How do I substitute selected contents despite any regex characters in vim? - regex

In following code:
int return_int_func() { return 0; }
float fv = return_int_func();
Obviously, compiler will warn me fv may lost precisions because of auto-casting. Face lots of those things, I want replace all stuffs with substitute command. In short, I want this:
float fv = static_cast<float>(return_int_func());
But real codes has lots of forms like that:
float fv = obj.int_field;
float fv = obj->load_int_field("name");
float fv = xx.yy->zz;
I want select my target (obj.int_field,obj->load_int_field("name") or xx.yy->zz) and replace it with static_cast<float>(\1). I tried this:
:'<,'>s/\%V/static_cast<float>(&)/g
But vim replaces all characters in selected word with static_cast... and that isn't what I want at all. Then I tried this:
:'<,'>s/\(\%V\)/static_cast<float>(\1)/g
Vim also do the same thing. I have no idea how to replace whole content (and despite any regex characters) with my pattern. Any suggestions?

The solution is almost too easy! Here it is.
:s/\%V.*\%V./static_cast<float>(&)/
This is actually almost the same as the example from the :help. We can take away from this that we should all just have looked up :h /\%V first thing in the morning ...
\%V is a zero-width atom that matches stuff that is selected in Visual mode. Here it can match at the start of the Visual area. .* then matches (greedily) as much as it can; its greediness is reined in by the final \%V., which requires the last character of the match also to lie within the Visual area.
Tip: If you need to make this change many times over many lines, define the following mappings (even better: put them in your vimrc permanently).
nnoremap & :&&<CR>
xnoremap & :&&<CR>
Then you can repeat the substitution shown above by simply selecting something, and then pressing & to perform the substitution.

Let me try to paraphrase your question: You would like to Visual select some text, and then perform a substitution, where the selected text is also part of the replacement text.
I think in this case a macro is a much better tool.
To create the macro, first select the first piece of text that you want to wrap in the static cast. For example, select return_int_func(). (For each step, I'll show what the buffer looks like.)
When you're ready, press qq to start recording into register q, then press c.
float fv = |;
Type the left part of your wrapper text, static_cast<float>(.
float fv = static_cast<float>(|;
Press CTRL-R " (Control-R followed by "): this will reinsert the original text.
float fv = static_cast<float>(return_int_func()|;
Type ) to complete the change, and then Escape to leave insert mode.
float fv = static_cast<float>(return_int_func()▉;
Finally, press q to stop recording.
At this point you have made the first change and also recorded it as a macro in register q.
For all remaining changes, simply select a target such as obj->load_int_field("name") and press #q to repeat the change.
Look up :help 10.1 for more information about macros.

The \%V facility is really not for acting on the selected text as a whole; more for searching inside of that text.
Assuming that you are going to be putting this into a function or maybe mapping this to a key combination, here is an alternative approach that does what you are looking for:
:exec 'normal! gv"adistatic_cast<float>('|exec 'normal! "apa)'
Note that this will use your a buffer, so if you want to use another buffer you can change the two instances of "a with "x, where x is the buffer you wish to use.
Basically this is going to programmatically yank the selected text, insert static_cast<float>(, paste the text that was selected, and then insert ).

You can try
:%s/float fv =(.*)$/float fv = static_cast<float>(\1)/g

if I understood you right, you want to do text substitution only on selected text. This is not so easy to do, at least no so easy as a :s command can do. Because, your visual selection can be in single line, can cross multi lines, also it could char-wise, line-wise, block-wise..
but it can be done with this function:
function! SubVisualText(pat, repl,flag)
try
let v_save = #v
normal! gv"vy
let s = #v
let s = substitute(s, a:pat, a:repl,a:flag)
call setreg('v',s,visualmode())
normal! gv"vp
finally
let #v = v_save
endtry
endfunction
you use it by:
source the function
visual select the area (could be done by v, V or Ctrl-V)
:<ctrl-u>call SubVisualText(pattern, replacement, flag)<Enter>
the <ctrl-u> is for removing the leading range, since the function doesn't need the range.
when you run it, it looks like:(I just tested with Ctrl-V selection)

