According to this page you can use perl compatible regular expression with
:perldo s/pattern/insert/g.
This works fine.
But, how can I get interactive search and replace with PCRE syntax in vim?
Since this does not work with perldo I search for a different solution.
Till the current release version of vim, there is no way to do :s/[perlRegex]/bar/c
So you are asking for a feature that doesn't exist.
You can do matching with verymagic, however it is not Perl Regex compatible flag. It is still using the vimregex engine, just changed the way of escaping regex patterns.
For example, in perl, you can do lookahead/behind (?<=foo)..., (?=foo), (?!foo).., you can use the handy \K : som.*ing\Kwhatever etc, you cannot use those syntax in vim directly, no matter which 'magic' level you have set. Vim has the same feature, but different syntax:
\#=
\#!
\#<=
and also the \zs \ze are very handy, even more powerful than perl's \K.
Vim is an Editor, with vim regex, you can not only do text matching, but also match base on visual selection, cursor position and so on.
If you really need to do complex pattern matching and really need do them in vim, learn vim regex! It is not difficult for you if you "know pcre very well"
Probably the closest you can get is:
:s/\vfoo/bar/gc
I suggest to try eregex plugin. It maps perl compatible regular expressions to VIM regex syntax.
In your example, s/pattern/insert/g is a perl command, not a Vim command using a perl-compatible regular expression syntax.
If perl doesn't have an equivalent of Vim's /c flag you will need to find an alternate method like… writing an actual perl script.
Related
I have a Base64 encoded string in a text file that is one line. In other words it contains no line breaks. I want to insert a line break every 78 characters.
None of the "wrap" extensions I have found do this, since they're geared for word wrapping and use word boundaries. Running any of these functions does nothing since the string contains no spaces or word boundaries.
I can do it on Unix using something like fold -w 78 but this seems like something that should exist in VS Code or at least an extension.
I'm not aware of an extension that does specifically what you're asking for, but what I would do is use the Edit with Shell Command extension, and use it to run fold -w 78 on the text in question from within VSCode. The extension even has a "quick command" feature you can use to save that command for quick use if it is something you do often.
I use that extension fairly often for one-off transformations with things like sort, sed, tr, and fmt. It's really handy when you know how to express the desired transformation as a shell command.
Hello I would like to master VIM as my primary IDE. I know there are many plugins etc. but I have one question: Is possible for VIM to understand the particular language (in case I wont plugin for my language) in which code is written? I mean some rules that I can define and then use e.g. for auto-completion, refactoring, folding etc. For example consider following two perl codes in which I want refactor variables. In first example there are variables with same names but one is scalar and another are array and hash, in second example same name of variable as was defined before was used in another scope (loop etc.). Thus refactoring using simple text matching is not elegant way I thing:
1st example:
my #same_name;
my $same_name; # How to refactor this variable without affecting all other variables?
my %same_name;
$same_name[0] = 5;
$same_name{"key"} = 5;
$same_name = 5;
2nd example:
my #array;
my $already_existing_variable; # How to refactor this variable
foreach my $already_existing_variable (#array){
print $already_existing_variable; # Without affecting this variable
}
print $already_existing_variable; # Variable that should be also refactorized
Or how can I achieve that when I type $arr {and hit TAB here} it will automatically extend to $array[ ? For this VIM must to know that when I prepend $ before variable which was declared with # I want to access array elements.
Another example would be: how to fold code using e.g. BEGIN and END keywords? Those was jut trivial examples but I think you get the point. I think it is similar task to write own compiler or something. Thank you.
I'm using vim with my perl scripts almost all days:
Rename variables
App::EditorTools gives you the function to rename variables like Padre.
install App::EditorTool $ cpanm App::EditorTools
install vim plugin $ editortools install-vim
move cursor on the variable name in vim.
type \pL
I'm not sure why it parses wrong scope in the 2nd example, but you can temporarily wrap the foreach brock with lambda and rename variables inside the code block first.
sub {
foreach my $already_existing_variable (#array){
print $already_existing_variable; # Without affecting this variable
}
}->();
Reformat script indent
Perl::Tidy has perltidy command line tool to format your script.
install Perl::Tidy $ cpanm Perl::Tidy
create ~/.perltidyrc according to your taste. like folowing:
-pbp
-pt=2
-sbt=2
-bt=2
-bbt=2
-ce
-nbbc
set equalprg to perltidy in your ~/.vimrc
au FileType perl setl ep=perltidy
type gg=G in your script.
Syntax Check
vim's built-in compiler plugin for perl does very well.
set :compilerto perl in your ~/.vimrc
au FileType perl :compiler perl
type :make in normal mode
type :copen to open up quickfix window
if you don't want warnings in your quickfix list, you can unset the flag for it.
let g:perl_compiler_force_warnings = 0
$PERL5LIB is also important while invoking :make, you can give specific directories in it.
let &l:path = './lib,./blib/lib,./blib/arch,' . &l:path
let $PERL5LIB = substitute(&l:path, ',', ':', 'g')
Completions
vim has nice completion tool out of the box. see :help comple-generic
type $ar in insert mode and press CTRL-N
You might be interested in ctags too.
I can already tell something is wrong when the first sentence is "Hello I would like to master VIM as my primary IDE."
Vim is not an IDE. Vim is a text editor. (You can use google to find out more)
If you feel that the features an IDE provides (refactoring, smarter auto-completion...) are more important than the features that vim provides (fast movement, never take your hands off the home row, programmable editor...) then you should use an IDE, or an IDE with a vim plugin. Usually, if you want to mix IDE features with vim features, you do it with a plugin to the IDE, not the other way around (there are some exceptions such as eclim).
In Perl, you call modules using :: as path separator. So, if you have a module on site/lib/GD/Image.pm, you call use GD::Image.
However, long time ago I found out that you can also call use GD'Image and things like my $img = new GD'Image;, and there are also modules on CPAN using that syntax on ther names/documentation.
What is the purpose or logic behind that? Is it maybe, as many things in Perl, just a feature intended to humanize sentences and allow you to create and use modules like Acme::Don't?
Does it have any other intention different to ::?
See perlmod for explanation:
The old package delimiter was a single quote, but double colon is now the preferred delimiter
So, the reason is history.
The single quote is an old ADA separator. However, it didn't play well with Emacs, so the double colon became used.
Good God! ADA? Emacs? I am old.
Is there a way to use perl-style regular expressions in Emacs? I use regexp-builder and query-replace-regexp pretty often but the Emacs-style regexes throw me off sometimes.
After a bit more searching I found a few similar questions:
Is it possible to change emacs' regexp syntax?
Elisp mechanism for converting PCRE regexps to emacs regexps
It looks like the solutions is to write an elisp function to convert PCRE's to Emacs-style regexes. This code on GitHub looks like a pretty good option.
In Komodo Edit, the tool interactively informs you of syntax errors, many of them are just a typo. For example, let's say that I have typed (in PERL):
use strict;
my ($name);
$name1="Schuller";
In this case, there should be a red underline under $name1, as we have used strict, and $name1 was undefined.
Thanks!
UltraEdit doesn't do true syntax checking, unfortunately.
It just does syntax highlighting based on a regex search of the file using rules defined in the word files (perl.uew for perl).
So, it will highlight certain keywords in a syntax specific manner, but won't tell you if are writing bad code.