Simple Regular Expressions

1.

Any character except a special character matches itself. Special characters are the regular expression delimiter plus \[. and sometimes ^*$.

2.

A . matches any character.

3.

A \ followed by any character except a digit or () {} <> matches that character.

4.

A nonempty string s bracketed [s] (or [^s]) forms a bracket expression that matches any character in (or not in) s. In s, \ has no special meaning, and ] may only appear as the first letter. A substring a-b, with a and b in ascending ASCII order, stands for the inclusive range of ASCII characters.

5.

A regular expression of form 1-4 followed by * matches a sequence of 0 or more matches of the regular expression.

6.

A regular expression of form 1-4 followed by \{m,n\} forms an interval expression that matches a sequence of m through n matches, inclusive, of the regular expression. The values of m and n must be non-negative and smaller than 256. The form \{m\} matches exactly m occurrences, \{m,\} matches at least m occurrences.

7.

The sequence \< forces the match to occur only at the beginning of a ``variable'' or ``word''; that is, either at the beginning of a line, or just before a letter, digit or underline and after a character not one of these.

8.

The sequence \> matches the end of a ``variable'' or ``word'', i.e. either the end of the line or before character which is neither a letter, nor a digit, nor the underline character.

9.

A regular expression, x, of form 1-11, parenthesized \(x\) is called a subexpression and matches what x matches.

10.

A \ followed by a digit n forms a backreference and matches a copy of the string that the parenthesized regular expression beginning with the nth \( matched.

11.

A regular expression of form 1-11, x, followed by a regular expression of form 1-10, y matches a match for x followed by a match for y, with the x match being as long as possible while still permitting a y match.

12.

A regular expression of form 1-11 preceded by ^ (or followed by $), is constrained to matches that begin at the left (or end at the right) end of a line (anchoring).

13.

A regular expression of form 1-12 picks out the longest among the leftmost matches in a line.

14.

An empty regular expression stands for a copy of the last regular expression encountered.

Basic Regular Expressions

Basic Regular Expressions add the following features to Simple Regular Expressions:

15.

The special character *, as described in 5., and the interval expressions described in 6. can also be applied to subexpressions as described in 9. For POSIX.1-2001 utilities such as /usr/5bin/posix2001/ed, subexpressions do not match the empty string if there is a possible longer match.

16.

In bracket expressions as described in 4., the following character sequences are considered special:

Character class expressions of the form [:class:]. In the C LC_CTYPE locale, the classes

[:alnum:]	[:cntrl:]	[:lower:]	[:space:]
[:alpha:]	[:digit:]	[:print:]	[:upper:]
[:blank:]	[:graph:]	[:punct:]	[:xdigit:]

are recognized; further locale-specific classes may be available. A character class expression matches any character that belongs to the given class in the current LC_CTYPE locale.

Collating symbol expressions of the form [.c.], where c is a collating symbol in the current LC_COLLATE locale. A collating symbol expression matches the specified collating symbol.

Equivalence class expressions of the form [=c=], where c is a collating symbol in the current LC_COLLATE locale. An equivalence class expression matches any character that has the same collating weight as c.

Regular expressions are used in addresses to specify lines and in one command (see s below) to specify a portion of a line which is to be replaced. If it is desired to use one of the regular expression metacharacters as an ordinary character, that character may be preceded by `\'. This also applies to the character bounding the regular expression (often `/') and to `\' itself.

To understand addressing in ed it is necessary to know that at any time there is a current line. Generally speaking, the current line is the last line affected by a command; however, the exact effect on the current line is discussed under the description of the command. Addresses are constructed as follows.

1.

The character `.' addresses the current line.

2.

The character `$' addresses the last line of the buffer.

3.

A decimal number n addresses the n-th line of the buffer.

4.

`′x' addresses the line marked with the name x, which must be a lower-case letter. Lines are marked with the k command described below.

5.

A regular expression enclosed in slashes `/' addresses the line found by searching forward from the current line and stopping at the first line containing a string that matches the regular expression. If necessary the search wraps around to the beginning of the buffer.

6.

A regular expression enclosed in queries `?' addresses the line found by searching backward from the current line and stopping at the first line containing a string that matches the regular expression. If necessary the search wraps around to the end of the buffer.

7.

An address followed by a plus sign `+' or a minus sign `-' followed by a decimal number specifies that address plus (resp. minus) the indicated number of lines. The plus sign may be omitted.

8.

If an address begins with `+' or `-' the addition or subtraction is taken with respect to the current line; e.g. `-5' is understood to mean `.-5'.

9.

If an address ends with `+' or `-', then 1 is added (resp. subtracted). As a consequence of this rule and rule 8, the address `-' refers to the line before the current line. Moreover, trailing `+' and `-' characters have cumulative effect, so `--' refers to the current line less 2.

