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**.
51 The following are the options that bc(1) accepts.
53 **-g**, **-\-global-stacks**
55 Turns the globals **ibase**, **obase**, and **scale** into stacks.
57 This has the effect that a copy of the current value of all three are pushed
58 onto a stack for every function call, as well as popped when every function
59 returns. This means that functions can assign to any and all of those
60 globals without worrying that the change will affect other functions.
61 Thus, a hypothetical function named **output(x,b)** that simply printed
62 **x** in base **b** could be written like this:
64 define void output(x, b) {
71 define void output(x, b) {
79 This makes writing functions much easier.
81 However, since using this flag means that functions cannot set **ibase**,
82 **obase**, or **scale** globally, functions that are made to do so cannot
83 work anymore. There are two possible use cases for that, and each has a
86 First, if a function is called on startup to turn bc(1) into a number
87 converter, it is possible to replace that capability with various shell
90 alias d2o="bc -e ibase=A -e obase=8"
91 alias h2b="bc -e ibase=G -e obase=2"
93 Second, if the purpose of a function is to set **ibase**, **obase**, or
94 **scale** globally for any other purpose, it could be split into one to
95 three functions (based on how many globals it sets) and each of those
96 functions could return the desired value for a global.
98 If the behavior of this option is desired for every run of bc(1), then users
99 could make sure to define **BC_ENV_ARGS** and include this option (see the
100 **ENVIRONMENT VARIABLES** section for more details).
102 If **-s**, **-w**, or any equivalents are used, this option is ignored.
104 This is a **non-portable extension**.
108 : Prints a usage message and quits.
110 **-i**, **-\-interactive**
112 : Forces interactive mode. (See the **INTERACTIVE MODE** section.)
114 This is a **non-portable extension**.
116 **-l**, **-\-mathlib**
118 : Sets **scale** (see the **SYNTAX** section) to **20** and loads the included
119 math library before running any code, including any expressions or files
120 specified on the command line.
122 To learn what is in the library, see the **LIBRARY** section.
124 **-P**, **-\-no-prompt**
126 : This option is a no-op.
128 This is a **non-portable extension**.
132 : This option is for compatibility with the [GNU bc(1)][2]; it is a no-op.
133 Without this option, GNU bc(1) prints a copyright header. This bc(1) only
134 prints the copyright header if one or more of the **-v**, **-V**, or
135 **-\-version** options are given.
137 This is a **non-portable extension**.
139 **-s**, **-\-standard**
141 : Process exactly the language defined by the [standard][1] and error if any
144 This is a **non-portable extension**.
146 **-v**, **-V**, **-\-version**
148 : Print the version information (copyright header) and exit.
150 This is a **non-portable extension**.
154 : Like **-s** and **-\-standard**, except that warnings (and not errors) are
155 printed for non-standard extensions and execution continues normally.
157 This is a **non-portable extension**.
159 **-e** *expr*, **-\-expression**=*expr*
161 : Evaluates *expr*. If multiple expressions are given, they are evaluated in
162 order. If files are given as well (see below), the expressions and files are
163 evaluated in the order given. This means that if a file is given before an
164 expression, the file is read in and evaluated first.
166 If this option is given on the command-line (i.e., not in **BC_ENV_ARGS**,
167 see the **ENVIRONMENT VARIABLES** section), then after processing all
168 expressions and files, bc(1) will exit, unless **-** (**stdin**) was given
169 as an argument at least once to **-f** or **-\-file**, whether on the
170 command-line or in **BC_ENV_ARGS**. However, if any other **-e**,
171 **-\-expression**, **-f**, or **-\-file** arguments are given after **-f-**
172 or equivalent is given, bc(1) will give a fatal error and exit.
174 This is a **non-portable extension**.
176 **-f** *file*, **-\-file**=*file*
178 : Reads in *file* and evaluates it, line by line, as though it were read
179 through **stdin**. If expressions are also given (see above), the
180 expressions are evaluated in the order given.
182 If this option is given on the command-line (i.e., not in **BC_ENV_ARGS**,
183 see the **ENVIRONMENT VARIABLES** section), then after processing all
184 expressions and files, bc(1) will exit, unless **-** (**stdin**) was given
185 as an argument at least once to **-f** or **-\-file**. However, if any other
186 **-e**, **-\-expression**, **-f**, or **-\-file** arguments are given after
187 **-f-** or equivalent is given, bc(1) will give a fatal error and exit.
