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** [**-ghilPqRsvVw**] [**-\-global-stacks**] [**-\-help**] [**-\-interactive**] [**-\-mathlib**] [**-\-no-prompt**] [**-\-no-read-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**.
50 This bc(1) is a drop-in replacement for *any* bc(1), including (and
51 especially) the GNU bc(1). It also has many extensions and extra features beyond
52 other implementations.
55 This bc(1) is a drop-in replacement for *any* bc(1), including (and
56 especially) the GNU bc(1).
61 The following are the options that bc(1) accepts.
63 **-g**, **-\-global-stacks**
65 {{ A H N P HN HP NP HNP }}
66 : Turns the globals **ibase**, **obase**, **scale**, and **seed** into stacks.
68 This has the effect that a copy of the current value of all four are pushed
70 {{ E EH EN EP EHN EHP ENP EHNP }}
71 Turns the globals **ibase**, **obase**, and **scale** into stacks.
73 This has the effect that a copy of the current value of all three are pushed
75 onto a stack for every function call, as well as popped when every function
76 returns. This means that functions can assign to any and all of those
77 globals without worrying that the change will affect other functions.
78 Thus, a hypothetical function named **output(x,b)** that simply printed
79 **x** in base **b** could be written like this:
81 define void output(x, b) {
88 define void output(x, b) {
96 This makes writing functions much easier.
98 {{ A H N P HN HP NP HNP }}
99 (**Note**: the function **output(x,b)** exists in the extended math library.
100 See the **LIBRARY** section.)
102 However, since using this flag means that functions cannot set **ibase**,
103 **obase**, **scale**, or **seed** globally, functions that are made to do so
104 cannot work anymore. There are two possible use cases for that, and each has
107 {{ E EH EN EP EHN EHP ENP EHNP }}
108 However, since using this flag means that functions cannot set **ibase**,
109 **obase**, or **scale** globally, functions that are made to do so cannot
110 work anymore. There are two possible use cases for that, and each has a
114 First, if a function is called on startup to turn bc(1) into a number
115 converter, it is possible to replace that capability with various shell
118 alias d2o="bc -e ibase=A -e obase=8"
119 alias h2b="bc -e ibase=G -e obase=2"
121 {{ A H N P HN HP NP HNP }}
122 Second, if the purpose of a function is to set **ibase**, **obase**,
123 **scale**, or **seed** globally for any other purpose, it could be split
124 into one to four functions (based on how many globals it sets) and each of
125 those functions could return the desired value for a global.
127 For functions that set **seed**, the value assigned to **seed** is not
128 propagated to parent functions. This means that the sequence of
129 pseudo-random numbers that they see will not be the same sequence of
130 pseudo-random numbers that any parent sees. This is only the case once
131 **seed** has been set.
133 If a function desires to not affect the sequence of pseudo-random numbers
134 of its parents, but wants to use the same **seed**, it can use the following
139 {{ E EH EN EP EHN EHP ENP EHNP }}
140 Second, if the purpose of a function is to set **ibase**, **obase**, or
141 **scale** globally for any other purpose, it could be split into one to
142 three functions (based on how many globals it sets) and each of those
143 functions could return the desired value for a global.
146 If the behavior of this option is desired for every run of bc(1), then users
147 could make sure to define **BC_ENV_ARGS** and include this option (see the
148 **ENVIRONMENT VARIABLES** section for more details).
150 If **-s**, **-w**, or any equivalents are used, this option is ignored.
152 This is a **non-portable extension**.
156 : Prints a usage message and quits.
158 **-i**, **-\-interactive**
160 : Forces interactive mode. (See the **INTERACTIVE MODE** section.)
162 This is a **non-portable extension**.
164 **-l**, **-\-mathlib**
166 : Sets **scale** (see the **SYNTAX** section) to **20** and loads the included
167 {{ A H N P HN HP NP HNP }}
168 math library and the extended math library before running any code,
169 including any expressions or files specified on the command line.
171 To learn what is in the libraries, see the **LIBRARY** section.
173 {{ E EH EN EP EHN EHP ENP EHNP }}
174 math library before running any code, including any expressions or files
175 specified on the command line.
177 To learn what is in the library, see the **LIBRARY** section.
180 **-P**, **-\-no-prompt**
182 {{ A E H N EH EN HN EHN }}
183 : Disables the prompt in TTY mode. (The prompt is only enabled in TTY mode.
184 See the **TTY MODE** section.) This is mostly for those users that do not
185 want a prompt or are not used to having them in bc(1). Most of those users
186 would want to put this option in **BC_ENV_ARGS** (see the
187 **ENVIRONMENT VARIABLES** section).
189 {{ P EP HP NP EHP ENP HNP EHNP }}
190 : This option is a no-op.
193 This is a **non-portable extension**.
195 **-R**, **-\-no-read-prompt**
197 {{ A E H N EH EN HN EHN }}
198 : Disables the read prompt in TTY mode. (The read prompt is only enabled in
199 TTY mode. See the **TTY MODE** section.) This is mostly for those users that
200 do not want a read prompt or are not used to having them in bc(1). Most of
201 those users would want to put this option in **BC_ENV_ARGS** (see the
202 **ENVIRONMENT VARIABLES** section). This option is also useful in hash bang
203 lines of bc(1) scripts that prompt for user input.
205 This option does not disable the regular prompt because the read prompt is
206 only used when the **read()** built-in function is called.
208 {{ P EP HP NP EHP ENP HNP EHNP }}
209 : Because bc(1) was built without support for prompts, this option is a no-op.
212 This is a **non-portable extension**.
216 : This option is for compatibility with the [GNU bc(1)][2]; it is a no-op.
217 Without this option, GNU bc(1) prints a copyright header. This bc(1) only
218 prints the copyright header if one or more of the **-v**, **-V**, or
219 **-\-version** options are given.
221 This is a **non-portable extension**.
223 **-s**, **-\-standard**
225 : Process exactly the language defined by the [standard][1] and error if any
228 This is a **non-portable extension**.
230 **-v**, **-V**, **-\-version**
232 : Print the version information (copyright header) and exit.
234 This is a **non-portable extension**.
238 : Like **-s** and **-\-standard**, except that warnings (and not errors) are
239 printed for non-standard extensions and execution continues normally.
241 This is a **non-portable extension**.
243 **-e** *expr*, **-\-expression**=*expr*
245 : Evaluates *expr*. If multiple expressions are given, they are evaluated in
246 order. If files are given as well (see below), the expressions and files are
247 evaluated in the order given. This means that if a file is given before an
248 expression, the file is read in and evaluated first.
250 If this option is given on the command-line (i.e., not in **BC_ENV_ARGS**,
251 see the **ENVIRONMENT VARIABLES** section), then after processing all
252 expressions and files, bc(1) will exit, unless **-** (**stdin**) was given
253 as an argument at least once to **-f** or **-\-file**, whether on the
254 command-line or in **BC_ENV_ARGS**. However, if any other **-e**,
255 **-\-expression**, **-f**, or **-\-file** arguments are given after **-f-**
256 or equivalent is given, bc(1) will give a fatal error and exit.