Related

Create custom commands for vim

I would like to create two commands for vim:
1 -
example code:
var a = "My string to be replaced";
So imagine that the cursor is on top of any letter inside the ". I would use the command ds" and it would do the equivalent of
T"
dt"
i
so it would delete everything inside the string and enter in insert mode automatically. If it is a function it would use the character used to call the command inside it, so dt" would delete everything inside " and dt'would delete everything inside ''.
2 -
example code:
class SomeClassNameHere {
so now imagine the cursor is somewhere on top of 'Name' and I would like to use a command that would put the cursor on the first uppercase character on the left of the current position and after that delete everything till the first uppercase character on the right (non including the character itself) and enter insert mode. This command is similar to the first one but it will look for an uppercase letter which probably is means the commands will need to use some regular expression like [A-Z]. After the command the code would look like that:
class SomeClassHere {
I would like those be some kind of map that I can add to my .vimrc file.
The first operation already exists as a text-object, it's i" and you are supposed to use it like this [operator][text-object]:
ci"
There's also i', i(, i{, etc. See :help text-objects.
The second operation could be described like this:
?[A-Z]<CR>
c/[A-Z]<CR>
and mapped like this:
nnoremap <key> ?[A-Z]<CR>c/[A-Z]<CR>
It would be even better to turn that in a proper text-object:
xnoremap iu ?[A-Z]<CR>o/[A-Z]<CR>h
onoremap iu :normal viu<CR>
that you could use like this:
ciu
diu
yiu
viu
Here is an improved version of romainl's answer:
xnoremap iu ?\u<cr>o/.\ze\([^a-z_]\<bar>$\)<cr>:<c-u>noh<cr>gv
onoremap iu :normal viu<cr>
It adds the ability to select or delete the last uppercase word, which was not possible, and hides the highlighting when done.

Any Characters Replace Increment Number

Is it possible when columns not align?
var[x];
var[x];
var[x];
var[x];
..
replace
var[0];
var[1];
var[2];
var[3];
Sezer, here is an easy way to do it in Notepad++
In the Replace tab, make sure you check the "Regular Expression" box at the bottom left
Search: (^\w+\[(\d+)\]=[^;]+;\w+\[)[^]]+
Replace: \1\2 or $1$2 (depending on your locale, you may have to choose one or the other.)
Hit "Replace All"
Note that this will work for two variables on the same line, regardless of their name, for instance firstvar[10]="asfa";somevar[x]="asdfas";
If you want to replace only when the first variable is called kelimeler, change the search to:
(^kelimeler\[(\d+)\]=[^;]+;\w+\[)[^]]+
If you want to replace only when the first variable is called kelimeler and the second variable is called specialvar, use:
(^kelimeler\[(\d+)\]=[^;]+;specialvar\[)[^]]+
You could just use a simple script, for example in JavaScript:
var x=10;
for (var i=0;i<x;i++) {
console.log('var['+i+'];');
}
with var x being the amount of rows you want to have
Then simply copy the logged output from your console.
Using such repetitive code does look awfully inefficient though. Almost every respectable programming/scripting language has some kind of incremental looping built in already. What is it you're trying to accomplish with this code? There is probably an easier way to do this.

