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*...] [*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**.
135 : This option is for compatibility with the [GNU bc(1)][2]; it is a no-op.
136 Without this option, GNU bc(1) prints a copyright header. This bc(1) only
137 prints the copyright header if one or more of the **-v**, **-V**, or
138 **-\-version** options are given.
140 This is a **non-portable extension**.
142 **-s**, **-\-standard**
144 : Process exactly the language defined by the [standard][1] and error if any
147 This is a **non-portable extension**.
149 **-v**, **-V**, **-\-version**
151 : Print the version information (copyright header) and exit.
153 This is a **non-portable extension**.
157 : Like **-s** and **-\-standard**, except that warnings (and not errors) are
158 printed for non-standard extensions and execution continues normally.
160 This is a **non-portable extension**.
162 **-e** *expr*, **-\-expression**=*expr*
164 : Evaluates *expr*. If multiple expressions are given, they are evaluated in
165 order. If files are given as well (see below), the expressions and files are
166 evaluated in the order given. This means that if a file is given before an
167 expression, the file is read in and evaluated first.
169 If this option is given on the command-line (i.e., not in **BC_ENV_ARGS**,
170 see the **ENVIRONMENT VARIABLES** section), then after processing all
171 expressions and files, bc(1) will exit, unless **-** (**stdin**) was given
172 as an argument at least once to **-f** or **-\-file**, whether on the
173 command-line or in **BC_ENV_ARGS**. However, if any other **-e**,
174 **-\-expression**, **-f**, or **-\-file** arguments are given after **-f-**
175 or equivalent is given, bc(1) will give a fatal error and exit.
177 This is a **non-portable extension**.
179 **-f** *file*, **-\-file**=*file*
181 : Reads in *file* and evaluates it, line by line, as though it were read
182 through **stdin**. If expressions are also given (see above), the
183 expressions are evaluated in the order given.
185 If this option is given on the command-line (i.e., not in **BC_ENV_ARGS**,
186 see the **ENVIRONMENT VARIABLES** section), then after processing all
187 expressions and files, bc(1) will exit, unless **-** (**stdin**) was given
188 as an argument at least once to **-f** or **-\-file**. However, if any other
189 **-e**, **-\-expression**, **-f**, or **-\-file** arguments are given after
190 **-f-** or equivalent is given, bc(1) will give a fatal error and exit.
192 This is a **non-portable extension**.
194 All long options are **non-portable extensions**.
198 Any non-error output is written to **stdout**. In addition, if history (see the
199 **HISTORY** section) and the prompt (see the **TTY MODE** section) are enabled,
200 both are output to **stdout**.
202 **Note**: Unlike other bc(1) implementations, this bc(1) will issue a fatal
203 error (see the **EXIT STATUS** section) if it cannot write to **stdout**, so if
204 **stdout** is closed, as in **bc <file> >&-**, it will quit with an error. This
205 is done so that bc(1) can report problems when **stdout** is redirected to a
208 If there are scripts that depend on the behavior of other bc(1) implementations,
209 it is recommended that those scripts be changed to redirect **stdout** to
214 Any error output is written to **stderr**.
216 **Note**: Unlike other bc(1) implementations, this bc(1) will issue a fatal
217 error (see the **EXIT STATUS** section) if it cannot write to **stderr**, so if
218 **stderr** is closed, as in **bc <file> 2>&-**, it will quit with an error. This
219 is done so that bc(1) can exit with an error code when **stderr** is redirected
222 If there are scripts that depend on the behavior of other bc(1) implementations,
223 it is recommended that those scripts be changed to redirect **stderr** to
228 The syntax for bc(1) programs is mostly C-like, with some differences. This
229 bc(1) follows the [POSIX standard][1], which is a much more thorough resource
230 for the language this bc(1) accepts. This section is meant to be a summary and a
231 listing of all the extensions to the standard.
233 In the sections below, **E** means expression, **S** means statement, and **I**
236 Identifiers (**I**) start with a lowercase letter and can be followed by any
237 number (up to **BC_NAME_MAX-1**) of lowercase letters (**a-z**), digits
238 (**0-9**), and underscores (**\_**). The regex is **\[a-z\]\[a-z0-9\_\]\***.
