3 SPDX-License-Identifier: BSD-2-Clause
5 Copyright (c) 2018-2020 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 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 : Disables the prompt in TTY mode. (The prompt is only enabled in TTY mode.
128 See the **TTY MODE** section) This is mostly for those users that do not
129 want a prompt or are not used to having them in bc(1). Most of those users
130 would want to put this option in **BC_ENV_ARGS** (see the
131 **ENVIRONMENT VARIABLES** section).
133 This is a **non-portable extension**.
137 : This option is for compatibility with the [GNU bc(1)][2]; it is a no-op.
138 Without this option, GNU bc(1) prints a copyright header. This bc(1) only
139 prints the copyright header if one or more of the **-v**, **-V**, or
140 **--version** options are given.
142 This is a **non-portable extension**.
144 **-s**, **--standard**
146 : Process exactly the language defined by the [standard][1] and error if any
149 This is a **non-portable extension**.
151 **-v**, **-V**, **--version**
153 : Print the version information (copyright header) and exit.
155 This is a **non-portable extension**.
159 : Like **-s** and **--standard**, except that warnings (and not errors) are
160 printed for non-standard extensions and execution continues normally.
162 This is a **non-portable extension**.
164 **-e** *expr*, **--expression**=*expr*
166 : Evaluates *expr*. If multiple expressions are given, they are evaluated in
167 order. If files are given as well (see below), the expressions and files are
168 evaluated in the order given. This means that if a file is given before an
169 expression, the file is read in and evaluated first.
171 After processing all expressions and files, bc(1) will exit, unless **-**
172 (**stdin**) was given as an argument at least once to **-f** or **--file**.
173 However, if any other **-e**, **--expression**, **-f**, or **--file**
174 arguments are given after that, bc(1) will give a fatal error and exit.
176 This is a **non-portable extension**.
178 **-f** *file*, **--file**=*file*
180 : Reads in *file* and evaluates it, line by line, as though it were read
181 through **stdin**. If expressions are also given (see above), the
182 expressions are evaluated in the order given.
184 After processing all expressions and files, bc(1) will exit, unless **-**
185 (**stdin**) was given as an argument at least once to **-f** or **--file**.
187 This is a **non-portable extension**.
189 All long options are **non-portable extensions**.
193 Any non-error output is written to **stdout**.
195 **Note**: Unlike other bc(1) implementations, this bc(1) will issue a fatal
196 error (see the **EXIT STATUS** section) if it cannot write to **stdout**, so if
197 **stdout** is closed, as in **bc <file> >&-**, it will quit with an error. This
198 is done so that bc(1) can report problems when **stdout** is redirected to a
201 If there are scripts that depend on the behavior of other bc(1) implementations,
202 it is recommended that those scripts be changed to redirect **stdout** to
207 Any error output is written to **stderr**.
209 **Note**: Unlike other bc(1) implementations, this bc(1) will issue a fatal
210 error (see the **EXIT STATUS** section) if it cannot write to **stderr**, so if
211 **stderr** is closed, as in **bc <file> 2>&-**, it will quit with an error. This
212 is done so that bc(1) can exit with an error code when **stderr** is redirected
215 If there are scripts that depend on the behavior of other bc(1) implementations,
216 it is recommended that those scripts be changed to redirect **stderr** to
221 The syntax for bc(1) programs is mostly C-like, with some differences. This
222 bc(1) follows the [POSIX standard][1], which is a much more thorough resource
223 for the language this bc(1) accepts. This section is meant to be a summary and a
224 listing of all the extensions to the standard.
226 In the sections below, **E** means expression, **S** means statement, and **I**
229 Identifiers (**I**) start with a lowercase letter and can be followed by any
230 number (up to **BC_NAME_MAX-1**) of lowercase letters (**a-z**), digits
231 (**0-9**), and underscores (**\_**). The regex is **\[a-z\]\[a-z0-9\_\]\***.
232 Identifiers with more than one character (letter) are a
233 **non-portable extension**.
235 **ibase** is a global variable determining how to interpret constant numbers. It
236 is the "input" base, or the number base used for interpreting input numbers.
