3 SPDX-License-Identifier: BSD-2-Clause
5 Copyright (c) 2018-2021 Gavin D. Howard and contributors.
7 Redistribution and use in source and binary forms, with or without
8 modification, are permitted provided that the following conditions are met:
10 * Redistributions of source code must retain the above copyright notice, this
11 list of conditions and the following disclaimer.
13 * Redistributions in binary form must reproduce the above copyright notice,
14 this list of conditions and the following disclaimer in the documentation
15 and/or other materials provided with the distribution.
17 THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
18 AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
19 IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
20 ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE
21 LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
22 CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
23 SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
24 INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
25 CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
26 ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
27 POSSIBILITY OF SUCH DAMAGE.
33 bc - arbitrary-precision decimal arithmetic language and calculator
37 **bc** [**-ghilPqRsvVw**] [**-\-global-stacks**] [**-\-help**] [**-\-interactive**] [**-\-mathlib**] [**-\-no-prompt**] [**-\-no-read-prompt**] [**-\-quiet**] [**-\-standard**] [**-\-warn**] [**-\-version**] [**-e** *expr*] [**-\-expression**=*expr*...] [**-f** *file*...] [**-\-file**=*file*...] [*file*...]
41 bc(1) is an interactive processor for a language first standardized in 1991 by
42 POSIX. (The current standard is [here][1].) The language provides unlimited
43 precision decimal arithmetic and is somewhat C-like, but there are differences.
44 Such differences will be noted in this document.
46 After parsing and handling options, this bc(1) reads any files given on the
47 command line and executes them before reading from **stdin**.
49 This bc(1) is a drop-in replacement for *any* bc(1), including (and
50 especially) the GNU bc(1).
54 The following are the options that bc(1) accepts.
56 **-g**, **-\-global-stacks**
58 Turns the globals **ibase**, **obase**, and **scale** into stacks.
60 This has the effect that a copy of the current value of all three are pushed
61 onto a stack for every function call, as well as popped when every function
62 returns. This means that functions can assign to any and all of those
63 globals without worrying that the change will affect other functions.
64 Thus, a hypothetical function named **output(x,b)** that simply printed
65 **x** in base **b** could be written like this:
67 define void output(x, b) {
74 define void output(x, b) {
82 This makes writing functions much easier.
84 However, since using this flag means that functions cannot set **ibase**,
85 **obase**, or **scale** globally, functions that are made to do so cannot
86 work anymore. There are two possible use cases for that, and each has a
89 First, if a function is called on startup to turn bc(1) into a number
90 converter, it is possible to replace that capability with various shell
93 alias d2o="bc -e ibase=A -e obase=8"
94 alias h2b="bc -e ibase=G -e obase=2"
96 Second, if the purpose of a function is to set **ibase**, **obase**, or
97 **scale** globally for any other purpose, it could be split into one to
98 three functions (based on how many globals it sets) and each of those
99 functions could return the desired value for a global.
101 If the behavior of this option is desired for every run of bc(1), then users
102 could make sure to define **BC_ENV_ARGS** and include this option (see the
103 **ENVIRONMENT VARIABLES** section for more details).
105 If **-s**, **-w**, or any equivalents are used, this option is ignored.
107 This is a **non-portable extension**.
111 : Prints a usage message and quits.
113 **-i**, **-\-interactive**
115 : Forces interactive mode. (See the **INTERACTIVE MODE** section.)
117 This is a **non-portable extension**.
119 **-l**, **-\-mathlib**
121 : Sets **scale** (see the **SYNTAX** section) to **20** and loads the included
122 math library before running any code, including any expressions or files
123 specified on the command line.
125 To learn what is in the library, see the **LIBRARY** section.
127 **-P**, **-\-no-prompt**
129 : This option is a no-op.
131 This is a **non-portable extension**.
133 **-R**, **-\-no-read-prompt**
135 : Because bc(1) was built without support for prompts, this option is a no-op.
137 This is a **non-portable extension**.
141 : This option is for compatibility with the [GNU bc(1)][2]; it is a no-op.
142 Without this option, GNU bc(1) prints a copyright header. This bc(1) only
143 prints the copyright header if one or more of the **-v**, **-V**, or
144 **-\-version** options are given.