189 This is a **non-portable extension**.
191 All long options are **non-portable extensions**.
195 Any non-error output is written to **stdout**. In addition, if history (see the
196 **HISTORY** section) and the prompt (see the **TTY MODE** section) are enabled,
197 both are output to **stdout**.
199 **Note**: Unlike other bc(1) implementations, this bc(1) will issue a fatal
200 error (see the **EXIT STATUS** section) if it cannot write to **stdout**, so if
201 **stdout** is closed, as in **bc <file> >&-**, it will quit with an error. This
202 is done so that bc(1) can report problems when **stdout** is redirected to a
205 If there are scripts that depend on the behavior of other bc(1) implementations,
206 it is recommended that those scripts be changed to redirect **stdout** to
211 Any error output is written to **stderr**.
213 **Note**: Unlike other bc(1) implementations, this bc(1) will issue a fatal
214 error (see the **EXIT STATUS** section) if it cannot write to **stderr**, so if
215 **stderr** is closed, as in **bc <file> 2>&-**, it will quit with an error. This
216 is done so that bc(1) can exit with an error code when **stderr** is redirected
219 If there are scripts that depend on the behavior of other bc(1) implementations,
220 it is recommended that those scripts be changed to redirect **stderr** to
225 The syntax for bc(1) programs is mostly C-like, with some differences. This
226 bc(1) follows the [POSIX standard][1], which is a much more thorough resource
227 for the language this bc(1) accepts. This section is meant to be a summary and a
228 listing of all the extensions to the standard.
230 In the sections below, **E** means expression, **S** means statement, and **I**
233 Identifiers (**I**) start with a lowercase letter and can be followed by any
234 number (up to **BC_NAME_MAX-1**) of lowercase letters (**a-z**), digits
235 (**0-9**), and underscores (**\_**). The regex is **\[a-z\]\[a-z0-9\_\]\***.
236 Identifiers with more than one character (letter) are a
237 **non-portable extension**.
239 **ibase** is a global variable determining how to interpret constant numbers. It
240 is the "input" base, or the number base used for interpreting input numbers.
241 **ibase** is initially **10**. If the **-s** (**-\-standard**) and **-w**
242 (**-\-warn**) flags were not given on the command line, the max allowable value
243 for **ibase** is **36**. Otherwise, it is **16**. The min allowable value for
244 **ibase** is **2**. The max allowable value for **ibase** can be queried in
245 bc(1) programs with the **maxibase()** built-in function.
247 **obase** is a global variable determining how to output results. It is the
248 "output" base, or the number base used for outputting numbers. **obase** is
249 initially **10**. The max allowable value for **obase** is **BC_BASE_MAX** and
250 can be queried in bc(1) programs with the **maxobase()** built-in function. The
251 min allowable value for **obase** is **2**. Values are output in the specified
254 The *scale* of an expression is the number of digits in the result of the
255 expression right of the decimal point, and **scale** is a global variable that
256 sets the precision of any operations, with exceptions. **scale** is initially
257 **0**. **scale** cannot be negative. The max allowable value for **scale** is
258 **BC_SCALE_MAX** and can be queried in bc(1) programs with the **maxscale()**
261 bc(1) has both *global* variables and *local* variables. All *local*
262 variables are local to the function; they are parameters or are introduced in
263 the **auto** list of a function (see the **FUNCTIONS** section). If a variable
264 is accessed which is not a parameter or in the **auto** list, it is assumed to
265 be *global*. If a parent function has a *local* variable version of a variable
266 that a child function considers *global*, the value of that *global* variable in
267 the child function is the value of the variable in the parent function, not the
268 value of the actual *global* variable.
270 All of the above applies to arrays as well.
272 The value of a statement that is an expression (i.e., any of the named
273 expressions or operands) is printed unless the lowest precedence operator is an
274 assignment operator *and* the expression is notsurrounded by parentheses.
276 The value that is printed is also assigned to the special variable **last**. A
277 single dot (**.**) may also be used as a synonym for **last**. These are
278 **non-portable extensions**.
280 Either semicolons or newlines may separate statements.
284 There are two kinds of comments:
286 1. Block comments are enclosed in **/\*** and **\*/**.
287 2. Line comments go from **#** until, and not including, the next newline. This
288 is a **non-portable extension**.