237 **ibase** is initially **10**. If the **-s** (**--standard**) and **-w**
238 (**--warn**) flags were not given on the command line, the max allowable value
239 for **ibase** is **36**. Otherwise, it is **16**. The min allowable value for
240 **ibase** is **2**. The max allowable value for **ibase** can be queried in
241 bc(1) programs with the **maxibase()** built-in function.
243 **obase** is a global variable determining how to output results. It is the
244 "output" base, or the number base used for outputting numbers. **obase** is
245 initially **10**. The max allowable value for **obase** is **BC_BASE_MAX** and
246 can be queried in bc(1) programs with the **maxobase()** built-in function. The
247 min allowable value for **obase** is **2**. Values are output in the specified
250 The *scale* of an expression is the number of digits in the result of the
251 expression right of the decimal point, and **scale** is a global variable that
252 sets the precision of any operations, with exceptions. **scale** is initially
253 **0**. **scale** cannot be negative. The max allowable value for **scale** is
254 **BC_SCALE_MAX** and can be queried in bc(1) programs with the **maxscale()**
257 bc(1) has both *global* variables and *local* variables. All *local*
258 variables are local to the function; they are parameters or are introduced in
259 the **auto** list of a function (see the **FUNCTIONS** section). If a variable
260 is accessed which is not a parameter or in the **auto** list, it is assumed to
261 be *global*. If a parent function has a *local* variable version of a variable
262 that a child function considers *global*, the value of that *global* variable in
263 the child function is the value of the variable in the parent function, not the
264 value of the actual *global* variable.
266 All of the above applies to arrays as well.
268 The value of a statement that is an expression (i.e., any of the named
269 expressions or operands) is printed unless the lowest precedence operator is an
270 assignment operator *and* the expression is notsurrounded by parentheses.
272 The value that is printed is also assigned to the special variable **last**. A
273 single dot (**.**) may also be used as a synonym for **last**. These are
274 **non-portable extensions**.
276 Either semicolons or newlines may separate statements.
280 There are two kinds of comments:
282 1. Block comments are enclosed in **/\*** and **\*/**.
283 2. Line comments go from **#** until, and not including, the next newline. This
284 is a **non-portable extension**.
288 The following are named expressions in bc(1):
291 2. Array Elements: **I[E]**
295 6. **last** or a single dot (**.**)
297 Number 6 is a **non-portable extension**.
299 Variables and arrays do not interfere; users can have arrays named the same as
300 variables. This also applies to functions (see the **FUNCTIONS** section), so a
301 user can have a variable, array, and function that all have the same name, and
302 they will not shadow each other, whether inside of functions or not.
304 Named expressions are required as the operand of **increment**/**decrement**
305 operators and as the left side of **assignment** operators (see the *Operators*
310 The following are valid operands in bc(1):
312 1. Numbers (see the *Numbers* subsection below).
313 2. Array indices (**I[E]**).
314 3. **(E)**: The value of **E** (used to change precedence).
315 4. **sqrt(E)**: The square root of **E**. **E** must be non-negative.
316 5. **length(E)**: The number of significant decimal digits in **E**.
317 6. **length(I[])**: The number of elements in the array **I**. This is a
318 **non-portable extension**.
319 7. **scale(E)**: The *scale* of **E**.
320 8. **abs(E)**: The absolute value of **E**. This is a **non-portable
322 9. **I()**, **I(E)**, **I(E, E)**, and so on, where **I** is an identifier for
323 a non-**void** function (see the *Void Functions* subsection of the
324 **FUNCTIONS** section). The **E** argument(s) may also be arrays of the form
325 **I[]**, which will automatically be turned into array references (see the
326 *Array References* subsection of the **FUNCTIONS** section) if the
327 corresponding parameter in the function definition is an array reference.
328 10. **read()**: Reads a line from **stdin** and uses that as an expression. The
329 result of that expression is the result of the **read()** operand. This is a
330 **non-portable extension**.
