TEACH VEDSEARCH Jonathan Meyer Oct 1992
If you are an experienced Ved user wanting to learn about the new Ved
search mechanism, skip to the section 'Notes for experienced Ved users'.
If you are an experienced VI or ED user, you might want to read the
section called 'Notes for VI, ED, GREP and AWK users' first.
CONTENTS - (Use <ENTER> g to access required sections)
-- Introduction
-- Searching forward: <ENTER> /
-- Non-embedded forward searching: <ENTER> "
-- Backward search <ENTER> \
-- Non-embedded backward search <ENTER> `
-- Empty search string: re-use last search string
-- Repeat forward search: <ESC> /
-- Repeat backward search: <ESC> \
-- Search pattern elements
-- Examples
-- Search options
-- Searching a region of the text
-- XVed users
-- s">The combined "search and substitute" command: <ENTER> s
-- Substitute pattern elements
-- Notes on string delimiters to be ignored on a first reading
-- Substitution options
-- Global substitutions: the '-ask' option
-- Substitution using regions
-- Shortcuts
-- Creating your own search or substitute commands
-- Notes for experienced Ved users
-- Notes for VI, ED, GREP and AWK users
-- See Also-- Introduction -------------------------------------------------------
Ved allows you to search forwards or backwards for a specified string.
You can either allow embedded strings to be found (e.g. 'the' in the
word 'their') or constrain the string to match only complete text items,
(e.g. prevent 'the' being found in 'their'). (See the section below on
the definition of "embedded".)
You can use special "pattern elements" in search strings to indicate
things like the beginning or end of a line, or an arbitrary character,
or control characters.
There is also a search-and-substitute command, for replacing one or more
occurrences of one string with another string. This can either be
interactive or work for the whole file, or just the current line or the
whole of a marked range (see HELP * MARK).
-- Searching forward: <ENTER> / ---------------------------------------
Suppose you want to move the cursor to a line containing the word
"define". You can do this by first going into command mode (press ENTER)
then type a "slash" i.e. "/" (not the "backslash" "\" higher up the
keyboard), then type the "search string" then press RETURN, e.g. the
command
<ENTER> /define
will search for "define". If you wish to find an occurrence followed by
some number of spaces, add the spaces and then another "/", e.g.
<ENTER> /define /
will find an occurrence of "define" followed by two spaces.
NOTE - searches "wrap around" to the top of the file. I.e. if the
required string is not found by searching down from the current
location, Ved will go to the top of the file and continue searching for
the first occurrence from there.
-- Non-embedded forward searching: <ENTER> " --------------------------
If you do not want to find an occurrence of the word embedded in a
larger word, use the command
<ENTER> "define
or, if you want trailing spaces type them followed by another ".
If the double quote character (") is used like this, Ved will not find a
line containing 'enddefine'. Embedded occurrences are ignored: strings
must begin and end on text item boundaries, though a string can straddle
more than one text item.
-- Backward search <ENTER> \ ------------------------------------------
<ENTER> \<string>
Will search backward for the last occurrence of <string> preceding he
current cursor location. NOTE - If the string is not found between the
current location and the top of the file it will wrap around and search
from the bottom.
The search string can be terminated with a second occurrence of "\".
This is essential for a string that ends with a space.
-- Non-embedded backward search <ENTER> ` -----------------------------
<ENTER> `<string>
This is exactly the same as <ENTER> \, except that it will not find
embedded occurrences of the string. I.e. it will find occurrences that
begin and end on text item boundaries, like <ENTER> ". <string> should
be terminated by "`" if it ends with a space.
-- Empty search string: re-use last search string ---------------------
If ever you give an empty string with any of the search or substitute
commands (described below), Ved will reuse the search string that you
supplied on the previous search.
-- Repeat forward search: <ESC> / --------------------------------------
The sequence <ESC> / can be used to repeat the last search. It will use
the last search string to search forward like <ENTER> / and <ENTER> ".
-- Repeat backward search: <ESC> \ ------------------------------------
This is like <ESC> / except that it searches backwards for the last
search string. On a long file, backward search using patterns (described
below) can be much slower than search without any pattern elements in
the string.
