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 decimal arithmetic language and calculator
37 **bc** [**-ghilPqsvVw**] [**--global-stacks**] [**--help**] [**--interactive**] [**--mathlib**] [**--no-prompt**] [**--quiet**] [**--standard**] [**--warn**] [**--version**] [**-e** *expr*] [**--expression**=*expr*...] [**-f** *file*...] [**-file**=*file*...]
42 bc(1) is an interactive processor for a language first standardized in 1991 by
43 POSIX. (The current standard is [here][1].) The language provides unlimited
44 precision decimal arithmetic and is somewhat C-like, but there are differences.
45 Such differences will be noted in this document.
47 After parsing and handling options, this bc(1) reads any files given on the
48 command line and executes them before reading from **stdin**.
52 The following are the options that bc(1) accepts.
54 **-g**, **--global-stacks**
56 Turns the globals **ibase**, **obase**, and **scale** into stacks.
58 This has the effect that a copy of the current value of all three are pushed
59 onto a stack for every function call, as well as popped when every function
60 returns. This means that functions can assign to any and all of those
61 globals without worrying that the change will affect other functions.
62 Thus, a hypothetical function named **output(x,b)** that simply printed
63 **x** in base **b** could be written like this:
65 define void output(x, b) {
72 define void output(x, b) {
80 This makes writing functions much easier.
82 However, since using this flag means that functions cannot set **ibase**,
83 **obase**, or **scale** globally, functions that are made to do so cannot
84 work anymore. There are two possible use cases for that, and each has a
87 First, if a function is called on startup to turn bc(1) into a number
88 converter, it is possible to replace that capability with various shell
91 alias d2o="bc -e ibase=A -e obase=8"
92 alias h2b="bc -e ibase=G -e obase=2"
94 Second, if the purpose of a function is to set **ibase**, **obase**, or
95 **scale** globally for any other purpose, it could be split into one to
96 three functions (based on how many globals it sets) and each of those
97 functions could return the desired value for a global.
99 If the behavior of this option is desired for every run of bc(1), then users
100 could make sure to define **BC_ENV_ARGS** and include this option (see the
101 **ENVIRONMENT VARIABLES** section for more details).
103 If **-s**, **-w**, or any equivalents are used, this option is ignored.
105 This is a **non-portable extension**.
109 : Prints a usage message and quits.
111 **-i**, **--interactive**
113 : Forces interactive mode. (See the **INTERACTIVE MODE** section.)
115 This is a **non-portable extension**.
117 **-l**, **--mathlib**
119 : Sets **scale** (see the **SYNTAX** section) to **20** and loads the included
120 math library before running any code, including any expressions or files
121 specified on the command line.
123 To learn what is in the library, see the **LIBRARY** section.
125 **-P**, **--no-prompt**
127 : This option is a no-op.
129 This is a **non-portable extension**.
133 : This option is for compatibility with the [GNU bc(1)][2]; it is a no-op.
134 Without this option, GNU bc(1) prints a copyright header. This bc(1) only
135 prints the copyright header if one or more of the **-v**, **-V**, or
136 **--version** options are given.
138 This is a **non-portable extension**.
140 **-s**, **--standard**
142 : Process exactly the language defined by the [standard][1] and error if any
145 This is a **non-portable extension**.
147 **-v**, **-V**, **--version**
149 : Print the version information (copyright header) and exit.
151 This is a **non-portable extension**.
155 : Like **-s** and **--standard**, except that warnings (and not errors) are
156 printed for non-standard extensions and execution continues normally.
158 This is a **non-portable extension**.
160 **-e** *expr*, **--expression**=*expr*
162 : Evaluates *expr*. If multiple expressions are given, they are evaluated in
163 order. If files are given as well (see below), the expressions and files are
164 evaluated in the order given. This means that if a file is given before an
165 expression, the file is read in and evaluated first.
167 After processing all expressions and files, bc(1) will exit, unless **-**
168 (**stdin**) was given as an argument at least once to **-f** or **--file**.
169 However, if any other **-e**, **--expression**, **-f**, or **--file**
170 arguments are given after that, bc(1) will give a fatal error and exit.
172 This is a **non-portable extension**.
174 **-f** *file*, **--file**=*file*
176 : Reads in *file* and evaluates it, line by line, as though it were read
177 through **stdin**. If expressions are also given (see above), the
178 expressions are evaluated in the order given.