239 Identifiers with more than one character (letter) are a
240 **non-portable extension**.
242 **ibase** is a global variable determining how to interpret constant numbers. It
243 is the "input" base, or the number base used for interpreting input numbers.
244 **ibase** is initially **10**. If the **-s** (**-\-standard**) and **-w**
245 (**-\-warn**) flags were not given on the command line, the max allowable value
246 for **ibase** is **36**. Otherwise, it is **16**. The min allowable value for
247 **ibase** is **2**. The max allowable value for **ibase** can be queried in
248 bc(1) programs with the **maxibase()** built-in function.
250 **obase** is a global variable determining how to output results. It is the
251 "output" base, or the number base used for outputting numbers. **obase** is
252 initially **10**. The max allowable value for **obase** is **BC_BASE_MAX** and
253 can be queried in bc(1) programs with the **maxobase()** built-in function. The
254 min allowable value for **obase** is **2**. Values are output in the specified
257 The *scale* of an expression is the number of digits in the result of the
258 expression right of the decimal point, and **scale** is a global variable that
259 sets the precision of any operations, with exceptions. **scale** is initially
260 **0**. **scale** cannot be negative. The max allowable value for **scale** is
261 **BC_SCALE_MAX** and can be queried in bc(1) programs with the **maxscale()**
264 bc(1) has both *global* variables and *local* variables. All *local*
265 variables are local to the function; they are parameters or are introduced in
266 the **auto** list of a function (see the **FUNCTIONS** section). If a variable
267 is accessed which is not a parameter or in the **auto** list, it is assumed to
268 be *global*. If a parent function has a *local* variable version of a variable
269 that a child function considers *global*, the value of that *global* variable in
270 the child function is the value of the variable in the parent function, not the
271 value of the actual *global* variable.
273 All of the above applies to arrays as well.
275 The value of a statement that is an expression (i.e., any of the named
276 expressions or operands) is printed unless the lowest precedence operator is an
277 assignment operator *and* the expression is notsurrounded by parentheses.
279 The value that is printed is also assigned to the special variable **last**. A
280 single dot (**.**) may also be used as a synonym for **last**. These are
281 **non-portable extensions**.
283 Either semicolons or newlines may separate statements.
287 There are two kinds of comments:
289 1. Block comments are enclosed in **/\*** and **\*/**.
290 2. Line comments go from **#** until, and not including, the next newline. This
291 is a **non-portable extension**.
295 The following are named expressions in bc(1):
298 2. Array Elements: **I[E]**
302 6. **last** or a single dot (**.**)
304 Number 6 is a **non-portable extension**.
306 Variables and arrays do not interfere; users can have arrays named the same as
307 variables. This also applies to functions (see the **FUNCTIONS** section), so a
308 user can have a variable, array, and function that all have the same name, and
309 they will not shadow each other, whether inside of functions or not.
311 Named expressions are required as the operand of **increment**/**decrement**
312 operators and as the left side of **assignment** operators (see the *Operators*
317 The following are valid operands in bc(1):
319 1. Numbers (see the *Numbers* subsection below).
320 2. Array indices (**I[E]**).
321 3. **(E)**: The value of **E** (used to change precedence).
322 4. **sqrt(E)**: The square root of **E**. **E** must be non-negative.
323 5. **length(E)**: The number of significant decimal digits in **E**.
324 6. **length(I[])**: The number of elements in the array **I**. This is a
325 **non-portable extension**.
326 7. **scale(E)**: The *scale* of **E**.
327 8. **abs(E)**: The absolute value of **E**. This is a **non-portable
329 9. **I()**, **I(E)**, **I(E, E)**, and so on, where **I** is an identifier for
330 a non-**void** function (see the *Void Functions* subsection of the
331 **FUNCTIONS** section). The **E** argument(s) may also be arrays of the form
332 **I[]**, which will automatically be turned into array references (see the
333 *Array References* subsection of the **FUNCTIONS** section) if the
334 corresponding parameter in the function definition is an array reference.
335 10. **read()**: Reads a line from **stdin** and uses that as an expression. The
336 result of that expression is the result of the **read()** operand. This is a
337 **non-portable extension**.
