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**.
51 This bc(1) is a drop-in replacement for *any* bc(1), including (and
52 especially) the GNU bc(1). It also has many extensions and extra features beyond
53 other implementations.
56 This bc(1) is a drop-in replacement for *any* bc(1), including (and
57 especially) the GNU bc(1).
62 The following are the options that bc(1) accepts.
64 **-g**, **--global-stacks**
66 {{ A H N P HN HP NP HNP }}
67 : Turns the globals **ibase**, **obase**, **scale**, and **seed** into stacks.
69 This has the effect that a copy of the current value of all four are pushed
71 {{ E EH EN EP EHN EHP ENP EHNP }}
72 Turns the globals **ibase**, **obase**, and **scale** into stacks.
74 This has the effect that a copy of the current value of all three are pushed
76 onto a stack for every function call, as well as popped when every function
77 returns. This means that functions can assign to any and all of those
78 globals without worrying that the change will affect other functions.
79 Thus, a hypothetical function named **output(x,b)** that simply printed
80 **x** in base **b** could be written like this:
82 define void output(x, b) {
89 define void output(x, b) {
97 This makes writing functions much easier.
99 {{ A H N P HN HP NP HNP }}
100 (**Note**: the function **output(x,b)** exists in the extended math library.
101 See the **LIBRARY** section.)
103 However, since using this flag means that functions cannot set **ibase**,
104 **obase**, **scale**, or **seed** globally, functions that are made to do so
105 cannot work anymore. There are two possible use cases for that, and each has
108 {{ E EH EN EP EHN EHP ENP EHNP }}
109 However, since using this flag means that functions cannot set **ibase**,
110 **obase**, or **scale** globally, functions that are made to do so cannot
111 work anymore. There are two possible use cases for that, and each has a
115 First, if a function is called on startup to turn bc(1) into a number
116 converter, it is possible to replace that capability with various shell
119 alias d2o="bc -e ibase=A -e obase=8"
120 alias h2b="bc -e ibase=G -e obase=2"
122 {{ A H N P HN HP NP HNP }}
123 Second, if the purpose of a function is to set **ibase**, **obase**,
124 **scale**, or **seed** globally for any other purpose, it could be split
125 into one to four functions (based on how many globals it sets) and each of
126 those functions could return the desired value for a global.
128 For functions that set **seed**, the value assigned to **seed** is not
129 propagated to parent functions. This means that the sequence of
130 pseudo-random numbers that they see will not be the same sequence of
131 pseudo-random numbers that any parent sees. This is only the case once
132 **seed** has been set.
134 If a function desires to not affect the sequence of pseudo-random numbers
135 of its parents, but wants to use the same **seed**, it can use the following
140 {{ E EH EN EP EHN EHP ENP EHNP }}
141 Second, if the purpose of a function is to set **ibase**, **obase**, or
142 **scale** globally for any other purpose, it could be split into one to
143 three functions (based on how many globals it sets) and each of those
144 functions could return the desired value for a global.
147 If the behavior of this option is desired for every run of bc(1), then users
148 could make sure to define **BC_ENV_ARGS** and include this option (see the
149 **ENVIRONMENT VARIABLES** section for more details).
151 If **-s**, **-w**, or any equivalents are used, this option is ignored.
153 This is a **non-portable extension**.
157 : Prints a usage message and quits.
159 **-i**, **--interactive**
161 : Forces interactive mode. (See the **INTERACTIVE MODE** section.)
163 This is a **non-portable extension**.
165 **-l**, **--mathlib**
167 : Sets **scale** (see the **SYNTAX** section) to **20** and loads the included
168 {{ A H N P HN HP NP HNP }}
169 math library and the extended math library before running any code,
170 including any expressions or files specified on the command line.
172 To learn what is in the libraries, see the **LIBRARY** section.
174 {{ E EH EN EP EHN EHP ENP EHNP }}
175 math library before running any code, including any expressions or files
176 specified on the command line.
178 To learn what is in the library, see the **LIBRARY** section.
181 **-P**, **--no-prompt**
183 {{ A E H N EH EN HN EHN }}
184 : Disables the prompt in TTY mode. (The prompt is only enabled in TTY mode.
185 See the **TTY MODE** section) This is mostly for those users that do not
186 want a prompt or are not used to having them in bc(1). Most of those users
187 would want to put this option in **BC_ENV_ARGS** (see the
188 **ENVIRONMENT VARIABLES** section).
190 {{ P EP HP NP EHP ENP HNP EHNP }}
191 : This option is a no-op.
194 This is a **non-portable extension**.
198 : This option is for compatibility with the [GNU bc(1)][2]; it is a no-op.
199 Without this option, GNU bc(1) prints a copyright header. This bc(1) only
200 prints the copyright header if one or more of the **-v**, **-V**, or
201 **--version** options are given.
203 This is a **non-portable extension**.
205 **-s**, **--standard**
207 : Process exactly the language defined by the [standard][1] and error if any
210 This is a **non-portable extension**.
212 **-v**, **-V**, **--version**
214 : Print the version information (copyright header) and exit.
216 This is a **non-portable extension**.
220 : Like **-s** and **--standard**, except that warnings (and not errors) are
221 printed for non-standard extensions and execution continues normally.
223 This is a **non-portable extension**.
225 **-e** *expr*, **--expression**=*expr*
227 : Evaluates *expr*. If multiple expressions are given, they are evaluated in
228 order. If files are given as well (see below), the expressions and files are
229 evaluated in the order given. This means that if a file is given before an
230 expression, the file is read in and evaluated first.
232 After processing all expressions and files, bc(1) will exit, unless **-**
233 (**stdin**) was given as an argument at least once to **-f** or **--file**.
234 However, if any other **-e**, **--expression**, **-f**, or **--file**
235 arguments are given after that, bc(1) will give a fatal error and exit.
237 This is a **non-portable extension**.
239 **-f** *file*, **--file**=*file*
241 : Reads in *file* and evaluates it, line by line, as though it were read
242 through **stdin**. If expressions are also given (see above), the
243 expressions are evaluated in the order given.
245 After processing all expressions and files, bc(1) will exit, unless **-**
246 (**stdin**) was given as an argument at least once to **-f** or **--file**.