180 After processing all expressions and files, bc(1) will exit, unless **-**
181 (**stdin**) was given as an argument at least once to **-f** or **--file**.
183 This is a **non-portable extension**.
185 All long options are **non-portable extensions**.
189 Any non-error output is written to **stdout**.
191 **Note**: Unlike other bc(1) implementations, this bc(1) will issue a fatal
192 error (see the **EXIT STATUS** section) if it cannot write to **stdout**, so if
193 **stdout** is closed, as in **bc <file> >&-**, it will quit with an error. This
194 is done so that bc(1) can report problems when **stdout** is redirected to a
197 If there are scripts that depend on the behavior of other bc(1) implementations,
198 it is recommended that those scripts be changed to redirect **stdout** to
203 Any error output is written to **stderr**.
205 **Note**: Unlike other bc(1) implementations, this bc(1) will issue a fatal
206 error (see the **EXIT STATUS** section) if it cannot write to **stderr**, so if
207 **stderr** is closed, as in **bc <file> 2>&-**, it will quit with an error. This
208 is done so that bc(1) can exit with an error code when **stderr** is redirected
211 If there are scripts that depend on the behavior of other bc(1) implementations,
212 it is recommended that those scripts be changed to redirect **stderr** to
217 The syntax for bc(1) programs is mostly C-like, with some differences. This
218 bc(1) follows the [POSIX standard][1], which is a much more thorough resource
219 for the language this bc(1) accepts. This section is meant to be a summary and a
220 listing of all the extensions to the standard.
222 In the sections below, **E** means expression, **S** means statement, and **I**
225 Identifiers (**I**) start with a lowercase letter and can be followed by any
226 number (up to **BC_NAME_MAX-1**) of lowercase letters (**a-z**), digits
227 (**0-9**), and underscores (**\_**). The regex is **\[a-z\]\[a-z0-9\_\]\***.
228 Identifiers with more than one character (letter) are a
229 **non-portable extension**.
231 **ibase** is a global variable determining how to interpret constant numbers. It
232 is the "input" base, or the number base used for interpreting input numbers.
233 **ibase** is initially **10**. If the **-s** (**--standard**) and **-w**
234 (**--warn**) flags were not given on the command line, the max allowable value
235 for **ibase** is **36**. Otherwise, it is **16**. The min allowable value for
236 **ibase** is **2**. The max allowable value for **ibase** can be queried in
237 bc(1) programs with the **maxibase()** built-in function.
239 **obase** is a global variable determining how to output results. It is the
240 "output" base, or the number base used for outputting numbers. **obase** is
241 initially **10**. The max allowable value for **obase** is **BC_BASE_MAX** and
242 can be queried in bc(1) programs with the **maxobase()** built-in function. The
243 min allowable value for **obase** is **2**. Values are output in the specified
246 The *scale* of an expression is the number of digits in the result of the
247 expression right of the decimal point, and **scale** is a global variable that
248 sets the precision of any operations, with exceptions. **scale** is initially
249 **0**. **scale** cannot be negative. The max allowable value for **scale** is
250 **BC_SCALE_MAX** and can be queried in bc(1) programs with the **maxscale()**
253 bc(1) has both *global* variables and *local* variables. All *local*
254 variables are local to the function; they are parameters or are introduced in
255 the **auto** list of a function (see the **FUNCTIONS** section). If a variable
256 is accessed which is not a parameter or in the **auto** list, it is assumed to
257 be *global*. If a parent function has a *local* variable version of a variable
258 that a child function considers *global*, the value of that *global* variable in
259 the child function is the value of the variable in the parent function, not the
260 value of the actual *global* variable.
262 All of the above applies to arrays as well.
264 The value of a statement that is an expression (i.e., any of the named
265 expressions or operands) is printed unless the lowest precedence operator is an
266 assignment operator *and* the expression is notsurrounded by parentheses.
268 The value that is printed is also assigned to the special variable **last**. A
269 single dot (**.**) may also be used as a synonym for **last**. These are
270 **non-portable extensions**.
272 Either semicolons or newlines may separate statements.
276 There are two kinds of comments:
278 1. Block comments are enclosed in **/\*** and **\*/**.
279 2. Line comments go from **#** until, and not including, the next newline. This
280 is a **non-portable extension**.
284 The following are named expressions in bc(1):
287 2. Array Elements: **I[E]**
291 6. **last** or a single dot (**.**)
293 Number 6 is a **non-portable extension**.
295 Variables and arrays do not interfere; users can have arrays named the same as
296 variables. This also applies to functions (see the **FUNCTIONS** section), so a
297 user can have a variable, array, and function that all have the same name, and
298 they will not shadow each other, whether inside of functions or not.
300 Named expressions are required as the operand of **increment**/**decrement**
301 operators and as the left side of **assignment** operators (see the *Operators*
306 The following are valid operands in bc(1):
308 1. Numbers (see the *Numbers* subsection below).
309 2. Array indices (**I[E]**).
310 3. **(E)**: The value of **E** (used to change precedence).
311 4. **sqrt(E)**: The square root of **E**. **E** must be non-negative.
312 5. **length(E)**: The number of significant decimal digits in **E**.
313 6. **length(I[])**: The number of elements in the array **I**. This is a
314 **non-portable extension**.
315 7. **scale(E)**: The *scale* of **E**.
316 8. **abs(E)**: The absolute value of **E**. This is a **non-portable
318 9. **I()**, **I(E)**, **I(E, E)**, and so on, where **I** is an identifier for
319 a non-**void** function (see the *Void Functions* subsection of the
320 **FUNCTIONS** section). The **E** argument(s) may also be arrays of the form
321 **I[]**, which will automatically be turned into array references (see the
322 *Array References* subsection of the **FUNCTIONS** section) if the
323 corresponding parameter in the function definition is an array reference.
324 10. **read()**: Reads a line from **stdin** and uses that as an expression. The
325 result of that expression is the result of the **read()** operand. This is a
326 **non-portable extension**.
327 11. **maxibase()**: The max allowable **ibase**. This is a **non-portable
329 12. **maxobase()**: The max allowable **obase**. This is a **non-portable
331 13. **maxscale()**: The max allowable **scale**. This is a **non-portable
336 Numbers are strings made up of digits, uppercase letters, and at most **1**
337 period for a radix. Numbers can have up to **BC_NUM_MAX** digits. Uppercase
338 letters are equal to **9** + their position in the alphabet (i.e., **A** equals
339 **10**, or **9+1**). If a digit or letter makes no sense with the current value
340 of **ibase**, they are set to the value of the highest valid digit in **ibase**.
342 Single-character numbers (i.e., **A** alone) take the value that they would have
343 if they were valid digits, regardless of the value of **ibase**. This means that
344 **A** alone always equals decimal **10** and **Z** alone always equals decimal
349 The following arithmetic and logical operators can be used. They are listed in
350 order of decreasing precedence. Operators in the same group have the same
355 : Type: Prefix and Postfix
359 Description: **increment**, **decrement**
367 Description: **negation**, **boolean not**
375 Description: **power**
383 Description: **multiply**, **divide**, **modulus**
391 Description: **add**, **subtract**
393 **=** **+=** **-=** **\*=** **/=** **%=** **\^=**
399 Description: **assignment**
401 **==** **\<=** **\>=** **!=** **\<** **\>**
407 Description: **relational**
415 Description: **boolean and**
423 Description: **boolean or**
425 The operators will be described in more detail below.
429 : The prefix and postfix **increment** and **decrement** operators behave
430 exactly like they would in C. They require a named expression (see the
431 *Named Expressions* subsection) as an operand.
433 The prefix versions of these operators are more efficient; use them where
438 : The **negation** operator returns **0** if a user attempts to negate any
439 expression with the value **0**. Otherwise, a copy of the expression with
440 its sign flipped is returned.
444 : The **boolean not** operator returns **1** if the expression is **0**, or
447 This is a **non-portable extension**.
451 : The **power** operator (not the **exclusive or** operator, as it would be in
452 C) takes two expressions and raises the first to the power of the value of
453 the second. The *scale* of the result is equal to **scale**.
455 The second expression must be an integer (no *scale*), and if it is
456 negative, the first value must be non-zero.
460 : The **multiply** operator takes two expressions, multiplies them, and
461 returns the product. If **a** is the *scale* of the first expression and
462 **b** is the *scale* of the second expression, the *scale* of the result is
463 equal to **min(a+b,max(scale,a,b))** where **min()** and **max()** return
468 : The **divide** operator takes two expressions, divides them, and returns the
469 quotient. The *scale* of the result shall be the value of **scale**.