338 11. **maxibase()**: The max allowable **ibase**. This is a **non-portable
340 12. **maxobase()**: The max allowable **obase**. This is a **non-portable
342 13. **maxscale()**: The max allowable **scale**. This is a **non-portable
347 Numbers are strings made up of digits, uppercase letters, and at most **1**
348 period for a radix. Numbers can have up to **BC_NUM_MAX** digits. Uppercase
349 letters are equal to **9** + their position in the alphabet (i.e., **A** equals
350 **10**, or **9+1**). If a digit or letter makes no sense with the current value
351 of **ibase**, they are set to the value of the highest valid digit in **ibase**.
353 Single-character numbers (i.e., **A** alone) take the value that they would have
354 if they were valid digits, regardless of the value of **ibase**. This means that
355 **A** alone always equals decimal **10** and **Z** alone always equals decimal
360 The following arithmetic and logical operators can be used. They are listed in
361 order of decreasing precedence. Operators in the same group have the same
366 : Type: Prefix and Postfix
370 Description: **increment**, **decrement**
378 Description: **negation**, **boolean not**
386 Description: **power**
394 Description: **multiply**, **divide**, **modulus**
402 Description: **add**, **subtract**
404 **=** **+=** **-=** **\*=** **/=** **%=** **\^=**
410 Description: **assignment**
412 **==** **\<=** **\>=** **!=** **\<** **\>**
418 Description: **relational**
426 Description: **boolean and**
434 Description: **boolean or**
436 The operators will be described in more detail below.
440 : The prefix and postfix **increment** and **decrement** operators behave
441 exactly like they would in C. They require a named expression (see the
442 *Named Expressions* subsection) as an operand.
444 The prefix versions of these operators are more efficient; use them where
449 : The **negation** operator returns **0** if a user attempts to negate any
450 expression with the value **0**. Otherwise, a copy of the expression with
451 its sign flipped is returned.
455 : The **boolean not** operator returns **1** if the expression is **0**, or
458 This is a **non-portable extension**.
462 : The **power** operator (not the **exclusive or** operator, as it would be in
463 C) takes two expressions and raises the first to the power of the value of
464 the second. The *scale* of the result is equal to **scale**.
466 The second expression must be an integer (no *scale*), and if it is
467 negative, the first value must be non-zero.
471 : The **multiply** operator takes two expressions, multiplies them, and
472 returns the product. If **a** is the *scale* of the first expression and
473 **b** is the *scale* of the second expression, the *scale* of the result is
474 equal to **min(a+b,max(scale,a,b))** where **min()** and **max()** return
479 : The **divide** operator takes two expressions, divides them, and returns the
480 quotient. The *scale* of the result shall be the value of **scale**.
482 The second expression must be non-zero.
486 : The **modulus** operator takes two expressions, **a** and **b**, and
487 evaluates them by 1) Computing **a/b** to current **scale** and 2) Using the
488 result of step 1 to calculate **a-(a/b)\*b** to *scale*
489 **max(scale+scale(b),scale(a))**.
491 The second expression must be non-zero.
495 : The **add** operator takes two expressions, **a** and **b**, and returns the
496 sum, with a *scale* equal to the max of the *scale*s of **a** and **b**.
500 : The **subtract** operator takes two expressions, **a** and **b**, and
501 returns the difference, with a *scale* equal to the max of the *scale*s of
504 **=** **+=** **-=** **\*=** **/=** **%=** **\^=**
506 : The **assignment** operators take two expressions, **a** and **b** where
507 **a** is a named expression (see the *Named Expressions* subsection).
509 For **=**, **b** is copied and the result is assigned to **a**. For all
510 others, **a** and **b** are applied as operands to the corresponding
511 arithmetic operator and the result is assigned to **a**.
513 **==** **\<=** **\>=** **!=** **\<** **\>**
515 : The **relational** operators compare two expressions, **a** and **b**, and
516 if the relation holds, according to C language semantics, the result is
517 **1**. Otherwise, it is **0**.
519 Note that unlike in C, these operators have a lower precedence than the
520 **assignment** operators, which means that **a=b\>c** is interpreted as
523 Also, unlike the [standard][1] requires, these operators can appear anywhere
524 any other expressions can be used. This allowance is a
525 **non-portable extension**.
