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** [**-ghilPqsvVw**] [**--global-stacks**] [**--help**] [**--interactive**] [**--mathlib**] [**--no-prompt**] [**--quiet**] [**--standard**] [**--warn**] [**--version**] [**-e** *expr*] [**--expression**=*expr*...] [**-f** *file*...] [**-file**=*file*...]
42 bc(1) is an interactive processor for a language first standardized in 1991 by
43 POSIX. (The current standard is [here][1].) The language provides unlimited
44 precision decimal arithmetic and is somewhat C-like, but there are differences.
45 Such differences will be noted in this document.
47 After parsing and handling options, this bc(1) reads any files given on the
48 command line and executes them before reading from **stdin**.
52 The following are the options that bc(1) accepts.
54 **-g**, **--global-stacks**
56 Turns the globals **ibase**, **obase**, and **scale** into stacks.
58 This has the effect that a copy of the current value of all three are pushed
59 onto a stack for every function call, as well as popped when every function
60 returns. This means that functions can assign to any and all of those
61 globals without worrying that the change will affect other functions.
62 Thus, a hypothetical function named **output(x,b)** that simply printed
63 **x** in base **b** could be written like this:
65 define void output(x, b) {
72 define void output(x, b) {
80 This makes writing functions much easier.
82 However, since using this flag means that functions cannot set **ibase**,
83 **obase**, or **scale** globally, functions that are made to do so cannot
84 work anymore. There are two possible use cases for that, and each has a
87 First, if a function is called on startup to turn bc(1) into a number
88 converter, it is possible to replace that capability with various shell
91 alias d2o="bc -e ibase=A -e obase=8"
92 alias h2b="bc -e ibase=G -e obase=2"
94 Second, if the purpose of a function is to set **ibase**, **obase**, or
95 **scale** globally for any other purpose, it could be split into one to
96 three functions (based on how many globals it sets) and each of those
97 functions could return the desired value for a global.
99 If the behavior of this option is desired for every run of bc(1), then users
100 could make sure to define **BC_ENV_ARGS** and include this option (see the
101 **ENVIRONMENT VARIABLES** section for more details).
103 If **-s**, **-w**, or any equivalents are used, this option is ignored.
105 This is a **non-portable extension**.
109 : Prints a usage message and quits.
111 **-i**, **--interactive**
113 : Forces interactive mode. (See the **INTERACTIVE MODE** section.)
115 This is a **non-portable extension**.
117 **-l**, **--mathlib**
119 : Sets **scale** (see the **SYNTAX** section) to **20** and loads the included
120 math library before running any code, including any expressions or files
121 specified on the command line.
123 To learn what is in the library, see the **LIBRARY** section.
125 **-P**, **--no-prompt**
127 : This option is a no-op.
129 This is a **non-portable extension**.
133 : This option is for compatibility with the [GNU bc(1)][2]; it is a no-op.
134 Without this option, GNU bc(1) prints a copyright header. This bc(1) only
135 prints the copyright header if one or more of the **-v**, **-V**, or
136 **--version** options are given.
138 This is a **non-portable extension**.
140 **-s**, **--standard**
142 : Process exactly the language defined by the [standard][1] and error if any
145 This is a **non-portable extension**.
147 **-v**, **-V**, **--version**
149 : Print the version information (copyright header) and exit.
151 This is a **non-portable extension**.
155 : Like **-s** and **--standard**, except that warnings (and not errors) are
156 printed for non-standard extensions and execution continues normally.
158 This is a **non-portable extension**.
160 **-e** *expr*, **--expression**=*expr*
162 : Evaluates *expr*. If multiple expressions are given, they are evaluated in
163 order. If files are given as well (see below), the expressions and files are
164 evaluated in the order given. This means that if a file is given before an
165 expression, the file is read in and evaluated first.
