more explicit convert; additonal tests; add GMP to docs

README.md

@@ -141,6 +141,7 @@ Julia uses the following external libraries, which are automatically downloaded
 - **[ARPACK]**              — a collection of subroutines designed to solve large, sparse eigenvalue problems.
 - **[FFTW]**                — library for computing fast Fourier transforms very quickly and efficiently.
 - **[PCRE]**                — Perl-compatible regular expressions library.
+- **[GMP]**                 - the GNU multiple precision arithmetic library, needed for bigint support
 - **[D3]**                  — JavaScript visualization library.
 
 [GNU make]:     http://www.gnu.org/software/make/
@@ -160,6 +161,7 @@ Julia uses the following external libraries, which are automatically downloaded
 [LLVM]:         http://www.llvm.org/
 [FemtoLisp]:    https://github.com/JeffBezanson/femtolisp
 [GNU readline]: http://cnswww.cns.cwru.edu/php/chet/readline/rltop.html
+[GMP]:          http://gmplib.org/
 [D3]:           http://mbostock.github.com/d3/
 
 <a name="Directories"/>

external/.gitignore

@@ -12,3 +12,4 @@
 /Clp-*
 /SuiteSparse-*
 /double-conversion-*
+/gmp-*

j/bigint.j

@@ -26,18 +26,21 @@ type BigInt <: Integer
 	end
 end
 
-#BigInt(x::Int64) = BigInt(string(x))
+convert(::Type{BigInt}, x::Int8) = BigInt(int(x))
+convert(::Type{BigInt}, x::Int16) = BigInt(int(x))
+convert(::Type{BigInt}, x::Int) = BigInt(x)
+
+macro define_bigint_convert ()
+	if WORD_SIZE == 64
+		:(convert(::Type{BigInt}, x::Int32) = BigInt(int(x)))
 
-#convert(::Type{BigInt}, x::Integer) = BigInt(string(x))
-function convert(::Type{BigInt}, x::Int64) 
-	if is(Int, Int64) 
-		BigInt(int(x))
 	else
-		BigInt(string(x))
+		:(convert(::Type{BigInt}, x::Int64) = BigInt(string(x)))
 	end
-
 end
 
+@define_bigint_convert
+
 promote_rule(::Type{BigInt}, ::Type{Int8}) = BigInt
 promote_rule(::Type{BigInt}, ::Type{Int16}) = BigInt
 promote_rule(::Type{BigInt}, ::Type{Int32}) = BigInt
@@ -83,11 +86,11 @@ function cmp(x::BigInt, y::BigInt)
 	ccall(dlsym(_jl_libgmp_wrapper, :_jl_mpz_cmp), Int, (Ptr{Void}, Ptr{Void}),x.mpz, y.mpz)
 end
 
-==(x::BigInt, y::BigInt)  = cmp(x,y) == 0 
+==(x::BigInt, y::BigInt) = cmp(x,y) == 0 
 <=(x::BigInt, y::BigInt) = cmp(x,y) <= 0 
->= (x::BigInt, y::BigInt) = cmp(x,y) >= 0 
-< (x::BigInt, y::BigInt) = cmp(x,y) < 0 
-> (x::BigInt, y::BigInt) = cmp(x,y) > 0 
+>=(x::BigInt, y::BigInt) = cmp(x,y) >= 0 
+<(x::BigInt, y::BigInt) = cmp(x,y) < 0 
+>(x::BigInt, y::BigInt) = cmp(x,y) > 0 
 
 function string(x::BigInt) 
 	s=ccall(dlsym(_jl_libgmp_wrapper, :_jl_mpz_printf), Ptr{Uint8}, (Ptr{Void},),x.mpz)

test/bigint.j

@@ -25,9 +25,9 @@ d = BigInt("-246913578024691357802469135780")
 @assert d == -c
 
 
-e=2^63-1
-@assert typeof(BigInt(e)) == BigInt
-@assert BigInt(e)+1 == BigInt("9223372036854775808")
+ee=2^63-1
+@assert typeof(BigInt(ee)) == BigInt
+@assert BigInt(ee)+1 == BigInt("9223372036854775808")
 
 #Multiple calls for sanity check, since we're doing direct memory manipulation
 @assert string(a) == "123456789012345678901234567890"
@@ -38,3 +38,10 @@ e=2^63-1
 
 @assert div(BigInt(3), BigInt(2)) == BigInt(1)
 @assert rem(BigInt(3), BigInt(2)) == BigInt(1)
+
+@assert a+int(1) == b
+@assert a+int8(1) == b
+@assert a+int16(1) == b
+@assert a+int32(1) == b
+@assert a+int64(1) == b
+