292 The following are named expressions in bc(1):
295 2. Array Elements: **I[E]**
299 6. **last** or a single dot (**.**)
301 Number 6 is a **non-portable extension**.
303 Variables and arrays do not interfere; users can have arrays named the same as
304 variables. This also applies to functions (see the **FUNCTIONS** section), so a
305 user can have a variable, array, and function that all have the same name, and
306 they will not shadow each other, whether inside of functions or not.
308 Named expressions are required as the operand of **increment**/**decrement**
309 operators and as the left side of **assignment** operators (see the *Operators*
314 The following are valid operands in bc(1):
316 1. Numbers (see the *Numbers* subsection below).
317 2. Array indices (**I[E]**).
318 3. **(E)**: The value of **E** (used to change precedence).
319 4. **sqrt(E)**: The square root of **E**. **E** must be non-negative.
320 5. **length(E)**: The number of significant decimal digits in **E**.
321 6. **length(I[])**: The number of elements in the array **I**. This is a
322 **non-portable extension**.
323 7. **scale(E)**: The *scale* of **E**.
324 8. **abs(E)**: The absolute value of **E**. This is a **non-portable
326 9. **I()**, **I(E)**, **I(E, E)**, and so on, where **I** is an identifier for
327 a non-**void** function (see the *Void Functions* subsection of the
328 **FUNCTIONS** section). The **E** argument(s) may also be arrays of the form
329 **I[]**, which will automatically be turned into array references (see the
330 *Array References* subsection of the **FUNCTIONS** section) if the
331 corresponding parameter in the function definition is an array reference.
332 10. **read()**: Reads a line from **stdin** and uses that as an expression. The
333 result of that expression is the result of the **read()** operand. This is a
334 **non-portable extension**.
335 11. **maxibase()**: The max allowable **ibase**. This is a **non-portable
337 12. **maxobase()**: The max allowable **obase**. This is a **non-portable
339 13. **maxscale()**: The max allowable **scale**. This is a **non-portable
344 Numbers are strings made up of digits, uppercase letters, and at most **1**
345 period for a radix. Numbers can have up to **BC_NUM_MAX** digits. Uppercase
346 letters are equal to **9** + their position in the alphabet (i.e., **A** equals
347 **10**, or **9+1**). If a digit or letter makes no sense with the current value
348 of **ibase**, they are set to the value of the highest valid digit in **ibase**.
350 Single-character numbers (i.e., **A** alone) take the value that they would have
351 if they were valid digits, regardless of the value of **ibase**. This means that
352 **A** alone always equals decimal **10** and **Z** alone always equals decimal
357 The following arithmetic and logical operators can be used. They are listed in
358 order of decreasing precedence. Operators in the same group have the same
363 : Type: Prefix and Postfix
367 Description: **increment**, **decrement**
375 Description: **negation**, **boolean not**
383 Description: **power**
391 Description: **multiply**, **divide**, **modulus**
399 Description: **add**, **subtract**
401 **=** **+=** **-=** **\*=** **/=** **%=** **\^=**
407 Description: **assignment**
409 **==** **\<=** **\>=** **!=** **\<** **\>**
415 Description: **relational**
423 Description: **boolean and**
431 Description: **boolean or**
433 The operators will be described in more detail below.
437 : The prefix and postfix **increment** and **decrement** operators behave
438 exactly like they would in C. They require a named expression (see the
439 *Named Expressions* subsection) as an operand.
441 The prefix versions of these operators are more efficient; use them where
446 : The **negation** operator returns **0** if a user attempts to negate any
447 expression with the value **0**. Otherwise, a copy of the expression with
448 its sign flipped is returned.
452 : The **boolean not** operator returns **1** if the expression is **0**, or
455 This is a **non-portable extension**.
459 : The **power** operator (not the **exclusive or** operator, as it would be in
460 C) takes two expressions and raises the first to the power of the value of
461 the second. The *scale* of the result is equal to **scale**.
463 The second expression must be an integer (no *scale*), and if it is
464 negative, the first value must be non-zero.
468 : The **multiply** operator takes two expressions, multiplies them, and
469 returns the product. If **a** is the *scale* of the first expression and
470 **b** is the *scale* of the second expression, the *scale* of the result is
471 equal to **min(a+b,max(scale,a,b))** where **min()** and **max()** return
476 : The **divide** operator takes two expressions, divides them, and returns the
477 quotient. The *scale* of the result shall be the value of **scale**.