529 : The **boolean and** operator takes two expressions and returns **1** if both
530 expressions are non-zero, **0** otherwise.
532 This is *not* a short-circuit operator.
534 This is a **non-portable extension**.
538 : The **boolean or** operator takes two expressions and returns **1** if one
539 of the expressions is non-zero, **0** otherwise.
541 This is *not* a short-circuit operator.
543 This is a **non-portable extension**.
547 The following items are statements:
550 2. **{** **S** **;** ... **;** **S** **}**
551 3. **if** **(** **E** **)** **S**
552 4. **if** **(** **E** **)** **S** **else** **S**
553 5. **while** **(** **E** **)** **S**
554 6. **for** **(** **E** **;** **E** **;** **E** **)** **S**
555 7. An empty statement
561 13. A string of characters, enclosed in double quotes
562 14. **print** **E** **,** ... **,** **E**
563 15. **I()**, **I(E)**, **I(E, E)**, and so on, where **I** is an identifier for
564 a **void** function (see the *Void Functions* subsection of the
565 **FUNCTIONS** section). The **E** argument(s) may also be arrays of the form
566 **I[]**, which will automatically be turned into array references (see the
567 *Array References* subsection of the **FUNCTIONS** section) if the
568 corresponding parameter in the function definition is an array reference.
570 Numbers 4, 9, 11, 12, 14, and 15 are **non-portable extensions**.
572 Also, as a **non-portable extension**, any or all of the expressions in the
573 header of a for loop may be omitted. If the condition (second expression) is
574 omitted, it is assumed to be a constant **1**.
576 The **break** statement causes a loop to stop iterating and resume execution
577 immediately following a loop. This is only allowed in loops.
579 The **continue** statement causes a loop iteration to stop early and returns to
580 the start of the loop, including testing the loop condition. This is only
583 The **if** **else** statement does the same thing as in C.
585 The **quit** statement causes bc(1) to quit, even if it is on a branch that will
586 not be executed (it is a compile-time command).
588 The **halt** statement causes bc(1) to quit, if it is executed. (Unlike **quit**
589 if it is on a branch of an **if** statement that is not executed, bc(1) does not
592 The **limits** statement prints the limits that this bc(1) is subject to. This
593 is like the **quit** statement in that it is a compile-time command.
595 An expression by itself is evaluated and printed, followed by a newline.
599 The "expressions" in a **print** statement may also be strings. If they are, there
600 are backslash escape sequences that are interpreted specially. What those
601 sequences are, and what they cause to be printed, are shown below:
615 Any other character following a backslash causes the backslash and character to
618 Any non-string expression in a print statement shall be assigned to **last**,
619 like any other expression that is printed.
621 ## Order of Evaluation
623 All expressions in a statment are evaluated left to right, except as necessary
624 to maintain order of operations. This means, for example, assuming that **i** is
625 equal to **0**, in the expression
629 the first (or 0th) element of **a** is set to **1**, and **i** is equal to **2**
630 at the end of the expression.
632 This includes function arguments. Thus, assuming **i** is equal to **0**, this
633 means that in the expression
637 the first argument passed to **x()** is **0**, and the second argument is **1**,
638 while **i** is equal to **2** before the function starts executing.
642 Function definitions are as follows:
652 Any **I** in the parameter list or **auto** list may be replaced with **I[]** to
653 make a parameter or **auto** var an array, and any **I** in the parameter list
654 may be replaced with **\*I[]** to make a parameter an array reference. Callers
655 of functions that take array references should not put an asterisk in the call;
656 they must be called with just **I[]** like normal array parameters and will be
657 automatically converted into references.
659 As a **non-portable extension**, the opening brace of a **define** statement may
660 appear on the next line.
662 As a **non-portable extension**, the return statement may also be in one of the
666 2. **return** **(** **)**
669 The first two, or not specifying a **return** statement, is equivalent to
670 **return (0)**, unless the function is a **void** function (see the *Void
671 Functions* subsection below).
675 Functions can also be **void** functions, defined as follows:
678 define void I(I,...,I){
685 They can only be used as standalone expressions, where such an expression would
686 be printed alone, except in a print statement.