146 This is a **non-portable extension**.
148 **-s**, **-\-standard**
150 : Process exactly the language defined by the [standard][1] and error if any
153 This is a **non-portable extension**.
155 **-v**, **-V**, **-\-version**
157 : Print the version information (copyright header) and exit.
159 This is a **non-portable extension**.
163 : Like **-s** and **-\-standard**, except that warnings (and not errors) are
164 printed for non-standard extensions and execution continues normally.
166 This is a **non-portable extension**.
168 **-e** *expr*, **-\-expression**=*expr*
170 : Evaluates *expr*. If multiple expressions are given, they are evaluated in
171 order. If files are given as well (see below), the expressions and files are
172 evaluated in the order given. This means that if a file is given before an
173 expression, the file is read in and evaluated first.
175 If this option is given on the command-line (i.e., not in **BC_ENV_ARGS**,
176 see the **ENVIRONMENT VARIABLES** section), then after processing all
177 expressions and files, bc(1) will exit, unless **-** (**stdin**) was given
178 as an argument at least once to **-f** or **-\-file**, whether on the
179 command-line or in **BC_ENV_ARGS**. However, if any other **-e**,
180 **-\-expression**, **-f**, or **-\-file** arguments are given after **-f-**
181 or equivalent is given, bc(1) will give a fatal error and exit.
183 This is a **non-portable extension**.
185 **-f** *file*, **-\-file**=*file*
187 : Reads in *file* and evaluates it, line by line, as though it were read
188 through **stdin**. If expressions are also given (see above), the
189 expressions are evaluated in the order given.
191 If this option is given on the command-line (i.e., not in **BC_ENV_ARGS**,
192 see the **ENVIRONMENT VARIABLES** section), then after processing all
193 expressions and files, bc(1) will exit, unless **-** (**stdin**) was given
194 as an argument at least once to **-f** or **-\-file**. However, if any other
195 **-e**, **-\-expression**, **-f**, or **-\-file** arguments are given after
196 **-f-** or equivalent is given, bc(1) will give a fatal error and exit.
198 This is a **non-portable extension**.
200 All long options are **non-portable extensions**.
204 Any non-error output is written to **stdout**. In addition, if history (see the
205 **HISTORY** section) and the prompt (see the **TTY MODE** section) are enabled,
206 both are output to **stdout**.
208 **Note**: Unlike other bc(1) implementations, this bc(1) will issue a fatal
209 error (see the **EXIT STATUS** section) if it cannot write to **stdout**, so if
210 **stdout** is closed, as in **bc <file> >&-**, it will quit with an error. This
211 is done so that bc(1) can report problems when **stdout** is redirected to a
214 If there are scripts that depend on the behavior of other bc(1) implementations,
215 it is recommended that those scripts be changed to redirect **stdout** to
220 Any error output is written to **stderr**.
222 **Note**: Unlike other bc(1) implementations, this bc(1) will issue a fatal
223 error (see the **EXIT STATUS** section) if it cannot write to **stderr**, so if
224 **stderr** is closed, as in **bc <file> 2>&-**, it will quit with an error. This
225 is done so that bc(1) can exit with an error code when **stderr** is redirected
228 If there are scripts that depend on the behavior of other bc(1) implementations,
229 it is recommended that those scripts be changed to redirect **stderr** to
234 The syntax for bc(1) programs is mostly C-like, with some differences. This
235 bc(1) follows the [POSIX standard][1], which is a much more thorough resource
236 for the language this bc(1) accepts. This section is meant to be a summary and a
237 listing of all the extensions to the standard.
239 In the sections below, **E** means expression, **S** means statement, and **I**
242 Identifiers (**I**) start with a lowercase letter and can be followed by any
243 number (up to **BC_NAME_MAX-1**) of lowercase letters (**a-z**), digits
244 (**0-9**), and underscores (**\_**). The regex is **\[a-z\]\[a-z0-9\_\]\***.
245 Identifiers with more than one character (letter) are a
246 **non-portable extension**.
248 **ibase** is a global variable determining how to interpret constant numbers. It
249 is the "input" base, or the number base used for interpreting input numbers.