331 11. **maxibase()**: The max allowable **ibase**. This is a **non-portable
333 12. **maxobase()**: The max allowable **obase**. This is a **non-portable
335 13. **maxscale()**: The max allowable **scale**. This is a **non-portable
340 Numbers are strings made up of digits, uppercase letters, and at most **1**
341 period for a radix. Numbers can have up to **BC_NUM_MAX** digits. Uppercase
342 letters are equal to **9** + their position in the alphabet (i.e., **A** equals
343 **10**, or **9+1**). If a digit or letter makes no sense with the current value
344 of **ibase**, they are set to the value of the highest valid digit in **ibase**.
346 Single-character numbers (i.e., **A** alone) take the value that they would have
347 if they were valid digits, regardless of the value of **ibase**. This means that
348 **A** alone always equals decimal **10** and **Z** alone always equals decimal
353 The following arithmetic and logical operators can be used. They are listed in
354 order of decreasing precedence. Operators in the same group have the same
359 : Type: Prefix and Postfix
363 Description: **increment**, **decrement**
371 Description: **negation**, **boolean not**
379 Description: **power**
387 Description: **multiply**, **divide**, **modulus**
395 Description: **add**, **subtract**
397 **=** **+=** **-=** **\*=** **/=** **%=** **\^=**
403 Description: **assignment**
405 **==** **\<=** **\>=** **!=** **\<** **\>**
411 Description: **relational**
419 Description: **boolean and**
427 Description: **boolean or**
429 The operators will be described in more detail below.
433 : The prefix and postfix **increment** and **decrement** operators behave
434 exactly like they would in C. They require a named expression (see the
435 *Named Expressions* subsection) as an operand.
437 The prefix versions of these operators are more efficient; use them where
442 : The **negation** operator returns **0** if a user attempts to negate any
443 expression with the value **0**. Otherwise, a copy of the expression with
444 its sign flipped is returned.
448 : The **boolean not** operator returns **1** if the expression is **0**, or
451 This is a **non-portable extension**.
455 : The **power** operator (not the **exclusive or** operator, as it would be in
456 C) takes two expressions and raises the first to the power of the value of
459 The second expression must be an integer (no *scale*), and if it is
460 negative, the first value must be non-zero.
464 : The **multiply** operator takes two expressions, multiplies them, and
465 returns the product. If **a** is the *scale* of the first expression and
466 **b** is the *scale* of the second expression, the *scale* of the result is
467 equal to **min(a+b,max(scale,a,b))** where **min()** and **max()** return
472 : The **divide** operator takes two expressions, divides them, and returns the
473 quotient. The *scale* of the result shall be the value of **scale**.
475 The second expression must be non-zero.
479 : The **modulus** operator takes two expressions, **a** and **b**, and
480 evaluates them by 1) Computing **a/b** to current **scale** and 2) Using the
481 result of step 1 to calculate **a-(a/b)\*b** to *scale*
482 **max(scale+scale(b),scale(a))**.
484 The second expression must be non-zero.
488 : The **add** operator takes two expressions, **a** and **b**, and returns the
489 sum, with a *scale* equal to the max of the *scale*s of **a** and **b**.
493 : The **subtract** operator takes two expressions, **a** and **b**, and
494 returns the difference, with a *scale* equal to the max of the *scale*s of
497 **=** **+=** **-=** **\*=** **/=** **%=** **\^=**
499 : The **assignment** operators take two expressions, **a** and **b** where
500 **a** is a named expression (see the *Named Expressions* subsection).
502 For **=**, **b** is copied and the result is assigned to **a**. For all
503 others, **a** and **b** are applied as operands to the corresponding
504 arithmetic operator and the result is assigned to **a**.
506 **==** **\<=** **\>=** **!=** **\<** **\>**
508 : The **relational** operators compare two expressions, **a** and **b**, and
509 if the relation holds, according to C language semantics, the result is
510 **1**. Otherwise, it is **0**.
512 Note that unlike in C, these operators have a lower precedence than the
513 **assignment** operators, which means that **a=b\>c** is interpreted as
516 Also, unlike the [standard][1] requires, these operators can appear anywhere
517 any other expressions can be used. This allowance is a
518 **non-portable extension**.
