Moved unit tests back into main directory

Makefile

@@ -1,27 +1,24 @@
-SRC = .
-TEST = ./test
-
 all: test      \
 	 benchmark \
      coverage
 
 format: 
-	gofmt -w=true ${SRC}/*.go
-	gofmt -w=true ${TEST}/*.go
+	gofmt -w=true *.go
+	gofmt -w=true *.go
 
 build: format
-	go build ${SRC}
+	go build
 
 test: build
-	go test ${TEST}
+	go test
 
 benchmark: build
-	go test ${TEST} -bench .
+	go test -bench .
 
 coverage: build
-	go test ${TEST} -cover .
+	go test -cover .
 
 debug: build
-	gofmt -w=true ${SRC}
-	go test ${TEST} -run TestCryptoHashBlocks
+	gofmt -w=true *.go
+	go test -run TestCryptoHashBlocks
 

doc.go

@@ -1,49 +1,13 @@
 /*
-tweetnacl-go is a port of Dan Bernstein's "crypto library in a 100 tweets" code to the Go language. 
+tweetnacl-go is a port of Dan Bernstein's "crypto library in a 100 tweets" code to the Go language.
 It is implemented as a wrapper around the original code to preserve the design and timing
 characteristics of the original implementation.
 
-Version Info
-------------
-tweetnacl: 20140427
-
-tweetnacl-go: 0.00.0
-
-Disclaimer
-----------
 The Go wrapper has been kept as 'thin' as possible to avoid compromising the careful design
-and coding of the original TweetNaCl implementation. However, cryptography being what it is, 
-the wrapper may have (entirely inadvertently) introduced non-obvious vulnerabilities (for 
-instance [How to zero a buffer][daemonology]). So ....
-
-**USE ENTIRELY AT YOUR OWN RISK !**
-
-Notes
------
-1. There is an existing port of TweetNaCl to Go ([go-tweetnacl][go-tweetnacl]) which reimplements the C code
-   in Go.
-2. The cgo FFI doesn't handle zero length byte arrays particularly elegantly (see discussion at
-   [Passing zero length Go byte array to a C function][golang-nuts]). The conversion to a (\*C.uchar) pointer
-   has been abstracted into a convenience function _makePtr_, which includes a commented out version
-   that returns nil for a zero length array if you prefer not to use the esoteric 'unsafe' version.
-
-
-References
-----------
-
-1.  http://tweetnacl.cr.yp.to
-2.  http://tweetnacl.cr.yp.to/tweetnacl-20131229.pdf
-3.  http://cr.yp.to/highspeed/naclcrypto-20090310.pdf
-4.  http://blog.skylable.com/2014/05/tweetnacl-carrybit-bug
-5.  http://cryptojedi.org/peter/data/tenerife-20130121.pdf
-6.  http://coderinaworldofcode.blogspot.com/2014/03/on-nacl.html
-7.  http://blog.regehr.org/archives/1063
-8.  http://www.daemonology.net/blog/2014-09-04-how-to-zero-a-buffer.html
-9.  http://www.daemonology.net/blog/2014-09-06-zeroing-buffers-is-insufficient.html
-10. http://www.daemonology.net/blog/2014-09-05-erratum.html
-11. https://github.com/keybase/go-tweetnacl
-12. http://www.metzdowd.com/pipermail/cryptography/2016-March/028824.html
-13. https://groups.google.com/forum/#!topic/golang-nuts/NNBdjztWquo
+and coding of the original TweetNaCl implementation. However, cryptography being what it is,
+the wrapper may have (entirely inadvertently) introduced non-obvious vulnerabilities (for
+instance.
 
+http://tweetnacl.cr.yp.to
 */
 package tweetnacl