471 The second expression must be non-zero.
475 : The **modulus** operator takes two expressions, **a** and **b**, and
476 evaluates them by 1) Computing **a/b** to current **scale** and 2) Using the
477 result of step 1 to calculate **a-(a/b)\*b** to *scale*
478 **max(scale+scale(b),scale(a))**.
480 The second expression must be non-zero.
484 : The **add** operator takes two expressions, **a** and **b**, and returns the
485 sum, with a *scale* equal to the max of the *scale*s of **a** and **b**.
489 : The **subtract** operator takes two expressions, **a** and **b**, and
490 returns the difference, with a *scale* equal to the max of the *scale*s of
493 **=** **+=** **-=** **\*=** **/=** **%=** **\^=**
495 : The **assignment** operators take two expressions, **a** and **b** where
496 **a** is a named expression (see the *Named Expressions* subsection).
498 For **=**, **b** is copied and the result is assigned to **a**. For all
499 others, **a** and **b** are applied as operands to the corresponding
500 arithmetic operator and the result is assigned to **a**.
502 **==** **\<=** **\>=** **!=** **\<** **\>**
504 : The **relational** operators compare two expressions, **a** and **b**, and
505 if the relation holds, according to C language semantics, the result is
506 **1**. Otherwise, it is **0**.
508 Note that unlike in C, these operators have a lower precedence than the
509 **assignment** operators, which means that **a=b\>c** is interpreted as
512 Also, unlike the [standard][1] requires, these operators can appear anywhere
513 any other expressions can be used. This allowance is a
514 **non-portable extension**.
518 : The **boolean and** operator takes two expressions and returns **1** if both
519 expressions are non-zero, **0** otherwise.
521 This is *not* a short-circuit operator.
523 This is a **non-portable extension**.
527 : The **boolean or** operator takes two expressions and returns **1** if one
528 of the expressions is non-zero, **0** otherwise.
530 This is *not* a short-circuit operator.
532 This is a **non-portable extension**.
536 The following items are statements:
539 2. **{** **S** **;** ... **;** **S** **}**
540 3. **if** **(** **E** **)** **S**
541 4. **if** **(** **E** **)** **S** **else** **S**
542 5. **while** **(** **E** **)** **S**
543 6. **for** **(** **E** **;** **E** **;** **E** **)** **S**
544 7. An empty statement
550 13. A string of characters, enclosed in double quotes
551 14. **print** **E** **,** ... **,** **E**
552 15. **I()**, **I(E)**, **I(E, E)**, and so on, where **I** is an identifier for
553 a **void** function (see the *Void Functions* subsection of the
554 **FUNCTIONS** section). The **E** argument(s) may also be arrays of the form
555 **I[]**, which will automatically be turned into array references (see the
556 *Array References* subsection of the **FUNCTIONS** section) if the
557 corresponding parameter in the function definition is an array reference.
559 Numbers 4, 9, 11, 12, 14, and 15 are **non-portable extensions**.
561 Also, as a **non-portable extension**, any or all of the expressions in the
562 header of a for loop may be omitted. If the condition (second expression) is
563 omitted, it is assumed to be a constant **1**.
565 The **break** statement causes a loop to stop iterating and resume execution
566 immediately following a loop. This is only allowed in loops.
568 The **continue** statement causes a loop iteration to stop early and returns to
569 the start of the loop, including testing the loop condition. This is only
572 The **if** **else** statement does the same thing as in C.
574 The **quit** statement causes bc(1) to quit, even if it is on a branch that will
575 not be executed (it is a compile-time command).
577 The **halt** statement causes bc(1) to quit, if it is executed. (Unlike **quit**
578 if it is on a branch of an **if** statement that is not executed, bc(1) does not
581 The **limits** statement prints the limits that this bc(1) is subject to. This
582 is like the **quit** statement in that it is a compile-time command.
584 An expression by itself is evaluated and printed, followed by a newline.
588 The "expressions" in a **print** statement may also be strings. If they are, there
589 are backslash escape sequences that are interpreted specially. What those
590 sequences are, and what they cause to be printed, are shown below:
604 Any other character following a backslash causes the backslash and character to
607 Any non-string expression in a print statement shall be assigned to **last**,
608 like any other expression that is printed.
610 ## Order of Evaluation
612 All expressions in a statment are evaluated left to right, except as necessary
613 to maintain order of operations. This means, for example, assuming that **i** is
614 equal to **0**, in the expression
618 the first (or 0th) element of **a** is set to **1**, and **i** is equal to **2**
619 at the end of the expression.