522 : The **boolean and** operator takes two expressions and returns **1** if both
523 expressions are non-zero, **0** otherwise.
525 This is *not* a short-circuit operator.
527 This is a **non-portable extension**.
531 : The **boolean or** operator takes two expressions and returns **1** if one
532 of the expressions is non-zero, **0** otherwise.
534 This is *not* a short-circuit operator.
536 This is a **non-portable extension**.
540 The following items are statements:
543 2. **{** **S** **;** ... **;** **S** **}**
544 3. **if** **(** **E** **)** **S**
545 4. **if** **(** **E** **)** **S** **else** **S**
546 5. **while** **(** **E** **)** **S**
547 6. **for** **(** **E** **;** **E** **;** **E** **)** **S**
548 7. An empty statement
554 13. A string of characters, enclosed in double quotes
555 14. **print** **E** **,** ... **,** **E**
556 15. **I()**, **I(E)**, **I(E, E)**, and so on, where **I** is an identifier for
557 a **void** function (see the *Void Functions* subsection of the
558 **FUNCTIONS** section). The **E** argument(s) may also be arrays of the form
559 **I[]**, which will automatically be turned into array references (see the
560 *Array References* subsection of the **FUNCTIONS** section) if the
561 corresponding parameter in the function definition is an array reference.
563 Numbers 4, 9, 11, 12, 14, and 15 are **non-portable extensions**.
565 Also, as a **non-portable extension**, any or all of the expressions in the
566 header of a for loop may be omitted. If the condition (second expression) is
567 omitted, it is assumed to be a constant **1**.
569 The **break** statement causes a loop to stop iterating and resume execution
570 immediately following a loop. This is only allowed in loops.
572 The **continue** statement causes a loop iteration to stop early and returns to
573 the start of the loop, including testing the loop condition. This is only
576 The **if** **else** statement does the same thing as in C.
578 The **quit** statement causes bc(1) to quit, even if it is on a branch that will
579 not be executed (it is a compile-time command).
581 The **halt** statement causes bc(1) to quit, if it is executed. (Unlike **quit**
582 if it is on a branch of an **if** statement that is not executed, bc(1) does not
585 The **limits** statement prints the limits that this bc(1) is subject to. This
586 is like the **quit** statement in that it is a compile-time command.
588 An expression by itself is evaluated and printed, followed by a newline.
592 The "expressions" in a **print** statement may also be strings. If they are, there
593 are backslash escape sequences that are interpreted specially. What those
594 sequences are, and what they cause to be printed, are shown below:
608 Any other character following a backslash causes the backslash and character to
611 Any non-string expression in a print statement shall be assigned to **last**,
612 like any other expression that is printed.
614 ## Order of Evaluation
616 All expressions in a statment are evaluated left to right, except as necessary
617 to maintain order of operations. This means, for example, assuming that **i** is
618 equal to **0**, in the expression
622 the first (or 0th) element of **a** is set to **1**, and **i** is equal to **2**
623 at the end of the expression.
625 This includes function arguments. Thus, assuming **i** is equal to **0**, this
626 means that in the expression
630 the first argument passed to **x()** is **0**, and the second argument is **1**,
631 while **i** is equal to **2** before the function starts executing.
635 Function definitions are as follows:
645 Any **I** in the parameter list or **auto** list may be replaced with **I[]** to
646 make a parameter or **auto** var an array, and any **I** in the parameter list
647 may be replaced with **\*I[]** to make a parameter an array reference. Callers
648 of functions that take array references should not put an asterisk in the call;
649 they must be called with just **I[]** like normal array parameters and will be
650 automatically converted into references.
652 As a **non-portable extension**, the opening brace of a **define** statement may
653 appear on the next line.
655 As a **non-portable extension**, the return statement may also be in one of the
659 2. **return** **(** **)**
662 The first two, or not specifying a **return** statement, is equivalent to
663 **return (0)**, unless the function is a **void** function (see the *Void
664 Functions* subsection below).
668 Functions can also be **void** functions, defined as follows:
671 define void I(I,...,I){
678 They can only be used as standalone expressions, where such an expression would
679 be printed alone, except in a print statement.
