
Pervasives(3) OCaml library Pervasives(3)
NAME
Pervasives  The initially opened module.
Module
Module Pervasives
Documentation
Module Pervasives
: sig end
The initially opened module.
This module provides the basic operations over the builtin types (numbers, booleans,
strings, exceptions, references, lists, arrays, inputoutput channels, ...).
This module is automatically opened at the beginning of each compilation. All components
of this module can therefore be referred by their short name, without prefixing them by
Pervasives .
=== Exceptions ===
val raise : exn > 'a
Raise the given exception value
val invalid_arg : string > 'a
Raise exception Invalid_argument with the given string.
val failwith : string > 'a
Raise exception Failure with the given string.
exception Exit
The Exit exception is not raised by any library function. It is provided for use in your
programs.
=== Comparisons ===
val (=) : 'a > 'a > bool
e1 = e2 tests for structural equality of e1 and e2 . Mutable structures (e.g. references
and arrays) are equal if and only if their current contents are structurally equal, even
if the two mutable objects are not the same physical object. Equality between functional
values raises Invalid_argument . Equality between cyclic data structures may not termi
nate.
val (<>) : 'a > 'a > bool
Negation of Pervasives.(=) .
val (<) : 'a > 'a > bool
See Pervasives.(>=) .
val (>) : 'a > 'a > bool
See Pervasives.(>=) .
val (<=) : 'a > 'a > bool
See Pervasives.(>=) .
val (>=) : 'a > 'a > bool
Structural ordering functions. These functions coincide with the usual orderings over
integers, characters, strings and floatingpoint numbers, and extend them to a total
ordering over all types. The ordering is compatible with ( = ) . As in the case of ( = )
, mutable structures are compared by contents. Comparison between functional values
raises Invalid_argument . Comparison between cyclic structures may not terminate.
val compare : 'a > 'a > int
compare x y returns 0 if x is equal to y , a negative integer if x is less than y , and a
positive integer if x is greater than y . The ordering implemented by compare is compati
ble with the comparison predicates = , < and > defined above, with one difference on the
treatment of the float value Pervasives.nan . Namely, the comparison predicates treat nan
as different from any other float value, including itself; while compare treats nan as
equal to itself and less than any other float value. This treatment of nan ensures that
compare defines a total ordering relation.
compare applied to functional values may raise Invalid_argument . compare applied to
cyclic structures may not terminate.
The compare function can be used as the comparison function required by the Set.Make and
Map.Make functors, as well as the List.sort and Array.sort functions.
val min : 'a > 'a > 'a
Return the smaller of the two arguments. The result is unspecified if one of the argu
ments contains the float value nan .
val max : 'a > 'a > 'a
Return the greater of the two arguments. The result is unspecified if one of the argu
ments contains the float value nan .
val (==) : 'a > 'a > bool
e1 == e2 tests for physical equality of e1 and e2 . On mutable types such as references,
arrays, strings, records with mutable fields and objects with mutable instance variables,
e1 == e2 is true if and only if physical modification of e1 also affects e2 . On
nonmutable types, the behavior of ( == ) is implementationdependent; however, it is
guaranteed that e1 == e2 implies compare e1 e2 = 0 .
val (!=) : 'a > 'a > bool
Negation of Pervasives.(==) .
=== Boolean operations ===
val not : bool > bool
The boolean negation.
val (&&) : bool > bool > bool
The boolean ``and''. Evaluation is sequential, lefttoright: in e1 && e2 , e1 is evalu
ated first, and if it returns false , e2 is not evaluated at all.
val (&) : bool > bool > bool
Deprecated.
Pervasives.(&&) should be used instead.
val () : bool > bool > bool
The boolean ``or''. Evaluation is sequential, lefttoright: in e1  e2 , e1 is evaluated
first, and if it returns true , e2 is not evaluated at all.
val (or) : bool > bool > bool
Deprecated.
Pervasives.() should be used instead.
=== Integer arithmetic ===
=== Integers are 31 bits wide (or 63 bits on 64bit processors). All operations are taken
modulo 2^{31} (or 2^{63}). They do not fail on overflow. ===
val (~) : int > int
Unary negation. You can also write  e instead of ~ e .
val (~+) : int > int
Unary addition. You can also write + e instead of ~+ e .