248 This is a **non-portable extension**.
250 All long options are **non-portable extensions**.
254 Any non-error output is written to **stdout**.
256 **Note**: Unlike other bc(1) implementations, this bc(1) will issue a fatal
257 error (see the **EXIT STATUS** section) if it cannot write to **stdout**, so if
258 **stdout** is closed, as in **bc <file> >&-**, it will quit with an error. This
259 is done so that bc(1) can report problems when **stdout** is redirected to a
262 If there are scripts that depend on the behavior of other bc(1) implementations,
263 it is recommended that those scripts be changed to redirect **stdout** to
268 Any error output is written to **stderr**.
270 **Note**: Unlike other bc(1) implementations, this bc(1) will issue a fatal
271 error (see the **EXIT STATUS** section) if it cannot write to **stderr**, so if
272 **stderr** is closed, as in **bc <file> 2>&-**, it will quit with an error. This
273 is done so that bc(1) can exit with an error code when **stderr** is redirected
276 If there are scripts that depend on the behavior of other bc(1) implementations,
277 it is recommended that those scripts be changed to redirect **stderr** to
282 The syntax for bc(1) programs is mostly C-like, with some differences. This
283 bc(1) follows the [POSIX standard][1], which is a much more thorough resource
284 for the language this bc(1) accepts. This section is meant to be a summary and a
285 listing of all the extensions to the standard.
287 In the sections below, **E** means expression, **S** means statement, and **I**
290 Identifiers (**I**) start with a lowercase letter and can be followed by any
291 number (up to **BC_NAME_MAX-1**) of lowercase letters (**a-z**), digits
292 (**0-9**), and underscores (**\_**). The regex is **\[a-z\]\[a-z0-9\_\]\***.
293 Identifiers with more than one character (letter) are a
294 **non-portable extension**.
296 **ibase** is a global variable determining how to interpret constant numbers. It
297 is the "input" base, or the number base used for interpreting input numbers.
298 **ibase** is initially **10**. If the **-s** (**--standard**) and **-w**
299 (**--warn**) flags were not given on the command line, the max allowable value
300 for **ibase** is **36**. Otherwise, it is **16**. The min allowable value for
301 **ibase** is **2**. The max allowable value for **ibase** can be queried in
302 bc(1) programs with the **maxibase()** built-in function.
304 **obase** is a global variable determining how to output results. It is the
305 "output" base, or the number base used for outputting numbers. **obase** is
306 initially **10**. The max allowable value for **obase** is **BC_BASE_MAX** and
307 can be queried in bc(1) programs with the **maxobase()** built-in function. The
308 {{ A H N P HN HP NP HNP }}
309 min allowable value for **obase** is **0**. If **obase** is **0**, values are
310 output in scientific notation, and if **obase** is **1**, values are output in
311 engineering notation. Otherwise, values are output in the specified base.
313 Outputting in scientific and engineering notations are **non-portable
316 {{ E EH EN EP EHN EHP ENP EHNP }}
317 min allowable value for **obase** is **2**. Values are output in the specified
321 The *scale* of an expression is the number of digits in the result of the
322 expression right of the decimal point, and **scale** is a global variable that
323 sets the precision of any operations, with exceptions. **scale** is initially
324 **0**. **scale** cannot be negative. The max allowable value for **scale** is
325 **BC_SCALE_MAX** and can be queried in bc(1) programs with the **maxscale()**
328 bc(1) has both *global* variables and *local* variables. All *local*
329 variables are local to the function; they are parameters or are introduced in
330 the **auto** list of a function (see the **FUNCTIONS** section). If a variable
331 is accessed which is not a parameter or in the **auto** list, it is assumed to
332 be *global*. If a parent function has a *local* variable version of a variable
333 that a child function considers *global*, the value of that *global* variable in
334 the child function is the value of the variable in the parent function, not the
335 value of the actual *global* variable.
337 All of the above applies to arrays as well.
339 The value of a statement that is an expression (i.e., any of the named
340 expressions or operands) is printed unless the lowest precedence operator is an
341 assignment operator *and* the expression is notsurrounded by parentheses.
343 The value that is printed is also assigned to the special variable **last**. A
344 single dot (**.**) may also be used as a synonym for **last**. These are
345 **non-portable extensions**.
347 Either semicolons or newlines may separate statements.
351 There are two kinds of comments:
353 1. Block comments are enclosed in **/\*** and **\*/**.
354 2. Line comments go from **#** until, and not including, the next newline. This
355 is a **non-portable extension**.
359 The following are named expressions in bc(1):
362 2. Array Elements: **I[E]**
366 {{ A H N P HN HP NP HNP }}
368 7. **last** or a single dot (**.**)
370 Numbers 6 and 7 are **non-portable extensions**.
372 The meaning of **seed** is dependent on the current pseudo-random number
373 generator but is guaranteed to not change except for new major versions.
375 The *scale* and sign of the value may be significant.
377 If a previously used **seed** value is assigned to **seed** and used again, the
378 pseudo-random number generator is guaranteed to produce the same sequence of
379 pseudo-random numbers as it did when the **seed** value was previously used.
381 The exact value assigned to **seed** is not guaranteed to be returned if
382 **seed** is queried again immediately. However, if **seed** *does* return a
383 different value, both values, when assigned to **seed**, are guaranteed to
384 produce the same sequence of pseudo-random numbers. This means that certain
385 values assigned to **seed** will *not* produce unique sequences of pseudo-random
386 numbers. The value of **seed** will change after any use of the **rand()** and
387 **irand(E)** operands (see the *Operands* subsection below), except if the
388 parameter passed to **irand(E)** is **0**, **1**, or negative.
390 There is no limit to the length (number of significant decimal digits) or
391 *scale* of the value that can be assigned to **seed**.
393 {{ E EH EN EP EHN EHP ENP EHNP }}
394 6. **last** or a single dot (**.**)
396 Number 6 is a **non-portable extension**.
399 Variables and arrays do not interfere; users can have arrays named the same as
400 variables. This also applies to functions (see the **FUNCTIONS** section), so a
401 user can have a variable, array, and function that all have the same name, and
402 they will not shadow each other, whether inside of functions or not.
404 Named expressions are required as the operand of **increment**/**decrement**
405 operators and as the left side of **assignment** operators (see the *Operators*
410 The following are valid operands in bc(1):
412 1. Numbers (see the *Numbers* subsection below).
413 2. Array indices (**I[E]**).
414 3. **(E)**: The value of **E** (used to change precedence).
415 4. **sqrt(E)**: The square root of **E**. **E** must be non-negative.
416 5. **length(E)**: The number of significant decimal digits in **E**.
417 6. **length(I[])**: The number of elements in the array **I**. This is a
418 **non-portable extension**.
419 7. **scale(E)**: The *scale* of **E**.
420 8. **abs(E)**: The absolute value of **E**. This is a **non-portable
422 9. **I()**, **I(E)**, **I(E, E)**, and so on, where **I** is an identifier for
423 a non-**void** function (see the *Void Functions* subsection of the
424 **FUNCTIONS** section). The **E** argument(s) may also be arrays of the form
425 **I[]**, which will automatically be turned into array references (see the
426 *Array References* subsection of the **FUNCTIONS** section) if the
427 corresponding parameter in the function definition is an array reference.
428 10. **read()**: Reads a line from **stdin** and uses that as an expression. The
429 result of that expression is the result of the **read()** operand. This is a
430 **non-portable extension**.