258 This is a **non-portable extension**.
260 **-f** *file*, **-\-file**=*file*
262 : Reads in *file* and evaluates it, line by line, as though it were read
263 through **stdin**. If expressions are also given (see above), the
264 expressions are evaluated in the order given.
266 If this option is given on the command-line (i.e., not in **BC_ENV_ARGS**,
267 see the **ENVIRONMENT VARIABLES** section), then after processing all
268 expressions and files, bc(1) will exit, unless **-** (**stdin**) was given
269 as an argument at least once to **-f** or **-\-file**. However, if any other
270 **-e**, **-\-expression**, **-f**, or **-\-file** arguments are given after
271 **-f-** or equivalent is given, bc(1) will give a fatal error and exit.
273 This is a **non-portable extension**.
275 All long options are **non-portable extensions**.
279 Any non-error output is written to **stdout**. In addition, if history (see the
280 **HISTORY** section) and the prompt (see the **TTY MODE** section) are enabled,
281 both are output to **stdout**.
283 **Note**: Unlike other bc(1) implementations, this bc(1) will issue a fatal
284 error (see the **EXIT STATUS** section) if it cannot write to **stdout**, so if
285 **stdout** is closed, as in **bc <file> >&-**, it will quit with an error. This
286 is done so that bc(1) can report problems when **stdout** is redirected to a
289 If there are scripts that depend on the behavior of other bc(1) implementations,
290 it is recommended that those scripts be changed to redirect **stdout** to
295 Any error output is written to **stderr**.
297 **Note**: Unlike other bc(1) implementations, this bc(1) will issue a fatal
298 error (see the **EXIT STATUS** section) if it cannot write to **stderr**, so if
299 **stderr** is closed, as in **bc <file> 2>&-**, it will quit with an error. This
300 is done so that bc(1) can exit with an error code when **stderr** is redirected
303 If there are scripts that depend on the behavior of other bc(1) implementations,
304 it is recommended that those scripts be changed to redirect **stderr** to
309 The syntax for bc(1) programs is mostly C-like, with some differences. This
310 bc(1) follows the [POSIX standard][1], which is a much more thorough resource
311 for the language this bc(1) accepts. This section is meant to be a summary and a
312 listing of all the extensions to the standard.
314 In the sections below, **E** means expression, **S** means statement, and **I**
317 Identifiers (**I**) start with a lowercase letter and can be followed by any
318 number (up to **BC_NAME_MAX-1**) of lowercase letters (**a-z**), digits
319 (**0-9**), and underscores (**\_**). The regex is **\[a-z\]\[a-z0-9\_\]\***.
320 Identifiers with more than one character (letter) are a
321 **non-portable extension**.
323 **ibase** is a global variable determining how to interpret constant numbers. It
324 is the "input" base, or the number base used for interpreting input numbers.
325 **ibase** is initially **10**. If the **-s** (**-\-standard**) and **-w**
326 (**-\-warn**) flags were not given on the command line, the max allowable value
327 for **ibase** is **36**. Otherwise, it is **16**. The min allowable value for
328 **ibase** is **2**. The max allowable value for **ibase** can be queried in
329 bc(1) programs with the **maxibase()** built-in function.
331 **obase** is a global variable determining how to output results. It is the
332 "output" base, or the number base used for outputting numbers. **obase** is
333 initially **10**. The max allowable value for **obase** is **BC_BASE_MAX** and
334 can be queried in bc(1) programs with the **maxobase()** built-in function. The
335 {{ A H N P HN HP NP HNP }}
336 min allowable value for **obase** is **0**. If **obase** is **0**, values are
337 output in scientific notation, and if **obase** is **1**, values are output in
338 engineering notation. Otherwise, values are output in the specified base.
340 Outputting in scientific and engineering notations are **non-portable
343 {{ E EH EN EP EHN EHP ENP EHNP }}
344 min allowable value for **obase** is **2**. Values are output in the specified
348 The *scale* of an expression is the number of digits in the result of the
349 expression right of the decimal point, and **scale** is a global variable that
350 sets the precision of any operations, with exceptions. **scale** is initially
351 **0**. **scale** cannot be negative. The max allowable value for **scale** is
352 **BC_SCALE_MAX** and can be queried in bc(1) programs with the **maxscale()**
355 bc(1) has both *global* variables and *local* variables. All *local*
356 variables are local to the function; they are parameters or are introduced in
357 the **auto** list of a function (see the **FUNCTIONS** section). If a variable
358 is accessed which is not a parameter or in the **auto** list, it is assumed to
359 be *global*. If a parent function has a *local* variable version of a variable
360 that a child function considers *global*, the value of that *global* variable in
361 the child function is the value of the variable in the parent function, not the
362 value of the actual *global* variable.
364 All of the above applies to arrays as well.
366 The value of a statement that is an expression (i.e., any of the named
367 expressions or operands) is printed unless the lowest precedence operator is an
368 assignment operator *and* the expression is notsurrounded by parentheses.
370 The value that is printed is also assigned to the special variable **last**. A
371 single dot (**.**) may also be used as a synonym for **last**. These are
372 **non-portable extensions**.
374 Either semicolons or newlines may separate statements.
378 There are two kinds of comments:
380 1. Block comments are enclosed in **/\*** and **\*/**.
381 2. Line comments go from **#** until, and not including, the next newline. This
382 is a **non-portable extension**.
386 The following are named expressions in bc(1):
389 2. Array Elements: **I[E]**
393 {{ A H N P HN HP NP HNP }}
395 7. **last** or a single dot (**.**)
397 Numbers 6 and 7 are **non-portable extensions**.
399 The meaning of **seed** is dependent on the current pseudo-random number
400 generator but is guaranteed to not change except for new major versions.
402 The *scale* and sign of the value may be significant.
404 If a previously used **seed** value is assigned to **seed** and used again, the
405 pseudo-random number generator is guaranteed to produce the same sequence of
406 pseudo-random numbers as it did when the **seed** value was previously used.
408 The exact value assigned to **seed** is not guaranteed to be returned if
409 **seed** is queried again immediately. However, if **seed** *does* return a
410 different value, both values, when assigned to **seed**, are guaranteed to
411 produce the same sequence of pseudo-random numbers. This means that certain
412 values assigned to **seed** will *not* produce unique sequences of pseudo-random
413 numbers. The value of **seed** will change after any use of the **rand()** and
414 **irand(E)** operands (see the *Operands* subsection below), except if the
415 parameter passed to **irand(E)** is **0**, **1**, or negative.
417 There is no limit to the length (number of significant decimal digits) or
418 *scale* of the value that can be assigned to **seed**.
420 {{ E EH EN EP EHN EHP ENP EHNP }}
421 6. **last** or a single dot (**.**)
423 Number 6 is a **non-portable extension**.
426 Variables and arrays do not interfere; users can have arrays named the same as
427 variables. This also applies to functions (see the **FUNCTIONS** section), so a
428 user can have a variable, array, and function that all have the same name, and
429 they will not shadow each other, whether inside of functions or not.