Since 3.12.0
val succ : int > int
succ x is x + 1 .
val pred : int > int
pred x is x  1 .
val (+) : int > int > int
Integer addition.
val () : int > int > int
Integer subtraction.
val ( * ) : int > int > int
Integer multiplication.
val (/) : int > int > int
Integer division. Raise Division_by_zero if the second argument is 0. Integer division
rounds the real quotient of its arguments towards zero. More precisely, if x >= 0 and y >
0 , x / y is the greatest integer less than or equal to the real quotient of x by y .
Moreover, ( x) / y = x / ( y) =  (x / y) .
val (mod) : int > int > int
Integer remainder. If y is not zero, the result of x mod y satisfies the following prop
erties: x = (x / y) * y + x mod y and abs(x mod y) <= abs(y)  1 . If y = 0 , x mod y
raises Division_by_zero . Note that x mod y is negative only if x < 0 . Raise Divi
sion_by_zero if y is zero.
val abs : int > int
Return the absolute value of the argument. Note that this may be negative if the argument
is min_int .
val max_int : int
The greatest representable integer.
val min_int : int
The smallest representable integer.
=== Bitwise operations ===
val (land) : int > int > int
Bitwise logical and.
val (lor) : int > int > int
Bitwise logical or.
val (lxor) : int > int > int
Bitwise logical exclusive or.
val lnot : int > int
Bitwise logical negation.
val (lsl) : int > int > int
n lsl m shifts n to the left by m bits. The result is unspecified if m < 0 or m >= bit
size , where bitsize is 32 on a 32bit platform and 64 on a 64bit platform.
val (lsr) : int > int > int
n lsr m shifts n to the right by m bits. This is a logical shift: zeroes are inserted
regardless of the sign of n . The result is unspecified if m < 0 or m >= bitsize .
val (asr) : int > int > int
n asr m shifts n to the right by m bits. This is an arithmetic shift: the sign bit of n
is replicated. The result is unspecified if m < 0 or m >= bitsize .
=== Floatingpoint arithmetic OCaml's floatingpoint numbers follow the IEEE 754 standard,
using double precision (64 bits) numbers. Floatingpoint operations never raise an excep
tion on overflow, underflow, division by zero, etc. Instead, special IEEE numbers are
returned as appropriate, such as infinity for 1.0 /. 0.0, neg_infinity for 1.0 /. 0.0,
and nan (``not a number'') for 0.0 /. 0.0. These special numbers then propagate through
floatingpoint computations as expected: for instance, 1.0 /. infinity is 0.0, and any
arithmetic operation with nan as argument returns nan as result. ===
val (~.) : float > float
Unary negation. You can also write . e instead of ~. e .
val (~+.) : float > float
Unary addition. You can also write +. e instead of ~+. e .
Since 3.12.0
val (+.) : float > float > float
Floatingpoint addition
val (.) : float > float > float
Floatingpoint subtraction
val ( *. ) : float > float > float
Floatingpoint multiplication
val (/.) : float > float > float
Floatingpoint division.
val ( ** ) : float > float > float
Exponentiation.
val sqrt : float > float
Square root.
val exp : float > float
Exponential.
val log : float > float
Natural logarithm.
val log10 : float > float
Base 10 logarithm.
val expm1 : float > float
expm1 x computes exp x . 1.0 , giving numericallyaccurate results even if x is close to
0.0 .
Since 3.12.0
val log1p : float > float
log1p x computes log(1.0 +. x) (natural logarithm), giving numericallyaccurate results
even if x is close to 0.0 .
Since 3.12.0
val cos : float > float
Cosine. Argument is in radians.
val sin : float > float
Sine. Argument is in radians.
val tan : float > float
Tangent. Argument is in radians.
val acos : float > float
Arc cosine. The argument must fall within the range [1.0, 1.0] . Result is in radians
and is between 0.0 and pi .
val asin : float > float
Arc sine. The argument must fall within the range [1.0, 1.0] . Result is in radians and
is between pi/2 and pi/2 .
val atan : float > float
Arc tangent. Result is in radians and is between pi/2 and pi/2 .
val atan2 : float > float > float
atan2 y x returns the arc tangent of y /. x . The signs of x and y are used to determine
the quadrant of the result. Result is in radians and is between pi and pi .
val hypot : float > float > float
hypot x y returns sqrt(x *. x + y *. y) , that is, the length of the hypotenuse of a
rightangled triangle with sides of length x and y , or, equivalently, the distance of the
point (x,y) to origin.