681 Void functions can only use the first two **return** statements listed above.
682 They can also omit the return statement entirely.
684 The word "void" is not treated as a keyword; it is still possible to have
685 variables, arrays, and functions named **void**. The word "void" is only
686 treated specially right after the **define** keyword.
688 This is a **non-portable extension**.
692 For any array in the parameter list, if the array is declared in the form
698 it is a **reference**. Any changes to the array in the function are reflected,
699 when the function returns, to the array that was passed in.
701 Other than this, all function arguments are passed by value.
703 This is a **non-portable extension**.
707 All of the functions below are available when the **-l** or **--mathlib**
708 command-line flags are given.
712 The [standard][1] defines the following functions for the math library:
716 : Returns the sine of **x**, which is assumed to be in radians.
718 This is a transcendental function (see the *Transcendental Functions*
723 : Returns the cosine of **x**, which is assumed to be in radians.
725 This is a transcendental function (see the *Transcendental Functions*
730 : Returns the arctangent of **x**, in radians.
732 This is a transcendental function (see the *Transcendental Functions*
737 : Returns the natural logarithm of **x**.
739 This is a transcendental function (see the *Transcendental Functions*
744 : Returns the mathematical constant **e** raised to the power of **x**.
746 This is a transcendental function (see the *Transcendental Functions*
751 : Returns the bessel integer order **n** (truncated) of **x**.
753 This is a transcendental function (see the *Transcendental Functions*
756 ## Transcendental Functions
758 All transcendental functions can return slightly inaccurate results (up to 1
759 [ULP][4]). This is unavoidable, and [this article][5] explains why it is
760 impossible and unnecessary to calculate exact results for the transcendental
763 Because of the possible inaccuracy, I recommend that users call those functions
764 with the precision (**scale**) set to at least 1 higher than is necessary. If
765 exact results are *absolutely* required, users can double the precision
766 (**scale**) and then truncate.
768 The transcendental functions in the standard math library are:
779 When bc(1) encounters an error or a signal that it has a non-default handler
780 for, it resets. This means that several things happen.
782 First, any functions that are executing are stopped and popped off the stack.
783 The behavior is not unlike that of exceptions in programming languages. Then
784 the execution point is set so that any code waiting to execute (after all
785 functions returned) is skipped.
787 Thus, when bc(1) resets, it skips any remaining code waiting to be executed.
788 Then, if it is interactive mode, and the error was not a fatal error (see the
789 **EXIT STATUS** section), it asks for more input; otherwise, it exits with the
790 appropriate return code.
792 Note that this reset behavior is different from the GNU bc(1), which attempts to
793 start executing the statement right after the one that caused an error.
797 Most bc(1) implementations use **char** types to calculate the value of **1**
798 decimal digit at a time, but that can be slow. This bc(1) does something
801 It uses large integers to calculate more than **1** decimal digit at a time. If
802 built in a environment where **BC_LONG_BIT** (see the **LIMITS** section) is
803 **64**, then each integer has **9** decimal digits. If built in an environment
804 where **BC_LONG_BIT** is **32** then each integer has **4** decimal digits. This
805 value (the number of decimal digits per large integer) is called
808 The actual values of **BC_LONG_BIT** and **BC_BASE_DIGS** can be queried with
809 the **limits** statement.
811 In addition, this bc(1) uses an even larger integer for overflow checking. This
812 integer type depends on the value of **BC_LONG_BIT**, but is always at least
813 twice as large as the integer type used to store digits.
817 The following are the limits on bc(1):
821 : The number of bits in the **long** type in the environment where bc(1) was
822 built. This determines how many decimal digits can be stored in a single
823 large integer (see the **PERFORMANCE** section).
827 : The number of decimal digits per large integer (see the **PERFORMANCE**
828 section). Depends on **BC_LONG_BIT**.