431 Named expressions are required as the operand of **increment**/**decrement**
432 operators and as the left side of **assignment** operators (see the *Operators*
437 The following are valid operands in bc(1):
439 1. Numbers (see the *Numbers* subsection below).
440 2. Array indices (**I[E]**).
441 3. **(E)**: The value of **E** (used to change precedence).
442 4. **sqrt(E)**: The square root of **E**. **E** must be non-negative.
443 5. **length(E)**: The number of significant decimal digits in **E**.
444 6. **length(I[])**: The number of elements in the array **I**. This is a
445 **non-portable extension**.
446 7. **scale(E)**: The *scale* of **E**.
447 8. **abs(E)**: The absolute value of **E**. This is a **non-portable
449 9. **I()**, **I(E)**, **I(E, E)**, and so on, where **I** is an identifier for
450 a non-**void** function (see the *Void Functions* subsection of the
451 **FUNCTIONS** section). The **E** argument(s) may also be arrays of the form
452 **I[]**, which will automatically be turned into array references (see the
453 *Array References* subsection of the **FUNCTIONS** section) if the
454 corresponding parameter in the function definition is an array reference.
455 10. **read()**: Reads a line from **stdin** and uses that as an expression. The
456 result of that expression is the result of the **read()** operand. This is a
457 **non-portable extension**.
458 11. **maxibase()**: The max allowable **ibase**. This is a **non-portable
460 12. **maxobase()**: The max allowable **obase**. This is a **non-portable
462 13. **maxscale()**: The max allowable **scale**. This is a **non-portable
464 {{ A H N P HN HP NP HNP }}
465 14. **rand()**: A pseudo-random integer between **0** (inclusive) and
466 **BC_RAND_MAX** (inclusive). Using this operand will change the value of
467 **seed**. This is a **non-portable extension**.
468 15. **irand(E)**: A pseudo-random integer between **0** (inclusive) and the
469 value of **E** (exclusive). If **E** is negative or is a non-integer
470 (**E**'s *scale* is not **0**), an error is raised, and bc(1) resets (see
471 the **RESET** section) while **seed** remains unchanged. If **E** is larger
472 than **BC_RAND_MAX**, the higher bound is honored by generating several
473 pseudo-random integers, multiplying them by appropriate powers of
474 **BC_RAND_MAX+1**, and adding them together. Thus, the size of integer that
475 can be generated with this operand is unbounded. Using this operand will
476 change the value of **seed**, unless the value of **E** is **0** or **1**.
477 In that case, **0** is returned, and **seed** is *not* changed. This is a
478 **non-portable extension**.
479 16. **maxrand()**: The max integer returned by **rand()**. This is a
480 **non-portable extension**.
482 The integers generated by **rand()** and **irand(E)** are guaranteed to be as
483 unbiased as possible, subject to the limitations of the pseudo-random number
486 **Note**: The values returned by the pseudo-random number generator with
487 **rand()** and **irand(E)** are guaranteed to *NOT* be cryptographically secure.
488 This is a consequence of using a seeded pseudo-random number generator. However,
489 they *are* guaranteed to be reproducible with identical **seed** values. This
490 means that the pseudo-random values from bc(1) should only be used where a
491 reproducible stream of pseudo-random numbers is *ESSENTIAL*. In any other case,
492 use a non-seeded pseudo-random number generator.
497 Numbers are strings made up of digits, uppercase letters, and at most **1**
498 period for a radix. Numbers can have up to **BC_NUM_MAX** digits. Uppercase
499 letters are equal to **9** + their position in the alphabet (i.e., **A** equals
500 **10**, or **9+1**). If a digit or letter makes no sense with the current value
501 of **ibase**, they are set to the value of the highest valid digit in **ibase**.
503 Single-character numbers (i.e., **A** alone) take the value that they would have
504 if they were valid digits, regardless of the value of **ibase**. This means that
505 **A** alone always equals decimal **10** and **Z** alone always equals decimal
508 {{ A H N P HN HP NP HNP }}
509 In addition, bc(1) accepts numbers in scientific notation. These have the form
510 **\<number\>e\<integer\>**. The exponent (the portion after the **e**) must be
511 an integer. An example is **1.89237e9**, which is equal to **1892370000**.
512 Negative exponents are also allowed, so **4.2890e-3** is equal to **0.0042890**.
514 Using scientific notation is an error or warning if the **-s** or **-w**,
515 respectively, command-line options (or equivalents) are given.
517 **WARNING**: Both the number and the exponent in scientific notation are
518 interpreted according to the current **ibase**, but the number is still
519 multiplied by **10\^exponent** regardless of the current **ibase**. For example,
520 if **ibase** is **16** and bc(1) is given the number string **FFeA**, the
521 resulting decimal number will be **2550000000000**, and if bc(1) is given the
522 number string **10e-4**, the resulting decimal number will be **0.0016**.
524 Accepting input as scientific notation is a **non-portable extension**.
529 The following arithmetic and logical operators can be used. They are listed in
530 order of decreasing precedence. Operators in the same group have the same
535 : Type: Prefix and Postfix
539 Description: **increment**, **decrement**
547 Description: **negation**, **boolean not**
549 {{ A H N P HN HP NP HNP }}
556 Description: **truncation**
564 Description: **set precision**
573 Description: **power**
581 Description: **multiply**, **divide**, **modulus**
589 Description: **add**, **subtract**
591 {{ A H N P HN HP NP HNP }}
598 Description: **shift left**, **shift right**
600 **=** **\<\<=** **\>\>=** **+=** **-=** **\*=** **/=** **%=** **\^=** **\@=**
602 {{ E EH EN EP EHN EHP ENP EHNP }}
603 **=** **+=** **-=** **\*=** **/=** **%=** **\^=**
610 Description: **assignment**
612 **==** **\<=** **\>=** **!=** **\<** **\>**
618 Description: **relational**
626 Description: **boolean and**
634 Description: **boolean or**
636 The operators will be described in more detail below.
640 : The prefix and postfix **increment** and **decrement** operators behave
641 exactly like they would in C. They require a named expression (see the
642 *Named Expressions* subsection) as an operand.
644 The prefix versions of these operators are more efficient; use them where
649 : The **negation** operator returns **0** if a user attempts to negate any
650 expression with the value **0**. Otherwise, a copy of the expression with
651 its sign flipped is returned.
655 : The **boolean not** operator returns **1** if the expression is **0**, or
658 This is a **non-portable extension**.
660 {{ A H N P HN HP NP HNP }}
663 : The **truncation** operator returns a copy of the given expression with all
664 of its *scale* removed.
666 This is a **non-portable extension**.
670 : The **set precision** operator takes two expressions and returns a copy of
671 the first with its *scale* equal to the value of the second expression. That
672 could either mean that the number is returned without change (if the
673 *scale* of the first expression matches the value of the second
674 expression), extended (if it is less), or truncated (if it is more).