Since 4.00.0
val cosh : float > float
Hyperbolic cosine. Argument is in radians.
val sinh : float > float
Hyperbolic sine. Argument is in radians.
val tanh : float > float
Hyperbolic tangent. Argument is in radians.
val ceil : float > float
Round above to an integer value. ceil f returns the least integer value greater than or
equal to f . The result is returned as a float.
val floor : float > float
Round below to an integer value. floor f returns the greatest integer value less than or
equal to f . The result is returned as a float.
val abs_float : float > float
abs_float f returns the absolute value of f .
val copysign : float > float > float
copysign x y returns a float whose absolute value is that of x and whose sign is that of y
. If x is nan , returns nan . If y is nan , returns either x or . x , but it is not
specified which.
Since 4.00.0
val mod_float : float > float > float
mod_float a b returns the remainder of a with respect to b . The returned value is a . n
*. b , where n is the quotient a /. b rounded towards zero to an integer.
val frexp : float > float * int
frexp f returns the pair of the significant and the exponent of f . When f is zero, the
significant x and the exponent n of f are equal to zero. When f is nonzero, they are
defined by f = x *. 2 ** n and 0.5 <= x < 1.0 .
val ldexp : float > int > float
ldexp x n returns x *. 2 ** n .
val modf : float > float * float
modf f returns the pair of the fractional and integral part of f .
val float : int > float
Same as Pervasives.float_of_int .
val float_of_int : int > float
Convert an integer to floatingpoint.
val truncate : float > int
Same as Pervasives.int_of_float .
val int_of_float : float > int
Truncate the given floatingpoint number to an integer. The result is unspecified if the
argument is nan or falls outside the range of representable integers.
val infinity : float
Positive infinity.
val neg_infinity : float
Negative infinity.
val nan : float
A special floatingpoint value denoting the result of an undefined operation such as 0.0
/. 0.0 . Stands for ``not a number''. Any floatingpoint operation with nan as argument
returns nan as result. As for floatingpoint comparisons, = , < , <= , > and >= return
false and <> returns true if one or both of their arguments is nan .
val max_float : float
The largest positive finite value of type float .
val min_float : float
The smallest positive, nonzero, nondenormalized value of type float .
val epsilon_float : float
The difference between 1.0 and the smallest exactly representable floatingpoint number
greater than 1.0 .
type fpclass =
 FP_normal (* Normal number, none of the below *)
 FP_subnormal (* Number very close to 0.0, has reduced precision *)
 FP_zero (* Number is 0.0 or 0.0 *)
 FP_infinite (* Number is positive or negative infinity *)
 FP_nan (* Not a number: result of an undefined operation *)
The five classes of floatingpoint numbers, as determined by the Pervasives.classify_float
function.
val classify_float : float > fpclass
Return the class of the given floatingpoint number: normal, subnormal, zero, infinite, or
not a number.
=== String operations More string operations are provided in module String. ===
val (^) : string > string > string
String concatenation.
=== Character operations More character operations are provided in module Char. ===
val int_of_char : char > int
Return the ASCII code of the argument.
val char_of_int : int > char
Return the character with the given ASCII code. Raise Invalid_argument char_of_int if the
argument is outside the range 0255.
=== Unit operations ===
val ignore : 'a > unit
Discard the value of its argument and return () . For instance, ignore(f x) discards the
result of the sideeffecting function f . It is equivalent to f x; () , except that the
latter may generate a compiler warning; writing ignore(f x) instead avoids the warning.
=== String conversion functions ===
val string_of_bool : bool > string
Return the string representation of a boolean. As the returned values may be shared, the
user should not modify them directly.
val bool_of_string : string > bool
Convert the given string to a boolean. Raise Invalid_argument bool_of_string if the
string is not true or false .
val string_of_int : int > string
Return the string representation of an integer, in decimal.
val int_of_string : string > int
Convert the given string to an integer. The string is read in decimal (by default) or in
hexadecimal (if it begins with 0x or 0X ), octal (if it begins with 0o or 0O ), or binary
(if it begins with 0b or 0B ). Raise Failure int_of_string if the given string is not a
valid representation of an integer, or if the integer represented exceeds the range of
integers representable in type int .
val string_of_float : float > string
Return the string representation of a floatingpoint number.
val float_of_string : string > float
Convert the given string to a float. Raise Failure float_of_string if the given string is
not a valid representation of a float.