-- Search pattern elements --------------------------------------------
To match things like control characters, Ved graphics characters,
or Ved special space characters, you can use all `backslash' sequences
that are recognised in Pop-11 strings. These are described in `Backslash
in Strings & Character Constants' in REF *ITEMISE. For example,
<ENTER> /\t
finds a `tab' character, whereas
<ENTER> /\St
would search for a `trailing space' character.
Note that if you specify character attributes (using \[...] or \{...} ),
these will be ignored in the search string. (However, they are not
ignored in substitute strings, see 'Substitution pattern elements'
below.)
The search string can also contain special 'regular expression' search
symbols, which control the search used in the above <ENTER> commands.
The following table summarises the search options that are available.
TEACH *REGEXP gives more examples of regular expressions. REF *REGEXP
gives full details of the Poplog regular expression matcher.
Pattern Meaning
------- -------
@. or @? match any single character
@* match zero or more occurrences of the previous character
@^ or @a constrains a match to the start of the line
@$ or @z constrains a match to the end of the line
@< constrains match to the start of a word boundary
@> constrains match to the end of a word boundary
@{N@} matches N occurrences of previous character, where N is
a positive integer.
@{N,@} matches N or more occurrences of the previous character
@{N,M@} match at least N, at most M occurrences of previous
character, where N and M are positive integers
@[...@] match any character between the brackets, where ... is
any characters e.g. abc or range of characters e.g. A-Z
@[^...@] match any character except those between the brackets
@(...@) denotes a sub-expression for use with @N
@N match a repeat of the N'th sub-expression where N is
a number 1-9
@C match C where C is the search delimiter or `@'
@i turns off case sensitivity for subsequent letters
@c turns on case sensitivity for subsequent letters
Examples using each pattern are given below.
@^ or @a -- anchor search at beginning of line.
To specify that the word must occur at the beginning of a line, use
the symbol "@^" before the word. E.g.
<ENTER> /@^define
finds lines that begin with 'define'.
You can also use @a, which is provided for backward compatibility.
@$ or @z -- anchor search at end of line
To search for "define" at the end of the line, use the symbol "@$"
after the word:
<ENTER> /define@$
Note that "@$@^" will match a line break.
You can also use @z, which is provided for backward compatibility.
@. or @? -- match any character
To search for strings containing some specified and some arbitrary
characters use @. to indicate each arbitrary character. E.g.
<ENTER> /@.at
will find all the following: cat hat mat tat rat ...
It will also find embedded forms, e.g. catch that smattering ...
This can be suppressed by using <ENTER> "
You can also use @?, which is provided for backward compatibility.
@* -- match zero or more occurrences of the previous pattern
To search for a character repeated zero or more times (up to the end
of a line) use the @* pattern. E.g.
<ENTER> /ap@*
will match 'a', 'ap', 'appp', 'appppp', etc.
More usefully, to search for any character repeated any number of
times, use @.@* E.g.
<ENTER> /define @.@*;
will find 'define ', followed by any number of characters, followed
by ';',
Note that you can constrain the start and end of the match to word
boundaries using <ENTER> "
@[ and @] -- match range of characters
To search for any one of a set of characters, place them between
@[ and @]. E.g.
<ENTER> /@[ABC@]nd
will find And, Bnd or Cnd. You can also use the '-' sign to indicate
a range of characters, E.g.
<ENTER> /@[A-Z@]ile
will match Tile, File, Zile, or anything else that starts with an
upper case letter followed by 'ile'. You can use the same technique for
numbers:
<ENTER> /@[1-9@]
will find, 1, 2, 3, ..., 9.
Use @[ and @] with @* to specify zero or more occurrences of any of
the characters. E.g.
<ENTER> /repeat @[0-9@]@*
will match 'repeat ', 'repeat 1', repeat 59', 'repeat 214', or
repeat followed by any other integer.
If the first character after the @[ is ^, the pattern will match
anything except the characters between @[ and @]. E.g.
<ENTER> /@[^\s\t@]
finds the next character that is not a space or a tab character.
@< -- anchor search to the start of a word
Use @< to search for a string which is the start of a word. E.g.
<ENTER> /@<mem
will find "member" but will not find "strmember" or "lmember".
@> -- anchor search to the end of a word
To search for a word that ends in a particular pattern, use @> E.g.
<ENTER> /or@>
will find "or", "endfor" but not "ordinal".