676 The second expression must be an integer (no *scale*) and non-negative.
678 This is a **non-portable extension**.
683 : The **power** operator (not the **exclusive or** operator, as it would be in
684 C) takes two expressions and raises the first to the power of the value of
685 the second. The *scale* of the result is equal to **scale**.
687 The second expression must be an integer (no *scale*), and if it is
688 negative, the first value must be non-zero.
692 : The **multiply** operator takes two expressions, multiplies them, and
693 returns the product. If **a** is the *scale* of the first expression and
694 **b** is the *scale* of the second expression, the *scale* of the result is
695 equal to **min(a+b,max(scale,a,b))** where **min()** and **max()** return
700 : The **divide** operator takes two expressions, divides them, and returns the
701 quotient. The *scale* of the result shall be the value of **scale**.
703 The second expression must be non-zero.
707 : The **modulus** operator takes two expressions, **a** and **b**, and
708 evaluates them by 1) Computing **a/b** to current **scale** and 2) Using the
709 result of step 1 to calculate **a-(a/b)\*b** to *scale*
710 **max(scale+scale(b),scale(a))**.
712 The second expression must be non-zero.
716 : The **add** operator takes two expressions, **a** and **b**, and returns the
717 sum, with a *scale* equal to the max of the *scale*s of **a** and **b**.
721 : The **subtract** operator takes two expressions, **a** and **b**, and
722 returns the difference, with a *scale* equal to the max of the *scale*s of
725 {{ A H N P HN HP NP HNP }}
728 : The **left shift** operator takes two expressions, **a** and **b**, and
729 returns a copy of the value of **a** with its decimal point moved **b**
732 The second expression must be an integer (no *scale*) and non-negative.
734 This is a **non-portable extension**.
738 : The **right shift** operator takes two expressions, **a** and **b**, and
739 returns a copy of the value of **a** with its decimal point moved **b**
742 The second expression must be an integer (no *scale*) and non-negative.
744 This is a **non-portable extension**.
747 {{ A H N P HN HP NP HNP }}
748 **=** **\<\<=** **\>\>=** **+=** **-=** **\*=** **/=** **%=** **\^=** **\@=**
750 {{ E EH EN EP EHN EHP ENP EHNP }}
751 **=** **+=** **-=** **\*=** **/=** **%=** **\^=**
754 : The **assignment** operators take two expressions, **a** and **b** where
755 **a** is a named expression (see the *Named Expressions* subsection).
757 For **=**, **b** is copied and the result is assigned to **a**. For all
758 others, **a** and **b** are applied as operands to the corresponding
759 arithmetic operator and the result is assigned to **a**.
761 {{ A H N P HN HP NP HNP }}
762 The **assignment** operators that correspond to operators that are
763 extensions are themselves **non-portable extensions**.
766 **==** **\<=** **\>=** **!=** **\<** **\>**
768 : The **relational** operators compare two expressions, **a** and **b**, and
769 if the relation holds, according to C language semantics, the result is
770 **1**. Otherwise, it is **0**.
772 Note that unlike in C, these operators have a lower precedence than the
773 **assignment** operators, which means that **a=b\>c** is interpreted as
776 Also, unlike the [standard][1] requires, these operators can appear anywhere
777 any other expressions can be used. This allowance is a
778 **non-portable extension**.
782 : The **boolean and** operator takes two expressions and returns **1** if both
783 expressions are non-zero, **0** otherwise.
785 This is *not* a short-circuit operator.
787 This is a **non-portable extension**.
791 : The **boolean or** operator takes two expressions and returns **1** if one
792 of the expressions is non-zero, **0** otherwise.
794 This is *not* a short-circuit operator.
796 This is a **non-portable extension**.
800 The following items are statements:
803 2. **{** **S** **;** ... **;** **S** **}**
804 3. **if** **(** **E** **)** **S**
805 4. **if** **(** **E** **)** **S** **else** **S**
806 5. **while** **(** **E** **)** **S**
807 6. **for** **(** **E** **;** **E** **;** **E** **)** **S**
808 7. An empty statement
814 13. A string of characters, enclosed in double quotes
815 14. **print** **E** **,** ... **,** **E**
816 15. **I()**, **I(E)**, **I(E, E)**, and so on, where **I** is an identifier for
817 a **void** function (see the *Void Functions* subsection of the
818 **FUNCTIONS** section). The **E** argument(s) may also be arrays of the form
819 **I[]**, which will automatically be turned into array references (see the
820 *Array References* subsection of the **FUNCTIONS** section) if the
821 corresponding parameter in the function definition is an array reference.
823 Numbers 4, 9, 11, 12, 14, and 15 are **non-portable extensions**.
825 Also, as a **non-portable extension**, any or all of the expressions in the
826 header of a for loop may be omitted. If the condition (second expression) is
827 omitted, it is assumed to be a constant **1**.
829 The **break** statement causes a loop to stop iterating and resume execution
830 immediately following a loop. This is only allowed in loops.
832 The **continue** statement causes a loop iteration to stop early and returns to
833 the start of the loop, including testing the loop condition. This is only
836 The **if** **else** statement does the same thing as in C.
838 The **quit** statement causes bc(1) to quit, even if it is on a branch that will
839 not be executed (it is a compile-time command).
841 The **halt** statement causes bc(1) to quit, if it is executed. (Unlike **quit**
842 if it is on a branch of an **if** statement that is not executed, bc(1) does not
845 The **limits** statement prints the limits that this bc(1) is subject to. This
846 is like the **quit** statement in that it is a compile-time command.
848 An expression by itself is evaluated and printed, followed by a newline.
850 {{ A H N P HN HP NP HNP }}
851 Both scientific notation and engineering notation are available for printing the
852 results of expressions. Scientific notation is activated by assigning **0** to
853 **obase**, and engineering notation is activated by assigning **1** to
854 **obase**. To deactivate them, just assign a different value to **obase**.
856 Scientific notation and engineering notation are disabled if bc(1) is run with
857 either the **-s** or **-w** command-line options (or equivalents).
859 Printing numbers in scientific notation and/or engineering notation is a
860 **non-portable extension**.
865 The "expressions" in a **print** statement may also be strings. If they are, there
866 are backslash escape sequences that are interpreted specially. What those
867 sequences are, and what they cause to be printed, are shown below:
881 Any other character following a backslash causes the backslash and character to
884 Any non-string expression in a print statement shall be assigned to **last**,
885 like any other expression that is printed.
887 ## Order of Evaluation
889 All expressions in a statment are evaluated left to right, except as necessary
890 to maintain order of operations. This means, for example, assuming that **i** is
891 equal to **0**, in the expression
895 the first (or 0th) element of **a** is set to **1**, and **i** is equal to **2**
896 at the end of the expression.
898 This includes function arguments. Thus, assuming **i** is equal to **0**, this
899 means that in the expression
903 the first argument passed to **x()** is **0**, and the second argument is **1**,
904 while **i** is equal to **2** before the function starts executing.
908 Function definitions are as follows:
918 Any **I** in the parameter list or **auto** list may be replaced with **I[]** to
919 make a parameter or **auto** var an array, and any **I** in the parameter list
920 may be replaced with **\*I[]** to make a parameter an array reference. Callers
921 of functions that take array references should not put an asterisk in the call;
922 they must be called with just **I[]** like normal array parameters and will be
923 automatically converted into references.