479 The second expression must be non-zero.
483 : The **modulus** operator takes two expressions, **a** and **b**, and
484 evaluates them by 1) Computing **a/b** to current **scale** and 2) Using the
485 result of step 1 to calculate **a-(a/b)\*b** to *scale*
486 **max(scale+scale(b),scale(a))**.
488 The second expression must be non-zero.
492 : The **add** operator takes two expressions, **a** and **b**, and returns the
493 sum, with a *scale* equal to the max of the *scale*s of **a** and **b**.
497 : The **subtract** operator takes two expressions, **a** and **b**, and
498 returns the difference, with a *scale* equal to the max of the *scale*s of
501 **=** **+=** **-=** **\*=** **/=** **%=** **\^=**
503 : The **assignment** operators take two expressions, **a** and **b** where
504 **a** is a named expression (see the *Named Expressions* subsection).
506 For **=**, **b** is copied and the result is assigned to **a**. For all
507 others, **a** and **b** are applied as operands to the corresponding
508 arithmetic operator and the result is assigned to **a**.
510 **==** **\<=** **\>=** **!=** **\<** **\>**
512 : The **relational** operators compare two expressions, **a** and **b**, and
513 if the relation holds, according to C language semantics, the result is
514 **1**. Otherwise, it is **0**.
516 Note that unlike in C, these operators have a lower precedence than the
517 **assignment** operators, which means that **a=b\>c** is interpreted as
520 Also, unlike the [standard][1] requires, these operators can appear anywhere
521 any other expressions can be used. This allowance is a
522 **non-portable extension**.
526 : The **boolean and** operator takes two expressions and returns **1** if both
527 expressions are non-zero, **0** otherwise.
529 This is *not* a short-circuit operator.
531 This is a **non-portable extension**.
535 : The **boolean or** operator takes two expressions and returns **1** if one
536 of the expressions is non-zero, **0** otherwise.
538 This is *not* a short-circuit operator.
540 This is a **non-portable extension**.
544 The following items are statements:
547 2. **{** **S** **;** ... **;** **S** **}**
548 3. **if** **(** **E** **)** **S**
549 4. **if** **(** **E** **)** **S** **else** **S**
550 5. **while** **(** **E** **)** **S**
551 6. **for** **(** **E** **;** **E** **;** **E** **)** **S**
552 7. An empty statement
558 13. A string of characters, enclosed in double quotes
559 14. **print** **E** **,** ... **,** **E**
560 15. **I()**, **I(E)**, **I(E, E)**, and so on, where **I** is an identifier for
561 a **void** function (see the *Void Functions* subsection of the
562 **FUNCTIONS** section). The **E** argument(s) may also be arrays of the form
563 **I[]**, which will automatically be turned into array references (see the
564 *Array References* subsection of the **FUNCTIONS** section) if the
565 corresponding parameter in the function definition is an array reference.
567 Numbers 4, 9, 11, 12, 14, and 15 are **non-portable extensions**.
569 Also, as a **non-portable extension**, any or all of the expressions in the
570 header of a for loop may be omitted. If the condition (second expression) is
571 omitted, it is assumed to be a constant **1**.
573 The **break** statement causes a loop to stop iterating and resume execution
574 immediately following a loop. This is only allowed in loops.
576 The **continue** statement causes a loop iteration to stop early and returns to
577 the start of the loop, including testing the loop condition. This is only
580 The **if** **else** statement does the same thing as in C.
582 The **quit** statement causes bc(1) to quit, even if it is on a branch that will
583 not be executed (it is a compile-time command).
585 The **halt** statement causes bc(1) to quit, if it is executed. (Unlike **quit**
586 if it is on a branch of an **if** statement that is not executed, bc(1) does not
589 The **limits** statement prints the limits that this bc(1) is subject to. This
590 is like the **quit** statement in that it is a compile-time command.
592 An expression by itself is evaluated and printed, followed by a newline.
596 The "expressions" in a **print** statement may also be strings. If they are, there
597 are backslash escape sequences that are interpreted specially. What those
598 sequences are, and what they cause to be printed, are shown below:
612 Any other character following a backslash causes the backslash and character to
615 Any non-string expression in a print statement shall be assigned to **last**,
616 like any other expression that is printed.
618 ## Order of Evaluation
620 All expressions in a statment are evaluated left to right, except as necessary
621 to maintain order of operations. This means, for example, assuming that **i** is
622 equal to **0**, in the expression
626 the first (or 0th) element of **a** is set to **1**, and **i** is equal to **2**
627 at the end of the expression.