Note that:
<ENTER> /@<define@>
is equivalent to
<ENTER> "define
@( and @) -- specify a sub-expression
If you want to find a string that is repeated more than once, place
@( and @) around string and then use @1 to repeat it. This can
save typing. E.g.
<ENTER> /@(echo@)@1@1@1
will match four repetitions of 'echo'. You can use patterns inside
the @( and @), E.g.
<ENTER> /@(@.cho@) @1
will match 'echo echo' or ' Echo Echo'. Note that @1 repeats the
same string of characters as was previously matched by the
expression enclosed in @( @), so the above pattern will not find
'echo Echo' or 'Echo echo'.
There can be up to nine @( @) expressions in a search. Use @2 to
repeat the second expression, @3 to repeat the third, etc.
@{ and @} -- match a range of occurrences
To find a range of occurrences of a pattern, use @{ and @}. E.g.
<ENTER> /p@{1,3@}
will match one, two or three p's. If you don't specify the second
number, as many matches as possible are done, up to the end of the
line. E.g.
<ENTER> /p@{1,@}
finds as many consecutive p's as possible. @{ @} combines with other
search patterns as well. E.g.
<ENTER> /@[A-Z@]@{1,5@}
will find up to five upper case letters next to each other.
@i -- make the following characters case-insensitive
Use @i to find both uppercase and lowercase characters. E.g.
<ENTER>/@ifile
finds 'File', 'fIlE', 'FILE', 'file', etc.
@c -- makes the following characters case-sensitive
Just as @i turns off case sensitivity, @c turns it back on again. So
if you want to ignore the case of some of the letters in the middle
of a string, surround them with an @i and @c E.g.
<ENTER>/define @ifoo@c with_props
finds 'define Foo with_props', 'define fOo with_props', etc.
@@ -- match the single character "@"
@C -- where C is one of "/", "\", """, or "`" - use C
In order to search for a string containing the current search
string delimiter precede it with "@".
-- Examples -----------------------------------------------------------
To find a string containing "d" and "e" separated by any other character
use the symbol "@." to match an arbitrary character, e.g.
<ENTER> /d@.e
this will search for the first occurrence of any of "dae" "dbe", etc.
Similarly
<ENTER> /d@.@.@.ne
will find "define" or "dryane" or "debone".
<ENTER> /@/usr@/lib
will search forward for an occurrence of the unix file path name
/usr/lib.
<ENTER> /@@@@
will search for a string containing "@@".
If you give an empty search string, e.g. press ENTER and type the
command:
/
then Ved will simply use the same search string as in your last search
command (or substitute command - see below).
If the search finds the wrong occurrence you can use the REDO key (see
HELP *VEDKEYS). This is the "repeat last command" key. The search will
be repeated.
WARNING: If Ved cannot find the string by searching to the end of the
file it goes to the top of the file and looks for the first occurrence
it can find from there. This 'wrap-around' feature is useful but can be
confusing. The line number of the cursor is always displayed on the
command line to keep you informed.
You can search backwards using the '\' symbol, thus: press enter, then
type:
\define
to search backwards for an occurrence of 'define'. Backward search will
'wrap around' from the top to the bottom of the file if it cannot find
the string.
-- Search options -----------------------------------------------------
After the second delimiter of a search command you can specify
additional parameters to control the search.
For example, to find 'define', but without wrapping over the end of the
file, do:
<ENTER> /define/ -wrap
The -wrap turns off 'wrap-around' for the search - the searching stops
when it reaches the end of the file.
CAUTION: The space before the '-wrap' can be omitted, but this may lead
to confusion when doing backward searches with <ENTER> \ so its better
to include a space.
Just as the minus sign turns off an option, you can use the plus sign to
force an option to be on. E.g.
<ENTER> /define/ +back
will make the search go backwards even though you used <ENTER> /
You can specify a number after the second delimiter to make the search
find a specific occurrence of a string. E.g.
<ENTER> /define/ 3
finds the third occurrence of 'define' forward from the current position.
You could also use:
<ENTER> /define/ -3
to find the third occurrence of 'define' before the cursor (looking
backwards).
The following table summarises search options:
Option Description
--------------------------------------------------------------------
back +back forces the search to go backwards. -back forces
the search to go forwards.
case +case makes the search case sensitive. Use -case to ignore
the case of letters.
embed +embed matches embedded occurrences. -embed constrains the
search to non-embedded items.
here +here starts the search from the cursor. -here always
searches from the start of the region.
wrap +wrap enables 'wrap-around'. -wrap makes the search stop
at the end of the search region.