925 As a **non-portable extension**, the opening brace of a **define** statement may
926 appear on the next line.
928 As a **non-portable extension**, the return statement may also be in one of the
932 2. **return** **(** **)**
935 The first two, or not specifying a **return** statement, is equivalent to
936 **return (0)**, unless the function is a **void** function (see the *Void
937 Functions* subsection below).
941 Functions can also be **void** functions, defined as follows:
944 define void I(I,...,I){
951 They can only be used as standalone expressions, where such an expression would
952 be printed alone, except in a print statement.
954 Void functions can only use the first two **return** statements listed above.
955 They can also omit the return statement entirely.
957 The word "void" is not treated as a keyword; it is still possible to have
958 variables, arrays, and functions named **void**. The word "void" is only
959 treated specially right after the **define** keyword.
961 This is a **non-portable extension**.
965 For any array in the parameter list, if the array is declared in the form
971 it is a **reference**. Any changes to the array in the function are reflected,
972 when the function returns, to the array that was passed in.
974 Other than this, all function arguments are passed by value.
976 This is a **non-portable extension**.
980 {{ A H N P HN HP NP HNP }}
981 All of the functions below, including the functions in the extended math
982 library (see the *Extended Library* subsection below), are available when the
983 **-l** or **-\-mathlib** command-line flags are given, except that the extended
984 math library is not available when the **-s** option, the **-w** option, or
985 equivalents are given.
987 {{ E EH EN EP EHN EHP ENP EHNP }}
988 All of the functions below are available when the **-l** or **-\-mathlib**
989 command-line flags are given.
994 The [standard][1] defines the following functions for the math library:
998 : Returns the sine of **x**, which is assumed to be in radians.
1000 This is a transcendental function (see the *Transcendental Functions*
1005 : Returns the cosine of **x**, which is assumed to be in radians.
1007 This is a transcendental function (see the *Transcendental Functions*
1012 : Returns the arctangent of **x**, in radians.
1014 This is a transcendental function (see the *Transcendental Functions*
1019 : Returns the natural logarithm of **x**.
1021 This is a transcendental function (see the *Transcendental Functions*
1026 : Returns the mathematical constant **e** raised to the power of **x**.
1028 This is a transcendental function (see the *Transcendental Functions*
1033 : Returns the bessel integer order **n** (truncated) of **x**.
1035 This is a transcendental function (see the *Transcendental Functions*
1038 {{ A H N P HN HP NP HNP }}
1041 The extended library is *not* loaded when the **-s**/**-\-standard** or
1042 **-w**/**-\-warn** options are given since they are not part of the library
1043 defined by the [standard][1].
1045 The extended library is a **non-portable extension**.
1049 : Calculates **x** to the power of **y**, even if **y** is not an integer, and
1050 returns the result to the current **scale**.
1052 It is an error if **y** is negative and **x** is **0**.
1054 This is a transcendental function (see the *Transcendental Functions*
1059 : Returns **x** rounded to **p** decimal places according to the rounding mode
1060 [round half away from **0**][3].
1064 : Returns **x** rounded to **p** decimal places according to the rounding mode
1065 [round away from **0**][6].
1069 : Returns the factorial of the truncated absolute value of **x**.
1073 : Returns the permutation of the truncated absolute value of **n** of the
1074 truncated absolute value of **k**, if **k \<= n**. If not, it returns **0**.
1078 : Returns the combination of the truncated absolute value of **n** of the
1079 truncated absolute value of **k**, if **k \<= n**. If not, it returns **0**.
1083 : Returns the logarithm base **2** of **x**.
1085 This is a transcendental function (see the *Transcendental Functions*
1090 : Returns the logarithm base **10** of **x**.
1092 This is a transcendental function (see the *Transcendental Functions*
1097 : Returns the logarithm base **b** of **x**.
1099 This is a transcendental function (see the *Transcendental Functions*
1104 : Returns the cube root of **x**.
1108 : Calculates the truncated value of **n**, **r**, and returns the **r**th root
1109 of **x** to the current **scale**.
1111 If **r** is **0** or negative, this raises an error and causes bc(1) to
1112 reset (see the **RESET** section). It also raises an error and causes bc(1)
1113 to reset if **r** is even and **x** is negative.
1117 : Returns **pi** to **p** decimal places.
1119 This is a transcendental function (see the *Transcendental Functions*
1124 : Returns the tangent of **x**, which is assumed to be in radians.
1126 This is a transcendental function (see the *Transcendental Functions*
1131 : Returns the arctangent of **y/x**, in radians. If both **y** and **x** are
1132 equal to **0**, it raises an error and causes bc(1) to reset (see the
1133 **RESET** section). Otherwise, if **x** is greater than **0**, it returns
1134 **a(y/x)**. If **x** is less than **0**, and **y** is greater than or equal
1135 to **0**, it returns **a(y/x)+pi**. If **x** is less than **0**, and **y**
1136 is less than **0**, it returns **a(y/x)-pi**. If **x** is equal to **0**,
1137 and **y** is greater than **0**, it returns **pi/2**. If **x** is equal to
1138 **0**, and **y** is less than **0**, it returns **-pi/2**.
1140 This function is the same as the **atan2()** function in many programming
1143 This is a transcendental function (see the *Transcendental Functions*
1148 : Returns the sine of **x**, which is assumed to be in radians.
1150 This is an alias of **s(x)**.
1152 This is a transcendental function (see the *Transcendental Functions*
1157 : Returns the cosine of **x**, which is assumed to be in radians.
1159 This is an alias of **c(x)**.
1161 This is a transcendental function (see the *Transcendental Functions*
1166 : Returns the tangent of **x**, which is assumed to be in radians.
1168 If **x** is equal to **1** or **-1**, this raises an error and causes bc(1)
1169 to reset (see the **RESET** section).
1171 This is an alias of **t(x)**.
1173 This is a transcendental function (see the *Transcendental Functions*
1178 : Returns the arctangent of **x**, in radians.
1180 This is an alias of **a(x)**.
1182 This is a transcendental function (see the *Transcendental Functions*
1187 : Returns the arctangent of **y/x**, in radians. If both **y** and **x** are
1188 equal to **0**, it raises an error and causes bc(1) to reset (see the
1189 **RESET** section). Otherwise, if **x** is greater than **0**, it returns
1190 **a(y/x)**. If **x** is less than **0**, and **y** is greater than or equal
1191 to **0**, it returns **a(y/x)+pi**. If **x** is less than **0**, and **y**
1192 is less than **0**, it returns **a(y/x)-pi**. If **x** is equal to **0**,
1193 and **y** is greater than **0**, it returns **pi/2**. If **x** is equal to
1194 **0**, and **y** is less than **0**, it returns **-pi/2**.
1196 This function is the same as the **atan2()** function in many programming
1199 This is an alias of **a2(y, x)**.
1201 This is a transcendental function (see the *Transcendental Functions*
1206 : Converts **x** from radians to degrees and returns the result.