10.

To maintain compatibility with earlier versions of the editor, the character `^' in addresses is equivalent to `-'.

Commands may require zero, one, or two addresses. Commands which require no addresses regard the presence of an address as an error. Commands which accept one or two addresses assume default addresses when insufficient are given. If more addresses are given than such a command requires, the last one or two (depending on what is accepted) are used.

Addresses are separated from each other typically by a comma `,'. They may also be separated by a semicolon `;'. In this case the current line `.' is set to the previous address before the next address is interpreted. This feature can be used to determine the starting line for forward and backward searches (`/', `?'). The second address of any two-address sequence must correspond to a line following the line corresponding to the first address.

Omission of the first address causes the first line to be used with `,', or the current line with `;', respectively; if the second address is also omitted, the last line of the buffer is used. Thus a single `,' specifies the entire contents of the buffer, and a single `;' specifies the contents ranging from the current line to the last one.

In the following list of ed commands, the default addresses are shown in parentheses. The parentheses are not part of the address, but are used to show that the given addresses are the default.

As mentioned, it is generally illegal for more than one command to appear on a line. However, most commands may be suffixed by `p', `l', or `n', in which case the current line is either printed, listed, or numbered respectively in the way discussed below.

(.)a

<text>

.

The append command reads the given text and appends it after the addressed line. `.' is left on the last line input, if there were any, otherwise at the addressed line. Address `0' is legal for this command; text is placed at the beginning of the buffer.

(.,.)c

<text>

.

The change command deletes the addressed lines, then accepts input text which replaces these lines. `.' is left at the last line input; if there were none, it is left at the line preceding the deleted lines. For /usr/5bin/posix2001/ed, a `0' as the first address is identical to `1'.

(.,.)d

The delete command deletes the addressed lines from the buffer. The line originally after the last line deleted becomes the current line; if the lines deleted were originally at the end, the new last line becomes the current line.

e filename

The edit command causes the entire contents of the buffer to be deleted, and then the named file to be read in. `.' is set to the last line of the buffer. The number of characters read is typed. `filename' is remembered for possible use as a default file name in a subsequent r or w command. If `filename' is missing, the remembered name is used. A `filename' starting with a `!' causes the output of the shell command following this character to be read in.

E filename

This command is the same as e, except that no diagnostic results when no w has been given since the last buffer alteration.

f filename

The filename command prints the currently remembered file name. If `filename' is given, the currently remembered file name is changed to `filename'.

(1,$)g/regular expression/command list

In the global command, the first step is to mark every line which matches the given regular expression. Then for every such line, the given command list is executed with `.' initially set to that line. A single command or the first of multiple commands appears on the same line with the global command. All lines of a multi-line list except the last line must be ended with `\'. A, i, and c commands and associated input are permitted; the `.' terminating input mode may be omitted if it would be on the last line of the command list. The commands g and v are not permitted in the command list.

(1,$)G/regular expression/

The interactive global command first marks every line matching the given regular expression. Then each line is printed and a command is read and executed for this line. A single newline character causes the line to remain unchanged, an isolated `&' repeats the command given for the previous line. The command can be terminated by an interrupt signal.

h

This command prints a verbose description for the last error encountered.

H

This command acts like the h command, but also causes verbose descriptions to be printed on all following error conditions. Another H turns verbose mode off.

(.)i

<text>

.

This command inserts the given text before the addressed line. `.' is left at the last line input, or, if there were none, at the line before the addressed line. This command differs from the a command only in the placement of the text. For /usr/5bin/posix2001/ed, a `0' as the first address is identical to `1'.

(.,.+1)j

This command joins the addressed lines into a single line; intermediate newlines simply disappear. `.' is left at the resulting line.

(.)kx

The mark command marks the addressed line with name x, which must be a lower-case letter. The address form `′x' then addresses this line.

(.,.)l

The list command prints the addressed lines in an unambiguous way: /usr/5bin/ed prints non-graphic control characters in three-digit octal; /usr/5bin/s42/ed, /usr/5bin/posix/ed, and /usr/5bin/posix2001/ed print control characters as C-style escape sequences or in three-digit octal. Long lines are folded. The l command may be placed on the same line after any non-i/o command.

(.,.)ma

The move command repositions the addressed lines after the line addressed by a. The last of the moved lines becomes the current line.

(.,.)n

This command prints lines preceded by their line numbers. It otherwise acts like the p command described below.

(.,.)p

The print command prints the addressed lines. `.' is left at the last line printed. The p command may be placed on the same line after any non-i/o command.