688 Void functions can only use the first two **return** statements listed above.
689 They can also omit the return statement entirely.
691 The word "void" is not treated as a keyword; it is still possible to have
692 variables, arrays, and functions named **void**. The word "void" is only
693 treated specially right after the **define** keyword.
695 This is a **non-portable extension**.
699 For any array in the parameter list, if the array is declared in the form
705 it is a **reference**. Any changes to the array in the function are reflected,
706 when the function returns, to the array that was passed in.
708 Other than this, all function arguments are passed by value.
710 This is a **non-portable extension**.
714 All of the functions below are available when the **-l** or **-\-mathlib**
715 command-line flags are given.
719 The [standard][1] defines the following functions for the math library:
723 : Returns the sine of **x**, which is assumed to be in radians.
725 This is a transcendental function (see the *Transcendental Functions*
730 : Returns the cosine of **x**, which is assumed to be in radians.
732 This is a transcendental function (see the *Transcendental Functions*
737 : Returns the arctangent of **x**, in radians.
739 This is a transcendental function (see the *Transcendental Functions*
744 : Returns the natural logarithm of **x**.
746 This is a transcendental function (see the *Transcendental Functions*
751 : Returns the mathematical constant **e** raised to the power of **x**.
753 This is a transcendental function (see the *Transcendental Functions*
758 : Returns the bessel integer order **n** (truncated) of **x**.
760 This is a transcendental function (see the *Transcendental Functions*
763 ## Transcendental Functions
765 All transcendental functions can return slightly inaccurate results (up to 1
766 [ULP][4]). This is unavoidable, and [this article][5] explains why it is
767 impossible and unnecessary to calculate exact results for the transcendental
770 Because of the possible inaccuracy, I recommend that users call those functions
771 with the precision (**scale**) set to at least 1 higher than is necessary. If
772 exact results are *absolutely* required, users can double the precision
773 (**scale**) and then truncate.
775 The transcendental functions in the standard math library are:
786 When bc(1) encounters an error or a signal that it has a non-default handler
787 for, it resets. This means that several things happen.
789 First, any functions that are executing are stopped and popped off the stack.
790 The behavior is not unlike that of exceptions in programming languages. Then
791 the execution point is set so that any code waiting to execute (after all
792 functions returned) is skipped.
794 Thus, when bc(1) resets, it skips any remaining code waiting to be executed.
795 Then, if it is interactive mode, and the error was not a fatal error (see the
796 **EXIT STATUS** section), it asks for more input; otherwise, it exits with the
797 appropriate return code.
799 Note that this reset behavior is different from the GNU bc(1), which attempts to
800 start executing the statement right after the one that caused an error.
804 Most bc(1) implementations use **char** types to calculate the value of **1**
805 decimal digit at a time, but that can be slow. This bc(1) does something
808 It uses large integers to calculate more than **1** decimal digit at a time. If
809 built in a environment where **BC_LONG_BIT** (see the **LIMITS** section) is
810 **64**, then each integer has **9** decimal digits. If built in an environment
811 where **BC_LONG_BIT** is **32** then each integer has **4** decimal digits. This
812 value (the number of decimal digits per large integer) is called
815 The actual values of **BC_LONG_BIT** and **BC_BASE_DIGS** can be queried with
816 the **limits** statement.
818 In addition, this bc(1) uses an even larger integer for overflow checking. This
819 integer type depends on the value of **BC_LONG_BIT**, but is always at least
820 twice as large as the integer type used to store digits.
824 The following are the limits on bc(1):
828 : The number of bits in the **long** type in the environment where bc(1) was
829 built. This determines how many decimal digits can be stored in a single
830 large integer (see the **PERFORMANCE** section).
834 : The number of decimal digits per large integer (see the **PERFORMANCE**
835 section). Depends on **BC_LONG_BIT**.
839 : The max decimal number that each large integer can store (see
840 **BC_BASE_DIGS**) plus **1**. Depends on **BC_BASE_DIGS**.
844 : The max number that the overflow type (see the **PERFORMANCE** section) can
845 hold. Depends on **BC_LONG_BIT**.
849 : The maximum output base. Set at **BC_BASE_POW**.
853 : The maximum size of arrays. Set at **SIZE_MAX-1**.
857 : The maximum **scale**. Set at **BC_OVERFLOW_MAX-1**.