167 If this option is given on the command-line (i.e., not in **BC_ENV_ARGS**,
168 see the **ENVIRONMENT VARIABLES** section), then after processing all
169 expressions and files, bc(1) will exit, unless **-** (**stdin**) was given
170 as an argument at least once to **-f** or **--file**, whether on the
171 command-line or in **BC_ENV_ARGS**. However, if any other **-e**,
172 **--expression**, **-f**, or **--file** arguments are given after **-f-** or
173 equivalent is given, bc(1) will give a fatal error and exit.
175 This is a **non-portable extension**.
177 **-f** *file*, **--file**=*file*
179 : Reads in *file* and evaluates it, line by line, as though it were read
180 through **stdin**. If expressions are also given (see above), the
181 expressions are evaluated in the order given.
183 If this option is given on the command-line (i.e., not in **BC_ENV_ARGS**,
184 see the **ENVIRONMENT VARIABLES** section), then after processing all
185 expressions and files, bc(1) will exit, unless **-** (**stdin**) was given
186 as an argument at least once to **-f** or **--file**. However, if any other
187 **-e**, **--expression**, **-f**, or **--file** arguments are given after
188 **-f-** or equivalent is given, bc(1) will give a fatal error and exit.
190 This is a **non-portable extension**.
192 All long options are **non-portable extensions**.
196 Any non-error output is written to **stdout**. In addition, if history (see the
197 **HISTORY** section) and the prompt (see the **TTY MODE** section) are enabled,
198 both are output to **stdout**.
200 **Note**: Unlike other bc(1) implementations, this bc(1) will issue a fatal
201 error (see the **EXIT STATUS** section) if it cannot write to **stdout**, so if
202 **stdout** is closed, as in **bc <file> >&-**, it will quit with an error. This
203 is done so that bc(1) can report problems when **stdout** is redirected to a
206 If there are scripts that depend on the behavior of other bc(1) implementations,
207 it is recommended that those scripts be changed to redirect **stdout** to
212 Any error output is written to **stderr**.
214 **Note**: Unlike other bc(1) implementations, this bc(1) will issue a fatal
215 error (see the **EXIT STATUS** section) if it cannot write to **stderr**, so if
216 **stderr** is closed, as in **bc <file> 2>&-**, it will quit with an error. This
217 is done so that bc(1) can exit with an error code when **stderr** is redirected
220 If there are scripts that depend on the behavior of other bc(1) implementations,
221 it is recommended that those scripts be changed to redirect **stderr** to
226 The syntax for bc(1) programs is mostly C-like, with some differences. This
227 bc(1) follows the [POSIX standard][1], which is a much more thorough resource
228 for the language this bc(1) accepts. This section is meant to be a summary and a
229 listing of all the extensions to the standard.
231 In the sections below, **E** means expression, **S** means statement, and **I**
234 Identifiers (**I**) start with a lowercase letter and can be followed by any
235 number (up to **BC_NAME_MAX-1**) of lowercase letters (**a-z**), digits
236 (**0-9**), and underscores (**\_**). The regex is **\[a-z\]\[a-z0-9\_\]\***.
237 Identifiers with more than one character (letter) are a
238 **non-portable extension**.
240 **ibase** is a global variable determining how to interpret constant numbers. It
241 is the "input" base, or the number base used for interpreting input numbers.
242 **ibase** is initially **10**. If the **-s** (**--standard**) and **-w**
243 (**--warn**) flags were not given on the command line, the max allowable value
244 for **ibase** is **36**. Otherwise, it is **16**. The min allowable value for
245 **ibase** is **2**. The max allowable value for **ibase** can be queried in
246 bc(1) programs with the **maxibase()** built-in function.
