Kill long dead docs

Docs/Design/RKCache.txt

@@ -1,92 +0,0 @@
-RestKit Cache Design
-====================
-
-This document outlines the design for a lightweight cache implementation for the RestKit framework. The goals
-of the cache are:
-* Provide support for offline access to data payloads
-* Provide storage of last modification times for etags
-* Provide cache policies at the client and request level
-* RKResponse should be able to answer - (BOOL)wasLoadedFromCache
-* Each RKClient has its own cache (path initialized by Base URL somehow?)
-* RKCache should have a storage policy. This should be settable to session (store for the run of the app) or permanently.
-
-Needs to store:
-* Key/values
-* Settable via NSURL (not resource path)
-
-Concerns:
-* Does this need to be harmonized with the managed object cache? Can it be?
-* Don't love how tightly coupled cacheing is in Three20. Want something more orthogonal.
-* How are we going to unit test this? Need to identify test cases...
-* Where does the cache loading logic go? Request queue? Client? Inside of RKRequest? New class?
-
-References:
-* TTURLCache
-* ASIDownloadCache
-
-ETags:
-* http://www.w3.org/Protocols/rfc2616/rfc2616-sec14.html
-* Comes across as either "ETag" or "Etag" header
-* Should be contained to RKRequest / RKResponse (don't bleed through into the cache)
-* Needs to be added on load via If-None-Match
-
-Request cache policies:
-* RKClient should have a default cache policy that gets passed through to requests
-* RKRequest will have a cache policy property. This determines when and how to utilize the cache.
-* You should be able to bitwise | these together, i.e. RKRequestCachePolicyLoadIfOffline | RKRequestCachePolicyEtag
-typedef enum {
-    RKRequestCachePolicyNone = 0,  // Never use the cache
-    RKRequestCachePolicyLoadIfOffline, // Load from the cache when we are offline
-    RKRequestCachePolicyLoadOnError, // Load from the cache if we encounter an error
-    RKRequestCachePolicyEtag, // Load from the cache if we have data stored and the server returns a 304 (not modified) response
-    RKRequestCachePolicyDefault = RKRequestCachePolicyEtag;
-} RKRequestCachePolicy;
-
-###################################################################################
-// Proposed Interface
-
-// Returns a cache key for a URL (md5 the full URL after coercing to a string)
-// This will let you work with the cache via URL's pretty easily
-- (NSString*)RKCacheKeyForURL:(NSURL*)URL;
-
-// Storage policy. Determines if we clear the cache out when the app is shut down. Cache
-// instance needs to register for 
-typedef enum {
-    RKCacheStoragePolicyDisabled, // The cache has been disabled. Attempts to store data will silently fail
-    RKCacheStoragePolicyForDurationOfSession, // Cache data for the length of the session. Clear cache at app exit.
-    RKCacheStoragePolicyPermanently // Cache data permanently, until explicitly expired or flushed
-} RKCacheStoragePolicy;
-
-@interface RKCache : NSObject {
-    NSString* _cachePath;
-    RKCacheStoragePolicy _storagePolicy;
-}
-
-@property (nonatomic, readonly) NSString* cachePath; // Full path to the cache
-@property (nonatomic, assign) RKCacheStoragePolicy storagePolicy; // User can change storage policy.
-
-// Should be initialized with the full path to store it. RKClient should
-// figure out the appropriate path when initializing.
-- (id)initWithCachePath:(NSString*)cachePath storagePolicy:(RKCacheStoragePolicy)storagePolicy;
-
-// Key/value storage
-- (NSString*)pathForKey:(NSString*)key;
-- (BOOL)hasDataForKey:(NSString*)key;
-- (void)storeData:(NSData*)data forKey:(NSString*)key;
-- (NSData*)dataForKey:(NSString*)key;
-- (NSData*)dataForKey:(NSString*)key expires:(NSTimeInterval)expirationAge
-           timestamp:(NSDate**)timestamp;
-
-// Cache Invalidation
-- (void)invalidateKey:(NSString*)key;
-- (void)invalidateAll;
-
-@end
-
-// RKRequest
-Get new properties:
-NSString* cacheKey; // When GET, defaults to RKCacheKeyForURL. For POST/PUT, md5 of the serialization data.
-RKRequestCachePolicy cachePolicy; // Cache policy inherited from RKClient.
-
-// RKResponse
-BOOL wasLoadedFromCache; // YES when the response was built from cached data

Docs/Design/RKRequestQueue.txt

@@ -1,25 +0,0 @@
-This document details an extension to the RKRequest queue to make it more general purpose:
-
-1) RestKit currently implements a hard limit on 5 concurrent requests on a queue basis. This should be exposed as a property called concurrentRequestsLimit
-2) The sendRequest: method is poorly named. It should probably be called addRequest: or enqueueRequest:
-3) You'll probably want delegate methods to keep track of what's going on in the Queue. I've sketched them below.
-4) We may want to make freshly alloc'd queues start in the suspended state.
-
-RKRequestQueue* queue = [[RKRequestQueue alloc] init];
-queue.concurrentRequestsLimit = 1;
-queue.delegate = self;
-RKObjectLoader* loader = [[RKObjectManager sharedManager] objectLoaderWithResourcePath:@"/whatever" delegate:self];
-[queue addRequest:loader];
-RKRequest* request = [[RKClient sharedClient] requestWithResourcePath:@"/another" delegate:self];
-[queue addRequest:request];
-[queue start];
-
-// Delegate methods
-@optional
-- (void)requestQueueDidStart:(RKRequestQueue*)queue; // Sent when queue starts running
-- (void)requestQueueDidFinish:(RKRequestQueue*)queue; // Sent when its emptied
-- (void)requestQueue:(RKRequestQueue*)queue willSendRequest:(RKRequest*)request;
-- (void)requestQueue:(RKRequestQueue*)queue didSendRequest:(RKRequest*)request;
-- (void)requestQueue:(RKRequestQueue*)queue didLoadResponse:(RKResponse*)response;
-- (void)requestQueue:(RKRequestQueue*)queue didCancelRequest:(RKRequest*)request;
-- (void)requestQueue:(RKRequestQueue*)queue didFailRequest:(RKRequest*)request withError:(NSError*)error;