Python Script editor.replace Bug? concerning the characters '(' and ')'

Robert Jablko

I was a little fast, operators work fine, but replace ‘(’ while working now, padds for every ‘(’. The original script hat functions that took the text, worked character by character and returned the final string, but doesn’t work with script. I have to find another way.

Mark Olson

You could replace your entire script with a sequence of regex-replaces (documentation here), which could then be saved as a macro and bound to a keyboard shortcut.

Regex-replace 1 (remove space after and before operators):

Open the find/replace form (Edit->Replace... from the Notepad++ main menu, or Ctrl+H with default keybindings)
Set the following options:
- Search mode set to Regular expressions
- Wrap around checked.
- Find what: \h*([+-*<>:=]=?|!=|//?|%|\*\*)\h*
- Replace with: ${1}
Hit Replace all

Regex-replace 2 (add exactly one space before and after any of {}[]())
The same as the previous steps, except in step 2,
Find what: (?<!\s)[(){}\[\]](?!\s)
Replace with: \x20${0}\x20 (note that \x20 is a way of rendering the literal space character so that it’s visible)

I haven’t tested either of these regexes, but I’m pretty sure they should do what you want.

By the way, your script’s runtime will be proportional to the square of the length of the text (i.e., will be extremely slow if you run it on a long file), because it does a global operation (editor.replace) for every character in the text. Whenever you’re writing a script in any language, always ask yourself if there’s a way to do a series of local modifications rather than a global modification. And remember: strings are immutable, so you can’t perform local modifications to a string in Python. That’s a lot of the reason why regular expressions are so handy: they avoid having to do potentially very slow and clunky changes to strings.

Robert Jablko

@Mark-Olson

Thank you :) I will try that.

But before, I solved my problem by creating a string, clear the editor and then add the string. It works. Don’t know if it’s fast. I am new to programming.

text=editor.getText()
################################################################################
''' compress operators, removes trailing and following space from operatios '''
operators=[ '+', '-', '*', ':', '/' '//', '%', '**', ]
assigners=[ '=', '+=', '-=', '*=', ':=', ]               
comparators=[ '==', '!=', '<', '>', '<=', '>=', ]                 
for character in operators+assigners+comparators:
    if ' '+character+' ' in text:
        editor.replace( ' '+character+' ', character )
################################################################################
''' padd brackets, adds space to brackets right of open, left of close '''
brackets_open='([{'
brackets_close=')]}'
result=''
for index, character in enumerate( text ): 
    
    ''' searching edge cases '''
    if index==0: # first character
        trailing_character=" "
        if len( text )>1:
            following_character=text[ index+1 ]
        else:
            following_character=" "
    elif index==len( text )-1:  # last character
        trailing_character=text[ index-1 ] 
        following_character=" "
    else:
        trailing_character=text[ index-1 ]
        following_character=text[ index+1 ]
    
    ''' replacing '''
    if character in brackets_open and following_character!=' ':
        if character=='(':
            result+='\('+' ' #needs escape '\('
needs escape '\('
        else:
            result+=character+' '
    elif character in brackets_close and trailing_character!=' ':
        if character=='\)':  
            result+=' '+'\(' #needs escape '\)'
        else:
            result+=' '+character
    else:
        result+=character
################################################################################
editor.clearAll( )
editor.addText( result )
################################################################################

Ekopalypse

@Robert-Jablko

Are you sure your script is doing what you expect it to do?

Let me recap what you do

you retrieve the current text content
you replace the operators_assigners_comparators in the current text
you iterate the retrieved text (retrieved text != current text) to create a new content and replace the current text.

Furthermore, if you use the replace method, you do not need to escape the (; this is only necessary if you use rereplace (note, rereplace, this is the regular expression replace form)

Instead of deleting and adding text, you can use setText in one go.

General advice:

There is no need to build strings like result+='\('+' '
use result+='\( '

use a script layout like

def main():
    here your code

main()

This gives you the option to return from a function, as the standard exit functions do what they should do, they end the process, in this case this would be notepad++.
Eg.

def main():
    if x == y:
        # oops something went wrong exit the script
        return

main()

vs.

def main():
    if x == y:
        # oops NOW NPP CRAHSES
        exit()

main()

Mark Olson

@Robert-Jablko
Your script’s runtime will still be proportional to the square of the length of the file, because the x += y operator (when applied to strings) creates a copy of x with y at the end.

This is actually not true. I tested the runtime of creating a string by +='ing characters one at a time, and it appears to be linear.

Since this is a Notepad++ forum, not a Python forum, I would encourage those interested to look at this StackOverflow post. I probably shouldn’t have even posted this in the first place, but I felt duty-bound because I originally believed that the performance implications of not saying something were too great.