248 **obase** is a global variable determining how to output results. It is the
249 "output" base, or the number base used for outputting numbers. **obase** is
250 initially **10**. The max allowable value for **obase** is **BC_BASE_MAX** and
251 can be queried in bc(1) programs with the **maxobase()** built-in function. The
252 min allowable value for **obase** is **2**. Values are output in the specified
255 The *scale* of an expression is the number of digits in the result of the
256 expression right of the decimal point, and **scale** is a global variable that
257 sets the precision of any operations, with exceptions. **scale** is initially
258 **0**. **scale** cannot be negative. The max allowable value for **scale** is
259 **BC_SCALE_MAX** and can be queried in bc(1) programs with the **maxscale()**
262 bc(1) has both *global* variables and *local* variables. All *local*
263 variables are local to the function; they are parameters or are introduced in
264 the **auto** list of a function (see the **FUNCTIONS** section). If a variable
265 is accessed which is not a parameter or in the **auto** list, it is assumed to
266 be *global*. If a parent function has a *local* variable version of a variable
267 that a child function considers *global*, the value of that *global* variable in
268 the child function is the value of the variable in the parent function, not the
269 value of the actual *global* variable.
271 All of the above applies to arrays as well.
273 The value of a statement that is an expression (i.e., any of the named
274 expressions or operands) is printed unless the lowest precedence operator is an
275 assignment operator *and* the expression is notsurrounded by parentheses.
277 The value that is printed is also assigned to the special variable **last**. A
278 single dot (**.**) may also be used as a synonym for **last**. These are
279 **non-portable extensions**.
281 Either semicolons or newlines may separate statements.
285 There are two kinds of comments:
287 1. Block comments are enclosed in **/\*** and **\*/**.
288 2. Line comments go from **#** until, and not including, the next newline. This
289 is a **non-portable extension**.
293 The following are named expressions in bc(1):
296 2. Array Elements: **I[E]**
300 6. **last** or a single dot (**.**)
302 Number 6 is a **non-portable extension**.
304 Variables and arrays do not interfere; users can have arrays named the same as
305 variables. This also applies to functions (see the **FUNCTIONS** section), so a
306 user can have a variable, array, and function that all have the same name, and
307 they will not shadow each other, whether inside of functions or not.
309 Named expressions are required as the operand of **increment**/**decrement**
310 operators and as the left side of **assignment** operators (see the *Operators*
315 The following are valid operands in bc(1):
317 1. Numbers (see the *Numbers* subsection below).
318 2. Array indices (**I[E]**).
319 3. **(E)**: The value of **E** (used to change precedence).
320 4. **sqrt(E)**: The square root of **E**. **E** must be non-negative.
321 5. **length(E)**: The number of significant decimal digits in **E**.
322 6. **length(I[])**: The number of elements in the array **I**. This is a
323 **non-portable extension**.
324 7. **scale(E)**: The *scale* of **E**.
325 8. **abs(E)**: The absolute value of **E**. This is a **non-portable
327 9. **I()**, **I(E)**, **I(E, E)**, and so on, where **I** is an identifier for
328 a non-**void** function (see the *Void Functions* subsection of the
329 **FUNCTIONS** section). The **E** argument(s) may also be arrays of the form
330 **I[]**, which will automatically be turned into array references (see the
331 *Array References* subsection of the **FUNCTIONS** section) if the
332 corresponding parameter in the function definition is an array reference.
333 10. **read()**: Reads a line from **stdin** and uses that as an expression. The
334 result of that expression is the result of the **read()** operand. This is a
335 **non-portable extension**.
336 11. **maxibase()**: The max allowable **ibase**. This is a **non-portable
338 12. **maxobase()**: The max allowable **obase**. This is a **non-portable
340 13. **maxscale()**: The max allowable **scale**. This is a **non-portable
345 Numbers are strings made up of digits, uppercase letters, and at most **1**
346 period for a radix. Numbers can have up to **BC_NUM_MAX** digits. Uppercase
347 letters are equal to **9** + their position in the alphabet (i.e., **A** equals
348 **10**, or **9+1**). If a digit or letter makes no sense with the current value
349 of **ibase**, they are set to the value of the highest valid digit in **ibase**.