250 **ibase** is initially **10**. If the **-s** (**-\-standard**) and **-w**
251 (**-\-warn**) flags were not given on the command line, the max allowable value
252 for **ibase** is **36**. Otherwise, it is **16**. The min allowable value for
253 **ibase** is **2**. The max allowable value for **ibase** can be queried in
254 bc(1) programs with the **maxibase()** built-in function.
256 **obase** is a global variable determining how to output results. It is the
257 "output" base, or the number base used for outputting numbers. **obase** is
258 initially **10**. The max allowable value for **obase** is **BC_BASE_MAX** and
259 can be queried in bc(1) programs with the **maxobase()** built-in function. The
260 min allowable value for **obase** is **2**. Values are output in the specified
263 The *scale* of an expression is the number of digits in the result of the
264 expression right of the decimal point, and **scale** is a global variable that
265 sets the precision of any operations, with exceptions. **scale** is initially
266 **0**. **scale** cannot be negative. The max allowable value for **scale** is
267 **BC_SCALE_MAX** and can be queried in bc(1) programs with the **maxscale()**
270 bc(1) has both *global* variables and *local* variables. All *local*
271 variables are local to the function; they are parameters or are introduced in
272 the **auto** list of a function (see the **FUNCTIONS** section). If a variable
273 is accessed which is not a parameter or in the **auto** list, it is assumed to
274 be *global*. If a parent function has a *local* variable version of a variable
275 that a child function considers *global*, the value of that *global* variable in
276 the child function is the value of the variable in the parent function, not the
277 value of the actual *global* variable.
279 All of the above applies to arrays as well.
281 The value of a statement that is an expression (i.e., any of the named
282 expressions or operands) is printed unless the lowest precedence operator is an
283 assignment operator *and* the expression is notsurrounded by parentheses.
285 The value that is printed is also assigned to the special variable **last**. A
286 single dot (**.**) may also be used as a synonym for **last**. These are
287 **non-portable extensions**.
289 Either semicolons or newlines may separate statements.
293 There are two kinds of comments:
295 1. Block comments are enclosed in **/\*** and **\*/**.
296 2. Line comments go from **#** until, and not including, the next newline. This
297 is a **non-portable extension**.
301 The following are named expressions in bc(1):
304 2. Array Elements: **I[E]**
308 6. **last** or a single dot (**.**)
310 Number 6 is a **non-portable extension**.
312 Variables and arrays do not interfere; users can have arrays named the same as
313 variables. This also applies to functions (see the **FUNCTIONS** section), so a
314 user can have a variable, array, and function that all have the same name, and
315 they will not shadow each other, whether inside of functions or not.
317 Named expressions are required as the operand of **increment**/**decrement**
318 operators and as the left side of **assignment** operators (see the *Operators*
323 The following are valid operands in bc(1):
325 1. Numbers (see the *Numbers* subsection below).
326 2. Array indices (**I[E]**).
327 3. **(E)**: The value of **E** (used to change precedence).
328 4. **sqrt(E)**: The square root of **E**. **E** must be non-negative.
329 5. **length(E)**: The number of significant decimal digits in **E**.
330 6. **length(I[])**: The number of elements in the array **I**. This is a
331 **non-portable extension**.
332 7. **scale(E)**: The *scale* of **E**.
333 8. **abs(E)**: The absolute value of **E**. This is a **non-portable
335 9. **I()**, **I(E)**, **I(E, E)**, and so on, where **I** is an identifier for
336 a non-**void** function (see the *Void Functions* subsection of the
337 **FUNCTIONS** section). The **E** argument(s) may also be arrays of the form
338 **I[]**, which will automatically be turned into array references (see the
339 *Array References* subsection of the **FUNCTIONS** section) if the
340 corresponding parameter in the function definition is an array reference.
341 10. **read()**: Reads a line from **stdin** and uses that as an expression. The
342 result of that expression is the result of the **read()** operand. This is a
343 **non-portable extension**.