629 This includes function arguments. Thus, assuming **i** is equal to **0**, this
630 means that in the expression
634 the first argument passed to **x()** is **0**, and the second argument is **1**,
635 while **i** is equal to **2** before the function starts executing.
639 Function definitions are as follows:
649 Any **I** in the parameter list or **auto** list may be replaced with **I[]** to
650 make a parameter or **auto** var an array, and any **I** in the parameter list
651 may be replaced with **\*I[]** to make a parameter an array reference. Callers
652 of functions that take array references should not put an asterisk in the call;
653 they must be called with just **I[]** like normal array parameters and will be
654 automatically converted into references.
656 As a **non-portable extension**, the opening brace of a **define** statement may
657 appear on the next line.
659 As a **non-portable extension**, the return statement may also be in one of the
663 2. **return** **(** **)**
666 The first two, or not specifying a **return** statement, is equivalent to
667 **return (0)**, unless the function is a **void** function (see the *Void
668 Functions* subsection below).
672 Functions can also be **void** functions, defined as follows:
675 define void I(I,...,I){
682 They can only be used as standalone expressions, where such an expression would
683 be printed alone, except in a print statement.
685 Void functions can only use the first two **return** statements listed above.
686 They can also omit the return statement entirely.
688 The word "void" is not treated as a keyword; it is still possible to have
689 variables, arrays, and functions named **void**. The word "void" is only
690 treated specially right after the **define** keyword.
692 This is a **non-portable extension**.
696 For any array in the parameter list, if the array is declared in the form
702 it is a **reference**. Any changes to the array in the function are reflected,
703 when the function returns, to the array that was passed in.
705 Other than this, all function arguments are passed by value.
707 This is a **non-portable extension**.
711 All of the functions below are available when the **-l** or **-\-mathlib**
712 command-line flags are given.
716 The [standard][1] defines the following functions for the math library:
720 : Returns the sine of **x**, which is assumed to be in radians.
722 This is a transcendental function (see the *Transcendental Functions*
727 : Returns the cosine of **x**, which is assumed to be in radians.
729 This is a transcendental function (see the *Transcendental Functions*
734 : Returns the arctangent of **x**, in radians.
736 This is a transcendental function (see the *Transcendental Functions*
741 : Returns the natural logarithm of **x**.
743 This is a transcendental function (see the *Transcendental Functions*
748 : Returns the mathematical constant **e** raised to the power of **x**.
750 This is a transcendental function (see the *Transcendental Functions*
755 : Returns the bessel integer order **n** (truncated) of **x**.
757 This is a transcendental function (see the *Transcendental Functions*
760 ## Transcendental Functions
762 All transcendental functions can return slightly inaccurate results (up to 1
763 [ULP][4]). This is unavoidable, and [this article][5] explains why it is
764 impossible and unnecessary to calculate exact results for the transcendental
767 Because of the possible inaccuracy, I recommend that users call those functions
768 with the precision (**scale**) set to at least 1 higher than is necessary. If
769 exact results are *absolutely* required, users can double the precision
770 (**scale**) and then truncate.
772 The transcendental functions in the standard math library are:
783 When bc(1) encounters an error or a signal that it has a non-default handler
784 for, it resets. This means that several things happen.
786 First, any functions that are executing are stopped and popped off the stack.
787 The behavior is not unlike that of exceptions in programming languages. Then
788 the execution point is set so that any code waiting to execute (after all
789 functions returned) is skipped.
791 Thus, when bc(1) resets, it skips any remaining code waiting to be executed.
792 Then, if it is interactive mode, and the error was not a fatal error (see the
793 **EXIT STATUS** section), it asks for more input; otherwise, it exits with the
794 appropriate return code.
796 Note that this reset behavior is different from the GNU bc(1), which attempts to
797 start executing the statement right after the one that caused an error.
801 Most bc(1) implementations use **char** types to calculate the value of **1**
802 decimal digit at a time, but that can be slow. This bc(1) does something
805 It uses large integers to calculate more than **1** decimal digit at a time. If
806 built in a environment where **BC_LONG_BIT** (see the **LIMITS** section) is
807 **64**, then each integer has **9** decimal digits. If built in an environment
808 where **BC_LONG_BIT** is **32** then each integer has **4** decimal digits. This
809 value (the number of decimal digits per large integer) is called
812 The actual values of **BC_LONG_BIT** and **BC_BASE_DIGS** can be queried with
813 the **limits** statement.