1208 This is a transcendental function (see the *Transcendental Functions*
1213 : Converts **x** from degrees to radians and returns the result.
1215 This is a transcendental function (see the *Transcendental Functions*
1220 : Generates a pseudo-random number between **0** (inclusive) and **1**
1221 (exclusive) with the number of decimal digits after the decimal point equal
1222 to the truncated absolute value of **p**. If **p** is not **0**, then
1223 calling this function will change the value of **seed**. If **p** is **0**,
1224 then **0** is returned, and **seed** is *not* changed.
1228 : Generates a pseudo-random number that is between **0** (inclusive) and the
1229 truncated absolute value of **i** (exclusive) with the number of decimal
1230 digits after the decimal point equal to the truncated absolute value of
1231 **p**. If the absolute value of **i** is greater than or equal to **2**, and
1232 **p** is not **0**, then calling this function will change the value of
1233 **seed**; otherwise, **0** is returned and **seed** is not changed.
1237 : Returns **x** with its sign flipped with probability **0.5**. In other
1238 words, it randomizes the sign of **x**.
1242 : Returns a random boolean value (either **0** or **1**).
1246 : Returns the numbers of unsigned integer bytes required to hold the truncated
1247 absolute value of **x**.
1251 : Returns the numbers of signed, two's-complement integer bytes required to
1252 hold the truncated value of **x**.
1256 : Outputs the hexadecimal (base **16**) representation of **x**.
1258 This is a **void** function (see the *Void Functions* subsection of the
1259 **FUNCTIONS** section).
1263 : Outputs the binary (base **2**) representation of **x**.
1265 This is a **void** function (see the *Void Functions* subsection of the
1266 **FUNCTIONS** section).
1270 : Outputs the base **b** representation of **x**.
1272 This is a **void** function (see the *Void Functions* subsection of the
1273 **FUNCTIONS** section).
1277 : Outputs the representation, in binary and hexadecimal, of **x** as an
1278 unsigned integer in as few power of two bytes as possible. Both outputs are
1279 split into bytes separated by spaces.
1281 If **x** is not an integer or is negative, an error message is printed
1282 instead, but bc(1) is not reset (see the **RESET** section).
1284 This is a **void** function (see the *Void Functions* subsection of the
1285 **FUNCTIONS** section).
1289 : Outputs the representation, in binary and hexadecimal, of **x** as a signed,
1290 two's-complement integer in as few power of two bytes as possible. Both
1291 outputs are split into bytes separated by spaces.
1293 If **x** is not an integer, an error message is printed instead, but bc(1)
1294 is not reset (see the **RESET** section).
1296 This is a **void** function (see the *Void Functions* subsection of the
1297 **FUNCTIONS** section).
1301 : Outputs the representation, in binary and hexadecimal, of **x** as an
1302 unsigned integer in **n** bytes. Both outputs are split into bytes separated
1305 If **x** is not an integer, is negative, or cannot fit into **n** bytes, an
1306 error message is printed instead, but bc(1) is not reset (see the **RESET**
1309 This is a **void** function (see the *Void Functions* subsection of the
1310 **FUNCTIONS** section).
1314 : Outputs the representation, in binary and hexadecimal, of **x** as a signed,
1315 two's-complement integer in **n** bytes. Both outputs are split into bytes
1316 separated by spaces.
1318 If **x** is not an integer or cannot fit into **n** bytes, an error message
1319 is printed instead, but bc(1) is not reset (see the **RESET** section).
1321 This is a **void** function (see the *Void Functions* subsection of the
1322 **FUNCTIONS** section).
1326 : Outputs the representation, in binary and hexadecimal, of **x** as an
1327 unsigned integer in **1** byte. Both outputs are split into bytes separated
1330 If **x** is not an integer, is negative, or cannot fit into **1** byte, an
1331 error message is printed instead, but bc(1) is not reset (see the **RESET**
1334 This is a **void** function (see the *Void Functions* subsection of the
1335 **FUNCTIONS** section).
1339 : Outputs the representation, in binary and hexadecimal, of **x** as a signed,
1340 two's-complement integer in **1** byte. Both outputs are split into bytes
1341 separated by spaces.
1343 If **x** is not an integer or cannot fit into **1** byte, an error message
1344 is printed instead, but bc(1) is not reset (see the **RESET** section).
1346 This is a **void** function (see the *Void Functions* subsection of the
1347 **FUNCTIONS** section).
1351 : Outputs the representation, in binary and hexadecimal, of **x** as an
1352 unsigned integer in **2** bytes. Both outputs are split into bytes separated
1355 If **x** is not an integer, is negative, or cannot fit into **2** bytes, an
1356 error message is printed instead, but bc(1) is not reset (see the **RESET**
1359 This is a **void** function (see the *Void Functions* subsection of the
1360 **FUNCTIONS** section).
1364 : Outputs the representation, in binary and hexadecimal, of **x** as a signed,
1365 two's-complement integer in **2** bytes. Both outputs are split into bytes
1366 separated by spaces.
1368 If **x** is not an integer or cannot fit into **2** bytes, an error message
1369 is printed instead, but bc(1) is not reset (see the **RESET** section).
1371 This is a **void** function (see the *Void Functions* subsection of the
1372 **FUNCTIONS** section).
1376 : Outputs the representation, in binary and hexadecimal, of **x** as an
1377 unsigned integer in **4** bytes. Both outputs are split into bytes separated
1380 If **x** is not an integer, is negative, or cannot fit into **4** bytes, an
1381 error message is printed instead, but bc(1) is not reset (see the **RESET**
1384 This is a **void** function (see the *Void Functions* subsection of the
1385 **FUNCTIONS** section).
1389 : Outputs the representation, in binary and hexadecimal, of **x** as a signed,
1390 two's-complement integer in **4** bytes. Both outputs are split into bytes
1391 separated by spaces.
1393 If **x** is not an integer or cannot fit into **4** bytes, an error message
1394 is printed instead, but bc(1) is not reset (see the **RESET** section).
1396 This is a **void** function (see the *Void Functions* subsection of the
1397 **FUNCTIONS** section).
1401 : Outputs the representation, in binary and hexadecimal, of **x** as an
1402 unsigned integer in **8** bytes. Both outputs are split into bytes separated
1405 If **x** is not an integer, is negative, or cannot fit into **8** bytes, an
1406 error message is printed instead, but bc(1) is not reset (see the **RESET**
1409 This is a **void** function (see the *Void Functions* subsection of the
1410 **FUNCTIONS** section).
1414 : Outputs the representation, in binary and hexadecimal, of **x** as a signed,
1415 two's-complement integer in **8** bytes. Both outputs are split into bytes
1416 separated by spaces.
1418 If **x** is not an integer or cannot fit into **8** bytes, an error message
1419 is printed instead, but bc(1) is not reset (see the **RESET** section).
1421 This is a **void** function (see the *Void Functions* subsection of the
1422 **FUNCTIONS** section).