=== Pair operations ===
val fst : 'a * 'b > 'a
Return the first component of a pair.
val snd : 'a * 'b > 'b
Return the second component of a pair.
=== List operations More list operations are provided in module List. ===
val (@) : 'a list > 'a list > 'a list
List concatenation.
=== Input/output Note: all input/output functions can raise Sys_error when the system
calls they invoke fail. ===
type in_channel
The type of input channel.
type out_channel
The type of output channel.
val stdin : in_channel
The standard input for the process.
val stdout : out_channel
The standard output for the process.
val stderr : out_channel
The standard error output for the process.
=== Output functions on standard output ===
val print_char : char > unit
Print a character on standard output.
val print_string : string > unit
Print a string on standard output.
val print_int : int > unit
Print an integer, in decimal, on standard output.
val print_float : float > unit
Print a floatingpoint number, in decimal, on standard output.
val print_endline : string > unit
Print a string, followed by a newline character, on standard output and flush standard
output.
val print_newline : unit > unit
Print a newline character on standard output, and flush standard output. This can be used
to simulate line buffering of standard output.
=== Output functions on standard error ===
val prerr_char : char > unit
Print a character on standard error.
val prerr_string : string > unit
Print a string on standard error.
val prerr_int : int > unit
Print an integer, in decimal, on standard error.
val prerr_float : float > unit
Print a floatingpoint number, in decimal, on standard error.
val prerr_endline : string > unit
Print a string, followed by a newline character on standard error and flush standard
error.
val prerr_newline : unit > unit
Print a newline character on standard error, and flush standard error.
=== Input functions on standard input ===
val read_line : unit > string
Flush standard output, then read characters from standard input until a newline character
is encountered. Return the string of all characters read, without the newline character at
the end.
val read_int : unit > int
Flush standard output, then read one line from standard input and convert it to an inte
ger. Raise Failure int_of_string if the line read is not a valid representation of an
integer.
val read_float : unit > float
Flush standard output, then read one line from standard input and convert it to a float
ingpoint number. The result is unspecified if the line read is not a valid representa
tion of a floatingpoint number.
=== General output functions ===
type open_flag =
 Open_rdonly (* open for reading. *)
 Open_wronly (* open for writing. *)
 Open_append (* open for appending: always write at end of file. *)
 Open_creat (* create the file if it does not exist. *)
 Open_trunc (* empty the file if it already exists. *)
 Open_excl (* fail if Open_creat and the file already exists. *)
 Open_binary (* open in binary mode (no conversion). *)
 Open_text (* open in text mode (may perform conversions). *)
 Open_nonblock (* open in nonblocking mode. *)
Opening modes for Pervasives.open_out_gen and Pervasives.open_in_gen .
val open_out : string > out_channel
Open the named file for writing, and return a new output channel on that file, positionned
at the beginning of the file. The file is truncated to zero length if it already exists.
It is created if it does not already exists. Raise Sys_error if the file could not be
opened.
val open_out_bin : string > out_channel
Same as Pervasives.open_out , but the file is opened in binary mode, so that no transla
tion takes place during writes. On operating systems that do not distinguish between text
mode and binary mode, this function behaves like Pervasives.open_out .
val open_out_gen : open_flag list > int > string > out_channel
open_out_gen mode perm filename opens the named file for writing, as described above. The
extra argument mode specify the opening mode. The extra argument perm specifies the file
permissions, in case the file must be created. Pervasives.open_out and Perva
sives.open_out_bin are special cases of this function.
val flush : out_channel > unit
Flush the buffer associated with the given output channel, performing all pending writes
on that channel. Interactive programs must be careful about flushing standard output and
standard error at the right time.
val flush_all : unit > unit
Flush all open output channels; ignore errors.
val output_char : out_channel > char > unit
Write the character on the given output channel.
val output_string : out_channel > string > unit
Write the string on the given output channel.
val output : out_channel > string > int > int > unit
output oc buf pos len writes len characters from string buf , starting at offset pos , to
the given output channel oc . Raise Invalid_argument output if pos and len do not desig
nate a valid substring of buf .
val output_byte : out_channel > int > unit
Write one 8bit integer (as the single character with that code) on the given output chan
nel. The given integer is taken modulo 256.
val output_binary_int : out_channel > int > unit
Write one integer in binary format (4 bytes, bigendian) on the given output channel. The
given integer is taken modulo 2^{32. The only reliable way to read it back is through the
Pervasives.input_binary_int function. The format is compatible across all machines for a
given version of OCaml.
val output_value : out_channel > 'a > unit
Write the representation of a structured value of any type to a channel. Circularities and
sharing inside the value are detected and preserved. The object can be read back, by the
function Pervasives.input_value . See the description of module Marshal for more informa
tion. Pervasives.output_value is equivalent to Marshal.to_channel with an empty list of
flags.
val seek_out : out_channel > int > unit
seek_out chan pos sets the current writing position to pos for channel chan . This works
only for regular files. On files of other kinds (such as terminals, pipes and sockets),
the behavior is unspecified.
val pos_out : out_channel > int
Return the current writing position for the given channel. Does not work on channels
opened with the Open_append flag (returns unspecified results).
val out_channel_length : out_channel > int
Return the size (number of characters) of the regular file on which the given channel is
opened. If the channel is opened on a file that is not a regular file, the result is
meaningless.
val close_out : out_channel > unit
Close the given channel, flushing all buffered write operations. Output functions raise a
Sys_error exception when they are applied to a closed output channel, except close_out and
flush , which do nothing when applied to an already closed channel. Note that close_out
may raise Sys_error if the operating system signals an error when flushing or closing.
val close_out_noerr : out_channel > unit
Same as close_out , but ignore all errors.
val set_binary_mode_out : out_channel > bool > unit
set_binary_mode_out oc true sets the channel oc to binary mode: no translations take place
during output. set_binary_mode_out oc false sets the channel oc to text mode: depending
on the operating system, some translations may take place during output. For instance,
under Windows, endoflines will be translated from \n to \r\n . This function has no
effect under operating systems that do not distinguish between text mode and binary mode.
=== General input functions ===
val open_in : string > in_channel
Open the named file for reading, and return a new input channel on that file, positionned
at the beginning of the file. Raise Sys_error if the file could not be opened.
val open_in_bin : string > in_channel
Same as Pervasives.open_in , but the file is opened in binary mode, so that no translation
takes place during reads. On operating systems that do not distinguish between text mode
and binary mode, this function behaves like Pervasives.open_in .
val open_in_gen : open_flag list > int > string > in_channel
open_in_gen mode perm filename opens the named file for reading, as described above. The
extra arguments mode and perm specify the opening mode and file permissions. Perva
sives.open_in and Pervasives.open_in_bin are special cases of this function.
val input_char : in_channel > char
Read one character from the given input channel. Raise End_of_file if there are no more
characters to read.
val input_line : in_channel > string
Read characters from the given input channel, until a newline character is encountered.
Return the string of all characters read, without the newline character at the end. Raise
End_of_file if the end of the file is reached at the beginning of line.
val input : in_channel > string > int > int > int
input ic buf pos len reads up to len characters from the given channel ic , storing them
in string buf , starting at character number pos . It returns the actual number of char
acters read, between 0 and len (inclusive). A return value of 0 means that the end of
file was reached. A return value between 0 and len exclusive means that not all requested
len characters were read, either because no more characters were available at that time,
or because the implementation found it convenient to do a partial read; input must be
called again to read the remaining characters, if desired. (See also Perva
sives.really_input for reading exactly len characters.) Exception Invalid_argument input
is raised if pos and len do not designate a valid substring of buf .
val really_input : in_channel > string > int > int > unit
really_input ic buf pos len reads len characters from channel ic , storing them in string
buf , starting at character number pos . Raise End_of_file if the end of file is reached
before len characters have been read. Raise Invalid_argument really_input if pos and len
do not designate a valid substring of buf .
val input_byte : in_channel > int
Same as Pervasives.input_char , but return the 8bit integer representing the character.