344 11. **maxibase()**: The max allowable **ibase**. This is a **non-portable
346 12. **maxobase()**: The max allowable **obase**. This is a **non-portable
348 13. **maxscale()**: The max allowable **scale**. This is a **non-portable
353 Numbers are strings made up of digits, uppercase letters, and at most **1**
354 period for a radix. Numbers can have up to **BC_NUM_MAX** digits. Uppercase
355 letters are equal to **9** + their position in the alphabet (i.e., **A** equals
356 **10**, or **9+1**). If a digit or letter makes no sense with the current value
357 of **ibase**, they are set to the value of the highest valid digit in **ibase**.
359 Single-character numbers (i.e., **A** alone) take the value that they would have
360 if they were valid digits, regardless of the value of **ibase**. This means that
361 **A** alone always equals decimal **10** and **Z** alone always equals decimal
366 The following arithmetic and logical operators can be used. They are listed in
367 order of decreasing precedence. Operators in the same group have the same
372 : Type: Prefix and Postfix
376 Description: **increment**, **decrement**
384 Description: **negation**, **boolean not**
392 Description: **power**
400 Description: **multiply**, **divide**, **modulus**
408 Description: **add**, **subtract**
410 **=** **+=** **-=** **\*=** **/=** **%=** **\^=**
416 Description: **assignment**
418 **==** **\<=** **\>=** **!=** **\<** **\>**
424 Description: **relational**
432 Description: **boolean and**
440 Description: **boolean or**
442 The operators will be described in more detail below.
446 : The prefix and postfix **increment** and **decrement** operators behave
447 exactly like they would in C. They require a named expression (see the
448 *Named Expressions* subsection) as an operand.
450 The prefix versions of these operators are more efficient; use them where
455 : The **negation** operator returns **0** if a user attempts to negate any
456 expression with the value **0**. Otherwise, a copy of the expression with
457 its sign flipped is returned.
461 : The **boolean not** operator returns **1** if the expression is **0**, or
464 This is a **non-portable extension**.
468 : The **power** operator (not the **exclusive or** operator, as it would be in
469 C) takes two expressions and raises the first to the power of the value of
470 the second. The *scale* of the result is equal to **scale**.
472 The second expression must be an integer (no *scale*), and if it is
473 negative, the first value must be non-zero.
477 : The **multiply** operator takes two expressions, multiplies them, and
478 returns the product. If **a** is the *scale* of the first expression and
479 **b** is the *scale* of the second expression, the *scale* of the result is
480 equal to **min(a+b,max(scale,a,b))** where **min()** and **max()** return
485 : The **divide** operator takes two expressions, divides them, and returns the
486 quotient. The *scale* of the result shall be the value of **scale**.
488 The second expression must be non-zero.
492 : The **modulus** operator takes two expressions, **a** and **b**, and
493 evaluates them by 1) Computing **a/b** to current **scale** and 2) Using the
494 result of step 1 to calculate **a-(a/b)\*b** to *scale*
495 **max(scale+scale(b),scale(a))**.
497 The second expression must be non-zero.
501 : The **add** operator takes two expressions, **a** and **b**, and returns the
502 sum, with a *scale* equal to the max of the *scale*s of **a** and **b**.
506 : The **subtract** operator takes two expressions, **a** and **b**, and
507 returns the difference, with a *scale* equal to the max of the *scale*s of
510 **=** **+=** **-=** **\*=** **/=** **%=** **\^=**
512 : The **assignment** operators take two expressions, **a** and **b** where
513 **a** is a named expression (see the *Named Expressions* subsection).
515 For **=**, **b** is copied and the result is assigned to **a**. For all
516 others, **a** and **b** are applied as operands to the corresponding
517 arithmetic operator and the result is assigned to **a**.
519 **==** **\<=** **\>=** **!=** **\<** **\>**
521 : The **relational** operators compare two expressions, **a** and **b**, and
522 if the relation holds, according to C language semantics, the result is
523 **1**. Otherwise, it is **0**.
525 Note that unlike in C, these operators have a lower precedence than the
526 **assignment** operators, which means that **a=b\>c** is interpreted as
529 Also, unlike the [standard][1] requires, these operators can appear anywhere
530 any other expressions can be used. This allowance is a
531 **non-portable extension**.
535 : The **boolean and** operator takes two expressions and returns **1** if both
536 expressions are non-zero, **0** otherwise.