431 11. **maxibase()**: The max allowable **ibase**. This is a **non-portable
433 12. **maxobase()**: The max allowable **obase**. This is a **non-portable
435 13. **maxscale()**: The max allowable **scale**. This is a **non-portable
437 {{ A H N P HN HP NP HNP }}
438 14. **rand()**: A pseudo-random integer between **0** (inclusive) and
439 **BC_RAND_MAX** (inclusive). Using this operand will change the value of
440 **seed**. This is a **non-portable extension**.
441 15. **irand(E)**: A pseudo-random integer between **0** (inclusive) and the
442 value of **E** (exclusive). If **E** is negative or is a non-integer
443 (**E**'s *scale* is not **0**), an error is raised, and bc(1) resets (see
444 the **RESET** section) while **seed** remains unchanged. If **E** is larger
445 than **BC_RAND_MAX**, the higher bound is honored by generating several
446 pseudo-random integers, multiplying them by appropriate powers of
447 **BC_RAND_MAX+1**, and adding them together. Thus, the size of integer that
448 can be generated with this operand is unbounded. Using this operand will
449 change the value of **seed**, unless the value of **E** is **0** or **1**.
450 In that case, **0** is returned, and **seed** is *not* changed. This is a
451 **non-portable extension**.
452 16. **maxrand()**: The max integer returned by **rand()**. This is a
453 **non-portable extension**.
455 The integers generated by **rand()** and **irand(E)** are guaranteed to be as
456 unbiased as possible, subject to the limitations of the pseudo-random number
459 **Note**: The values returned by the pseudo-random number generator with
460 **rand()** and **irand(E)** are guaranteed to *NOT* be cryptographically secure.
461 This is a consequence of using a seeded pseudo-random number generator. However,
462 they *are* guaranteed to be reproducible with identical **seed** values.
467 Numbers are strings made up of digits, uppercase letters, and at most **1**
468 period for a radix. Numbers can have up to **BC_NUM_MAX** digits. Uppercase
469 letters are equal to **9** + their position in the alphabet (i.e., **A** equals
470 **10**, or **9+1**). If a digit or letter makes no sense with the current value
471 of **ibase**, they are set to the value of the highest valid digit in **ibase**.
473 Single-character numbers (i.e., **A** alone) take the value that they would have
474 if they were valid digits, regardless of the value of **ibase**. This means that
475 **A** alone always equals decimal **10** and **Z** alone always equals decimal
478 {{ A H N P HN HP NP HNP }}
479 In addition, bc(1) accepts numbers in scientific notation. These have the form
480 **\<number\>e\<integer\>**. The exponent (the portion after the **e**) must be
481 an integer. An example is **1.89237e9**, which is equal to **1892370000**.
482 Negative exponents are also allowed, so **4.2890e-3** is equal to **0.0042890**.
484 Using scientific notation is an error or warning if the **-s** or **-w**,
485 respectively, command-line options (or equivalents) are given.
487 **WARNING**: Both the number and the exponent in scientific notation are
488 interpreted according to the current **ibase**, but the number is still
489 multiplied by **10\^exponent** regardless of the current **ibase**. For example,
490 if **ibase** is **16** and bc(1) is given the number string **FFeA**, the
491 resulting decimal number will be **2550000000000**, and if bc(1) is given the
492 number string **10e-4**, the resulting decimal number will be **0.0016**.
494 Accepting input as scientific notation is a **non-portable extension**.
499 The following arithmetic and logical operators can be used. They are listed in
500 order of decreasing precedence. Operators in the same group have the same
505 : Type: Prefix and Postfix
509 Description: **increment**, **decrement**
517 Description: **negation**, **boolean not**
519 {{ A H N P HN HP NP HNP }}
526 Description: **truncation**
534 Description: **set precision**
543 Description: **power**
551 Description: **multiply**, **divide**, **modulus**
559 Description: **add**, **subtract**
561 {{ A H N P HN HP NP HNP }}
568 Description: **shift left**, **shift right**
570 **=** **\<\<=** **\>\>=** **+=** **-=** **\*=** **/=** **%=** **\^=** **\@=**
572 {{ E EH EN EP EHN EHP ENP EHNP }}
573 **=** **+=** **-=** **\*=** **/=** **%=** **\^=**
580 Description: **assignment**
582 **==** **\<=** **\>=** **!=** **\<** **\>**
588 Description: **relational**
596 Description: **boolean and**
604 Description: **boolean or**
606 The operators will be described in more detail below.
610 : The prefix and postfix **increment** and **decrement** operators behave
611 exactly like they would in C. They require a named expression (see the
612 *Named Expressions* subsection) as an operand.
614 The prefix versions of these operators are more efficient; use them where
619 : The **negation** operator returns **0** if a user attempts to negate any
620 expression with the value **0**. Otherwise, a copy of the expression with
621 its sign flipped is returned.
625 : The **boolean not** operator returns **1** if the expression is **0**, or
628 This is a **non-portable extension**.
630 {{ A H N P HN HP NP HNP }}
633 : The **truncation** operator returns a copy of the given expression with all
634 of its *scale* removed.
636 This is a **non-portable extension**.
640 : The **set precision** operator takes two expressions and returns a copy of
641 the first with its *scale* equal to the value of the second expression. That
642 could either mean that the number is returned without change (if the
643 *scale* of the first expression matches the value of the second
644 expression), extended (if it is less), or truncated (if it is more).
646 The second expression must be an integer (no *scale*) and non-negative.
648 This is a **non-portable extension**.
653 : The **power** operator (not the **exclusive or** operator, as it would be in
654 C) takes two expressions and raises the first to the power of the value of
655 the second. The *scale* of the result is equal to **scale**.
657 The second expression must be an integer (no *scale*), and if it is
658 negative, the first value must be non-zero.
662 : The **multiply** operator takes two expressions, multiplies them, and
663 returns the product. If **a** is the *scale* of the first expression and
664 **b** is the *scale* of the second expression, the *scale* of the result is
665 equal to **min(a+b,max(scale,a,b))** where **min()** and **max()** return
670 : The **divide** operator takes two expressions, divides them, and returns the
671 quotient. The *scale* of the result shall be the value of **scale**.
673 The second expression must be non-zero.
677 : The **modulus** operator takes two expressions, **a** and **b**, and
678 evaluates them by 1) Computing **a/b** to current **scale** and 2) Using the
679 result of step 1 to calculate **a-(a/b)\*b** to *scale*
680 **max(scale+scale(b),scale(a))**.
682 The second expression must be non-zero.
686 : The **add** operator takes two expressions, **a** and **b**, and returns the
687 sum, with a *scale* equal to the max of the *scale*s of **a** and **b**.