1426 : Outputs the representation of the truncated absolute value of **x** as an
1427 unsigned integer in hexadecimal using **n** bytes. Not all of the value will
1428 be output if **n** is too small.
1430 This is a **void** function (see the *Void Functions* subsection of the
1431 **FUNCTIONS** section).
1433 **binary_uint(x, n)**
1435 : Outputs the representation of the truncated absolute value of **x** as an
1436 unsigned integer in binary using **n** bytes. Not all of the value will be
1437 output if **n** is too small.
1439 This is a **void** function (see the *Void Functions* subsection of the
1440 **FUNCTIONS** section).
1442 **output_uint(x, n)**
1444 : Outputs the representation of the truncated absolute value of **x** as an
1445 unsigned integer in the current **obase** (see the **SYNTAX** section) using
1446 **n** bytes. Not all of the value will be output if **n** is too small.
1448 This is a **void** function (see the *Void Functions* subsection of the
1449 **FUNCTIONS** section).
1451 **output_byte(x, i)**
1453 : Outputs byte **i** of the truncated absolute value of **x**, where **0** is
1454 the least significant byte and **number_of_bytes - 1** is the most
1457 This is a **void** function (see the *Void Functions* subsection of the
1458 **FUNCTIONS** section).
1461 ## Transcendental Functions
1463 All transcendental functions can return slightly inaccurate results (up to 1
1464 [ULP][4]). This is unavoidable, and [this article][5] explains why it is
1465 impossible and unnecessary to calculate exact results for the transcendental
1468 Because of the possible inaccuracy, I recommend that users call those functions
1469 with the precision (**scale**) set to at least 1 higher than is necessary. If
1470 exact results are *absolutely* required, users can double the precision
1471 (**scale**) and then truncate.
1473 The transcendental functions in the standard math library are:
1482 {{ A H N P HN HP NP HNP }}
1483 The transcendental functions in the extended math library are:
1502 When bc(1) encounters an error or a signal that it has a non-default handler
1503 for, it resets. This means that several things happen.
1505 First, any functions that are executing are stopped and popped off the stack.
1506 The behavior is not unlike that of exceptions in programming languages. Then
1507 the execution point is set so that any code waiting to execute (after all
1508 functions returned) is skipped.
1510 Thus, when bc(1) resets, it skips any remaining code waiting to be executed.
1511 Then, if it is interactive mode, and the error was not a fatal error (see the
1512 **EXIT STATUS** section), it asks for more input; otherwise, it exits with the
1513 appropriate return code.
1515 Note that this reset behavior is different from the GNU bc(1), which attempts to
1516 start executing the statement right after the one that caused an error.
1520 Most bc(1) implementations use **char** types to calculate the value of **1**
1521 decimal digit at a time, but that can be slow. This bc(1) does something
1524 It uses large integers to calculate more than **1** decimal digit at a time. If
1525 built in a environment where **BC_LONG_BIT** (see the **LIMITS** section) is
1526 **64**, then each integer has **9** decimal digits. If built in an environment
1527 where **BC_LONG_BIT** is **32** then each integer has **4** decimal digits. This
1528 value (the number of decimal digits per large integer) is called
1531 The actual values of **BC_LONG_BIT** and **BC_BASE_DIGS** can be queried with
1532 the **limits** statement.
1534 In addition, this bc(1) uses an even larger integer for overflow checking. This
1535 integer type depends on the value of **BC_LONG_BIT**, but is always at least
1536 twice as large as the integer type used to store digits.
1540 The following are the limits on bc(1):
1544 : The number of bits in the **long** type in the environment where bc(1) was
1545 built. This determines how many decimal digits can be stored in a single
1546 large integer (see the **PERFORMANCE** section).
1550 : The number of decimal digits per large integer (see the **PERFORMANCE**
1551 section). Depends on **BC_LONG_BIT**.
1555 : The max decimal number that each large integer can store (see
1556 **BC_BASE_DIGS**) plus **1**. Depends on **BC_BASE_DIGS**.
1560 : The max number that the overflow type (see the **PERFORMANCE** section) can
1561 hold. Depends on **BC_LONG_BIT**.
1565 : The maximum output base. Set at **BC_BASE_POW**.
1569 : The maximum size of arrays. Set at **SIZE_MAX-1**.
1573 : The maximum **scale**. Set at **BC_OVERFLOW_MAX-1**.
1577 : The maximum length of strings. Set at **BC_OVERFLOW_MAX-1**.
1581 : The maximum length of identifiers. Set at **BC_OVERFLOW_MAX-1**.
1585 : The maximum length of a number (in decimal digits), which includes digits
1586 after the decimal point. Set at **BC_OVERFLOW_MAX-1**.
1588 {{ A H N P HN HP NP HNP }}
1591 : The maximum integer (inclusive) returned by the **rand()** operand. Set at
1592 **2\^BC_LONG_BIT-1**.
1597 : The maximum allowable exponent (positive or negative). Set at
1598 **BC_OVERFLOW_MAX**.
1602 : The maximum number of vars/arrays. Set at **SIZE_MAX-1**.
1604 The actual values can be queried with the **limits** statement.
1606 These limits are meant to be effectively non-existent; the limits are so large
1607 (at least on 64-bit machines) that there should not be any point at which they
1608 become a problem. In fact, memory should be exhausted before these limits should
1611 # ENVIRONMENT VARIABLES
1613 bc(1) recognizes the following environment variables:
1617 : If this variable exists (no matter the contents), bc(1) behaves as if
1618 the **-s** option was given.
1622 : This is another way to give command-line arguments to bc(1). They should be
1623 in the same format as all other command-line arguments. These are always
1624 processed first, so any files given in **BC_ENV_ARGS** will be processed
1625 before arguments and files given on the command-line. This gives the user
1626 the ability to set up "standard" options and files to be used at every
1627 invocation. The most useful thing for such files to contain would be useful
1628 functions that the user might want every time bc(1) runs.
1630 The code that parses **BC_ENV_ARGS** will correctly handle quoted arguments,
1631 but it does not understand escape sequences. For example, the string
1632 **"/home/gavin/some bc file.bc"** will be correctly parsed, but the string
1633 **"/home/gavin/some \"bc\" file.bc"** will include the backslashes.
1635 The quote parsing will handle either kind of quotes, **'** or **"**. Thus,
1636 if you have a file with any number of single quotes in the name, you can use
1637 double quotes as the outside quotes, as in **"some 'bc' file.bc"**, and vice
1638 versa if you have a file with double quotes. However, handling a file with
1639 both kinds of quotes in **BC_ENV_ARGS** is not supported due to the
1640 complexity of the parsing, though such files are still supported on the
1641 command-line where the parsing is done by the shell.
1645 : If this environment variable exists and contains an integer that is greater
1646 than **1** and is less than **UINT16_MAX** (**2\^16-1**), bc(1) will output
1647 lines to that length, including the backslash (**\\**). The default line
1652 bc(1) returns the following exit statuses:
1660 : A math error occurred. This follows standard practice of using **1** for
1661 expected errors, since math errors will happen in the process of normal
1664 {{ A H N P HN HP NP HNP }}
1665 Math errors include divide by **0**, taking the square root of a negative
1666 number, using a negative number as a bound for the pseudo-random number
1667 generator, attempting to convert a negative number to a hardware integer,
1668 overflow when converting a number to a hardware integer, and attempting to
1669 use a non-integer where an integer is required.