815 In addition, this bc(1) uses an even larger integer for overflow checking. This
816 integer type depends on the value of **BC_LONG_BIT**, but is always at least
817 twice as large as the integer type used to store digits.
821 The following are the limits on bc(1):
825 : The number of bits in the **long** type in the environment where bc(1) was
826 built. This determines how many decimal digits can be stored in a single
827 large integer (see the **PERFORMANCE** section).
831 : The number of decimal digits per large integer (see the **PERFORMANCE**
832 section). Depends on **BC_LONG_BIT**.
836 : The max decimal number that each large integer can store (see
837 **BC_BASE_DIGS**) plus **1**. Depends on **BC_BASE_DIGS**.
841 : The max number that the overflow type (see the **PERFORMANCE** section) can
842 hold. Depends on **BC_LONG_BIT**.
846 : The maximum output base. Set at **BC_BASE_POW**.
850 : The maximum size of arrays. Set at **SIZE_MAX-1**.
854 : The maximum **scale**. Set at **BC_OVERFLOW_MAX-1**.
858 : The maximum length of strings. Set at **BC_OVERFLOW_MAX-1**.
862 : The maximum length of identifiers. Set at **BC_OVERFLOW_MAX-1**.
866 : The maximum length of a number (in decimal digits), which includes digits
867 after the decimal point. Set at **BC_OVERFLOW_MAX-1**.
871 : The maximum allowable exponent (positive or negative). Set at
876 : The maximum number of vars/arrays. Set at **SIZE_MAX-1**.
878 The actual values can be queried with the **limits** statement.
880 These limits are meant to be effectively non-existent; the limits are so large
881 (at least on 64-bit machines) that there should not be any point at which they
882 become a problem. In fact, memory should be exhausted before these limits should
885 # ENVIRONMENT VARIABLES
887 bc(1) recognizes the following environment variables:
891 : If this variable exists (no matter the contents), bc(1) behaves as if
892 the **-s** option was given.
896 : This is another way to give command-line arguments to bc(1). They should be
897 in the same format as all other command-line arguments. These are always
898 processed first, so any files given in **BC_ENV_ARGS** will be processed
899 before arguments and files given on the command-line. This gives the user
900 the ability to set up "standard" options and files to be used at every
901 invocation. The most useful thing for such files to contain would be useful
902 functions that the user might want every time bc(1) runs.
904 The code that parses **BC_ENV_ARGS** will correctly handle quoted arguments,
905 but it does not understand escape sequences. For example, the string
906 **"/home/gavin/some bc file.bc"** will be correctly parsed, but the string
907 **"/home/gavin/some \"bc\" file.bc"** will include the backslashes.
909 The quote parsing will handle either kind of quotes, **'** or **"**. Thus,
910 if you have a file with any number of single quotes in the name, you can use
911 double quotes as the outside quotes, as in **"some 'bc' file.bc"**, and vice
912 versa if you have a file with double quotes. However, handling a file with
913 both kinds of quotes in **BC_ENV_ARGS** is not supported due to the
914 complexity of the parsing, though such files are still supported on the
915 command-line where the parsing is done by the shell.
919 : If this environment variable exists and contains an integer that is greater
920 than **1** and is less than **UINT16_MAX** (**2\^16-1**), bc(1) will output
921 lines to that length, including the backslash (**\\**). The default line
926 bc(1) returns the following exit statuses:
934 : A math error occurred. This follows standard practice of using **1** for
935 expected errors, since math errors will happen in the process of normal
938 Math errors include divide by **0**, taking the square root of a negative
939 number, attempting to convert a negative number to a hardware integer,
940 overflow when converting a number to a hardware integer, and attempting to
941 use a non-integer where an integer is required.
943 Converting to a hardware integer happens for the second operand of the power
944 (**\^**) operator and the corresponding assignment operator.
948 : A parse error occurred.
950 Parse errors include unexpected **EOF**, using an invalid character, failing
951 to find the end of a string or comment, using a token where it is invalid,
952 giving an invalid expression, giving an invalid print statement, giving an
953 invalid function definition, attempting to assign to an expression that is
954 not a named expression (see the *Named Expressions* subsection of the
955 **SYNTAX** section), giving an invalid **auto** list, having a duplicate
956 **auto**/function parameter, failing to find the end of a code block,
957 attempting to return a value from a **void** function, attempting to use a
958 variable as a reference, and using any extensions when the option **-s** or
959 any equivalents were given.