N an integer N finds the N'th occurrence of STRING. If N
is negative, the direction of the search is reversed.
See REF * VED_/ for full details of search options.
-- Searching a region of the text -------------------------------------
You can restrict a search to a specified 'region' of the file. You
do this by typing the name of the region after the last delimiter of the
search command and before any options. E.g.
<ENTER> /if/ range
will find 'if' within the marked range (see HELP *MARK). Other regions
you can use include "line", "procedure", and "word", E.g.
<ENTER> /e/ word
finds the letter 'e' in the word that your cursor is positioned over.
See REF * VED_/ for a list of all of the regions and options that are
available.
-- XVed users ---------------------------------------------------------
If you are using XVed, you can set XVed so that after a search the
matching text is highlighted. To do this, either do:
<ENTER> xved SearchDoesSelect true
or add the following line to your .Xdefaults:
XVed*SearchDoesSelect: true
When you see this highlighted text you have three options: If you hit
the DEL key the highlighted text will be deleted. If you start typing in
some characters, the highlighted text will be replaced by what you type.
If you move the cursor, the highlight is cleared and the text is left
intact.
s">-- The combined "search and substitute" command: <ENTER> s ------------
Suppose you wish to replace an occurrence of "deffine" with "define". You
can type ENTER, then:
s/deffine/define
and press RETURN.
Ved will search for an occurrence of "deffine" and wiggle the cursor to
catch your eye. You then have five options:
(a) press "y" to do the substitution and stop.
(b) press RETURN to do the substitution and go on to the next
occurrence of the search string.
(c) press DEL to leave this occurrence and then go on to the next
one
(d) press "n" to stop without doing the substitution.
(e) press "g" to do the substitution and continue to do it globally
throughout the rest of the region.
Option (b) is specially useful, since it allows you to go through the
file changing lots of occurrences of a string, checking each case.
(See also <ENTER> gs, below)
If you want to ensure that it does not find an embedded occurrence of
"deffine" use '"' instead of '/', i.e.
<ENTER> s"deffine"define
then <RETURN>. This will not replace 'enddeffine'.
-- Substitute pattern elements ----------------------------------------
Just as you can use special search pattern elements in the search
string, there are a few special symbols that you can occur in the
substitute string:
@n -- substitute in a line break
E.g. to put a line break before occurrences of "rabbit", do
<ENTER> s/rabbit/@nrabbit
(You can also use \n for the same purpose.)
@& -- substitute in the character string matched by the search.
If you want to repeat the search string in the replacement string
use "@&" to save retyping. E.g. to replace "rabbit" with "rabbits",
type the command
<ENTER> s/rabbit/@&s
@p -- pastes in the contents of the clipboard (XVed only)
If you are running XVed, you can use the '@p' pattern to substitute
in whatever is currently held on the X clipboard (The CLIPBOARD
selection, not the PRIMARY selection).
This allows you to select some text, copy it to the clipboard, and
then use the text as part of a substitution command.
@N -- where N is a digit 1-9, substitute in the N'th sub-expression
If you have placed an @( and @) around part of the search pattern,
you can substitute in the text that matched between these brackets
using @1, @2, @3, etc. E.g.
<ENTER> s/define @(foo@)/@1
Replaces 'define foo' with 'foo'. A more sophisticated example is:
<ENTER> s/;;; @(@.@*@)/ @/* @1 *@/
which finds ';;; ' followed by any text (to the end of the line)
and replaces it with /* <text> */, changing ';;;' style comments
into /* */ style comments.
As for the search string, the substitute string may also contain any
`backslash' sequence recognised in a Pop-11 string.
[You might want to skip the rest of this section on first reading.]
Unlike the search string, however, character attributes specified
with \[...] or \{...} (see REF *ITEMISE) are not ignored, i.e. they
are substituted in. For example:
s/HELP/\{b}HELP
Changes occurrences of 'HELP' to bold 'HELP'.
When you are using @&, @p, or @N (where N is a digit), you can place a
character attribute specification before the @ sign. Any attributes
given to the `@' character are added to those of the relevant text
before it is placed in the buffer.
