diff options
Diffstat (limited to 'doc/regsub.n')
| -rw-r--r-- | doc/regsub.n | 107 |
1 files changed, 71 insertions, 36 deletions
diff --git a/doc/regsub.n b/doc/regsub.n index 4e5e88d..5adfd61 100644 --- a/doc/regsub.n +++ b/doc/regsub.n @@ -13,40 +13,41 @@ .SH NAME regsub \- Perform substitutions based on regular expression pattern matching .SH SYNOPSIS -.VS 8.4 \fBregsub \fR?\fIswitches\fR? \fIexp string subSpec \fR?\fIvarName\fR? -.VE 8.4 .BE - .SH DESCRIPTION .PP This command matches the regular expression \fIexp\fR against \fIstring\fR, -.VS 8.4 and either copies \fIstring\fR to the variable whose name is given by \fIvarName\fR or returns \fIstring\fR if \fIvarName\fR is not present. -.VE 8.4 (Regular expression matching is described in the \fBre_syntax\fR reference page.) If there is a match, then while copying \fIstring\fR to \fIvarName\fR -.VS 8.4 (or to the result of this command if \fIvarName\fR is not present) -.VE 8.4 the portion of \fIstring\fR that matched \fIexp\fR is replaced with \fIsubSpec\fR. -If \fIsubSpec\fR contains a ``&'' or ``\e0'', then it is replaced -in the substitution with the portion of \fIstring\fR that -matched \fIexp\fR. -If \fIsubSpec\fR contains a ``\e\fIn\fR'', where \fIn\fR is a digit +If \fIsubSpec\fR contains a +.QW & +or +.QW \e0 , +then it is replaced in the substitution with the portion of +\fIstring\fR that matched \fIexp\fR. +If \fIsubSpec\fR contains a +.QW \e\fIn\fR , +where \fIn\fR is a digit between 1 and 9, then it is replaced in the substitution with -the portion of \fIstring\fR that matched the \fIn\fR-th +the portion of \fIstring\fR that matched the \fIn\fR'th parenthesized subexpression of \fIexp\fR. Additional backslashes may be used in \fIsubSpec\fR to prevent special -interpretation of ``&'' or ``\e0'' or ``\e\fIn\fR'' or -backslash. +interpretation of +.QW & , +.QW \e0 , +.QW \e\fIn\fR +and backslashes. The use of backslashes in \fIsubSpec\fR tends to interact badly -with the Tcl parser's use of backslashes, so it's generally +with the Tcl parser's use of backslashes, so it is generally safest to enclose \fIsubSpec\fR in braces if it includes backslashes. .LP @@ -59,7 +60,10 @@ All ranges in \fIstring\fR that match \fIexp\fR are found and substitution is performed for each of these ranges. Without this switch only the first matching range is found and substituted. -If \fB\-all\fR is specified, then ``&'' and ``\e\fIn\fR'' +If \fB\-all\fR is specified, then +.QW & +and +.QW \e\fIn\fR sequences are handled for each substitution using the information from the corresponding match. .TP 15 @@ -70,21 +74,37 @@ the \fB(?x)\fR embedded option (see the \fBre_syntax\fR manual page). .TP 15 \fB\-line\fR Enables newline-sensitive matching. By default, newline is a -completely ordinary character with no special meaning. With this -flag, `[^' bracket expressions and `.' never match newline, `^' +completely ordinary character with no special meaning. With this flag, +.QW [^ +bracket expressions and +.QW . +never match newline, +.QW ^ matches an empty string after any newline in addition to its normal -function, and `$' matches an empty string before any newline in +function, and +.QW $ +matches an empty string before any newline in addition to its normal function. This flag is equivalent to specifying both \fB\-linestop\fR and \fB\-lineanchor\fR, or the \fB(?n)\fR embedded option (see the \fBre_syntax\fR manual page). .TP 15 \fB\-linestop\fR -Changes the behavior of `[^' bracket expressions and `.' so that they +Changes the behavior of +.QW [^ +bracket expressions and +.QW . +so that they stop at newlines. This is the same as specifying the \fB(?p)\fR embedded option (see the \fBre_syntax\fR manual page). .TP 15 \fB\-lineanchor\fR -Changes the behavior of `^' and `$' (the ``anchors'') so they match the +Changes the behavior of +.QW ^ +and +.QW $ +(the +.QW anchors ) +so they match the beginning and end of a line respectively. This is the same as specifying the \fB(?w)\fR embedded option (see the \fBre_syntax\fR manual page). @@ -96,8 +116,14 @@ by \fIsubSpec\fR use the original unconverted form of \fIstring\fR. .TP 10 \fB\-start\fR \fIindex\fR Specifies a character index offset into the string to start -matching the regular expression at. When using this switch, `^' -will not match the beginning of the line, and \\A will still +matching the regular expression at. +.VS 8.5 +The \fIindex\fR value is interpreted in the same manner +as the \fIindex\fR argument to \fBstring index\fR. +.VE 8.5 +When using this switch, +.QW ^ +will not match the beginning of the line, and \eA will still match the start of the string at \fIindex\fR. \fIindex\fR will be constrained to the bounds of the input string. .TP 10 @@ -105,42 +131,51 @@ match the start of the string at \fIindex\fR. Marks the end of switches. The argument following this one will be treated as \fIexp\fR even if it starts with a \fB\-\fR. .PP -.VS 8.4 If \fIvarName\fR is supplied, the command returns a count of the number of matching ranges that were found and replaced, otherwise the string after replacement is returned. -.VE 8.4 See the manual entry for \fBregexp\fR for details on the interpretation of regular expressions. .SH EXAMPLES +.PP Replace (in the string in variable \fIstring\fR) every instance of \fBfoo\fR which is a word by itself with \fBbar\fR: .CS -\fBregsub\fR -all {\e<foo\e>} $string bar string +\fBregsub\fR -all {\emfoo\eM} $string bar string +.CE +or (using the +.QW "basic regular expression" +syntax): +.CS +\fBregsub\fR -all {(?b)\e<foo\e>} $string bar string .CE .PP Insert double-quotes around the first instance of the word -\fBinteresting\fR, however it is capitalised. +\fBinteresting\fR, however it is capitalized. .CS -\fBregsub\fR -nocase {\e<interesting\e>} $string {"&"} string +\fBregsub\fR -nocase {\eyinteresting\ey} $string {"&"} string .CE .PP Convert all non-ASCII and Tcl-significant characters into \eu escape sequences by using \fBregsub\fR and \fBsubst\fR in combination: .CS -# This RE is just a character class for everything "bad" -set RE {[][{}\e$\es\eu0100-\euffff]} +# This RE is just a character class for almost everything "bad" +set RE {[][{};#\e\e\e$ \er\et\eu0080-\euffff]} # We will substitute with a fragment of Tcl script in brackets set substitution {[format \e\e\e\eu%04x [scan "\e\e&" %c]]} # Now we apply the substitution to get a subst-string that -# will perform the computational parts of the conversion. -set quoted [subst [\fBregsub\fR -all $RE $string $substitution]] +# will perform the computational parts of the conversion. Note +# that newline is handled specially through \fBstring map\fR since +# backslash-newline is a special sequence. +set quoted [subst [string map {\en {\e\eu000a}} \e + [\fBregsub\fR -all $RE $string $substitution]]] .CE - .SH "SEE ALSO" -regexp(n), re_syntax(n), subst(n) - +regexp(n), re_syntax(n), subst(n), +.VS 8.5 +string(n) +.VE .SH KEYWORDS -match, pattern, regular expression, substitute +match, pattern, quoting, regular expression, substitute |