963 : A runtime error occurred.
965 Runtime errors include assigning an invalid number to **ibase**, **obase**,
966 or **scale**; give a bad expression to a **read()** call, calling **read()**
967 inside of a **read()** call, type errors, passing the wrong number of
968 arguments to functions, attempting to call an undefined function, and
969 attempting to use a **void** function call as a value in an expression.
973 : A fatal error occurred.
975 Fatal errors include memory allocation errors, I/O errors, failing to open
976 files, attempting to use files that do not have only ASCII characters (bc(1)
977 only accepts ASCII characters), attempting to open a directory as a file,
978 and giving invalid command-line options.
980 The exit status **4** is special; when a fatal error occurs, bc(1) always exits
981 and returns **4**, no matter what mode bc(1) is in.
983 The other statuses will only be returned when bc(1) is not in interactive mode
984 (see the **INTERACTIVE MODE** section), since bc(1) resets its state (see the
985 **RESET** section) and accepts more input when one of those errors occurs in
986 interactive mode. This is also the case when interactive mode is forced by the
987 **-i** flag or **-\-interactive** option.
989 These exit statuses allow bc(1) to be used in shell scripting with error
990 checking, and its normal behavior can be forced by using the **-i** flag or
991 **-\-interactive** option.
995 Per the [standard][1], bc(1) has an interactive mode and a non-interactive mode.
996 Interactive mode is turned on automatically when both **stdin** and **stdout**
997 are hooked to a terminal, but the **-i** flag and **-\-interactive** option can
998 turn it on in other cases.
1000 In interactive mode, bc(1) attempts to recover from errors (see the **RESET**
1001 section), and in normal execution, flushes **stdout** as soon as execution is
1002 done for the current input.
1006 If **stdin**, **stdout**, and **stderr** are all connected to a TTY, bc(1) turns
1009 TTY mode is different from interactive mode because interactive mode is required
1010 in the [bc(1) specification][1], and interactive mode requires only **stdin**
1011 and **stdout** to be connected to a terminal.
1015 Sending a **SIGINT** will cause bc(1) to stop execution of the current input. If
1016 bc(1) is in TTY mode (see the **TTY MODE** section), it will reset (see the
1017 **RESET** section). Otherwise, it will clean up and exit.
1019 Note that "current input" can mean one of two things. If bc(1) is processing
1020 input from **stdin** in TTY mode, it will ask for more input. If bc(1) is
1021 processing input from a file in TTY mode, it will stop processing the file and
1022 start processing the next file, if one exists, or ask for input from **stdin**
1023 if no other file exists.
1025 This means that if a **SIGINT** is sent to bc(1) as it is executing a file, it
1026 can seem as though bc(1) did not respond to the signal since it will immediately
1027 start executing the next file. This is by design; most files that users execute
1028 when interacting with bc(1) have function definitions, which are quick to parse.
1029 If a file takes a long time to execute, there may be a bug in that file. The
1030 rest of the files could still be executed without problem, allowing the user to
1033 **SIGTERM** and **SIGQUIT** cause bc(1) to clean up and exit, and it uses the
1034 default handler for all other signals.
1042 bc(1) is compliant with the [IEEE Std 1003.1-2017 (“POSIX.1-2017”)][1]
1043 specification. The flags **-efghiqsvVw**, all long options, and the extensions
1044 noted above are extensions to that specification.
1046 Note that the specification explicitly says that bc(1) only accepts numbers that
1047 use a period (**.**) as a radix point, regardless of the value of
1052 None are known. Report bugs at https://git.yzena.com/gavin/bc.
1056 Gavin D. Howard <gavin@yzena.com> and contributors.
1058 [1]: https://pubs.opengroup.org/onlinepubs/9699919799/utilities/bc.html
1059 [2]: https://www.gnu.org/software/bc/
1060 [3]: https://en.wikipedia.org/wiki/Rounding#Round_half_away_from_zero
1061 [4]: https://en.wikipedia.org/wiki/Unit_in_the_last_place
1062 [5]: https://people.eecs.berkeley.edu/~wkahan/LOG10HAF.TXT
1063 [6]: https://en.wikipedia.org/wiki/Rounding#Rounding_away_from_zero