Robert Jablko

@Ekopalypse

Thank you for the advice. My code perfectly explained.

I realize that I replace the brackets with the editor method, and then create a new string. I have changed that to replace within the string and script the editor method. That way I don’t need any escape characters.
Talking about escape characters. I really do not understand the difference between replace and rereplace, but what I do know, is that when I used the replace method in the first iteration, ‘(’ was gone and that was the whole reason to seek help. Alan Kilborn told me that I need the escape character ‘(’ and it worked with replace. Now that I use addText, e.g. SetText (Thank you), it’s actually not an issue anymore.
result+=‘(’+’ ’ is actually on purpose, I find it clearer and reusable, I have made it a point to add the variable pad, although in this case it doesn’t make a difference.
Layout issues. In my first try I used my original script with definitions, and for some reason, it would not run. I don’t know why and I thought it’s an issue with Python Script, so I made a code without functions. Thanks to you, I have changed that back to my original definitions. Here is the new code. What do you think?

def compress_operators( text, pad_operator=' ' ):
    print( "text in compress:", text , )
    
    ''' removes trailing and following pad from operators '''
    operators=[ '+', '-', '*', ':', '/' '//', '%', '**', ]
    assigners=[ '=', '+=', '-=', '*=', ':=', ]               
    comparators=[ '==', '!=', '<', '>', '<=', '>=', ]                 

    for character in operators+assigners+comparators:
        if pad_operator+character+pad_operator in text: # padded both sides '4 = 5' -> '4=5'
            text=text.replace( pad_operator+character+pad_operator, character )
        elif pad_operator+character in text: # padded left '4 =5' -> '4=5'
            text=text.replace( pad_operator+character, character ) 
        elif character+pad_operator in text: # padded right '4= 5' -> '4=5'
            text=text.replace( character+pad_operator, character )
            
    return text
################################################################################
def padd_brackets( text, pad_bracket=' ' ):
    ''' padds brackets, adds pad to brackets right of open, left of close 
        pad needs to be one character in this code, length_pad=1 '''
    print( "text in brackets:", text , )
    brackets_open='([{'
    brackets_close=')]}'
    
    result=''
    for index, character in enumerate( text ): 
        
        ''' searching edge cases '''
        if index==0: # first character
            trailing_character=pad_bracket
            if len( text )>1:
                following_character=text[ index+1 ]
            else:
                following_character=pad_bracket
        elif index==len( text )-1:  # last character
            trailing_character=text[ index-1 ] 
            following_character=pad_bracket
        else: # all characters in between
            trailing_character=text[ index-1 ]
            following_character=text[ index+1 ]
        
        ''' padding the brackets '''
        if character in brackets_open and following_character!=pad_bracket:
            result+=character+pad_bracket  # pad open bracket '(x' -> '( x'
        elif character in brackets_close and trailing_character!=pad_bracket:
            result+=pad_bracket+character # pad closing bracket 'x)' -> 'x )'
        else:
            result+=character
    
    return result
################################################################################    
def MyPyFormat( text, pad_operator=' ', pad_bracket=' ' ):
    ''' converts standard python code into my invididual format '''
    
    ''' 1st remove trailing and following pad from operators '''
    text_co=compress_operators( text, pad_operator=' ' )
    
    ''' 2nd padd brackets, add pad to brackets right of open, left of close 
        pad needs to be one character in this code, length_pad=1 '''
    text_pb=padd_brackets( text_co, pad_bracket=' ' )
    
    return text_pb
################################################################################
editor.setText( MyPyFormat( editor.getText( ), pad_operator=' ', pad_bracket=' ' ) )

Robert Jablko

@Mark-Olson

Thank you for testing. How do you test?

Why not posting it here? The Only problem is see, is that the link doesn’t work. :)

Mark Olson

@Robert-Jablko

Discussing how to performance test Python applications is outside the scope of this discussion, so I’ll keep it brief: I use time.perf_counter to measure time elapsed and matplotlib.pyplot to plot lines. Please don’t ask for further details, so as to avoid derailing the discussion.

On a more NPP-related note, your script is looking better now. I still think you would benefit from learning regular expressions, because the regex-based approach is much simpler and doesn’t require plugins.

Ekopalypse

@Robert-Jablko said in Python Script editor.replace Bug? concerning the characters '(' and ')':

I really do not understand the difference between replace and rereplace

With regular expressions, there are certain characters such as ( that have a meaning for the regex syntax. So to find them literally, you have to escape them.