Comment/Uncomment in Vim using Regex

I'm trying to use :'<,'>s!^!// ! to comment out highlighted code in visual mode. Is it possible to edit the regex so that it uncomments the code if it is already commented?
Some of the other answers mention plugins. I currently use NERD commenter, which provides a ci (comment invert) command that does what you want.
For starters, here are a couple of ways to avoid commenting out lines that are already commented. For simplicity, I will assume that the comment characters are all in the first column. Either one accepts a range, such as the Visual range '<,'> in your question.
:s#^\(// \)\#!#// #
:v#^// #s!^!// !
OK, if you really want a single command that toggles whether a line is commented out, then you have to capture whether the line is initially commented and then replace with an expression:
:%s!^\(// \|\)!\=strpart('// ', strlen(submatch(1)))
:help /\(
:help /\#!
:help :g
:help sub-replace-expression
A dirty but very versatile trick I use is to exploit the :normal command, that will emulate keystroke sequence you give on the given range.
For instance, to comment, select your whole lines in visual mode (using V) and enter the command :
:normal i//
This is quiet self-explanatory : i open edit mode, and // writes it (because you entered edit mode); the command will be applied to every selected line, with the cursor positioned to the beginning of the line (because we selected whole lines with V and not v; it gets tricky otherwise). With the same idea, you can remove the commenting on a visual selection using :
:normal xx
(This does exactly what it says : deletes the first two characters, as you would type on your keyboard).
There is several pros for this approach, including :
If you use several different languages with different comment signs ({ , /*, //, #, %) , it's just a matter of changing a single strokes.
You can really get creative with the normal command, since you can exploit marks for the command's range, and literally do anything :
:normal A //Hello world
Again, it feels very natural because you are simply feeding a command with the keystroke you would have used on your selection. You can even switch between modes using this trick.
well, I doubt you can do it using only one regex, and that's why Tim Pope has written an extension exactly for that:
https://github.com/tpope/vim-commentary
then all you have to do is gc to comment a line or a visual selection, and gc to uncomment it!
There are several plugins that work to comment and uncomment code across multiple languages. That being said, you would probably want to use two separate regexes in this case. For example, say I have some code...
var x = 5;
// Does stuff 5 times
for ( var i = 0; i < x; i++ ) {
// stuff
}
When I run said regex on this I don't want the already commented lines to be uncommented. So it's better to do something similar, but in reverse (e.g. :'<,'>s!^// !!).
Also, the \ze and \zs operators can be useful here to preserve whitespace. And :* is a shortcut for :'<,'>. For instance I could do
:*s!^\s*\zs// !!
This will remove only the characters // from lines that start with them or with whitespace before them.

Vim: Invert string (by words)