351 Single-character numbers (i.e., **A** alone) take the value that they would have
352 if they were valid digits, regardless of the value of **ibase**. This means that
353 **A** alone always equals decimal **10** and **Z** alone always equals decimal
358 The following arithmetic and logical operators can be used. They are listed in
359 order of decreasing precedence. Operators in the same group have the same
364 : Type: Prefix and Postfix
368 Description: **increment**, **decrement**
376 Description: **negation**, **boolean not**
384 Description: **power**
392 Description: **multiply**, **divide**, **modulus**
400 Description: **add**, **subtract**
402 **=** **+=** **-=** **\*=** **/=** **%=** **\^=**
408 Description: **assignment**
410 **==** **\<=** **\>=** **!=** **\<** **\>**
416 Description: **relational**
424 Description: **boolean and**
432 Description: **boolean or**
434 The operators will be described in more detail below.
438 : The prefix and postfix **increment** and **decrement** operators behave
439 exactly like they would in C. They require a named expression (see the
440 *Named Expressions* subsection) as an operand.
442 The prefix versions of these operators are more efficient; use them where
447 : The **negation** operator returns **0** if a user attempts to negate any
448 expression with the value **0**. Otherwise, a copy of the expression with
449 its sign flipped is returned.
453 : The **boolean not** operator returns **1** if the expression is **0**, or
456 This is a **non-portable extension**.
460 : The **power** operator (not the **exclusive or** operator, as it would be in
461 C) takes two expressions and raises the first to the power of the value of
462 the second. The *scale* of the result is equal to **scale**.
464 The second expression must be an integer (no *scale*), and if it is
465 negative, the first value must be non-zero.
469 : The **multiply** operator takes two expressions, multiplies them, and
470 returns the product. If **a** is the *scale* of the first expression and
471 **b** is the *scale* of the second expression, the *scale* of the result is
472 equal to **min(a+b,max(scale,a,b))** where **min()** and **max()** return
477 : The **divide** operator takes two expressions, divides them, and returns the
478 quotient. The *scale* of the result shall be the value of **scale**.
480 The second expression must be non-zero.
484 : The **modulus** operator takes two expressions, **a** and **b**, and
485 evaluates them by 1) Computing **a/b** to current **scale** and 2) Using the
486 result of step 1 to calculate **a-(a/b)\*b** to *scale*
487 **max(scale+scale(b),scale(a))**.
489 The second expression must be non-zero.
493 : The **add** operator takes two expressions, **a** and **b**, and returns the
494 sum, with a *scale* equal to the max of the *scale*s of **a** and **b**.
498 : The **subtract** operator takes two expressions, **a** and **b**, and
499 returns the difference, with a *scale* equal to the max of the *scale*s of
502 **=** **+=** **-=** **\*=** **/=** **%=** **\^=**
504 : The **assignment** operators take two expressions, **a** and **b** where
505 **a** is a named expression (see the *Named Expressions* subsection).
507 For **=**, **b** is copied and the result is assigned to **a**. For all
508 others, **a** and **b** are applied as operands to the corresponding
509 arithmetic operator and the result is assigned to **a**.
511 **==** **\<=** **\>=** **!=** **\<** **\>**
513 : The **relational** operators compare two expressions, **a** and **b**, and
514 if the relation holds, according to C language semantics, the result is
515 **1**. Otherwise, it is **0**.
517 Note that unlike in C, these operators have a lower precedence than the
518 **assignment** operators, which means that **a=b\>c** is interpreted as
521 Also, unlike the [standard][1] requires, these operators can appear anywhere
522 any other expressions can be used. This allowance is a
523 **non-portable extension**.
527 : The **boolean and** operator takes two expressions and returns **1** if both
528 expressions are non-zero, **0** otherwise.
530 This is *not* a short-circuit operator.
532 This is a **non-portable extension**.
536 : The **boolean or** operator takes two expressions and returns **1** if one
537 of the expressions is non-zero, **0** otherwise.