538 This is *not* a short-circuit operator.
540 This is a **non-portable extension**.
544 : The **boolean or** operator takes two expressions and returns **1** if one
545 of the expressions is non-zero, **0** otherwise.
547 This is *not* a short-circuit operator.
549 This is a **non-portable extension**.
553 The following items are statements:
556 2. **{** **S** **;** ... **;** **S** **}**
557 3. **if** **(** **E** **)** **S**
558 4. **if** **(** **E** **)** **S** **else** **S**
559 5. **while** **(** **E** **)** **S**
560 6. **for** **(** **E** **;** **E** **;** **E** **)** **S**
561 7. An empty statement
567 13. A string of characters, enclosed in double quotes
568 14. **print** **E** **,** ... **,** **E**
569 15. **I()**, **I(E)**, **I(E, E)**, and so on, where **I** is an identifier for
570 a **void** function (see the *Void Functions* subsection of the
571 **FUNCTIONS** section). The **E** argument(s) may also be arrays of the form
572 **I[]**, which will automatically be turned into array references (see the
573 *Array References* subsection of the **FUNCTIONS** section) if the
574 corresponding parameter in the function definition is an array reference.
576 Numbers 4, 9, 11, 12, 14, and 15 are **non-portable extensions**.
578 Also, as a **non-portable extension**, any or all of the expressions in the
579 header of a for loop may be omitted. If the condition (second expression) is
580 omitted, it is assumed to be a constant **1**.
582 The **break** statement causes a loop to stop iterating and resume execution
583 immediately following a loop. This is only allowed in loops.
585 The **continue** statement causes a loop iteration to stop early and returns to
586 the start of the loop, including testing the loop condition. This is only
589 The **if** **else** statement does the same thing as in C.
591 The **quit** statement causes bc(1) to quit, even if it is on a branch that will
592 not be executed (it is a compile-time command).
594 The **halt** statement causes bc(1) to quit, if it is executed. (Unlike **quit**
595 if it is on a branch of an **if** statement that is not executed, bc(1) does not
598 The **limits** statement prints the limits that this bc(1) is subject to. This
599 is like the **quit** statement in that it is a compile-time command.
601 An expression by itself is evaluated and printed, followed by a newline.
605 The "expressions" in a **print** statement may also be strings. If they are, there
606 are backslash escape sequences that are interpreted specially. What those
607 sequences are, and what they cause to be printed, are shown below:
621 Any other character following a backslash causes the backslash and character to
624 Any non-string expression in a print statement shall be assigned to **last**,
625 like any other expression that is printed.
627 ## Order of Evaluation
629 All expressions in a statment are evaluated left to right, except as necessary
630 to maintain order of operations. This means, for example, assuming that **i** is
631 equal to **0**, in the expression
635 the first (or 0th) element of **a** is set to **1**, and **i** is equal to **2**
636 at the end of the expression.
638 This includes function arguments. Thus, assuming **i** is equal to **0**, this
639 means that in the expression
643 the first argument passed to **x()** is **0**, and the second argument is **1**,
644 while **i** is equal to **2** before the function starts executing.
648 Function definitions are as follows:
658 Any **I** in the parameter list or **auto** list may be replaced with **I[]** to
659 make a parameter or **auto** var an array, and any **I** in the parameter list
660 may be replaced with **\*I[]** to make a parameter an array reference. Callers
661 of functions that take array references should not put an asterisk in the call;
662 they must be called with just **I[]** like normal array parameters and will be
663 automatically converted into references.
665 As a **non-portable extension**, the opening brace of a **define** statement may
666 appear on the next line.
668 As a **non-portable extension**, the return statement may also be in one of the
672 2. **return** **(** **)**
675 The first two, or not specifying a **return** statement, is equivalent to
676 **return (0)**, unless the function is a **void** function (see the *Void
677 Functions* subsection below).
681 Functions can also be **void** functions, defined as follows:
684 define void I(I,...,I){
691 They can only be used as standalone expressions, where such an expression would
692 be printed alone, except in a print statement.