This is my string:
"this is my sentence"
I would like to have this output:
"sentence my is this"
I would like to select a few words on a line (in a buffer) and reverse it word by word.
Can anyone help me?
It's not totally clear what the context is here: you could be talking about text in a line in a buffer or about a string stored in a VimScript variable.
note: Different interpretations of the question led to various approaches and solutions.
There are some "old updates" that start about halfway through that have been rendered more or less obsolete by a plugin mentioned just above that section. I've left them in because they may provide useful info for some people.
full line replacement
So to store the text from the current line in the current buffer in a vimscript variable, you do
let words = getline('.')
And then to reverse their order, you just do
let words = join(reverse(split(words)))
If you want to replace the current line with the reversed words, you do
call setline('.', words)
You can do it in one somewhat inscrutable line with
call setline('.', join(reverse(split(getline('.')))))
or even define a command that does that with
command! ReverseLine call setline('.', join(reverse(split(getline('.')))))
partial-line (character-wise) selections
As explained down in the "old updates" section, running general commands on a character- or block-wise visual selection — the former being what the OP wants to do here — can be pretty complicated. Ex commands like :substitute will be run on entire lines even if only part of the line is selected using a character-wise visual select (as initiated with an unshifted v).
I realized after the OP commented below that reversing the words in a partial-line character-wise selection can be accomplished fairly easily with
:s/\%V.*\%V./\=join(reverse(split(submatch(0))))/
wherein
\%V within the RE matches some part of the visual selection. Apparently this does not extend after the last character in the selection: leaving out the final . will exclude the last selected character.
\= at the beginning of the replacement indicates that it is to be evaluated as a vimscript expression, with some differences.
submatch(0) returns the entire match. This works a bit like perl's $&, $1, etc., except that it is only available when evaluating the replacement. I think this means that it can only be used in a :substitute command or in a call to substitute()
So if you want to do a substitution on a single-line selection, this will work quite well. You can even pipe the selection through a system command using ...\=system(submatch(0)).
multiple-line character-wise selections
This seems to also work on a multiple-line character-wise selection, but you have to be careful to delete the range (the '<,'> that vim puts at the beginning of a command when coming from visual mode). You want to run the command on just the line where your visual selection starts. You'll also have to use \_.* instead of .* in order to match across newlines.
block-wise selections
For block-wise selections, I don't think there's a reasonably convenient way to manipulate them. I have written a plugin that can be used to make these sorts of edits less painful, by providing a way to run arbitrary Ex commands on any visual selection as though it were the entire buffer contents.
It is available at https://github.com/intuited/visdo. Currently there's no packaging, and it is not yet available on vim.org, but you can just git clone it and copy the contents (minus the README file) into your vimdir.
If you use vim-addon-manager, just clone visdo in your vim-addons directory and you'll subsequently be able to ActivateAddons visdo. I've put in a request to have it added to the VAM addons repository, so at some point you will be able to dispense with the cloning and just do ActivateAddons visdo.
The plugin adds a :VisDo command that is meant to be prefixed to another command (similarly to the way that :tab or :silent work). Running a command with VisDo prepended will cause that command to run on a buffer containing only the current contents of the visual selection. After the command completes, the buffer's contents are pasted into the original buffer's visual selection, and the temp buffer is deleted.
So to complete the OP's goal with VisDo, you would do (with the words to be reversed selected, and with the above-defined ReverseLine command available):
:'<,'>VisDo ReverseLine
old updates
...previous updates follow ... warning: verbose, somewhat obselete, and mostly unnecessary...
The OP's edit makes it more clear that the goal here is to be able to reverse the words contained in a visual selection, and specifically a character-wise visual selection.
This is decidedly not a simple task. The fact that vim does not make this sort of thing easy really confused me when I first started using it. I guess this is because its roots are still very much in the line-oriented editing functionality of ed and its predecessors and descendants. For example, the substitute command :'<,'>s/.../.../ will always work on entire lines even if you are in character-wise or block-wise (ctrlv) visual selection mode.