621 This includes function arguments. Thus, assuming **i** is equal to **0**, this
622 means that in the expression
626 the first argument passed to **x()** is **0**, and the second argument is **1**,
627 while **i** is equal to **2** before the function starts executing.
631 Function definitions are as follows:
641 Any **I** in the parameter list or **auto** list may be replaced with **I[]** to
642 make a parameter or **auto** var an array, and any **I** in the parameter list
643 may be replaced with **\*I[]** to make a parameter an array reference. Callers
644 of functions that take array references should not put an asterisk in the call;
645 they must be called with just **I[]** like normal array parameters and will be
646 automatically converted into references.
648 As a **non-portable extension**, the opening brace of a **define** statement may
649 appear on the next line.
651 As a **non-portable extension**, the return statement may also be in one of the
655 2. **return** **(** **)**
658 The first two, or not specifying a **return** statement, is equivalent to
659 **return (0)**, unless the function is a **void** function (see the *Void
660 Functions* subsection below).
664 Functions can also be **void** functions, defined as follows:
667 define void I(I,...,I){
674 They can only be used as standalone expressions, where such an expression would
675 be printed alone, except in a print statement.
677 Void functions can only use the first two **return** statements listed above.
678 They can also omit the return statement entirely.
680 The word "void" is not treated as a keyword; it is still possible to have
681 variables, arrays, and functions named **void**. The word "void" is only
682 treated specially right after the **define** keyword.
684 This is a **non-portable extension**.
688 For any array in the parameter list, if the array is declared in the form
694 it is a **reference**. Any changes to the array in the function are reflected,
695 when the function returns, to the array that was passed in.
697 Other than this, all function arguments are passed by value.
699 This is a **non-portable extension**.
703 All of the functions below are available when the **-l** or **--mathlib**
704 command-line flags are given.
708 The [standard][1] defines the following functions for the math library:
712 : Returns the sine of **x**, which is assumed to be in radians.
714 This is a transcendental function (see the *Transcendental Functions*
719 : Returns the cosine of **x**, which is assumed to be in radians.
721 This is a transcendental function (see the *Transcendental Functions*
726 : Returns the arctangent of **x**, in radians.
728 This is a transcendental function (see the *Transcendental Functions*
733 : Returns the natural logarithm of **x**.
735 This is a transcendental function (see the *Transcendental Functions*
740 : Returns the mathematical constant **e** raised to the power of **x**.
742 This is a transcendental function (see the *Transcendental Functions*
747 : Returns the bessel integer order **n** (truncated) of **x**.
749 This is a transcendental function (see the *Transcendental Functions*
752 ## Transcendental Functions
754 All transcendental functions can return slightly inaccurate results (up to 1
755 [ULP][4]). This is unavoidable, and [this article][5] explains why it is
756 impossible and unnecessary to calculate exact results for the transcendental
759 Because of the possible inaccuracy, I recommend that users call those functions
760 with the precision (**scale**) set to at least 1 higher than is necessary. If
761 exact results are *absolutely* required, users can double the precision
762 (**scale**) and then truncate.
764 The transcendental functions in the standard math library are:
775 When bc(1) encounters an error or a signal that it has a non-default handler
776 for, it resets. This means that several things happen.
778 First, any functions that are executing are stopped and popped off the stack.
779 The behavior is not unlike that of exceptions in programming languages. Then
780 the execution point is set so that any code waiting to execute (after all
781 functions returned) is skipped.
783 Thus, when bc(1) resets, it skips any remaining code waiting to be executed.
784 Then, if it is interactive mode, and the error was not a fatal error (see the
785 **EXIT STATUS** section), it asks for more input; otherwise, it exits with the
786 appropriate return code.
788 Note that this reset behavior is different from the GNU bc(1), which attempts to
789 start executing the statement right after the one that caused an error.
793 Most bc(1) implementations use **char** types to calculate the value of **1**
794 decimal digit at a time, but that can be slow. This bc(1) does something
797 It uses large integers to calculate more than **1** decimal digit at a time. If
798 built in a environment where **BC_LONG_BIT** (see the **LIMITS** section) is
799 **64**, then each integer has **9** decimal digits. If built in an environment
800 where **BC_LONG_BIT** is **32** then each integer has **4** decimal digits. This
801 value (the number of decimal digits per large integer) is called
804 The actual values of **BC_LONG_BIT** and **BC_BASE_DIGS** can be queried with
805 the **limits** statement.