694 Void functions can only use the first two **return** statements listed above.
695 They can also omit the return statement entirely.
697 The word "void" is not treated as a keyword; it is still possible to have
698 variables, arrays, and functions named **void**. The word "void" is only
699 treated specially right after the **define** keyword.
701 This is a **non-portable extension**.
705 For any array in the parameter list, if the array is declared in the form
711 it is a **reference**. Any changes to the array in the function are reflected,
712 when the function returns, to the array that was passed in.
714 Other than this, all function arguments are passed by value.
716 This is a **non-portable extension**.
720 All of the functions below are available when the **-l** or **-\-mathlib**
721 command-line flags are given.
725 The [standard][1] defines the following functions for the math library:
729 : Returns the sine of **x**, which is assumed to be in radians.
731 This is a transcendental function (see the *Transcendental Functions*
736 : Returns the cosine of **x**, which is assumed to be in radians.
738 This is a transcendental function (see the *Transcendental Functions*
743 : Returns the arctangent of **x**, in radians.
745 This is a transcendental function (see the *Transcendental Functions*
750 : Returns the natural logarithm of **x**.
752 This is a transcendental function (see the *Transcendental Functions*
757 : Returns the mathematical constant **e** raised to the power of **x**.
759 This is a transcendental function (see the *Transcendental Functions*
764 : Returns the bessel integer order **n** (truncated) of **x**.
766 This is a transcendental function (see the *Transcendental Functions*
769 ## Transcendental Functions
771 All transcendental functions can return slightly inaccurate results (up to 1
772 [ULP][4]). This is unavoidable, and [this article][5] explains why it is
773 impossible and unnecessary to calculate exact results for the transcendental
776 Because of the possible inaccuracy, I recommend that users call those functions
777 with the precision (**scale**) set to at least 1 higher than is necessary. If
778 exact results are *absolutely* required, users can double the precision
779 (**scale**) and then truncate.
781 The transcendental functions in the standard math library are:
792 When bc(1) encounters an error or a signal that it has a non-default handler
793 for, it resets. This means that several things happen.
795 First, any functions that are executing are stopped and popped off the stack.
796 The behavior is not unlike that of exceptions in programming languages. Then
797 the execution point is set so that any code waiting to execute (after all
798 functions returned) is skipped.
800 Thus, when bc(1) resets, it skips any remaining code waiting to be executed.
801 Then, if it is interactive mode, and the error was not a fatal error (see the
802 **EXIT STATUS** section), it asks for more input; otherwise, it exits with the
803 appropriate return code.
805 Note that this reset behavior is different from the GNU bc(1), which attempts to
806 start executing the statement right after the one that caused an error.
810 Most bc(1) implementations use **char** types to calculate the value of **1**
811 decimal digit at a time, but that can be slow. This bc(1) does something
814 It uses large integers to calculate more than **1** decimal digit at a time. If
815 built in a environment where **BC_LONG_BIT** (see the **LIMITS** section) is
816 **64**, then each integer has **9** decimal digits. If built in an environment
817 where **BC_LONG_BIT** is **32** then each integer has **4** decimal digits. This
818 value (the number of decimal digits per large integer) is called
821 The actual values of **BC_LONG_BIT** and **BC_BASE_DIGS** can be queried with
822 the **limits** statement.
824 In addition, this bc(1) uses an even larger integer for overflow checking. This
825 integer type depends on the value of **BC_LONG_BIT**, but is always at least
826 twice as large as the integer type used to store digits.
830 The following are the limits on bc(1):
834 : The number of bits in the **long** type in the environment where bc(1) was
835 built. This determines how many decimal digits can be stored in a single
836 large integer (see the **PERFORMANCE** section).
840 : The number of decimal digits per large integer (see the **PERFORMANCE**
841 section). Depends on **BC_LONG_BIT**.
845 : The max decimal number that each large integer can store (see
846 **BC_BASE_DIGS**) plus **1**. Depends on **BC_BASE_DIGS**.
850 : The max number that the overflow type (see the **PERFORMANCE** section) can
851 hold. Depends on **BC_LONG_BIT**.
855 : The maximum output base. Set at **BC_BASE_POW**.
859 : The maximum size of arrays. Set at **SIZE_MAX-1**.