861 : The maximum length of strings. Set at **BC_OVERFLOW_MAX-1**.
865 : The maximum length of identifiers. Set at **BC_OVERFLOW_MAX-1**.
869 : The maximum length of a number (in decimal digits), which includes digits
870 after the decimal point. Set at **BC_OVERFLOW_MAX-1**.
874 : The maximum allowable exponent (positive or negative). Set at
879 : The maximum number of vars/arrays. Set at **SIZE_MAX-1**.
881 The actual values can be queried with the **limits** statement.
883 These limits are meant to be effectively non-existent; the limits are so large
884 (at least on 64-bit machines) that there should not be any point at which they
885 become a problem. In fact, memory should be exhausted before these limits should
888 # ENVIRONMENT VARIABLES
890 bc(1) recognizes the following environment variables:
894 : If this variable exists (no matter the contents), bc(1) behaves as if
895 the **-s** option was given.
899 : This is another way to give command-line arguments to bc(1). They should be
900 in the same format as all other command-line arguments. These are always
901 processed first, so any files given in **BC_ENV_ARGS** will be processed
902 before arguments and files given on the command-line. This gives the user
903 the ability to set up "standard" options and files to be used at every
904 invocation. The most useful thing for such files to contain would be useful
905 functions that the user might want every time bc(1) runs.
907 The code that parses **BC_ENV_ARGS** will correctly handle quoted arguments,
908 but it does not understand escape sequences. For example, the string
909 **"/home/gavin/some bc file.bc"** will be correctly parsed, but the string
910 **"/home/gavin/some \"bc\" file.bc"** will include the backslashes.
912 The quote parsing will handle either kind of quotes, **'** or **"**. Thus,
913 if you have a file with any number of single quotes in the name, you can use
914 double quotes as the outside quotes, as in **"some 'bc' file.bc"**, and vice
915 versa if you have a file with double quotes. However, handling a file with
916 both kinds of quotes in **BC_ENV_ARGS** is not supported due to the
917 complexity of the parsing, though such files are still supported on the
918 command-line where the parsing is done by the shell.
922 : If this environment variable exists and contains an integer that is greater
923 than **1** and is less than **UINT16_MAX** (**2\^16-1**), bc(1) will output
924 lines to that length, including the backslash (**\\**). The default line
929 bc(1) returns the following exit statuses:
937 : A math error occurred. This follows standard practice of using **1** for
938 expected errors, since math errors will happen in the process of normal
941 Math errors include divide by **0**, taking the square root of a negative
942 number, attempting to convert a negative number to a hardware integer,
943 overflow when converting a number to a hardware integer, and attempting to
944 use a non-integer where an integer is required.
946 Converting to a hardware integer happens for the second operand of the power
947 (**\^**) operator and the corresponding assignment operator.
951 : A parse error occurred.
953 Parse errors include unexpected **EOF**, using an invalid character, failing
954 to find the end of a string or comment, using a token where it is invalid,
955 giving an invalid expression, giving an invalid print statement, giving an
956 invalid function definition, attempting to assign to an expression that is
957 not a named expression (see the *Named Expressions* subsection of the
958 **SYNTAX** section), giving an invalid **auto** list, having a duplicate
959 **auto**/function parameter, failing to find the end of a code block,
960 attempting to return a value from a **void** function, attempting to use a
961 variable as a reference, and using any extensions when the option **-s** or
962 any equivalents were given.
966 : A runtime error occurred.
968 Runtime errors include assigning an invalid number to **ibase**, **obase**,
969 or **scale**; give a bad expression to a **read()** call, calling **read()**
970 inside of a **read()** call, type errors, passing the wrong number of
971 arguments to functions, attempting to call an undefined function, and
972 attempting to use a **void** function call as a value in an expression.
976 : A fatal error occurred.
978 Fatal errors include memory allocation errors, I/O errors, failing to open
979 files, attempting to use files that do not have only ASCII characters (bc(1)
980 only accepts ASCII characters), attempting to open a directory as a file,
981 and giving invalid command-line options.
983 The exit status **4** is special; when a fatal error occurs, bc(1) always exits
984 and returns **4**, no matter what mode bc(1) is in.