807 In addition, this bc(1) uses an even larger integer for overflow checking. This
808 integer type depends on the value of **BC_LONG_BIT**, but is always at least
809 twice as large as the integer type used to store digits.
813 The following are the limits on bc(1):
817 : The number of bits in the **long** type in the environment where bc(1) was
818 built. This determines how many decimal digits can be stored in a single
819 large integer (see the **PERFORMANCE** section).
823 : The number of decimal digits per large integer (see the **PERFORMANCE**
824 section). Depends on **BC_LONG_BIT**.
828 : The max decimal number that each large integer can store (see
829 **BC_BASE_DIGS**) plus **1**. Depends on **BC_BASE_DIGS**.
833 : The max number that the overflow type (see the **PERFORMANCE** section) can
834 hold. Depends on **BC_LONG_BIT**.
838 : The maximum output base. Set at **BC_BASE_POW**.
842 : The maximum size of arrays. Set at **SIZE_MAX-1**.
846 : The maximum **scale**. Set at **BC_OVERFLOW_MAX-1**.
850 : The maximum length of strings. Set at **BC_OVERFLOW_MAX-1**.
854 : The maximum length of identifiers. Set at **BC_OVERFLOW_MAX-1**.
858 : The maximum length of a number (in decimal digits), which includes digits
859 after the decimal point. Set at **BC_OVERFLOW_MAX-1**.
863 : The maximum allowable exponent (positive or negative). Set at
868 : The maximum number of vars/arrays. Set at **SIZE_MAX-1**.
870 The actual values can be queried with the **limits** statement.
872 These limits are meant to be effectively non-existent; the limits are so large
873 (at least on 64-bit machines) that there should not be any point at which they
874 become a problem. In fact, memory should be exhausted before these limits should
877 # ENVIRONMENT VARIABLES
879 bc(1) recognizes the following environment variables:
883 : If this variable exists (no matter the contents), bc(1) behaves as if
884 the **-s** option was given.
888 : This is another way to give command-line arguments to bc(1). They should be
889 in the same format as all other command-line arguments. These are always
890 processed first, so any files given in **BC_ENV_ARGS** will be processed
891 before arguments and files given on the command-line. This gives the user
892 the ability to set up "standard" options and files to be used at every
893 invocation. The most useful thing for such files to contain would be useful
894 functions that the user might want every time bc(1) runs.
896 The code that parses **BC_ENV_ARGS** will correctly handle quoted arguments,
897 but it does not understand escape sequences. For example, the string
898 **"/home/gavin/some bc file.bc"** will be correctly parsed, but the string
899 **"/home/gavin/some \"bc\" file.bc"** will include the backslashes.
901 The quote parsing will handle either kind of quotes, **'** or **"**. Thus,
902 if you have a file with any number of single quotes in the name, you can use
903 double quotes as the outside quotes, as in **"some 'bc' file.bc"**, and vice
904 versa if you have a file with double quotes. However, handling a file with
905 both kinds of quotes in **BC_ENV_ARGS** is not supported due to the
906 complexity of the parsing, though such files are still supported on the
907 command-line where the parsing is done by the shell.
911 : If this environment variable exists and contains an integer that is greater
912 than **1** and is less than **UINT16_MAX** (**2\^16-1**), bc(1) will output
913 lines to that length, including the backslash (**\\**). The default line
918 bc(1) returns the following exit statuses:
926 : A math error occurred. This follows standard practice of using **1** for
927 expected errors, since math errors will happen in the process of normal
930 Math errors include divide by **0**, taking the square root of a negative
931 number, attempting to convert a negative number to a hardware integer,
932 overflow when converting a number to a hardware integer, and attempting to
933 use a non-integer where an integer is required.
935 Converting to a hardware integer happens for the second operand of the power
936 (**\^**) operator and the corresponding assignment operator.
940 : A parse error occurred.
942 Parse errors include unexpected **EOF**, using an invalid character, failing
943 to find the end of a string or comment, using a token where it is invalid,
944 giving an invalid expression, giving an invalid print statement, giving an
945 invalid function definition, attempting to assign to an expression that is
946 not a named expression (see the *Named Expressions* subsection of the
947 **SYNTAX** section), giving an invalid **auto** list, having a duplicate
948 **auto**/function parameter, failing to find the end of a code block,
949 attempting to return a value from a **void** function, attempting to use a
950 variable as a reference, and using any extensions when the option **-s** or
951 any equivalents were given.