I do know, is that when I used the replace method in the first iteration, ‘(’ was gone

I would be very surprised if it was a bug in PS, but if the problem occurs again, you should report it here so it can be investigated.

result+=‘(’+’ ’ is actually on purpose

Yes, what is important is what is important to you here.

What do you think?

It looks much nicer to me ;-)

Robert Jablko

@Mark-Olson

As I said earlier concerning regex, I will try it :) and I did. I found a horrible unintuitive syntax that does miracle things. I changed my whole code using regex and it works like a charm in python.

Unfortunately it throws a SyntaxError here in the Python script. Can anybody help?

SyntaxError: invalid syntax
  File "C:\npp\plugins\PythonScript\scripts\PimpMyPy\MyPyFormat npp Re 01.py", line 20
    text=re.sub( rf'({characters})(\s)', r'\1', text ) # ( operator, space )

Here is the full code I used:

import re 
###############################################################################
def compress_operators( text ):
    ''' removes spaces before and after operators and returns text '''
    print( text )

    # lists of operators, assigners and comparators
    operators=[ '\+', '-', ':', '\*', '\*\*', '/', '//', '%', ] # Mathematische Operatoren
    assigners=[ '=' ]+[ operator+'=' for operator in operators ]
    comparators=[ '==', '!=', '<', '>', '<=', '>=' ]
    bitwise_comparators=[ '&', '\|', '^', '~', '<<', '>>' ]
    
    # combine all operators in one search-string for RegEx
    characters='' 
    for character in operators+assigners+comparators+bitwise_comparators:
        characters+=character+'|'
    characters=characters[ :-1 ] # remove last '|'

    # substitute with RegEx
    text=re.sub( rf'({characters})(\s)', r'\1', text ) # ( operator, space )
    text=re.sub( rf'(\s)({characters})', r'\2', text ) # ( space, operator )
    
    return text
###############################################################################
def padd_brackets( text ):
    ''' add space after opening and before closing bracket, if not present '''
    
    # list of brackets
    opening_brackets=[ '(', '[', '{', ] 
    closing_brackets=[ ')', ']', '}', ]
    
    # create search_string depending on type of bracket and substitute with RegEx
    for bracket in opening_brackets+closing_brackets: 
        if bracket in opening_brackets:
            search=rf'([\{bracket}])(\S)' # ( bracket, any character )
            text=re.sub( search, r'\1 \2', text )  
        elif bracket in closing_brackets:
            search=rf'(\S)([\{bracket}])' # ( any character, bracket )
            text=re.sub( search, r'\1 \2', text ) 
   
    return text 
################################################################################
def my_py_format( text ):
    ''' creates individual format of python code
        compressed operators and padded brackets  '''
    
    text=compress_operators( text )
    text=padd_brackets( text )
    
    return text 
###############################################################################
if __name__=='__main__':
    editor.setText( my_py_format( editor.getText( ) )

Mark Olson

@Robert-Jablko

First of all, if you’re using regular expressions, you don’t need to use PythonScript; you can just use the Notepad++ find/replace form. That was the whole point of my suggestion.

As far as writing a regular expression to do what you want, I already wrote one above (EDIT: except for Regex-replace 1, I would change the Find what: to \h*([+-*<>:=]=?|!=|//?|[&|%^]|<<|>>|\*\*)\h*). I don’t have the time right now to explain syntax, but you clearly understand how to read documentation, so I’ll just let you figure out what my regexes do by consulting the handy-dandy user manual.

EDIT2: the more Pythonic (and much faster) way to render

blah = ''
for x in some_list:
     blah += x + sep # sep is a separator string (in your case '|')
blah = blah[:-len(sep)]

would be

blah = sep.join(some_list)

Ekopalypse

@Robert-Jablko said in Python Script editor.replace Bug? concerning the characters '(' and ')':

Unfortunately it throws a SyntaxError here in the Python script. Can anybody help?

It sounds like you are using a different Python version.
Could it be that your PythonScript (check the about menu item)

9b462b88-c8d7-4836-a193-91879c6dd744-{94D23D6D-E401-4545-8D60-1D3570F5140A}.png

is using Python 2.7, whereas your local python installation is Python 3.X.

If so, install the PS 3 version from the Github repository.
Note: Only recommended if you only work with utf-8 encoded files.

5c05f5ad-665a-4f73-bb69-9b7566fcc69a-{19AD1BD1-9FFE-499B-9C67-30B3DC672B08}.png

One additional thing about using regular expressions.
If you are using a Python script, you can either use the Python re module or the boost::regex library, which is also used by notepad++. You are currently using the Python module in your code. So please note that the syntax you find here in this community does not always work with the Python re module.
If you use the editor functions editor.rereplace and editor.research, use the Boost version.