539 This is *not* a short-circuit operator.
541 This is a **non-portable extension**.
545 The following items are statements:
548 2. **{** **S** **;** ... **;** **S** **}**
549 3. **if** **(** **E** **)** **S**
550 4. **if** **(** **E** **)** **S** **else** **S**
551 5. **while** **(** **E** **)** **S**
552 6. **for** **(** **E** **;** **E** **;** **E** **)** **S**
553 7. An empty statement
559 13. A string of characters, enclosed in double quotes
560 14. **print** **E** **,** ... **,** **E**
561 15. **I()**, **I(E)**, **I(E, E)**, and so on, where **I** is an identifier for
562 a **void** function (see the *Void Functions* subsection of the
563 **FUNCTIONS** section). The **E** argument(s) may also be arrays of the form
564 **I[]**, which will automatically be turned into array references (see the
565 *Array References* subsection of the **FUNCTIONS** section) if the
566 corresponding parameter in the function definition is an array reference.
568 Numbers 4, 9, 11, 12, 14, and 15 are **non-portable extensions**.
570 Also, as a **non-portable extension**, any or all of the expressions in the
571 header of a for loop may be omitted. If the condition (second expression) is
572 omitted, it is assumed to be a constant **1**.
574 The **break** statement causes a loop to stop iterating and resume execution
575 immediately following a loop. This is only allowed in loops.
577 The **continue** statement causes a loop iteration to stop early and returns to
578 the start of the loop, including testing the loop condition. This is only
581 The **if** **else** statement does the same thing as in C.
583 The **quit** statement causes bc(1) to quit, even if it is on a branch that will
584 not be executed (it is a compile-time command).
586 The **halt** statement causes bc(1) to quit, if it is executed. (Unlike **quit**
587 if it is on a branch of an **if** statement that is not executed, bc(1) does not
590 The **limits** statement prints the limits that this bc(1) is subject to. This
591 is like the **quit** statement in that it is a compile-time command.
593 An expression by itself is evaluated and printed, followed by a newline.
597 The "expressions" in a **print** statement may also be strings. If they are, there
598 are backslash escape sequences that are interpreted specially. What those
599 sequences are, and what they cause to be printed, are shown below:
613 Any other character following a backslash causes the backslash and character to
616 Any non-string expression in a print statement shall be assigned to **last**,
617 like any other expression that is printed.
619 ## Order of Evaluation
621 All expressions in a statment are evaluated left to right, except as necessary
622 to maintain order of operations. This means, for example, assuming that **i** is
623 equal to **0**, in the expression
627 the first (or 0th) element of **a** is set to **1**, and **i** is equal to **2**
628 at the end of the expression.
630 This includes function arguments. Thus, assuming **i** is equal to **0**, this
631 means that in the expression
635 the first argument passed to **x()** is **0**, and the second argument is **1**,
636 while **i** is equal to **2** before the function starts executing.
640 Function definitions are as follows:
650 Any **I** in the parameter list or **auto** list may be replaced with **I[]** to
651 make a parameter or **auto** var an array, and any **I** in the parameter list
652 may be replaced with **\*I[]** to make a parameter an array reference. Callers
653 of functions that take array references should not put an asterisk in the call;
654 they must be called with just **I[]** like normal array parameters and will be
655 automatically converted into references.
657 As a **non-portable extension**, the opening brace of a **define** statement may
658 appear on the next line.
660 As a **non-portable extension**, the return statement may also be in one of the
664 2. **return** **(** **)**
667 The first two, or not specifying a **return** statement, is equivalent to
668 **return (0)**, unless the function is a **void** function (see the *Void
669 Functions* subsection below).
673 Functions can also be **void** functions, defined as follows:
676 define void I(I,...,I){
683 They can only be used as standalone expressions, where such an expression would
684 be printed alone, except in a print statement.
686 Void functions can only use the first two **return** statements listed above.
687 They can also omit the return statement entirely.
689 The word "void" is not treated as a keyword; it is still possible to have
690 variables, arrays, and functions named **void**. The word "void" is only
691 treated specially right after the **define** keyword.