832 : The max decimal number that each large integer can store (see
833 **BC_BASE_DIGS**) plus **1**. Depends on **BC_BASE_DIGS**.
837 : The max number that the overflow type (see the **PERFORMANCE** section) can
838 hold. Depends on **BC_LONG_BIT**.
842 : The maximum output base. Set at **BC_BASE_POW**.
846 : The maximum size of arrays. Set at **SIZE_MAX-1**.
850 : The maximum **scale**. Set at **BC_OVERFLOW_MAX-1**.
854 : The maximum length of strings. Set at **BC_OVERFLOW_MAX-1**.
858 : The maximum length of identifiers. Set at **BC_OVERFLOW_MAX-1**.
862 : The maximum length of a number (in decimal digits), which includes digits
863 after the decimal point. Set at **BC_OVERFLOW_MAX-1**.
867 : The maximum allowable exponent (positive or negative). Set at
872 : The maximum number of vars/arrays. Set at **SIZE_MAX-1**.
874 The actual values can be queried with the **limits** statement.
876 These limits are meant to be effectively non-existent; the limits are so large
877 (at least on 64-bit machines) that there should not be any point at which they
878 become a problem. In fact, memory should be exhausted before these limits should
881 # ENVIRONMENT VARIABLES
883 bc(1) recognizes the following environment variables:
887 : If this variable exists (no matter the contents), bc(1) behaves as if
888 the **-s** option was given.
892 : This is another way to give command-line arguments to bc(1). They should be
893 in the same format as all other command-line arguments. These are always
894 processed first, so any files given in **BC_ENV_ARGS** will be processed
895 before arguments and files given on the command-line. This gives the user
896 the ability to set up "standard" options and files to be used at every
897 invocation. The most useful thing for such files to contain would be useful
898 functions that the user might want every time bc(1) runs.
900 The code that parses **BC_ENV_ARGS** will correctly handle quoted arguments,
901 but it does not understand escape sequences. For example, the string
902 **"/home/gavin/some bc file.bc"** will be correctly parsed, but the string
903 **"/home/gavin/some \"bc\" file.bc"** will include the backslashes.
905 The quote parsing will handle either kind of quotes, **'** or **"**. Thus,
906 if you have a file with any number of single quotes in the name, you can use
907 double quotes as the outside quotes, as in **"some 'bc' file.bc"**, and vice
908 versa if you have a file with double quotes. However, handling a file with
909 both kinds of quotes in **BC_ENV_ARGS** is not supported due to the
910 complexity of the parsing, though such files are still supported on the
911 command-line where the parsing is done by the shell.
915 : If this environment variable exists and contains an integer that is greater
916 than **1** and is less than **UINT16_MAX** (**2\^16-1**), bc(1) will output
917 lines to that length, including the backslash (**\\**). The default line
922 bc(1) returns the following exit statuses:
930 : A math error occurred. This follows standard practice of using **1** for
931 expected errors, since math errors will happen in the process of normal
934 Math errors include divide by **0**, taking the square root of a negative
935 number, attempting to convert a negative number to a hardware integer,
936 overflow when converting a number to a hardware integer, and attempting to
937 use a non-integer where an integer is required.
939 Converting to a hardware integer happens for the second operand of the power
940 (**\^**) operator and the corresponding assignment operator.
944 : A parse error occurred.
946 Parse errors include unexpected **EOF**, using an invalid character, failing
947 to find the end of a string or comment, using a token where it is invalid,
948 giving an invalid expression, giving an invalid print statement, giving an
949 invalid function definition, attempting to assign to an expression that is
950 not a named expression (see the *Named Expressions* subsection of the
951 **SYNTAX** section), giving an invalid **auto** list, having a duplicate
952 **auto**/function parameter, failing to find the end of a code block,
953 attempting to return a value from a **void** function, attempting to use a
954 variable as a reference, and using any extensions when the option **-s** or
955 any equivalents were given.
959 : A runtime error occurred.