863 : The maximum **scale**. Set at **BC_OVERFLOW_MAX-1**.
867 : The maximum length of strings. Set at **BC_OVERFLOW_MAX-1**.
871 : The maximum length of identifiers. Set at **BC_OVERFLOW_MAX-1**.
875 : The maximum length of a number (in decimal digits), which includes digits
876 after the decimal point. Set at **BC_OVERFLOW_MAX-1**.
880 : The maximum allowable exponent (positive or negative). Set at
885 : The maximum number of vars/arrays. Set at **SIZE_MAX-1**.
887 The actual values can be queried with the **limits** statement.
889 These limits are meant to be effectively non-existent; the limits are so large
890 (at least on 64-bit machines) that there should not be any point at which they
891 become a problem. In fact, memory should be exhausted before these limits should
894 # ENVIRONMENT VARIABLES
896 bc(1) recognizes the following environment variables:
900 : If this variable exists (no matter the contents), bc(1) behaves as if
901 the **-s** option was given.
905 : This is another way to give command-line arguments to bc(1). They should be
906 in the same format as all other command-line arguments. These are always
907 processed first, so any files given in **BC_ENV_ARGS** will be processed
908 before arguments and files given on the command-line. This gives the user
909 the ability to set up "standard" options and files to be used at every
910 invocation. The most useful thing for such files to contain would be useful
911 functions that the user might want every time bc(1) runs.
913 The code that parses **BC_ENV_ARGS** will correctly handle quoted arguments,
914 but it does not understand escape sequences. For example, the string
915 **"/home/gavin/some bc file.bc"** will be correctly parsed, but the string
916 **"/home/gavin/some \"bc\" file.bc"** will include the backslashes.
918 The quote parsing will handle either kind of quotes, **'** or **"**. Thus,
919 if you have a file with any number of single quotes in the name, you can use
920 double quotes as the outside quotes, as in **"some 'bc' file.bc"**, and vice
921 versa if you have a file with double quotes. However, handling a file with
922 both kinds of quotes in **BC_ENV_ARGS** is not supported due to the
923 complexity of the parsing, though such files are still supported on the
924 command-line where the parsing is done by the shell.
928 : If this environment variable exists and contains an integer that is greater
929 than **1** and is less than **UINT16_MAX** (**2\^16-1**), bc(1) will output
930 lines to that length, including the backslash (**\\**). The default line
935 bc(1) returns the following exit statuses:
943 : A math error occurred. This follows standard practice of using **1** for
944 expected errors, since math errors will happen in the process of normal
947 Math errors include divide by **0**, taking the square root of a negative
948 number, attempting to convert a negative number to a hardware integer,
949 overflow when converting a number to a hardware integer, and attempting to
950 use a non-integer where an integer is required.
952 Converting to a hardware integer happens for the second operand of the power
953 (**\^**) operator and the corresponding assignment operator.
957 : A parse error occurred.
959 Parse errors include unexpected **EOF**, using an invalid character, failing
960 to find the end of a string or comment, using a token where it is invalid,
961 giving an invalid expression, giving an invalid print statement, giving an
962 invalid function definition, attempting to assign to an expression that is
963 not a named expression (see the *Named Expressions* subsection of the
964 **SYNTAX** section), giving an invalid **auto** list, having a duplicate
965 **auto**/function parameter, failing to find the end of a code block,
966 attempting to return a value from a **void** function, attempting to use a
967 variable as a reference, and using any extensions when the option **-s** or
968 any equivalents were given.
972 : A runtime error occurred.
974 Runtime errors include assigning an invalid number to **ibase**, **obase**,
975 or **scale**; give a bad expression to a **read()** call, calling **read()**
976 inside of a **read()** call, type errors, passing the wrong number of
977 arguments to functions, attempting to call an undefined function, and
978 attempting to use a **void** function call as a value in an expression.
982 : A fatal error occurred.
984 Fatal errors include memory allocation errors, I/O errors, failing to open
985 files, attempting to use files that do not have only ASCII characters (bc(1)
986 only accepts ASCII characters), attempting to open a directory as a file,
987 and giving invalid command-line options.
