22:26:05 +07:00
Karl Seguin 7316f99bd9 replace on layeredcache 2014-11-13 22:23:52 +07:00
Karl Seguin 65573a0cb6 helper makefile 2014-11-13 22:20:23 +07:00
Karl Seguin cc0395a391 added replace method 2014-11-13 22:20:12 +07:00
Karl Seguin 7e08960075 update readme 2014-11-04 17:24:07 +07:00
Karl Seguin 5e131cc17c Buckets must be a power of 2. Move from % to & for determining the bucket. 2014-11-02 18:09:49 +07:00
Karl Seguin 624c03cd3e delete and deleteall return boolean to indicate if delete found the item 2014-10-27 08:30:48 +07:00
Karl Seguin 0fddc964ec added extend 2014-10-27 08:27:26 +07:00
Karl Seguin 77765a3f11 Get now returns the *Item rather than the item's value. Get no longer actively
purges stale items.

Combining these two changes, CCache can now be used to implement both of
Varnish's grace and saint mode.
2014-10-25 17:15:47 +07:00
Karl Seguin 3a00ce8f0a fixed possible nil panic when item is deleted immediately after being added 2014-10-25 12:24:52 +07:00
Karl Seguin b0e3fca0f6 fixed formatting 2014-10-25 12:21:10 +07:00