961 Runtime errors include assigning an invalid number to **ibase**, **obase**,
962 or **scale**; give a bad expression to a **read()** call, calling **read()**
963 inside of a **read()** call, type errors, passing the wrong number of
964 arguments to functions, attempting to call an undefined function, and
965 attempting to use a **void** function call as a value in an expression.
969 : A fatal error occurred.
971 Fatal errors include memory allocation errors, I/O errors, failing to open
972 files, attempting to use files that do not have only ASCII characters (bc(1)
973 only accepts ASCII characters), attempting to open a directory as a file,
974 and giving invalid command-line options.
976 The exit status **4** is special; when a fatal error occurs, bc(1) always exits
977 and returns **4**, no matter what mode bc(1) is in.
979 The other statuses will only be returned when bc(1) is not in interactive mode
980 (see the **INTERACTIVE MODE** section), since bc(1) resets its state (see the
981 **RESET** section) and accepts more input when one of those errors occurs in
982 interactive mode. This is also the case when interactive mode is forced by the
983 **-i** flag or **--interactive** option.
985 These exit statuses allow bc(1) to be used in shell scripting with error
986 checking, and its normal behavior can be forced by using the **-i** flag or
987 **--interactive** option.
991 Per the [standard][1], bc(1) has an interactive mode and a non-interactive mode.
992 Interactive mode is turned on automatically when both **stdin** and **stdout**
993 are hooked to a terminal, but the **-i** flag and **--interactive** option can
994 turn it on in other cases.
996 In interactive mode, bc(1) attempts to recover from errors (see the **RESET**
997 section), and in normal execution, flushes **stdout** as soon as execution is
998 done for the current input.
1002 If **stdin**, **stdout**, and **stderr** are all connected to a TTY, bc(1) turns
1005 The prompt is enabled in TTY mode.
1007 TTY mode is different from interactive mode because interactive mode is required
1008 in the [bc(1) specification][1], and interactive mode requires only **stdin**
1009 and **stdout** to be connected to a terminal.
1013 Sending a **SIGINT** will cause bc(1) to stop execution of the current input. If
1014 bc(1) is in TTY mode (see the **TTY MODE** section), it will reset (see the
1015 **RESET** section). Otherwise, it will clean up and exit.
1017 Note that "current input" can mean one of two things. If bc(1) is processing
1018 input from **stdin** in TTY mode, it will ask for more input. If bc(1) is
1019 processing input from a file in TTY mode, it will stop processing the file and
1020 start processing the next file, if one exists, or ask for input from **stdin**
1021 if no other file exists.
1023 This means that if a **SIGINT** is sent to bc(1) as it is executing a file, it
1024 can seem as though bc(1) did not respond to the signal since it will immediately
1025 start executing the next file. This is by design; most files that users execute
1026 when interacting with bc(1) have function definitions, which are quick to parse.
1027 If a file takes a long time to execute, there may be a bug in that file. The
1028 rest of the files could still be executed without problem, allowing the user to
1031 **SIGTERM** and **SIGQUIT** cause bc(1) to clean up and exit, and it uses the
1032 default handler for all other signals.
1040 bc(1) is compliant with the [IEEE Std 1003.1-2017 (“POSIX.1-2017”)][1]
1041 specification. The flags **-efghiqsvVw**, all long options, and the extensions
1042 noted above are extensions to that specification.
1044 Note that the specification explicitly says that bc(1) only accepts numbers that
1045 use a period (**.**) as a radix point, regardless of the value of
1050 None are known. Report bugs at https://git.yzena.com/gavin/bc.
1054 Gavin D. Howard <yzena.tech@gmail.com> and contributors.
1056 [1]: https://pubs.opengroup.org/onlinepubs/9699919799/utilities/bc.html
1057 [2]: https://www.gnu.org/software/bc/
1058 [3]: https://en.wikipedia.org/wiki/Rounding#Round_half_away_from_zero
1059 [4]: https://en.wikipedia.org/wiki/Unit_in_the_last_place
1060 [5]: https://people.eecs.berkeley.edu/~wkahan/LOG10HAF.TXT
1061 [6]: https://en.wikipedia.org/wiki/Rounding#Rounding_away_from_zero