691 : The **subtract** operator takes two expressions, **a** and **b**, and
692 returns the difference, with a *scale* equal to the max of the *scale*s of
695 {{ A H N P HN HP NP HNP }}
698 : The **left shift** operator takes two expressions, **a** and **b**, and
699 returns a copy of the value of **a** with its decimal point moved **b**
702 The second expression must be an integer (no *scale*) and non-negative.
704 This is a **non-portable extension**.
708 : The **right shift** operator takes two expressions, **a** and **b**, and
709 returns a copy of the value of **a** with its decimal point moved **b**
712 The second expression must be an integer (no *scale*) and non-negative.
714 This is a **non-portable extension**.
717 {{ A H N P HN HP NP HNP }}
718 **=** **\<\<=** **\>\>=** **+=** **-=** **\*=** **/=** **%=** **\^=** **\@=**
720 {{ E EH EN EP EHN EHP ENP EHNP }}
721 **=** **+=** **-=** **\*=** **/=** **%=** **\^=**
724 : The **assignment** operators take two expressions, **a** and **b** where
725 **a** is a named expression (see the *Named Expressions* subsection).
727 For **=**, **b** is copied and the result is assigned to **a**. For all
728 others, **a** and **b** are applied as operands to the corresponding
729 arithmetic operator and the result is assigned to **a**.
731 {{ A H N P HN HP NP HNP }}
732 The **assignment** operators that correspond to operators that are
733 extensions are themselves **non-portable extensions**.
736 **==** **\<=** **\>=** **!=** **\<** **\>**
738 : The **relational** operators compare two expressions, **a** and **b**, and
739 if the relation holds, according to C language semantics, the result is
740 **1**. Otherwise, it is **0**.
742 Note that unlike in C, these operators have a lower precedence than the
743 **assignment** operators, which means that **a=b\>c** is interpreted as
746 Also, unlike the [standard][1] requires, these operators can appear anywhere
747 any other expressions can be used. This allowance is a
748 **non-portable extension**.
752 : The **boolean and** operator takes two expressions and returns **1** if both
753 expressions are non-zero, **0** otherwise.
755 This is *not* a short-circuit operator.
757 This is a **non-portable extension**.
761 : The **boolean or** operator takes two expressions and returns **1** if one
762 of the expressions is non-zero, **0** otherwise.
764 This is *not* a short-circuit operator.
766 This is a **non-portable extension**.
770 The following items are statements:
773 2. **{** **S** **;** ... **;** **S** **}**
774 3. **if** **(** **E** **)** **S**
775 4. **if** **(** **E** **)** **S** **else** **S**
776 5. **while** **(** **E** **)** **S**
777 6. **for** **(** **E** **;** **E** **;** **E** **)** **S**
778 7. An empty statement
784 13. A string of characters, enclosed in double quotes
785 14. **print** **E** **,** ... **,** **E**
786 15. **I()**, **I(E)**, **I(E, E)**, and so on, where **I** is an identifier for
787 a **void** function (see the *Void Functions* subsection of the
788 **FUNCTIONS** section). The **E** argument(s) may also be arrays of the form
789 **I[]**, which will automatically be turned into array references (see the
790 *Array References* subsection of the **FUNCTIONS** section) if the
791 corresponding parameter in the function definition is an array reference.
793 Numbers 4, 9, 11, 12, 14, and 15 are **non-portable extensions**.
795 Also, as a **non-portable extension**, any or all of the expressions in the
796 header of a for loop may be omitted. If the condition (second expression) is
797 omitted, it is assumed to be a constant **1**.
799 The **break** statement causes a loop to stop iterating and resume execution
800 immediately following a loop. This is only allowed in loops.
802 The **continue** statement causes a loop iteration to stop early and returns to
803 the start of the loop, including testing the loop condition. This is only
806 The **if** **else** statement does the same thing as in C.
808 The **quit** statement causes bc(1) to quit, even if it is on a branch that will
809 not be executed (it is a compile-time command).
811 The **halt** statement causes bc(1) to quit, if it is executed. (Unlike **quit**
812 if it is on a branch of an **if** statement that is not executed, bc(1) does not
815 The **limits** statement prints the limits that this bc(1) is subject to. This
816 is like the **quit** statement in that it is a compile-time command.
818 An expression by itself is evaluated and printed, followed by a newline.
820 {{ A H N P HN HP NP HNP }}
821 Both scientific notation and engineering notation are available for printing the
822 results of expressions. Scientific notation is activated by assigning **0** to
823 **obase**, and engineering notation is activated by assigning **1** to
824 **obase**. To deactivate them, just assign a different value to **obase**.
826 Scientific notation and engineering notation are disabled if bc(1) is run with
827 either the **-s** or **-w** command-line options (or equivalents).
829 Printing numbers in scientific notation and/or engineering notation is a
830 **non-portable extension**.
835 The "expressions" in a **print** statement may also be strings. If they are, there
836 are backslash escape sequences that are interpreted specially. What those
837 sequences are, and what they cause to be printed, are shown below:
851 Any other character following a backslash causes the backslash and character to
854 Any non-string expression in a print statement shall be assigned to **last**,
855 like any other expression that is printed.
857 ## Order of Evaluation
859 All expressions in a statment are evaluated left to right, except as necessary
860 to maintain order of operations. This means, for example, assuming that **i** is
861 equal to **0**, in the expression
865 the first (or 0th) element of **a** is set to **1**, and **i** is equal to **2**
866 at the end of the expression.
868 This includes function arguments. Thus, assuming **i** is equal to **0**, this
869 means that in the expression
873 the first argument passed to **x()** is **0**, and the second argument is **1**,
874 while **i** is equal to **2** before the function starts executing.
878 Function definitions are as follows:
888 Any **I** in the parameter list or **auto** list may be replaced with **I[]** to
889 make a parameter or **auto** var an array, and any **I** in the parameter list
890 may be replaced with **\*I[]** to make a parameter an array reference. Callers
891 of functions that take array references should not put an asterisk in the call;
892 they must be called with just **I[]** like normal array parameters and will be
893 automatically converted into references.
895 As a **non-portable extension**, the opening brace of a **define** statement may
896 appear on the next line.
898 As a **non-portable extension**, the return statement may also be in one of the
902 2. **return** **(** **)**
905 The first two, or not specifying a **return** statement, is equivalent to
906 **return (0)**, unless the function is a **void** function (see the *Void
907 Functions* subsection below).
911 Functions can also be **void** functions, defined as follows:
914 define void I(I,...,I){
921 They can only be used as standalone expressions, where such an expression would
922 be printed alone, except in a print statement.
924 Void functions can only use the first two **return** statements listed above.
925 They can also omit the return statement entirely.
927 The word "void" is not treated as a keyword; it is still possible to have
928 variables, arrays, and functions named **void**. The word "void" is only
929 treated specially right after the **define** keyword.