You can specify other attribute combination modes other than adding by
using either +, -, ~ or = immediately after the @, e.g. @-& @~p or @=1.
They have the following meanings:
+ the attributes of the @ are added to the text
- the attributes of the @ are removed from the text
= the text is shown using the same attributes as the @
~ toggles the attributes set on the @.
In other words, @+& means 'the text that was matched by the search
including its character attributes' and @-& means 'the text that was
matched by the search excluding its character attributes'. E.g.
<ENTER> s/HELP/@-&
removes all character attributes from occurrences of 'HELP'.
So
<ENTER> s/rabbit/\{b}@&
changes 'rabbit' to bold 'rabbit' and underlined 'rabbit' to bold
and underlined 'rabbit'.
whereas
<ENTER> s/rabbit/\{b}@-&
replaces both 'rabbit' and underlined 'rabbit' with bold 'rabbit'.
This is very useful when using the @( and @) in conjunction with @N
to repeat the text. E.g.
<ENTER> s/;;; @(@.@*@)/;;; \{b}@1
finds ';;; ' followed by any number of characters up to the end of
the line, and makes all the characters following the ';;;' appear in
bold.
On the other hand:
<ENTER> s/;;; @(@.@*@)/;;; @-1
finds ';;; ' followed by any text up to the end of the line, and
substitutes it with ';;; ' followed by the same text, but with no
character attributes - undoing the changes made by the previous example.
-- Notes on string delimiters to be ignored on a first reading --------
As with search commands, you can use \ or ` instead of / or " to look
backwards for strings to substitute.
However you should avoid using \ as a delimiter in substitutions since
it is very hard to tell whether the \ is acting as a delimiter or the
start of a backslash string sequence. Instead you can use a / as the
delimiter and specify the "+back" option (see below).
Also, instead of one of the above delimiters, you can use any other
"string delimiter", which can be any non-alphanumeric character, e.g.
"." or "," or ";", as in
<ENTER> s.cat.cats
<ENTER> s can also be given an empty search string, meaning "use the
previous search string, e.g.
<ENTER> s//bunny
The two slashes, without a space indicate an empty search string, and
will cause the previous search string to be used.
If you type <ENTER> s without a search or substitute string then it will
use the last search string (whether used by S or by the search command)
and also the last substitute string. E.g.
<ENTER> s
will again look for 'rabbit' and replace it with 'bunny'.
-- Substitution options -----------------------------------------------
After the third delimiter of an <ENTER> s substitution command you can
specify additional parameters to control the search and substitution.
<ENTER> s accepts all of the options available to the search commands.
In addition, you can use the following options:
-ask
turns off 'asking' - substitutions are done without asking
your permission before each replace.
-every
turns off 'every' - instead of replacing every match on a line,
only the first (forward search) or last (backward search) match
on each line is replaced, and the rest of the line is ignored.
-verbose
turns off 'verbose' - when used with '-ask', the substitutions
are done without recording progress on the status line. This
can save time when there are hundreds of substitutions to
perform.
Note that with <ENTER> s if you specify a number after the second
delimiter, Ved does at most that many substitutions. For example:
<ENTER> s/rabbit/bunny/ 5
will replace the next five occurrences of rabbit with bunny, asking your
permission before each replace.
-- Global substitutions: the '-ask' option ----------------------------
You can use the '-ask' (or 'without asking') option to <ENTER> s to make
Ved replace ALL occurrences of a string globally in the file, without
asking your permission for each one. E.g. press ENTER then
s/rabbit/bunny/ -ask
then <RETURN> to replace all occurrences of 'rabbit' with 'bunny'. You
can use the symbol '"' instead of '/' to mean don't replace an embedded
substring. E.g. to replace "the", with "that", but not inside "there"
etc. do
<ENTER> s"the"that" -ask
When you do a global substitution, Ved will first copy the text as it
stands into -vveddump-. If you then realise that you have incorrectly
substituted some text, you can use <ENTER> yank to recover the lines of
the buffer as they were before the substitution.
If you don't want to copy the buffer before each global substitution
(e.g. if a file is very large), you can turn this feature off by
assigning -false- to vedsafesubstitute.