693 This is a **non-portable extension**.
697 For any array in the parameter list, if the array is declared in the form
703 it is a **reference**. Any changes to the array in the function are reflected,
704 when the function returns, to the array that was passed in.
706 Other than this, all function arguments are passed by value.
708 This is a **non-portable extension**.
712 All of the functions below are available when the **-l** or **--mathlib**
713 command-line flags are given.
717 The [standard][1] defines the following functions for the math library:
721 : Returns the sine of **x**, which is assumed to be in radians.
723 This is a transcendental function (see the *Transcendental Functions*
728 : Returns the cosine of **x**, which is assumed to be in radians.
730 This is a transcendental function (see the *Transcendental Functions*
735 : Returns the arctangent of **x**, in radians.
737 This is a transcendental function (see the *Transcendental Functions*
742 : Returns the natural logarithm of **x**.
744 This is a transcendental function (see the *Transcendental Functions*
749 : Returns the mathematical constant **e** raised to the power of **x**.
751 This is a transcendental function (see the *Transcendental Functions*
756 : Returns the bessel integer order **n** (truncated) of **x**.
758 This is a transcendental function (see the *Transcendental Functions*
761 ## Transcendental Functions
763 All transcendental functions can return slightly inaccurate results (up to 1
764 [ULP][4]). This is unavoidable, and [this article][5] explains why it is
765 impossible and unnecessary to calculate exact results for the transcendental
768 Because of the possible inaccuracy, I recommend that users call those functions
769 with the precision (**scale**) set to at least 1 higher than is necessary. If
770 exact results are *absolutely* required, users can double the precision
771 (**scale**) and then truncate.
773 The transcendental functions in the standard math library are:
784 When bc(1) encounters an error or a signal that it has a non-default handler
785 for, it resets. This means that several things happen.
787 First, any functions that are executing are stopped and popped off the stack.
788 The behavior is not unlike that of exceptions in programming languages. Then
789 the execution point is set so that any code waiting to execute (after all
790 functions returned) is skipped.
792 Thus, when bc(1) resets, it skips any remaining code waiting to be executed.
793 Then, if it is interactive mode, and the error was not a fatal error (see the
794 **EXIT STATUS** section), it asks for more input; otherwise, it exits with the
795 appropriate return code.
797 Note that this reset behavior is different from the GNU bc(1), which attempts to
798 start executing the statement right after the one that caused an error.
802 Most bc(1) implementations use **char** types to calculate the value of **1**
803 decimal digit at a time, but that can be slow. This bc(1) does something
806 It uses large integers to calculate more than **1** decimal digit at a time. If
807 built in a environment where **BC_LONG_BIT** (see the **LIMITS** section) is
808 **64**, then each integer has **9** decimal digits. If built in an environment
809 where **BC_LONG_BIT** is **32** then each integer has **4** decimal digits. This
810 value (the number of decimal digits per large integer) is called
813 The actual values of **BC_LONG_BIT** and **BC_BASE_DIGS** can be queried with
814 the **limits** statement.
816 In addition, this bc(1) uses an even larger integer for overflow checking. This
817 integer type depends on the value of **BC_LONG_BIT**, but is always at least
818 twice as large as the integer type used to store digits.
822 The following are the limits on bc(1):
826 : The number of bits in the **long** type in the environment where bc(1) was
827 built. This determines how many decimal digits can be stored in a single
828 large integer (see the **PERFORMANCE** section).
832 : The number of decimal digits per large integer (see the **PERFORMANCE**
833 section). Depends on **BC_LONG_BIT**.
837 : The max decimal number that each large integer can store (see
838 **BC_BASE_DIGS**) plus **1**. Depends on **BC_BASE_DIGS**.
842 : The max number that the overflow type (see the **PERFORMANCE** section) can
843 hold. Depends on **BC_LONG_BIT**.
847 : The maximum output base. Set at **BC_BASE_POW**.
851 : The maximum size of arrays. Set at **SIZE_MAX-1**.