Raise End_of_file if an end of file was reached.
val input_binary_int : in_channel > int
Read an integer encoded in binary format (4 bytes, bigendian) from the given input chan
nel. See Pervasives.output_binary_int . Raise End_of_file if an end of file was reached
while reading the integer.
val input_value : in_channel > 'a
Read the representation of a structured value, as produced by Pervasives.output_value ,
and return the corresponding value. This function is identical to Marshal.from_channel ;
see the description of module Marshal for more information, in particular concerning the
lack of type safety.
val seek_in : in_channel > int > unit
seek_in chan pos sets the current reading position to pos for channel chan . This works
only for regular files. On files of other kinds, the behavior is unspecified.
val pos_in : in_channel > int
Return the current reading position for the given channel.
val in_channel_length : in_channel > int
Return the size (number of characters) of the regular file on which the given channel is
opened. If the channel is opened on a file that is not a regular file, the result is
meaningless. The returned size does not take into account the endofline translations
that can be performed when reading from a channel opened in text mode.
val close_in : in_channel > unit
Close the given channel. Input functions raise a Sys_error exception when they are
applied to a closed input channel, except close_in , which does nothing when applied to an
already closed channel. Note that close_in may raise Sys_error if the operating system
signals an error.
val close_in_noerr : in_channel > unit
Same as close_in , but ignore all errors.
val set_binary_mode_in : in_channel > bool > unit
set_binary_mode_in ic true sets the channel ic to binary mode: no translations take place
during input. set_binary_mode_out ic false sets the channel ic to text mode: depending on
the operating system, some translations may take place during input. For instance, under
Windows, endoflines will be translated from \r\n to \n . This function has no effect
under operating systems that do not distinguish between text mode and binary mode.
=== Operations on large files ===
module LargeFile : sig end
Operations on large files. This submodule provides 64bit variants of the channel func
tions that manipulate file positions and file sizes. By representing positions and sizes
by 64bit integers (type int64 ) instead of regular integers (type int ), these alternate
functions allow operating on files whose sizes are greater than max_int .
=== References ===
type 'a ref = {
mutable contents : 'a ;
}
The type of references (mutable indirection cells) containing a value of type 'a .
val ref : 'a > 'a ref
Return a fresh reference containing the given value.
val (!) : 'a ref > 'a
!r returns the current contents of reference r . Equivalent to fun r > r.contents .
val (:=) : 'a ref > 'a > unit
r := a stores the value of a in reference r . Equivalent to fun r v > r.contents < v .
val incr : int ref > unit
Increment the integer contained in the given reference. Equivalent to fun r > r := succ
!r .
val decr : int ref > unit
Decrement the integer contained in the given reference. Equivalent to fun r > r := pred
!r .
=== Operations on format strings ===
=== Format strings are used to read and print data using formatted input functions in mod
ule Scanf and formatted output in modules Printf and Format. ===
type ('a, 'b, 'c, 'd) format4 = ('a, 'b, 'c, 'c, 'c, 'd) format6
Format strings have a general and highly polymorphic type ('a, 'b, 'c, 'd, 'e, 'f) format6
. Type format6 is built in. The two simplified types, format and format4 below are
included for backward compatibility with earlier releases of OCaml. 'a is the type of the
parameters of the format, 'b is the type of the first argument given to %a and %t printing
functions, 'c is the type of the result of the %a and %t functions, and also the type of
the argument transmitted to the first argument of kprintf style functions, 'd is the
result type for the scanf style functions, 'e is the type of the receiver function for
the scanf style functions, 'f is the result type for the printf style function.
type ('a, 'b, 'c) format = ('a, 'b, 'c, 'c) format4
val string_of_format : ('a, 'b, 'c, 'd, 'e, 'f) format6 > string
Converts a format string into a string.
val format_of_string : ('a, 'b, 'c, 'd, 'e, 'f) format6 > ('a, 'b, 'c, 'd, 'e, 'f) for
mat6
format_of_string s returns a format string read from the string literal s .
val (^^) : ('a, 'b, 'c, 'd, 'e, 'f) format6 > ('f, 'b, 'c, 'e, 'g, 'h) format6 > ('a,
'b, 'c, 'd, 'g, 'h) format6
f1 ^^ f2 catenates formats f1 and f2 . The result is a format that accepts arguments from
f1 , then arguments from f2 .
=== Program termination ===
val exit : int > 'a
Terminate the process, returning the given status code to the operating system: usually 0
to indicate no errors, and a small positive integer to indicate failure. All open output
channels are flushed with flush_all . An implicit exit 0 is performed each time a program
terminates normally. An implicit exit 2 is performed if the program terminates early
because of an uncaught exception.
val at_exit : (unit > unit) > unit
Register the given function to be called at program termination time. The functions regis
tered with at_exit will be called when the program executes Pervasives.exit , or termi
nates, either normally or because of an uncaught exception. The functions are called in
``last in, first out'' order: the function most recently added with at_exit is called
first.
OCamldoc 20140609 Pervasives(3) 