955 : A runtime error occurred.
957 Runtime errors include assigning an invalid number to **ibase**, **obase**,
958 or **scale**; give a bad expression to a **read()** call, calling **read()**
959 inside of a **read()** call, type errors, passing the wrong number of
960 arguments to functions, attempting to call an undefined function, and
961 attempting to use a **void** function call as a value in an expression.
965 : A fatal error occurred.
967 Fatal errors include memory allocation errors, I/O errors, failing to open
968 files, attempting to use files that do not have only ASCII characters (bc(1)
969 only accepts ASCII characters), attempting to open a directory as a file,
970 and giving invalid command-line options.
972 The exit status **4** is special; when a fatal error occurs, bc(1) always exits
973 and returns **4**, no matter what mode bc(1) is in.
975 The other statuses will only be returned when bc(1) is not in interactive mode
976 (see the **INTERACTIVE MODE** section), since bc(1) resets its state (see the
977 **RESET** section) and accepts more input when one of those errors occurs in
978 interactive mode. This is also the case when interactive mode is forced by the
979 **-i** flag or **--interactive** option.
981 These exit statuses allow bc(1) to be used in shell scripting with error
982 checking, and its normal behavior can be forced by using the **-i** flag or
983 **--interactive** option.
987 Per the [standard][1], bc(1) has an interactive mode and a non-interactive mode.
988 Interactive mode is turned on automatically when both **stdin** and **stdout**
989 are hooked to a terminal, but the **-i** flag and **--interactive** option can
990 turn it on in other cases.
992 In interactive mode, bc(1) attempts to recover from errors (see the **RESET**
993 section), and in normal execution, flushes **stdout** as soon as execution is
994 done for the current input.
998 If **stdin**, **stdout**, and **stderr** are all connected to a TTY, bc(1) turns
1001 TTY mode is different from interactive mode because interactive mode is required
1002 in the [bc(1) specification][1], and interactive mode requires only **stdin**
1003 and **stdout** to be connected to a terminal.
1007 Sending a **SIGINT** will cause bc(1) to stop execution of the current input. If
1008 bc(1) is in TTY mode (see the **TTY MODE** section), it will reset (see the
1009 **RESET** section). Otherwise, it will clean up and exit.
1011 Note that "current input" can mean one of two things. If bc(1) is processing
1012 input from **stdin** in TTY mode, it will ask for more input. If bc(1) is
1013 processing input from a file in TTY mode, it will stop processing the file and
1014 start processing the next file, if one exists, or ask for input from **stdin**
1015 if no other file exists.
1017 This means that if a **SIGINT** is sent to bc(1) as it is executing a file, it
1018 can seem as though bc(1) did not respond to the signal since it will immediately
1019 start executing the next file. This is by design; most files that users execute
1020 when interacting with bc(1) have function definitions, which are quick to parse.
1021 If a file takes a long time to execute, there may be a bug in that file. The
1022 rest of the files could still be executed without problem, allowing the user to
1025 **SIGTERM** and **SIGQUIT** cause bc(1) to clean up and exit, and it uses the
1026 default handler for all other signals.
1034 bc(1) is compliant with the [IEEE Std 1003.1-2017 (“POSIX.1-2017”)][1]
1035 specification. The flags **-efghiqsvVw**, all long options, and the extensions
1036 noted above are extensions to that specification.
1038 Note that the specification explicitly says that bc(1) only accepts numbers that
1039 use a period (**.**) as a radix point, regardless of the value of
1044 None are known. Report bugs at https://git.yzena.com/gavin/bc.
1048 Gavin D. Howard <gavin@yzena.com> and contributors.
1050 [1]: https://pubs.opengroup.org/onlinepubs/9699919799/utilities/bc.html
1051 [2]: https://www.gnu.org/software/bc/
1052 [3]: https://en.wikipedia.org/wiki/Rounding#Round_half_away_from_zero
1053 [4]: https://en.wikipedia.org/wiki/Unit_in_the_last_place
1054 [5]: https://people.eecs.berkeley.edu/~wkahan/LOG10HAF.TXT
1055 [6]: https://en.wikipedia.org/wiki/Rounding#Rounding_away_from_zero