989 The exit status **4** is special; when a fatal error occurs, bc(1) always exits
990 and returns **4**, no matter what mode bc(1) is in.
992 The other statuses will only be returned when bc(1) is not in interactive mode
993 (see the **INTERACTIVE MODE** section), since bc(1) resets its state (see the
994 **RESET** section) and accepts more input when one of those errors occurs in
995 interactive mode. This is also the case when interactive mode is forced by the
996 **-i** flag or **-\-interactive** option.
998 These exit statuses allow bc(1) to be used in shell scripting with error
999 checking, and its normal behavior can be forced by using the **-i** flag or
1000 **-\-interactive** option.
1004 Per the [standard][1], bc(1) has an interactive mode and a non-interactive mode.
1005 Interactive mode is turned on automatically when both **stdin** and **stdout**
1006 are hooked to a terminal, but the **-i** flag and **-\-interactive** option can
1007 turn it on in other cases.
1009 In interactive mode, bc(1) attempts to recover from errors (see the **RESET**
1010 section), and in normal execution, flushes **stdout** as soon as execution is
1011 done for the current input.
1015 If **stdin**, **stdout**, and **stderr** are all connected to a TTY, bc(1) turns
1018 TTY mode is required for history to be enabled (see the **COMMAND LINE HISTORY**
1019 section). It is also required to enable special handling for **SIGINT** signals.
1021 TTY mode is different from interactive mode because interactive mode is required
1022 in the [bc(1) specification][1], and interactive mode requires only **stdin**
1023 and **stdout** to be connected to a terminal.
1027 Sending a **SIGINT** will cause bc(1) to stop execution of the current input. If
1028 bc(1) is in TTY mode (see the **TTY MODE** section), it will reset (see the
1029 **RESET** section). Otherwise, it will clean up and exit.
1031 Note that "current input" can mean one of two things. If bc(1) is processing
1032 input from **stdin** in TTY mode, it will ask for more input. If bc(1) is
1033 processing input from a file in TTY mode, it will stop processing the file and
1034 start processing the next file, if one exists, or ask for input from **stdin**
1035 if no other file exists.
1037 This means that if a **SIGINT** is sent to bc(1) as it is executing a file, it
1038 can seem as though bc(1) did not respond to the signal since it will immediately
1039 start executing the next file. This is by design; most files that users execute
1040 when interacting with bc(1) have function definitions, which are quick to parse.
1041 If a file takes a long time to execute, there may be a bug in that file. The
1042 rest of the files could still be executed without problem, allowing the user to
1045 **SIGTERM** and **SIGQUIT** cause bc(1) to clean up and exit, and it uses the
1046 default handler for all other signals. The one exception is **SIGHUP**; in that
1047 case, when bc(1) is in TTY mode, a **SIGHUP** will cause bc(1) to clean up and
1050 # COMMAND LINE HISTORY
1052 bc(1) supports interactive command-line editing. If bc(1) is in TTY mode (see
1053 the **TTY MODE** section), history is enabled. Previous lines can be recalled
1054 and edited with the arrow keys.
1056 **Note**: tabs are converted to 8 spaces.
1064 bc(1) is compliant with the [IEEE Std 1003.1-2017 (“POSIX.1-2017”)][1]
1065 specification. The flags **-efghiqsvVw**, all long options, and the extensions
1066 noted above are extensions to that specification.
1068 Note that the specification explicitly says that bc(1) only accepts numbers that
1069 use a period (**.**) as a radix point, regardless of the value of
1074 None are known. Report bugs at https://git.yzena.com/gavin/bc.
1078 Gavin D. Howard <gavin@yzena.com> and contributors.
1080 [1]: https://pubs.opengroup.org/onlinepubs/9699919799/utilities/bc.html
1081 [2]: https://www.gnu.org/software/bc/
1082 [3]: https://en.wikipedia.org/wiki/Rounding#Round_half_away_from_zero
1083 [4]: https://en.wikipedia.org/wiki/Unit_in_the_last_place
1084 [5]: https://people.eecs.berkeley.edu/~wkahan/LOG10HAF.TXT
1085 [6]: https://en.wikipedia.org/wiki/Rounding#Rounding_away_from_zero