855 : The maximum **scale**. Set at **BC_OVERFLOW_MAX-1**.
859 : The maximum length of strings. Set at **BC_OVERFLOW_MAX-1**.
863 : The maximum length of identifiers. Set at **BC_OVERFLOW_MAX-1**.
867 : The maximum length of a number (in decimal digits), which includes digits
868 after the decimal point. Set at **BC_OVERFLOW_MAX-1**.
872 : The maximum allowable exponent (positive or negative). Set at
877 : The maximum number of vars/arrays. Set at **SIZE_MAX-1**.
879 The actual values can be queried with the **limits** statement.
881 These limits are meant to be effectively non-existent; the limits are so large
882 (at least on 64-bit machines) that there should not be any point at which they
883 become a problem. In fact, memory should be exhausted before these limits should
886 # ENVIRONMENT VARIABLES
888 bc(1) recognizes the following environment variables:
892 : If this variable exists (no matter the contents), bc(1) behaves as if
893 the **-s** option was given.
897 : This is another way to give command-line arguments to bc(1). They should be
898 in the same format as all other command-line arguments. These are always
899 processed first, so any files given in **BC_ENV_ARGS** will be processed
900 before arguments and files given on the command-line. This gives the user
901 the ability to set up "standard" options and files to be used at every
902 invocation. The most useful thing for such files to contain would be useful
903 functions that the user might want every time bc(1) runs.
905 The code that parses **BC_ENV_ARGS** will correctly handle quoted arguments,
906 but it does not understand escape sequences. For example, the string
907 **"/home/gavin/some bc file.bc"** will be correctly parsed, but the string
908 **"/home/gavin/some \"bc\" file.bc"** will include the backslashes.
910 The quote parsing will handle either kind of quotes, **'** or **"**. Thus,
911 if you have a file with any number of single quotes in the name, you can use
912 double quotes as the outside quotes, as in **"some 'bc' file.bc"**, and vice
913 versa if you have a file with double quotes. However, handling a file with
914 both kinds of quotes in **BC_ENV_ARGS** is not supported due to the
915 complexity of the parsing, though such files are still supported on the
916 command-line where the parsing is done by the shell.
920 : If this environment variable exists and contains an integer that is greater
921 than **1** and is less than **UINT16_MAX** (**2\^16-1**), bc(1) will output
922 lines to that length, including the backslash (**\\**). The default line
927 bc(1) returns the following exit statuses:
935 : A math error occurred. This follows standard practice of using **1** for
936 expected errors, since math errors will happen in the process of normal
939 Math errors include divide by **0**, taking the square root of a negative
940 number, attempting to convert a negative number to a hardware integer,
941 overflow when converting a number to a hardware integer, and attempting to
942 use a non-integer where an integer is required.
944 Converting to a hardware integer happens for the second operand of the power
945 (**\^**) operator and the corresponding assignment operator.
949 : A parse error occurred.
951 Parse errors include unexpected **EOF**, using an invalid character, failing
952 to find the end of a string or comment, using a token where it is invalid,
953 giving an invalid expression, giving an invalid print statement, giving an
954 invalid function definition, attempting to assign to an expression that is
955 not a named expression (see the *Named Expressions* subsection of the
956 **SYNTAX** section), giving an invalid **auto** list, having a duplicate
957 **auto**/function parameter, failing to find the end of a code block,
958 attempting to return a value from a **void** function, attempting to use a
959 variable as a reference, and using any extensions when the option **-s** or
960 any equivalents were given.
964 : A runtime error occurred.
966 Runtime errors include assigning an invalid number to **ibase**, **obase**,
967 or **scale**; give a bad expression to a **read()** call, calling **read()**
968 inside of a **read()** call, type errors, passing the wrong number of
969 arguments to functions, attempting to call an undefined function, and
970 attempting to use a **void** function call as a value in an expression.
974 : A fatal error occurred.