P

This command causes a prompt to be printed before following commands are read. The default prompt is a `*' character, but can be set with the -p command line option (which also enables the prompt). Another P disables the prompt.

q

The quit command causes ed to exit. No automatic write of a file is done.

Q

This command is the same as q, except that no diagnostic results when no w has been given since the last buffer alteration.

($)r filename

The read command reads in the given file after the addressed line. If no file name is given, the remembered file name, if any, is used (see e and f commands). The file name is remembered if there was no remembered file name already. Address `0' is legal for r and causes the file to be read at the beginning of the buffer. If the read is successful, the number of characters read is typed. `.' is left at the last line read in from the file. A `filename' starting with a `!' causes the output of the shell command following this character to be read in.

(.,.)s/regular expression/replacement/ or,

(.,.)s/regular expression/replacement/g or,

(.,.)s/regular expression/replacement/number

The substitute command searches each addressed line for an occurrence of the specified regular expression. On each line in which a match is found, all matched strings are replaced by the replacement specified, if the global replacement indicator `g' appears after the command. If the global indicator does not appear, only the first occurrence of the matched string is replaced; if the number indicator is given, the numbered occurrence is replaced. It is an error for the substitution to fail on all addressed lines. Any character other than space or new-line may be used instead of `/' to delimit the regular expression and the replacement. `.' is left at the last line substituted.

An ampersand `&' appearing in the replacement is replaced by the string matching the regular expression. The special meaning of `&' in this context may be suppressed by preceding it by `\'. The characters `\n' where n is a digit, are replaced by the text matched by the n-th regular subexpression enclosed between `\(' and `\)'. When nested, parenthesized subexpressions are present, n is determined by counting occurrences of `\(' starting from the left.

A substitution string consisting of a single `%' causes the string given on the previous substitution to be re-used.

Lines may be split by substituting new-line characters into them. The new-line in the replacement string must be escaped by preceding it by `\'.

(.,.)ta

This command acts just like the m command, except that a copy of the addressed lines is placed after address a (which may be 0). `.' is left on the last line of the copy.

u

The undo command restores the contents of the buffer before the last command was executed. If the undo command is given twice, the current state is restored.

(1,$)v/regular expression/command list

This command is the same as the global command g except that the command list is executed g with `.' initially set to every line except those matching the regular expression.

(1,$)V/regular expression/

This command is the same as the interactive global command G except that the commands are read g with `.' initially set to every line except those matching the regular expression.

(1,$)w filename

The write command writes the addressed lines onto the given file. If the file does not exist, it is created mode 666 (readable and writable by everyone). The file name is remembered if there was no remembered file name already. If no file name is given, the remembered file name, if any, is used (see e and f commands). `.' is unchanged. If the command is successful, the number of characters written is printed. A `filename' starting with a `!' causes the string following this character to be executed as a shell command with the addressed lines as standard input.

(1,$)W filename

This command is the same as w, except that the addressed lines are appended to the file.

($)=

The line number of the addressed line is typed. `.' is unchanged by this command.

!<shell command>

The remainder of the line after the `!' is sent to sh(1) to be interpreted as a command. `.' is unchanged. If the command starts with a `!', the previous command is inserted. A `%' causes the current file name to be inserted.

(.+1)<newline>

An address alone on a line causes the addressed line to be printed. A blank line alone is equivalent to `.+1p'; it is useful for stepping through text.

The following commands are extensions:

(.)b[count]

Prints a screenful of lines, starting at the addressed one, and browses forward in the buffer by this amount. With the optional count argument, the screen size for this and following b commands is set to the given number of lines.

help

Causes a summary of ed commands along with short descriptions to be printed on the terminal.

N

Makes the p command behave like the n command and vice-versa. If given a second time, the original semantics are restored.

(.)o[count]

Prints a screenful of lines centered around the addressed one. The current line is not changed. With the optional count argument, the amount of lines printed above and below for this and following o commands is set to the given number.

z

Performs the same actions as a w command followed by a q command.

If an interrupt signal is sent, ed prints a `?' and returns to its command level.

An input line that consists exactly of the two characters `\.' causes a period `.' to be inserted with the a, c, and i commands in /usr/5bin/ed and /usr/5bin/s42/ed.

Some size limitations: The maximum number of bytes in the buffer corresponds to the address size; on machines with 32-bit addressing, it is 2 G bytes, with 64-bit addressing, it is 9 E bytes. The limit on the number of lines depends on the amount of core: each line takes 2 words.

If a line contains a NUL character, regular expressions cannot match beyond this character. A substitute command deletes a NUL and all following characters on the line. NUL characters in command input are discarded. If an input file does not end with a newline, ed prints a message and appends one.

Omission of the `/' character following the regular expression or the replacement string to the global and substitute commands causes the affected lines to be printed. Thus the following commands have the same effect: