These changesets represents the changes on the squid/modio development branch of Squid. If there is any questions please contact squid-dev@squid-cache.org.
| Date | Author | Size | Description |
|---|
| 2001/03/18 22:51:21 | adri | +17 -13 | Fix merging stuff that wasn't picked up automagically. |
| 2001/03/18 22:19:05 | adri | +7 -10 | Fix up some evil logic that I broke. the http->out.offset is actually |
| 2001/02/21 18:46:33 | adri | +32 -434 | Remove the current range request implementation. This deletes the |
| 2001/02/20 16:15:25 | adri | +5 -5 | Fix the latest merge code to the modio state of mind |
| 2001/02/07 13:54:38 | adri | +7 -61 | Remove the icon loading. We'll handle this in a different fashion |
| 2001/02/06 10:10:36 | adri | +7 -2 | When doing a HEAD request, we can actually do a GET in the store and |
| 2001/02/06 10:08:03 | adri | +83 -26 | * add request_t to be a cbdata type. This means we can lock/unlock on it. |
| 2001/01/12 08:36:02 | adri | +2 -2 | mark the callback as invalid *after* we've done the callback in |
| 2001/01/10 23:29:16 | adri | +5 -6 | Bring back httpMemFree, lost in the merge. |
| 2001/01/10 23:06:46 | adri | +5 -7 | Convert the hash directory structure to the New World Order(tm). |
| 2001/01/10 20:41:13 | adri | +21 -5 | * kill ENTRY_VALIDATED. Its a per-fS issue, not a general store issue |
| 2001/01/08 23:34:11 | adri | +17 -101 | Since we are pushing index management into the store dirs, we can |
| 2001/01/07 23:34:52 | adri | +10 -7 | After careful consideration I've discovered that yes, I *can* increment |
| 2001/01/07 23:22:47 | adri | +3 -2 | Actually free the data correctly once its been passed off to the network. |
| 2001/01/06 14:45:16 | adri | +71 -14 | Split up some of the client request handling to support storeLookup(). |
| 2001/01/06 14:35:57 | adri | +13 -3 | More lobotomies - temporary ones this time, until I replace them with |
| 2001/01/06 13:11:02 | adri | +21 -11 | Split up clientPurgeRequest() into two functions - clientStartPurgeRequest() |
| 2001/01/06 13:09:52 | adri | +3 -104 | Lobotomise more UFS code out of the hash fs |
| 2001/01/05 15:33:10 | adri | +88 -46 | * Pass a new field to storeCreateEntry - the reply_obj_t. This tells us |
| 2001/01/05 14:28:21 | adri | +69 -1036 | The initial skeletal implementation of storeLookup() - the async |
| 2001/01/05 12:27:48 | adri | +3 -5 | Merge another storeKeyPublic() / storeGet() sequence with storeGetPublic(). |
| 2001/01/05 12:16:51 | adri | +3 -4 | Merge a storeKeyPublic() / storeGet() sequence with storeGetPublic(). |
| 2001/01/04 16:29:40 | adri | +303 -3 | add the beginnings of a network reply object. Its the same as the internal |
| 2001/01/04 14:23:12 | adri | +48 -37 | Initial commit of my quest to kill seen_offset: |
| 2001/01/03 12:11:32 | adri | +3 -2 | Framework for killing seen_offset: |
| 2000/12/27 13:15:21 | adri | +50 -26 | * fix an unused variable whilst dumping storeentry information. |
| 2000/12/27 11:04:05 | adri | +97 -131 | Massive Change Time. |
| 2000/12/21 14:33:04 | adri | +4 -3 | * make storeUnlockObject() return -1 if we've deallocated the storeentry, |
| 2000/12/21 12:51:00 | adri | +13 -10 | Again, I was smoking crack. |
| 2000/12/20 15:43:32 | adri | +7 -1 | oops! |
| 2000/12/20 15:39:22 | adri | +50 -92 | * Always attach a MemObject to a StoreEntry, rather than only if |
| 2000/12/20 15:27:39 | adri | +2 -2 | oops, tag version with modio so we can easily tell whether people are |
| 2000/12/20 12:41:21 | adri | +44 -44 | rename storeClientListAdd() and storeUnregister() to |
| 2000/12/20 11:48:41 | adri | +71 -94 | remove all references to the store_client struct outside of the |
| 2000/12/19 15:27:28 | adri | +103 -196 | * Move the store client list into StoreEntry as a single client |
| 2000/12/19 12:53:04 | adri | +64 -130 | more tidyups: |
| 2000/12/18 22:14:49 | adri | +37 -59 | More stuff |
| 2000/12/18 19:10:58 | adri | +456 -246 | My first crack at the concept of a 'reply object'. |
| 2000/12/18 12:39:35 | adri | +29 -33 | I'm smoking crack. |
| 2000/12/18 12:29:26 | adri | +17 -7 | Kill more uses of storeMemHiOffset() |
| 2000/12/18 11:10:32 | adri | +16 -22 | Delete the uses of storeMemHiOffset in the asn code - |
| 2000/12/16 09:42:13 | adri | +2 -2 | assert(foo == 0) doesn't become assert(funcWrappingAroundFoo()); :-) |
| 2000/12/16 09:29:03 | adri | +2072 -0 | the 'hash' fs type will be used to implement the next level of FS |
| 2000/12/13 09:45:02 | rbcollins | +102 -83 | implemented examination functions for the mem_obj data items inmem_lo & hi. This is to allow future work to move these i |