986 The other statuses will only be returned when bc(1) is not in interactive mode
987 (see the **INTERACTIVE MODE** section), since bc(1) resets its state (see the
988 **RESET** section) and accepts more input when one of those errors occurs in
989 interactive mode. This is also the case when interactive mode is forced by the
990 **-i** flag or **-\-interactive** option.
992 These exit statuses allow bc(1) to be used in shell scripting with error
993 checking, and its normal behavior can be forced by using the **-i** flag or
994 **-\-interactive** option.
998 Per the [standard][1], bc(1) has an interactive mode and a non-interactive mode.
999 Interactive mode is turned on automatically when both **stdin** and **stdout**
1000 are hooked to a terminal, but the **-i** flag and **-\-interactive** option can
1001 turn it on in other cases.
1003 In interactive mode, bc(1) attempts to recover from errors (see the **RESET**
1004 section), and in normal execution, flushes **stdout** as soon as execution is
1005 done for the current input.
1009 If **stdin**, **stdout**, and **stderr** are all connected to a TTY, bc(1) turns
1012 TTY mode is required for history to be enabled (see the **COMMAND LINE HISTORY**
1013 section). It is also required to enable special handling for **SIGINT** signals.
1015 TTY mode is different from interactive mode because interactive mode is required
1016 in the [bc(1) specification][1], and interactive mode requires only **stdin**
1017 and **stdout** to be connected to a terminal.
1021 Sending a **SIGINT** will cause bc(1) to stop execution of the current input. If
1022 bc(1) is in TTY mode (see the **TTY MODE** section), it will reset (see the
1023 **RESET** section). Otherwise, it will clean up and exit.
1025 Note that "current input" can mean one of two things. If bc(1) is processing
1026 input from **stdin** in TTY mode, it will ask for more input. If bc(1) is
1027 processing input from a file in TTY mode, it will stop processing the file and
1028 start processing the next file, if one exists, or ask for input from **stdin**
1029 if no other file exists.
1031 This means that if a **SIGINT** is sent to bc(1) as it is executing a file, it
1032 can seem as though bc(1) did not respond to the signal since it will immediately
1033 start executing the next file. This is by design; most files that users execute
1034 when interacting with bc(1) have function definitions, which are quick to parse.
1035 If a file takes a long time to execute, there may be a bug in that file. The
1036 rest of the files could still be executed without problem, allowing the user to
1039 **SIGTERM** and **SIGQUIT** cause bc(1) to clean up and exit, and it uses the
1040 default handler for all other signals. The one exception is **SIGHUP**; in that
1041 case, when bc(1) is in TTY mode, a **SIGHUP** will cause bc(1) to clean up and
1044 # COMMAND LINE HISTORY
1046 bc(1) supports interactive command-line editing. If bc(1) is in TTY mode (see
1047 the **TTY MODE** section), history is enabled. Previous lines can be recalled
1048 and edited with the arrow keys.
1050 **Note**: tabs are converted to 8 spaces.
1054 This bc(1) ships with support for adding error messages for different locales
1055 and thus, supports **LC_MESSAGES**.
1063 bc(1) is compliant with the [IEEE Std 1003.1-2017 (“POSIX.1-2017”)][1]
1064 specification. The flags **-efghiqsvVw**, all long options, and the extensions
1065 noted above are extensions to that specification.
1067 Note that the specification explicitly says that bc(1) only accepts numbers that
1068 use a period (**.**) as a radix point, regardless of the value of
1071 This bc(1) supports error messages for different locales, and thus, it supports
1076 None are known. Report bugs at https://git.yzena.com/gavin/bc.
1080 Gavin D. Howard <gavin@yzena.com> and contributors.
1082 [1]: https://pubs.opengroup.org/onlinepubs/9699919799/utilities/bc.html
1083 [2]: https://www.gnu.org/software/bc/
1084 [3]: https://en.wikipedia.org/wiki/Rounding#Round_half_away_from_zero
1085 [4]: https://en.wikipedia.org/wiki/Unit_in_the_last_place
1086 [5]: https://people.eecs.berkeley.edu/~wkahan/LOG10HAF.TXT
1087 [6]: https://en.wikipedia.org/wiki/Rounding#Rounding_away_from_zero