Vimscript does make it possible to find the column number of any 'mark', including the beginning of the visual selection ('<) and the end of the visual selection ('>). That is, as far as I can tell, the limit of its support. There is no direct way to get the contents of the visual selection, and there is no way to replace the visual selection with something else. Of course, you can do both of those things using normal-mode commands (y and p), but this clobbers registers and is kind of messy. But then you can save the initial registers and then restore them after the paste...
So basically you have to go to sort of extreme lengths to do with parts of lines what can easily done with entire lines. I suspect that the best way to do this is to write a command that copies the visual selection into a new buffer, runs some other command on it, and then replaces the original buffer's visual selection with the results, deleting the temp buffer. This approach should theoretically work for both character-wise and block-wise selections, as well as for the already-supported linewise selections. However, I haven't done that yet.
This 40-line code chunk declares a command ReverseCharwiseVisualWords which can be called from visual mode. It will only work if the character-wise visual selection is entirely on a single line. It works by getting the entire line containing the visual selection (using getline()) running a parameterized transformation function (ReverseWords) on the selected part of it, and pasting the whole partly-transformed line back. In retrospect, I think it's probably worth going the y/p route for anything more featureful.
" Return 1-based column numbers for the start and end of the visual selection.
function! GetVisualCols()
return [getpos("'<")[2], getpos("'>")[2]]
endfunction
" Convert a 0-based string index to an RE atom using 1-based column index
" :help /\%c
function! ColAtom(index)
return '\%' . string(a:index + 1) . 'c'
endfunction
" Replace the substring of a:str from a:start to a:end (inclusive)
" with a:repl
function! StrReplace(str, start, end, repl)
let regexp = ColAtom(a:start) . '.*' . ColAtom(a:end + 1)
return substitute(a:str, regexp, a:repl, '')
endfunction
" Replace the character-wise visual selection
" with the result of running a:Transform on it.
" Only works if the visual selection is on a single line.
function! TransformCharwiseVisual(Transform)
let [startCol, endCol] = GetVisualCols()
" Column numbers are 1-based; string indexes are 0-based
let [startIndex, endIndex] = [startCol - 1, endCol - 1]
let line = getline("'<")
let visualSelection = line[startIndex : endIndex]
let transformed = a:Transform(visualSelection)
let transformed_line = StrReplace(line, startIndex, endIndex, transformed)
call setline("'<", transformed_line)
endfunction
function! ReverseWords(words)
return join(reverse(split(a:words)))
endfunction
" Use -range to allow range to be passed
" as by default for commands initiated from visual mode,
" then ignore it.
command! -range ReverseCharwiseVisualWords
\ call TransformCharwiseVisual(function('ReverseWords'))
update 2
It turns out that doing things with y and p is a lot simpler, so I thought I'd post that too. Caveat: I didn't test this all too thoroughly, so there may be edge cases.
This function replaces TransformCharwiseVisual (and some of its dependencies) in the previous code block. It should theoretically work for block-wise selections too — it's up to the passed Transform function to do appropriate things with line delimiters.
function! TransformYankPasteVisual(Transform)
let original_unnamed = [getreg('"'), getregtype('"')]
try
" Reactivate and yank the current visual selection.
normal gvy
let #" = a:Transform(#")
normal gvp
finally
call call(function('setreg'), ['"'] + original_unnamed)
endtry
endfunction
So then you can just add a second command declaration
command! -range ReverseVisualWords
\ call TransformYankPasteVisual(function('ReverseWords'))
tangentially related gory detail
Note that the utility of a higher-level function like the ones used here is somewhat limited by the fact that there is no (easy or established) way to declare an inline function or block of code in vimscript. This wouldn't be such a limitation if the language weren't meant to be used interactively. You could write a function which substitutes its string argument into a dictionary function declaration and returns the function. However, dictionary functions cannot be called using the normal invocation syntax and have to be passed to call call(dictfunct, args, {}).
note: A more recent update, given above, obsoletes the above code. See the various sections preceding old updates for a cleaner way to do this.
Maybe:
:s/\v(.*) (.*) (.*) (.*)/\4 \3 \2 \1/
Of course you probably need to be more specific in the first part to find that particular sentence. Generally you can refer to match groups as \number with \0 being the whole match.
Here's a way to do by calling out to Ruby. After selecting the line you want to reverse, you can do this in command mode to replace it:
!ruby -e 'puts ARGF.read.strip.split(/\b/).reverse.join'
I found the solution myself thank to your answers and a lot of trying :)
This works:
function! Test()
exe 'normal ' . 'gv"ay'
let r = join(reverse(split(getreg('a'))))
let #a = r
exe 'normal ' . 'gv"ap'
endfunction
Didn't thought that I was enable to write such a function :)