1671 Converting to a hardware integer happens for the second operand of the power
1672 (**\^**), places (**\@**), left shift (**\<\<**), and right shift (**\>\>**)
1673 operators and their corresponding assignment operators.
1675 {{ E EH EN EP EHN EHP ENP EHNP }}
1676 Math errors include divide by **0**, taking the square root of a negative
1677 number, attempting to convert a negative number to a hardware integer,
1678 overflow when converting a number to a hardware integer, and attempting to
1679 use a non-integer where an integer is required.
1681 Converting to a hardware integer happens for the second operand of the power
1682 (**\^**) operator and the corresponding assignment operator.
1687 : A parse error occurred.
1689 Parse errors include unexpected **EOF**, using an invalid character, failing
1690 to find the end of a string or comment, using a token where it is invalid,
1691 giving an invalid expression, giving an invalid print statement, giving an
1692 invalid function definition, attempting to assign to an expression that is
1693 not a named expression (see the *Named Expressions* subsection of the
1694 **SYNTAX** section), giving an invalid **auto** list, having a duplicate
1695 **auto**/function parameter, failing to find the end of a code block,
1696 attempting to return a value from a **void** function, attempting to use a
1697 variable as a reference, and using any extensions when the option **-s** or
1698 any equivalents were given.
1702 : A runtime error occurred.
1704 Runtime errors include assigning an invalid number to **ibase**, **obase**,
1705 or **scale**; give a bad expression to a **read()** call, calling **read()**
1706 inside of a **read()** call, type errors, passing the wrong number of
1707 arguments to functions, attempting to call an undefined function, and
1708 attempting to use a **void** function call as a value in an expression.
1712 : A fatal error occurred.
1714 Fatal errors include memory allocation errors, I/O errors, failing to open
1715 files, attempting to use files that do not have only ASCII characters (bc(1)
1716 only accepts ASCII characters), attempting to open a directory as a file,
1717 and giving invalid command-line options.
1719 The exit status **4** is special; when a fatal error occurs, bc(1) always exits
1720 and returns **4**, no matter what mode bc(1) is in.
1722 The other statuses will only be returned when bc(1) is not in interactive mode
1723 (see the **INTERACTIVE MODE** section), since bc(1) resets its state (see the
1724 **RESET** section) and accepts more input when one of those errors occurs in
1725 interactive mode. This is also the case when interactive mode is forced by the
1726 **-i** flag or **-\-interactive** option.
1728 These exit statuses allow bc(1) to be used in shell scripting with error
1729 checking, and its normal behavior can be forced by using the **-i** flag or
1730 **-\-interactive** option.
1734 Per the [standard][1], bc(1) has an interactive mode and a non-interactive mode.
1735 Interactive mode is turned on automatically when both **stdin** and **stdout**
1736 are hooked to a terminal, but the **-i** flag and **-\-interactive** option can
1737 turn it on in other cases.
1739 In interactive mode, bc(1) attempts to recover from errors (see the **RESET**
1740 section), and in normal execution, flushes **stdout** as soon as execution is
1741 done for the current input.
1745 If **stdin**, **stdout**, and **stderr** are all connected to a TTY, bc(1) turns
1748 {{ A E N P EN EP NP ENP }}
1749 TTY mode is required for history to be enabled (see the **COMMAND LINE HISTORY**
1750 section). It is also required to enable special handling for **SIGINT** signals.
1753 {{ A E H N EH EN HN EHN }}
1754 The prompt is enabled in TTY mode.
1757 TTY mode is different from interactive mode because interactive mode is required
1758 in the [bc(1) specification][1], and interactive mode requires only **stdin**
1759 and **stdout** to be connected to a terminal.
1763 Sending a **SIGINT** will cause bc(1) to stop execution of the current input. If
1764 bc(1) is in TTY mode (see the **TTY MODE** section), it will reset (see the
1765 **RESET** section). Otherwise, it will clean up and exit.
1767 Note that "current input" can mean one of two things. If bc(1) is processing
1768 input from **stdin** in TTY mode, it will ask for more input. If bc(1) is
1769 processing input from a file in TTY mode, it will stop processing the file and
1770 start processing the next file, if one exists, or ask for input from **stdin**
1771 if no other file exists.
1773 This means that if a **SIGINT** is sent to bc(1) as it is executing a file, it
1774 can seem as though bc(1) did not respond to the signal since it will immediately
1775 start executing the next file. This is by design; most files that users execute
1776 when interacting with bc(1) have function definitions, which are quick to parse.
1777 If a file takes a long time to execute, there may be a bug in that file. The
1778 rest of the files could still be executed without problem, allowing the user to
1781 **SIGTERM** and **SIGQUIT** cause bc(1) to clean up and exit, and it uses the
1782 {{ A E N P EN EP NP ENP }}
1783 default handler for all other signals. The one exception is **SIGHUP**; in that
1784 case, when bc(1) is in TTY mode, a **SIGHUP** will cause bc(1) to clean up and
1787 {{ H EH HN HP EHN EHP HNP EHNP }}
1788 default handler for all other signals.
1791 {{ A E N P EN EP NP ENP }}
1792 # COMMAND LINE HISTORY
1794 bc(1) supports interactive command-line editing. If bc(1) is in TTY mode (see
1795 the **TTY MODE** section), history is enabled. Previous lines can be recalled
1796 and edited with the arrow keys.
1798 **Note**: tabs are converted to 8 spaces.
1801 {{ A E H P EH EP HP EHP }}
1804 This bc(1) ships with support for adding error messages for different locales
1805 and thus, supports **LC_MESSAGES**.
1814 bc(1) is compliant with the [IEEE Std 1003.1-2017 (“POSIX.1-2017”)][1]
1815 specification. The flags **-efghiqsvVw**, all long options, and the extensions
1816 noted above are extensions to that specification.
1818 Note that the specification explicitly says that bc(1) only accepts numbers that
1819 use a period (**.**) as a radix point, regardless of the value of
1822 {{ A E H P EH EP HP EHP }}
1823 This bc(1) supports error messages for different locales, and thus, it supports
1829 None are known. Report bugs at https://git.yzena.com/gavin/bc.
1833 Gavin D. Howard <gavin@yzena.com> and contributors.
1835 [1]: https://pubs.opengroup.org/onlinepubs/9699919799/utilities/bc.html
1836 [2]: https://www.gnu.org/software/bc/
1837 [3]: https://en.wikipedia.org/wiki/Rounding#Round_half_away_from_zero
1838 [4]: https://en.wikipedia.org/wiki/Unit_in_the_last_place
1839 [5]: https://people.eecs.berkeley.edu/~wkahan/LOG10HAF.TXT
1840 [6]: https://en.wikipedia.org/wiki/Rounding#Rounding_away_from_zero