-- Substitution using regions -----------------------------------------
You can restrict a substitute to a specified 'region' of the file. You
do this by typing the name of the region after the last delimiter of the
substitute command and before any options. For example:
<ENTER> s/defien/define/ range
will change occurrences of 'defien' to 'define' in the marked range.
(see HELP *VEDMARKED).
Other regions that you can use are "line", "procedure" and "paragraph".
-- Shortcuts ----------------------------------------------------------
Options to the search and substitute commands can be abbreviated to
one or more letters, to save typing.
Thus, instead of using '-wrap', you could simply type '-w'. E.g.
<ENTER> /define/ -w
Its worth using the full names at first to help you remember the option
names.
You can also abbreviate region names to one or more letters. E.g.
<ENTER> s/deffine/define/ r -a
means substitute 'deffine' for 'define' in the marked range without
asking.
In cases where two region start with the same letter, you can type two
or more letters to disambiguate them.
Ved provides a set of pre-defined commonly used substitute commands
which are shorthands for typing <ENTER> s commands with certain options.
These all use the same syntax as the <ENTER> s commands (though they
ignore any options typed after the final delimiter). E.g.
<ENTER> gs/deffine/define
performs a 'global' substitution of 'deffine' for 'define', replacing
all occurrences without asking the user first. This is the same as
typing:
<ENTER> s/deffine/define/ -ask
Other shorthand commands are:
<ENTER> gsr -- global substitute range
substitutes all occurrences of a string within the marked range.
See HELP *MARK.
<ENTER> gsp -- global substitute procedure
substitutes all occurrences of a string within the current
procedure. See HELP *ved_mcp.
<ENTER> gsl -- global substitute line
substitutes all occurrences of a string on the current line.
<ENTER> sgs -- silent global substitute
like <ENTER> gs, except that substitutions are done silently.
<ENTER> sgsr -- silent global substitute range
like <ENTER> gsr, except that substitutions are done silently.
-- Creating your own search or substitute commands --------------------
All Ved search and substitute commands are in fact implemented as
closures of a single procedure, -ved_search_or_substitute-. You can
add your own customised search commands by creating closures of this
procedure.
For example, to define an <ENTER> fwr which finds a non-embedded word in the
marked range, you can do:
define global ved_fwr
= ved_search_or_substitute(%'range -embed', false%);
enddefine;
The first argument to ved_search_or_substitute is a string containing
options to use for the search. If this is -false-, then the command uses
options typed by the user after the final delimiter.
The second argument is -false- if the command is only a search, and
-true- if the command is a search and substitute.
You can use the newly defined <ENTER> fwr to look for the word "define"
in the marked range:
<ENTER> fwr/define
or to look backward for the word "if" in the marked range:
<ENTER> fwr\if
As a second example, define an <ENTER> bcs which performs a backwards
caseless search:
define global ved_bcs
= ved_search_or_substitute(%'+back -case', false%);
enddefine;
You can use it to look backwards for "file", "File", "fIlE", etc.:
<ENTER> bcs/file
Finally, to define a substitute command which globally substitutes
characters in the current word, use:
define global ved_gsw
= ved_search_or_substitute(%'word -ask', true%);
enddefine;
Now try it:
<ENTER>gsw/a/b
changes all 'a' characters to 'b' characters in the current word.
-- Notes for experienced Ved users ------------------------------------
If you are familiar with the Ved search facilities that existed in
Poplog 14 you should read this section to learn about the differences
between the old Ved search mechanism and the new mechanism...
You should find that all of the search/substitute commands that you are
familiar with continue to work in the same way, with the following
changes/additions:
1) Extended search patterns
There is a new and much larger set of `@' patterns that you can now use.
These include @*, @[ ... @] and others. The following table lists the
new @ patterns:
Pattern Meaning
------- -------
@* match zero or more occurrences of the previous character
@< constrains match to the start of a word boundary
@> constrains match to the end of a word boundary
@{N@} matches N occurrences of previous character, where N is
a positive integer.
@{N,@} matches N or more occurrences of the previous character
@{N,M@} match at least N, at most M occurrences of previous
character, where N and M are positive integers
@[...@] match any character between the brackets, where ... is
any characters e.g. abc or range of characters e.g. A-Z
@[^...@] match any character except those between the brackets
@(...@) denotes a sub-expression for use with @N
@N match a repeat of the N'th sub-expression where N is
a number 1-9
@i turns off case sensitivity for subsequent letters
@c turns on case sensitivity for subsequent letters
Some examples are:
<ENTER> /@[a-z@]
finds the next lower case letter.