931 This is a **non-portable extension**.
935 For any array in the parameter list, if the array is declared in the form
941 it is a **reference**. Any changes to the array in the function are reflected,
942 when the function returns, to the array that was passed in.
944 Other than this, all function arguments are passed by value.
946 This is a **non-portable extension**.
950 {{ A H N P HN HP NP HNP }}
951 All of the functions below, including the functions in the extended math
952 library (see the *Extended Library* subsection below), are available when the
953 **-l** or **--mathlib** command-line flags are given, except that the extended
954 math library is not available when the **-s** option, the **-w** option, or
955 equivalents are given.
957 {{ E EH EN EP EHN EHP ENP EHNP }}
958 All of the functions below are available when the **-l** or **--mathlib**
959 command-line flags are given.
964 The [standard][1] defines the following functions for the math library:
968 : Returns the sine of **x**, which is assumed to be in radians.
970 This is a transcendental function (see the *Transcendental Functions*
975 : Returns the cosine of **x**, which is assumed to be in radians.
977 This is a transcendental function (see the *Transcendental Functions*
982 : Returns the arctangent of **x**, in radians.
984 This is a transcendental function (see the *Transcendental Functions*
989 : Returns the natural logarithm of **x**.
991 This is a transcendental function (see the *Transcendental Functions*
996 : Returns the mathematical constant **e** raised to the power of **x**.
998 This is a transcendental function (see the *Transcendental Functions*
1003 : Returns the bessel integer order **n** (truncated) of **x**.
1005 This is a transcendental function (see the *Transcendental Functions*
1008 {{ A H N P HN HP NP HNP }}
1011 The extended library is *not* loaded when the **-s**/**--standard** or
1012 **-w**/**--warn** options are given since they are not part of the library
1013 defined by the [standard][1].
1015 The extended library is a **non-portable extension**.
1019 : Calculates **x** to the power of **y**, even if **y** is not an integer, and
1020 returns the result to the current **scale**.
1022 It is an error if **y** is negative and **x** is **0**.
1024 This is a transcendental function (see the *Transcendental Functions*
1029 : Returns **x** rounded to **p** decimal places according to the rounding mode
1030 [round half away from **0**][3].
1034 : Returns **x** rounded to **p** decimal places according to the rounding mode
1035 [round away from **0**][6].
1039 : Returns the factorial of the truncated absolute value of **x**.
1043 : Returns the permutation of the truncated absolute value of **n** of the
1044 truncated absolute value of **k**, if **k \<= n**. If not, it returns **0**.
1048 : Returns the combination of the truncated absolute value of **n** of the
1049 truncated absolute value of **k**, if **k \<= n**. If not, it returns **0**.
1053 : Returns the logarithm base **2** of **x**.
1055 This is a transcendental function (see the *Transcendental Functions*
1060 : Returns the logarithm base **10** of **x**.
1062 This is a transcendental function (see the *Transcendental Functions*
1067 : Returns the logarithm base **b** of **x**.
1069 This is a transcendental function (see the *Transcendental Functions*
1074 : Returns the cube root of **x**.
1078 : Calculates the truncated value of **n**, **r**, and returns the **r**th root
1079 of **x** to the current **scale**.
1081 If **r** is **0** or negative, this raises an error and causes bc(1) to
1082 reset (see the **RESET** section). It also raises an error and causes bc(1)
1083 to reset if **r** is even and **x** is negative.
1087 : Returns **pi** to **p** decimal places.
1089 This is a transcendental function (see the *Transcendental Functions*
1094 : Returns the tangent of **x**, which is assumed to be in radians.
1096 This is a transcendental function (see the *Transcendental Functions*
1101 : Returns the arctangent of **y/x**, in radians. If both **y** and **x** are
1102 equal to **0**, it raises an error and causes bc(1) to reset (see the
1103 **RESET** section). Otherwise, if **x** is greater than **0**, it returns
1104 **a(y/x)**. If **x** is less than **0**, and **y** is greater than or equal
1105 to **0**, it returns **a(y/x)+pi**. If **x** is less than **0**, and **y**
1106 is less than **0**, it returns **a(y/x)-pi**. If **x** is equal to **0**,
1107 and **y** is greater than **0**, it returns **pi/2**. If **x** is equal to
1108 **0**, and **y** is less than **0**, it returns **-pi/2**.
1110 This function is the same as the **atan2()** function in many programming
1113 This is a transcendental function (see the *Transcendental Functions*
1118 : Returns the sine of **x**, which is assumed to be in radians.
1120 This is an alias of **s(x)**.
1122 This is a transcendental function (see the *Transcendental Functions*
1127 : Returns the cosine of **x**, which is assumed to be in radians.
1129 This is an alias of **c(x)**.
1131 This is a transcendental function (see the *Transcendental Functions*
1136 : Returns the tangent of **x**, which is assumed to be in radians.
1138 If **x** is equal to **1** or **-1**, this raises an error and causes bc(1)
1139 to reset (see the **RESET** section).
1141 This is an alias of **t(x)**.
1143 This is a transcendental function (see the *Transcendental Functions*
1148 : Returns the arctangent of **x**, in radians.
1150 This is an alias of **a(x)**.
1152 This is a transcendental function (see the *Transcendental Functions*
1157 : Returns the arctangent of **y/x**, in radians. If both **y** and **x** are
1158 equal to **0**, it raises an error and causes bc(1) to reset (see the
1159 **RESET** section). Otherwise, if **x** is greater than **0**, it returns
1160 **a(y/x)**. If **x** is less than **0**, and **y** is greater than or equal
1161 to **0**, it returns **a(y/x)+pi**. If **x** is less than **0**, and **y**
1162 is less than **0**, it returns **a(y/x)-pi**. If **x** is equal to **0**,
1163 and **y** is greater than **0**, it returns **pi/2**. If **x** is equal to
1164 **0**, and **y** is less than **0**, it returns **-pi/2**.
1166 This function is the same as the **atan2()** function in many programming
1169 This is an alias of **a2(y, x)**.
1171 This is a transcendental function (see the *Transcendental Functions*
1176 : Converts **x** from radians to degrees and returns the result.
1178 This is a transcendental function (see the *Transcendental Functions*
1183 : Converts **x** from degrees to radians and returns the result.
1185 This is a transcendental function (see the *Transcendental Functions*
1190 : Generates a pseudo-random number between **0** (inclusive) and **1**
1191 (exclusive) with the number of decimal digits after the decimal point equal
1192 to the truncated absolute value of **p**. If **p** is not **0**, then
1193 calling this function will change the value of **seed**. If **p** is **0**,
1194 then **0** is returned, and **seed** is *not* changed.
1198 : Generates a pseudo-random number that is between **0** (inclusive) and the
1199 truncated absolute value of **i** (exclusive) with the number of decimal
1200 digits after the decimal point equal to the truncated absolute value of
1201 **p**. If the absolute value of **i** is greater than or equal to **2**, and
1202 **p** is not **0**, then calling this function will change the value of
1203 **seed**; otherwise, **0** is returned and **seed** is not changed.
1207 : Returns **x** with its sign flipped with probability **0.5**. In other
1208 words, it randomizes the sign of **x**.
1212 : Returns a random boolean value (either **0** or **1**).
1216 : Returns the numbers of unsigned integer bytes required to hold the truncated
1217 absolute value of **x**.
1221 : Returns the numbers of signed, two's-complement integer bytes required to
1222 hold the truncated value of **x**.
1226 : Outputs the hexadecimal (base **16**) representation of **x**.
1228 This is a **void** function (see the *Void Functions* subsection of the
1229 **FUNCTIONS** section).
1233 : Outputs the binary (base **2**) representation of **x**.
1235 This is a **void** function (see the *Void Functions* subsection of the
1236 **FUNCTIONS** section).
1240 : Outputs the base **b** representation of **x**.
1242 This is a **void** function (see the *Void Functions* subsection of the
1243 **FUNCTIONS** section).
1247 : Outputs the representation, in binary and hexadecimal, of **x** as an
1248 unsigned integer in as few power of two bytes as possible. Both outputs are
1249 split into bytes separated by spaces.
1251 If **x** is not an integer or is negative, an error message is printed
1252 instead, but bc(1) is not reset (see the **RESET** section).
1254 This is a **void** function (see the *Void Functions* subsection of the
1255 **FUNCTIONS** section).
1259 : Outputs the representation, in binary and hexadecimal, of **x** as a signed,
1260 two's-complement integer in as few power of two bytes as possible. Both
1261 outputs are split into bytes separated by spaces.
1263 If **x** is not an integer, an error message is printed instead, but bc(1)
1264 is not reset (see the **RESET** section).
1266 This is a **void** function (see the *Void Functions* subsection of the
1267 **FUNCTIONS** section).
1271 : Outputs the representation, in binary and hexadecimal, of **x** as an
1272 unsigned integer in **n** bytes. Both outputs are split into bytes separated
1275 If **x** is not an integer, is negative, or cannot fit into **n** bytes, an
1276 error message is printed instead, but bc(1) is not reset (see the **RESET**
1279 This is a **void** function (see the *Void Functions* subsection of the
1280 **FUNCTIONS** section).
1284 : Outputs the representation, in binary and hexadecimal, of **x** as a signed,
1285 two's-complement integer in **n** bytes. Both outputs are split into bytes
1286 separated by spaces.
1288 If **x** is not an integer or cannot fit into **n** bytes, an error message
1289 is printed instead, but bc(1) is not reset (see the **RESET** section).
1291 This is a **void** function (see the *Void Functions* subsection of the
1292 **FUNCTIONS** section).
1296 : Outputs the representation, in binary and hexadecimal, of **x** as an
1297 unsigned integer in **1** byte. Both outputs are split into bytes separated
1300 If **x** is not an integer, is negative, or cannot fit into **1** byte, an
1301 error message is printed instead, but bc(1) is not reset (see the **RESET**
1304 This is a **void** function (see the *Void Functions* subsection of the
1305 **FUNCTIONS** section).
1309 : Outputs the representation, in binary and hexadecimal, of **x** as a signed,
1310 two's-complement integer in **1** byte. Both outputs are split into bytes
1311 separated by spaces.
1313 If **x** is not an integer or cannot fit into **1** byte, an error message
1314 is printed instead, but bc(1) is not reset (see the **RESET** section).
1316 This is a **void** function (see the *Void Functions* subsection of the
1317 **FUNCTIONS** section).
1321 : Outputs the representation, in binary and hexadecimal, of **x** as an
1322 unsigned integer in **2** bytes. Both outputs are split into bytes separated
1325 If **x** is not an integer, is negative, or cannot fit into **2** bytes, an
1326 error message is printed instead, but bc(1) is not reset (see the **RESET**
1329 This is a **void** function (see the *Void Functions* subsection of the
1330 **FUNCTIONS** section).
1334 : Outputs the representation, in binary and hexadecimal, of **x** as a signed,
1335 two's-complement integer in **2** bytes. Both outputs are split into bytes
1336 separated by spaces.
1338 If **x** is not an integer or cannot fit into **2** bytes, an error message
1339 is printed instead, but bc(1) is not reset (see the **RESET** section).
1341 This is a **void** function (see the *Void Functions* subsection of the
1342 **FUNCTIONS** section).
1346 : Outputs the representation, in binary and hexadecimal, of **x** as an
1347 unsigned integer in **4** bytes. Both outputs are split into bytes separated
1350 If **x** is not an integer, is negative, or cannot fit into **4** bytes, an
1351 error message is printed instead, but bc(1) is not reset (see the **RESET**
1354 This is a **void** function (see the *Void Functions* subsection of the
1355 **FUNCTIONS** section).
1359 : Outputs the representation, in binary and hexadecimal, of **x** as a signed,
1360 two's-complement integer in **4** bytes. Both outputs are split into bytes
1361 separated by spaces.
1363 If **x** is not an integer or cannot fit into **4** bytes, an error message
1364 is printed instead, but bc(1) is not reset (see the **RESET** section).
1366 This is a **void** function (see the *Void Functions* subsection of the
1367 **FUNCTIONS** section).
1371 : Outputs the representation, in binary and hexadecimal, of **x** as an
1372 unsigned integer in **8** bytes. Both outputs are split into bytes separated
1375 If **x** is not an integer, is negative, or cannot fit into **8** bytes, an
1376 error message is printed instead, but bc(1) is not reset (see the **RESET**
1379 This is a **void** function (see the *Void Functions* subsection of the
1380 **FUNCTIONS** section).
1384 : Outputs the representation, in binary and hexadecimal, of **x** as a signed,
1385 two's-complement integer in **8** bytes. Both outputs are split into bytes
1386 separated by spaces.
1388 If **x** is not an integer or cannot fit into **8** bytes, an error message
1389 is printed instead, but bc(1) is not reset (see the **RESET** section).
1391 This is a **void** function (see the *Void Functions* subsection of the
1392 **FUNCTIONS** section).
1396 : Outputs the representation of the truncated absolute value of **x** as an
1397 unsigned integer in hexadecimal using **n** bytes. Not all of the value will
1398 be output if **n** is too small.
1400 This is a **void** function (see the *Void Functions* subsection of the
1401 **FUNCTIONS** section).
1403 **binary_uint(x, n)**
1405 : Outputs the representation of the truncated absolute value of **x** as an
1406 unsigned integer in binary using **n** bytes. Not all of the value will be
1407 output if **n** is too small.
1409 This is a **void** function (see the *Void Functions* subsection of the
1410 **FUNCTIONS** section).
1412 **output_uint(x, n)**
1414 : Outputs the representation of the truncated absolute value of **x** as an
1415 unsigned integer in the current **obase** (see the **SYNTAX** section) using
1416 **n** bytes. Not all of the value will be output if **n** is too small.
1418 This is a **void** function (see the *Void Functions* subsection of the
1419 **FUNCTIONS** section).
1421 **output_byte(x, i)**
1423 : Outputs byte **i** of the truncated absolute value of **x**, where **0** is
1424 the least significant byte and **number_of_bytes - 1** is the most
1427 This is a **void** function (see the *Void Functions* subsection of the
1428 **FUNCTIONS** section).
1431 ## Transcendental Functions
1433 All transcendental functions can return slightly inaccurate results (up to 1
1434 [ULP][4]). This is unavoidable, and [this article][5] explains why it is
1435 impossible and unnecessary to calculate exact results for the transcendental
1438 Because of the possible inaccuracy, I recommend that users call those functions
1439 with the precision (**scale**) set to at least 1 higher than is necessary. If
1440 exact results are *absolutely* required, users can double the precision
1441 (**scale**) and then truncate.
1443 The transcendental functions in the standard math library are:
1452 {{ A H N P HN HP NP HNP }}
1453 The transcendental functions in the extended math library are:
1472 When bc(1) encounters an error or a signal that it has a non-default handler
1473 for, it resets. This means that several things happen.
1475 First, any functions that are executing are stopped and popped off the stack.
1476 The behavior is not unlike that of exceptions in programming languages. Then
1477 the execution point is set so that any code waiting to execute (after all
1478 functions returned) is skipped.
1480 Thus, when bc(1) resets, it skips any remaining code waiting to be executed.
1481 Then, if it is interactive mode, and the error was not a fatal error (see the
1482 **EXIT STATUS** section), it asks for more input; otherwise, it exits with the
1483 appropriate return code.
1485 Note that this reset behavior is different from the GNU bc(1), which attempts to
1486 start executing the statement right after the one that caused an error.
1490 Most bc(1) implementations use **char** types to calculate the value of **1**
1491 decimal digit at a time, but that can be slow. This bc(1) does something
1494 It uses large integers to calculate more than **1** decimal digit at a time. If
1495 built in a environment where **BC_LONG_BIT** (see the **LIMITS** section) is
1496 **64**, then each integer has **9** decimal digits. If built in an environment
1497 where **BC_LONG_BIT** is **32** then each integer has **4** decimal digits. This
1498 value (the number of decimal digits per large integer) is called
1501 The actual values of **BC_LONG_BIT** and **BC_BASE_DIGS** can be queried with
1502 the **limits** statement.
1504 In addition, this bc(1) uses an even larger integer for overflow checking. This
1505 integer type depends on the value of **BC_LONG_BIT**, but is always at least
1506 twice as large as the integer type used to store digits.
1510 The following are the limits on bc(1):
1514 : The number of bits in the **long** type in the environment where bc(1) was
1515 built. This determines how many decimal digits can be stored in a single
1516 large integer (see the **PERFORMANCE** section).
1520 : The number of decimal digits per large integer (see the **PERFORMANCE**
1521 section). Depends on **BC_LONG_BIT**.
1525 : The max decimal number that each large integer can store (see
1526 **BC_BASE_DIGS**) plus **1**. Depends on **BC_BASE_DIGS**.
1530 : The max number that the overflow type (see the **PERFORMANCE** section) can
1531 hold. Depends on **BC_LONG_BIT**.
1535 : The maximum output base. Set at **BC_BASE_POW**.
1539 : The maximum size of arrays. Set at **SIZE_MAX-1**.
1543 : The maximum **scale**. Set at **BC_OVERFLOW_MAX-1**.
1547 : The maximum length of strings. Set at **BC_OVERFLOW_MAX-1**.
1551 : The maximum length of identifiers. Set at **BC_OVERFLOW_MAX-1**.
1555 : The maximum length of a number (in decimal digits), which includes digits
1556 after the decimal point. Set at **BC_OVERFLOW_MAX-1**.
1558 {{ A H N P HN HP NP HNP }}
1561 : The maximum integer (inclusive) returned by the **rand()** operand. Set at
1562 **2\^BC_LONG_BIT-1**.
1567 : The maximum allowable exponent (positive or negative). Set at
1568 **BC_OVERFLOW_MAX**.
1572 : The maximum number of vars/arrays. Set at **SIZE_MAX-1**.
1574 The actual values can be queried with the **limits** statement.
1576 These limits are meant to be effectively non-existent; the limits are so large
1577 (at least on 64-bit machines) that there should not be any point at which they
1578 become a problem. In fact, memory should be exhausted before these limits should
1581 # ENVIRONMENT VARIABLES
1583 bc(1) recognizes the following environment variables:
1587 : If this variable exists (no matter the contents), bc(1) behaves as if
1588 the **-s** option was given.
1592 : This is another way to give command-line arguments to bc(1). They should be
1593 in the same format as all other command-line arguments. These are always
1594 processed first, so any files given in **BC_ENV_ARGS** will be processed
1595 before arguments and files given on the command-line. This gives the user
1596 the ability to set up "standard" options and files to be used at every
1597 invocation. The most useful thing for such files to contain would be useful
1598 functions that the user might want every time bc(1) runs.
1600 The code that parses **BC_ENV_ARGS** will correctly handle quoted arguments,
1601 but it does not understand escape sequences. For example, the string
1602 **"/home/gavin/some bc file.bc"** will be correctly parsed, but the string
1603 **"/home/gavin/some \"bc\" file.bc"** will include the backslashes.
1605 The quote parsing will handle either kind of quotes, **'** or **"**. Thus,
1606 if you have a file with any number of single quotes in the name, you can use
1607 double quotes as the outside quotes, as in **"some 'bc' file.bc"**, and vice
1608 versa if you have a file with double quotes. However, handling a file with
1609 both kinds of quotes in **BC_ENV_ARGS** is not supported due to the
1610 complexity of the parsing, though such files are still supported on the
1611 command-line where the parsing is done by the shell.
1615 : If this environment variable exists and contains an integer that is greater
1616 than **1** and is less than **UINT16_MAX** (**2\^16-1**), bc(1) will output
1617 lines to that length, including the backslash (**\\**). The default line
1622 bc(1) returns the following exit statuses:
1630 : A math error occurred. This follows standard practice of using **1** for
1631 expected errors, since math errors will happen in the process of normal
1634 {{ A H N P HN HP NP HNP }}
1635 Math errors include divide by **0**, taking the square root of a negative
1636 number, using a negative number as a bound for the pseudo-random number
1637 generator, attempting to convert a negative number to a hardware integer,
1638 overflow when converting a number to a hardware integer, and attempting to
1639 use a non-integer where an integer is required.
1641 Converting to a hardware integer happens for the second operand of the power
1642 (**\^**), places (**\@**), left shift (**\<\<**), and right shift (**\>\>**)
1643 operators and their corresponding assignment operators.
1645 {{ E EH EN EP EHN EHP ENP EHNP }}
1646 Math errors include divide by **0**, taking the square root of a negative
1647 number, attempting to convert a negative number to a hardware integer,
1648 overflow when converting a number to a hardware integer, and attempting to
1649 use a non-integer where an integer is required.
1651 Converting to a hardware integer happens for the second operand of the power
1652 (**\^**) operator and the corresponding assignment operator.
1657 : A parse error occurred.
1659 Parse errors include unexpected **EOF**, using an invalid character, failing
1660 to find the end of a string or comment, using a token where it is invalid,
1661 giving an invalid expression, giving an invalid print statement, giving an
1662 invalid function definition, attempting to assign to an expression that is
1663 not a named expression (see the *Named Expressions* subsection of the
1664 **SYNTAX** section), giving an invalid **auto** list, having a duplicate
1665 **auto**/function parameter, failing to find the end of a code block,
1666 attempting to return a value from a **void** function, attempting to use a
1667 variable as a reference, and using any extensions when the option **-s** or
1668 any equivalents were given.
1672 : A runtime error occurred.
1674 Runtime errors include assigning an invalid number to **ibase**, **obase**,
1675 or **scale**; give a bad expression to a **read()** call, calling **read()**
1676 inside of a **read()** call, type errors, passing the wrong number of
1677 arguments to functions, attempting to call an undefined function, and
1678 attempting to use a **void** function call as a value in an expression.
1682 : A fatal error occurred.
1684 Fatal errors include memory allocation errors, I/O errors, failing to open
1685 files, attempting to use files that do not have only ASCII characters (bc(1)
1686 only accepts ASCII characters), attempting to open a directory as a file,
1687 and giving invalid command-line options.
1689 The exit status **4** is special; when a fatal error occurs, bc(1) always exits
1690 and returns **4**, no matter what mode bc(1) is in.
1692 The other statuses will only be returned when bc(1) is not in interactive mode
1693 (see the **INTERACTIVE MODE** section), since bc(1) resets its state (see the
1694 **RESET** section) and accepts more input when one of those errors occurs in
1695 interactive mode. This is also the case when interactive mode is forced by the
1696 **-i** flag or **--interactive** option.
1698 These exit statuses allow bc(1) to be used in shell scripting with error
1699 checking, and its normal behavior can be forced by using the **-i** flag or
1700 **--interactive** option.
1704 Per the [standard][1], bc(1) has an interactive mode and a non-interactive mode.
1705 Interactive mode is turned on automatically when both **stdin** and **stdout**
1706 are hooked to a terminal, but the **-i** flag and **--interactive** option can
1707 turn it on in other cases.
1709 In interactive mode, bc(1) attempts to recover from errors (see the **RESET**
1710 section), and in normal execution, flushes **stdout** as soon as execution is
1711 done for the current input.
1715 If **stdin**, **stdout**, and **stderr** are all connected to a TTY, bc(1) turns
1718 {{ A E N P EN EP NP ENP }}
1719 TTY mode is required for history to be enabled (see the **COMMAND LINE HISTORY**
1720 section). It is also required to enable special handling for **SIGINT** signals.
1723 {{ A E H N EH EN HN EHN }}
1724 The prompt is enabled in TTY mode.
1727 TTY mode is different from interactive mode because interactive mode is required
1728 in the [bc(1) specification][1], and interactive mode requires only **stdin**
1729 and **stdout** to be connected to a terminal.
1733 Sending a **SIGINT** will cause bc(1) to stop execution of the current input. If
1734 bc(1) is in TTY mode (see the **TTY MODE** section), it will reset (see the
1735 **RESET** section). Otherwise, it will clean up and exit.
1737 Note that "current input" can mean one of two things. If bc(1) is processing
1738 input from **stdin** in TTY mode, it will ask for more input. If bc(1) is
1739 processing input from a file in TTY mode, it will stop processing the file and
1740 start processing the next file, if one exists, or ask for input from **stdin**
1741 if no other file exists.
1743 This means that if a **SIGINT** is sent to bc(1) as it is executing a file, it
1744 can seem as though bc(1) did not respond to the signal since it will immediately
1745 start executing the next file. This is by design; most files that users execute
1746 when interacting with bc(1) have function definitions, which are quick to parse.
1747 If a file takes a long time to execute, there may be a bug in that file. The
1748 rest of the files could still be executed without problem, allowing the user to
1751 **SIGTERM** and **SIGQUIT** cause bc(1) to clean up and exit, and it uses the
1752 {{ A E N P EN EP NP ENP }}
1753 default handler for all other signals. The one exception is **SIGHUP**; in that
1754 case, when bc(1) is in TTY mode, a **SIGHUP** will cause bc(1) to clean up and
1757 {{ H EH HN HP EHN EHP HNP EHNP }}
1758 default handler for all other signals.
1761 {{ A E N P EN EP NP ENP }}
1762 # COMMAND LINE HISTORY
1764 bc(1) supports interactive command-line editing. If bc(1) is in TTY mode (see
1765 the **TTY MODE** section), history is enabled. Previous lines can be recalled
1766 and edited with the arrow keys.
1768 **Note**: tabs are converted to 8 spaces.
1771 {{ A E H P EH EP HP EHP }}
1774 This bc(1) ships with support for adding error messages for different locales
1775 and thus, supports **LC_MESSAGES**.
1784 bc(1) is compliant with the [IEEE Std 1003.1-2017 (“POSIX.1-2017”)][1]
1785 specification. The flags **-efghiqsvVw**, all long options, and the extensions
1786 noted above are extensions to that specification.
1788 Note that the specification explicitly says that bc(1) only accepts numbers that
1789 use a period (**.**) as a radix point, regardless of the value of
1792 {{ A E H P EH EP HP EHP }}
1793 This bc(1) supports error messages for different locales, and thus, it supports
1799 None are known. Report bugs at https://git.yzena.com/gavin/bc.
1803 Gavin D. Howard <gavin@yzena.com> and contributors.
1805 [1]: https://pubs.opengroup.org/onlinepubs/9699919799/utilities/bc.html
1806 [2]: https://www.gnu.org/software/bc/
1807 [3]: https://en.wikipedia.org/wiki/Rounding#Round_half_away_from_zero
1808 [4]: https://en.wikipedia.org/wiki/Unit_in_the_last_place
1809 [5]: https://people.eecs.berkeley.edu/~wkahan/LOG10HAF.TXT
1810 [6]: https://en.wikipedia.org/wiki/Rounding#Rounding_away_from_zero