Use cases for regular expression find/replace

I recently discussed editors with a co-worker. He uses one of the less popular editors and I use another (I won't say which ones since it's not relevant and I want to avoid an editor flame war). I was saying that I didn't like his editor as much because it doesn't let you do find/replace with regular expressions.
He said he's never wanted to do that, which was surprising since it's something I find myself doing all the time. However, off the top of my head I wasn't able to come up with more than one or two examples. Can anyone here offer some examples of times when they've found regex find/replace useful in their editor? Here's what I've been able to come up with since then as examples of things that I've actually had to do:
Strip the beginning of a line off of every line in a file that looks like:
Line 25634 :
Line 632157 :
Taking a few dozen files with a standard header which is slightly different for each file and stripping the first 19 lines from all of them all at once.
Piping the result of a MySQL select statement into a text file, then removing all of the formatting junk and reformatting it as a Python dictionary for use in a simple script.
In a CSV file with no escaped commas, replace the first character of the 8th column of each row with a capital A.
Given a bunch of GDB stack traces with lines like
#3 0x080a6d61 in _mvl_set_req_done (req=0x82624a4, result=27158) at ../../mvl/src/mvl_serv.c:850
strip out everything from each line except the function names.
Does anyone else have any real-life examples? The next time this comes up, I'd like to be more prepared to list good examples of why this feature is useful.
Just last week, I used regex find/replace to convert a CSV file to an XML file.
Simple enough to do really, just chop up each field (luckily it didn't have any escaped commas) and push it back out with the appropriate tags in place of the commas.
Regex make it easy to replace whole words using word boundaries.
(\b\w+\b)
So you can replace unwanted words in your file without disturbing words like Scunthorpe
Yesterday I took a create table statement I made for an Oracle table and converted the fields to setString() method calls using JDBC and PreparedStatements. The table's field names were mapped to my class properties, so regex search and replace was the perfect fit.
Create Table text:
...
field_1 VARCHAR2(100) NULL,
field_2 VARCHAR2(10) NULL,
field_3 NUMBER(8) NULL,
field_4 VARCHAR2(100) NULL,
....
My Regex Search:
/([a-z_])+ .*?,?/
My Replacement:
pstmt.setString(1, \1);
The result:
...
pstmt.setString(1, field_1);
pstmt.setString(1, field_2);
pstmt.setString(1, field_3);
pstmt.setString(1, field_4);
....
I then went through and manually set the position int for each call and changed the method to setInt() (and others) where necessary, but that worked handy for me. I actually used it three or four times for similar field to method call conversions.
I like to use regexps to reformat lists of items like this:
int item1
double item2
to
public void item1(int item1){
}
public void item2(double item2){
}
This can be a big time saver.
I use it all the time when someone sends me a list of patient visit numbers in a column (say 100-200) and I need them in a '0000000444','000000004445' format. works wonders for me!
I also use it to pull out email addresses in an email. I send out group emails often and all the bounced returns come back in one email. So, I regex to pull them all out and then drop them into a string var to remove from the database.
I even wrote a little dialog prog to apply regex to my clipboard. It grabs the contents applies the regex and then loads it back into the clipboard.
One thing I use it for in web development all the time is stripping some text of its HTML tags. This might need to be done to sanitize user input for security, or for displaying a preview of a news article. For example, if you have an article with lots of HTML tags for formatting, you can't just do LEFT(article_text,100) + '...' (plus a "read more" link) and render that on a page at the risk of breaking the page by splitting apart an HTML tag.
Also, I've had to strip img tags in database records that link to images that no longer exist. And let's not forget web form validation. If you want to make a user has entered a correct email address (syntactically speaking) into a web form this is about the only way of checking it thoroughly.
I've just pasted a long character sequence into a string literal, and now I want to break it up into a concatenation of shorter string literals so it doesn't wrap. I also want it to be readable, so I want to break only after spaces. I select the whole string (minus the quotation marks) and do an in-selection-only replace-all with this regex:
/.{20,60} /
...and this replacement:
/$0"¶ + "/
...where the pilcrow is an actual newline, and the number of spaces varies from one incident to the next. Result:
String s = "I recently discussed editors with a co-worker. He uses one "
+ "of the less popular editors and I use another (I won't say "
+ "which ones since it's not relevant and I want to avoid an "
+ "editor flame war). I was saying that I didn't like his "
+ "editor as much because it doesn't let you do find/replace "
+ "with regular expressions.";
The first thing I do with any editor is try to figure out it's Regex oddities. I use it all the time. Nothing really crazy, but it's handy when you've got to copy/paste stuff between different types of text - SQL <-> PHP is the one I do most often - and you don't want to fart around making the same change 500 times.
Regex is very handy any time I am trying to replace a value that spans multiple lines. Or when I want to replace a value with something that contains a line break.
I also like that you can match things in a regular expression and not replace the full match using the $# syntax to output the portion of the match you want to maintain.
I agree with you on points 3, 4, and 5 but not necessarily points 1 and 2.
In some cases 1 and 2 are easier to achieve using a anonymous keyboard macro.
By this I mean doing the following:
Position the cursor on the first line
Start a keyboard macro recording
Modify the first line
Position the cursor on the next line
Stop record.
Now all that is needed to modify the next line is to repeat the macro.
I could live with out support for regex but could not live without anonymous keyboard macros.