<ENTER> /define @.@*
finds 'define ' followed by any number of characters to the end of the
line.
<ENTER> /@<for
finds words starting with 'for', e.g. 'for', 'foreach', 'forevery',
but not 'endfor'.
<ENTER> /the @ived@c editor
finds 'the Ved editor', 'the VeD editor', 'the Ved editor', etc.
For more details, use <ENTER> g to go to the section:
-- Search pattern elements
2) Extended search parameters
You can now provide extra search parameters after the second delimiter
of a search command. These parameters control attributes such as what
region of the buffer to search, which direction to search, whether the
search wraps around the end of the search region, and whether the search
should be case sensitive or case insensitive.
Some examples are:
<ENTER> /define/ range
looks for 'define' in the marked range.
<ENTER> /e/ word
looks for the letter 'e' in the current word.
<ENTER> "if" 3
looks for the third occurrence of "if".
For more details, use <ENTER> g to go to the section:
-- Search options
3) Backward searches
Note that backward searches now wrap around the top of the file and
continue searching at the end of the file. You can use the '-wrap'
option to turn this off.
4) Extended substitution patterns
There are a few more substitution patterns:
Pattern Meaning
------- -------
@p pastes in the contents of the clipboard (XVed only).
@N where N is 1-9, substitute the N'th sub-expression.
For more details and examples, use <ENTER> g to go to the section:
-- Substitute pattern elements
5) Extended substitution options
The <ENTER> s command can now also take extra parameters after the final
delimiter, and processes them in the same way as the search commands do.
There are a few additional options that are available only to <ENTER> s
commands.
Some examples are:
<ENTER> s/e/X/ word -ask
substitutes the letter 'e' for 'X' in the current word without asking
first.
<ENTER> s/'/"/ -every
Replaces only the first occurrence of ' on each line with ".
<ENTER> s/alice/jane/ 5
Replaces the next 5 occurrences of 'alice' with 'jane'.
For more details and examples, use <ENTER> g to go to the section:
-- Substitution options
6) 'Safe' substitution
When you do a global substitution, Ved will first copy the text as it
stands into -vveddump-. If you then realise that you have incorrectly
substituted some text, you can use <ENTER> yank to recover the lines of
the buffer as they were before the substitution.
For global substitutes in the marked range or in any other region, only
those lines that might be effected by the substitution are copied.
This copying can take time if the buffer is very large, and it is not
compatible with earlier versions of Ved. If you don't want to copy the
buffer before each global substitution you can turn this feature off by
doing:
false -> vedsafesubstitute;
-- Notes for VI, ED, GREP and AWK users -------------------------------
If you are familiar with the regular expression search patterns
found in VI or ED, or are a GREP or AWK, read this section to
learn about the differences between their search patterns and the
search patterns used in Ved.
Note that this topic is discussed in more detail in REF * REGEXP in the
section 'Poplog Regular Expressions'.
1) The Escape Character
Instead of using `\' as the character to specify search wildcards, Ved
uses the `@' character. This is because `\' is already used by the
Poplog string itemiser to represent control characters and other special
ASCII characters.
Thus, in Ved, instead of typing:
<ENTER> /\<int\>
you type:
<ENTER> /@<int@>
2) The patterns ^ $ . * [ and ]
In Ved, for backward compatibility with earlier search facilities, if
you want to use any of the special wildcards above, you should precede
them by the pattern escape character `@'. So instead of writing:
c.*[a-z]
You use:
c@.@*@[a-z@]
-- See Also -----------------------------------------------------------
REF * VEDSEARCH - Regular expression searching in Ved.
HELP * ASCII
HELP * VED - overview of Ved facilities and HELP files.
HELP * VED_G, *ENTER_G - Searching for section headings
REF * REGEXP - The regular expression matcher
REF * VEDCOMMS - Other ENTER commands
REF * VEDPROCS - Ved system procedures
REF * VEDVARS - Ved global variables
HELP * VED - Overview of other Ved facilities
TEACH *VEDEXPAND - Command line substitution mechanism
--- C.all/teach/vedsearch
--- Copyright University of Sussex 1995. All rights reserved.