976 Fatal errors include memory allocation errors, I/O errors, failing to open
977 files, attempting to use files that do not have only ASCII characters (bc(1)
978 only accepts ASCII characters), attempting to open a directory as a file,
979 and giving invalid command-line options.
981 The exit status **4** is special; when a fatal error occurs, bc(1) always exits
982 and returns **4**, no matter what mode bc(1) is in.
984 The other statuses will only be returned when bc(1) is not in interactive mode
985 (see the **INTERACTIVE MODE** section), since bc(1) resets its state (see the
986 **RESET** section) and accepts more input when one of those errors occurs in
987 interactive mode. This is also the case when interactive mode is forced by the
988 **-i** flag or **--interactive** option.
990 These exit statuses allow bc(1) to be used in shell scripting with error
991 checking, and its normal behavior can be forced by using the **-i** flag or
992 **--interactive** option.
996 Per the [standard][1], bc(1) has an interactive mode and a non-interactive mode.
997 Interactive mode is turned on automatically when both **stdin** and **stdout**
998 are hooked to a terminal, but the **-i** flag and **--interactive** option can
999 turn it on in other cases.
1001 In interactive mode, bc(1) attempts to recover from errors (see the **RESET**
1002 section), and in normal execution, flushes **stdout** as soon as execution is
1003 done for the current input.
1007 If **stdin**, **stdout**, and **stderr** are all connected to a TTY, bc(1) turns
1010 TTY mode is different from interactive mode because interactive mode is required
1011 in the [bc(1) specification][1], and interactive mode requires only **stdin**
1012 and **stdout** to be connected to a terminal.
1016 Sending a **SIGINT** will cause bc(1) to stop execution of the current input. If
1017 bc(1) is in TTY mode (see the **TTY MODE** section), it will reset (see the
1018 **RESET** section). Otherwise, it will clean up and exit.
1020 Note that "current input" can mean one of two things. If bc(1) is processing
1021 input from **stdin** in TTY mode, it will ask for more input. If bc(1) is
1022 processing input from a file in TTY mode, it will stop processing the file and
1023 start processing the next file, if one exists, or ask for input from **stdin**
1024 if no other file exists.
1026 This means that if a **SIGINT** is sent to bc(1) as it is executing a file, it
1027 can seem as though bc(1) did not respond to the signal since it will immediately
1028 start executing the next file. This is by design; most files that users execute
1029 when interacting with bc(1) have function definitions, which are quick to parse.
1030 If a file takes a long time to execute, there may be a bug in that file. The
1031 rest of the files could still be executed without problem, allowing the user to
1034 **SIGTERM** and **SIGQUIT** cause bc(1) to clean up and exit, and it uses the
1035 default handler for all other signals.
1039 This bc(1) ships with support for adding error messages for different locales
1040 and thus, supports **LC_MESSAGES**.
1048 bc(1) is compliant with the [IEEE Std 1003.1-2017 (“POSIX.1-2017”)][1]
1049 specification. The flags **-efghiqsvVw**, all long options, and the extensions
1050 noted above are extensions to that specification.
1052 Note that the specification explicitly says that bc(1) only accepts numbers that
1053 use a period (**.**) as a radix point, regardless of the value of
1056 This bc(1) supports error messages for different locales, and thus, it supports
1061 None are known. Report bugs at https://git.yzena.com/gavin/bc.
1065 Gavin D. Howard <gavin@yzena.com> and contributors.
1067 [1]: https://pubs.opengroup.org/onlinepubs/9699919799/utilities/bc.html
1068 [2]: https://www.gnu.org/software/bc/
1069 [3]: https://en.wikipedia.org/wiki/Rounding#Round_half_away_from_zero
1070 [4]: https://en.wikipedia.org/wiki/Unit_in_the_last_place
1071 [5]: https://people.eecs.berkeley.edu/~wkahan/LOG10HAF.TXT
1072 [6]: https://en.wikipedia.org/wiki/Rounding#Rounding_away_from_zero