Vi 101 - Searching For Hope (
Page 7 of 9 )
Vi also comes with
powerful search and replace features, both based largely on regular expressions.
Unfortunately, regular expressions are not part of this course of study; they
will be covered in "Reg-ex 101", and so I'm going to restrict myself to an
explanation of vi's string search functions, which can be accessed with the /
and ? commands.
Let's say that you've written a treatise on the role
played by Mark Hamil in "Star Wars: A New Hope" - except that, when
proof-reading your golden prose, you find that you've mistakenly spelt the word
"hope" as "dope" in different places in your document. Obviously, you can't
ignore this error - it would lead to a completely new interpretation of the
planet's favourite sci-fi flick. What do you do?
/dope
The /pattern command tells vi to search the document for the
specified pattern; once it finds it, it will position the cursor at the first
character of the string, and await further commands.
To search backwards,
use the ?pattern command, like this:
?dope
It's unlikely - not to mention naïve on your part - to assume
that this error would only have occurred once. You can repeat the last search by
pressing
n
or, in vi lingo, "next match".
It isn't enough to just
identify your mistake - the next step, according to most major religious
doctrines and television preachers, is to repent and make sure it doesn't happen
again. Vi can help you there too - although it will involve memorizing a command
which, at first glance, is almost enough to make you wonder if there really is a
God...
:1,$s/pattern_to_find/pattern_to_replace/g
In case you're wondering where the "s" came from - it's
vi-talk for "substitute". The 1 and the $ are symbols indicating the range
within which the substitution is to take place - in this case, from the
beginning of the document [line 1] to the end [$]. The "g" is a flag indicating
that *all* matches should be replaced.
And so, in the above example, the
command to replace the word "dope" with "hope" would be
:1,$s/dope/hope/g
Now, if only getting Leia's phone number was this easy...
This article
copyright 2000.
All rights reserved.{mospagebreak title=Bookmarks, Buffers and Beach
Bunnies}
Vi also allows you to define "bookmarks" within your document -
this makes it possible to easily jump to and between specific sections of large
files. As an example, consider the following blocks of text:
...this
document outlines the proposed corporate structure for Muscle Cars, Inc., a
company engaged in the business of manufacturing and selling...[page
2]
...the CEO and President of Muscle Cars, Inc., James van Hausen, who,
in his misbegotten childhood, was best-known for wrapping expensive sports cars
around trees, has succeeded in...[page 137]
So let's set a couple of
bookmarks, shall we? On page 2, move your cursor to the word "document", and
type
ma
This sets a bookmark, identified by the letter "a", at the
cursor position. Now find your way to page 137, and set a bookmark named "b" at
the word "cars". To switch back to page 2, all you now need to do is type
`a
and to get to page 137, simply type
`b
Any lowercase letter can be used as a bookmark identifier.
Obviously, you're limited to a total of 26 bookmarks per file - but in most
cases, that's more than enough. If you need more than that, I strongly suggest
you try therapy.
In addition to multiple bookmarks, vi also supports
multiple buffers. If you've been paying attention, you know that all the
material you delete or copy finds its way to a temporary buffer until it's
either replaced or removed. But vi also has "named buffers" - similar to a set
of storage lockers, these buffers can be used to store different blocks of text,
and make them available for insertion whenever required.
As an example,
take a look at this section of an essay handed in to me by one of my former
students:
...beach bunnies seem to have a pretty good life of it. All
they seem to do is lounge around on the beach, drinking pina coladas and
slathering suntan lotion all over themselves. Once in a while, they get up and
toss a volleyball around. Sounds like fun - where do I sign up?...
I'm
sure you understand why I said "former student" now...
Now let's suppose
that I wanted to transpose the phrases "drinking pina coladas" and "slathering
suntan lotion". Here's what I would do:
1. Move my cursor to the beginning of
the word "drinking"
2. Delete the three words, and move them to a buffer
named "p"
"p3dw
3. Move my cursor to the beginning of the word
"slathering"
4. Delete those three words, and move them to a buffer named
"s"
"s3dw
5. Move my cursor back to before the word "and"
6.
Insert ["put"] the contents of buffer "s"
"sp
7. Move my cursor to after the word "and"
8. Insert
the contents of buffer "p"
"pp
9. Congratulate myself loudly and exuberantly, as I'm doing
right now!
As you can see, to use a buffer, simply precede the regular
copy/delete/replace command with the buffer identifier. A buffer identifier
consists of a double-quote mark, followed by a lowercase letter. In the example
above, both "s and "p are buffer identifiers.
I should add here that
neither bookmarks nor buffer contents are retained once you exit the editor.
This article
copyright 2000.
All rights reserved.
