How to see hex value of character next to cursor?

PeterJones

How do I find the hex value of a character to the right of my cursor?

Some time ago, in response to another question, @Scott-Sumner used the PythonScript plugin (*) to create a callback which will constantly show the current character’s character-code in hex and decimal.

Using the same script, but skipping the editor callback, would allow you to do it on-demand rather than “live”.

At one time, I had tried to start getting it to work with UTF-8 unicode characters, but I had moved on from there without remembering that I’d never finished it. Who knows, as I find time in the near future, I may go back to playing with that

(*: If you need help installing PythonScript, see @Meta-Chuh’s Guide)

PeterJones

@PeterJones said:

Using the same script, but skipping the editor callback, would allow you to do it on-demand rather than “live”.

Technically, you would have to replace the line

editor.callback(callback_sci_UPDATEUI, [SCINTILLANOTIFICATION.UPDATEUI])

with

callback_sci_UPDATEUI(None)

PeterJones

Okay, I’m addicted. I got it somewhat debugged. This will work up to U+FFFF as a one-shot. (Or comment out the “on-demand” line and uncomment the “editor.callback” line to get it “live”)

# encoding=utf-8

def callback_sci_UPDATEUI(args):
    c = editor.getCharAt(editor.getCurrentPos())
    if c < 1 or c > 255:
        p = editor.getCurrentPos()
        q = editor.positionAfter(p)
        s = editor.getTextRange(p,q).decode('utf-8')
        try:
            c = ord(s)
        except:
            txt = "'{}' = {} char: ".format(s.encode('utf-8'), len(s))
            for ch in s:
                c = ord(ch)
                txt = txt + " HEX:0x{0:04X} DEC:{0} '{1}'".format(c, unichr(c).encode('utf-8') if c not in [13, 10, 0] else 'LINE-ENDING' if c != 0 else 'END-OF-FILE')
            notepad.setStatusBar(STATUSBARSECTION.DOCTYPE, txt)
            return
    try:
        info = "HEX:0x{0:04X} DEC:{0} '{1}'".format(c, unichr(c).encode('utf-8') if c not in [13, 10, 0] else 'LINE-ENDING' if c != 0 else 'END-OF-FILE')
    except ValueError:
        info = "HEX:?? DEC:?"
    notepad.setStatusBar(STATUSBARSECTION.DOCTYPE, info)

callback_sci_UPDATEUI(None)     # per https://notepad-plus-plus.org/community/topic/17799/, want on-demand
# editor.callback(callback_sci_UPDATEUI, [SCINTILLANOTIFICATION.UPDATEUI]) # per https://notepad-plus-plus.org/community/topic/14767/, want live

PeterJones

Okay, I found the way to make it compatible with U+10000 and above:

# encoding=utf-8

def get_wide_ordinal(char):
    '''https://stackoverflow.com/a/7291240/5508606'''
    if len(char) != 2:
        return ord(char)
    return 0x10000 + (ord(char[0]) - 0xD800) * 0x400 + (ord(char[1]) - 0xDC00)

def callback_sci_UPDATEUI(args):
    c = editor.getCharAt(editor.getCurrentPos())
    if c < 1 or c > 255:
        p = editor.getCurrentPos()
        q = editor.positionAfter(p)
        s = editor.getTextRange(p,q).decode('utf-8')
        c = get_wide_ordinal(s)
    else:
        s = unichr(c)

    try:
        info = "'{1}' = HEX:0x{0:04X} = DEC:{0} ".format(c, s.encode('utf-8') if c not in [13, 10, 0] else 'LINE-ENDING' if c != 0 else 'END-OF-FILE')
    except ValueError:
        info = "HEX:?? DEC:?"
    notepad.setStatusBar(STATUSBARSECTION.DOCTYPE, info)

callback_sci_UPDATEUI(None)     # per https://notepad-plus-plus.org/community/topic/17799/, want on-demand
# editor.callback(callback_sci_UPDATEUI, [SCINTILLANOTIFICATION.UPDATEUI]) # per https://notepad-plus-plus.org/community/topic/14767/, want live

Alan Kilborn

Addicted. Definitely addicted. :)

guy038

Hello, @PeterJones, and All,

I’ve tried your script, on a text, containing Unicode characters over the BMP => Just perfect !

Just a question : In which case the END-OF-FILE string is displayed in the status bar ? I initially thought it could be when opening a new-1 empty file but not !

Cheers,

guy038

Ekopalypse

@guy038

there is no end-of-file (EOF) string/char.
EOF is more like a status like in reading a file.
If the filehandle comes to the end of a file it sets it status EOF to true
to inform that the end is reached.

Or did I misunderstand your question?

PeterJones

@guy038 said:

In which case the END-OF-FILE string is displayed in the status bar ?

In @Scott-Sumner’s original, END-OF-FILE would show if you were at the last character of the file, or if the file were empty. However, my changes to try to trap the unicode characters caused the EOF-result to give errors.

To fix that bug, change

if c < 1 or c > 255:

to

if c < 0 or c > 255:

PeterJones

@Ekopalypse said:

there is no end-of-file (EOF) string/char.

@guy038 was talking about the else 'END-OF-FILE' that should go to the status bar – ie, output of the program, not input from the file.

The editor.getCharAt() apparently returns 0 if the getCurrentPos() is at the end of the file. But because I introduced the bug of c<1 instead of c<0 after that line, the 0 condition in the info = ... was never hit, so my code didn’t recognized the end of the file. (It was actually printing the untrapped error to the PythonScript console, but I hadn’t seen it, because I never tried testing the end-of-file condition while making my unicode-capable version of the script.)

Ekopalypse

@PeterJones

ahh, I was looking to your code searching if there is this statement but missed that I have to scroll horizontally to see it, then I thought @guy038 was really talking about EOF status - I should have known better :-) - SORRY @guy038 and thank you Peter for clarifying.

guy038

Hello, @PeterJones, @ekopalypse, and All,

So, finally, using your modification :

    if c < 0 or c > 255:

and your line, slightly modified :

        info = "  {1}    0x{0:04X}  -  {0} ".format(c, s.encode('utf-8') if c not in [13, 10, 0] else 'LINE-END' if c != 0 else 'FILE-END')

Everything is OK ! And I’ve never met the error exception :-))

    except ValueError:
        info = "HEX:?? DEC:?"

For instance :

Open a new file
Running the script shows | FILE-END 0x0000 - 0', in the status bar
Add a € Euro character
With cursor right before the currency symbol, I get | € 0x20AC - 8364'
With cursor right after the currency symbol, I get, again, | FILE-END 0x0000 - 0'
Now, hit the Enter key
Move the caret right after the € => This time, it answers | LINE-END 0x000D - 13'
Hit the Down arrow key. Again, we get | FILE-END 0x0000 - 0'

An other example :

In a new file, I added two characters, which give the same resulting glyph é !

é   //  LATIN SMALL LETTER E WITH ACUTE
é  //  LATIN SMALL LETTER E + DIACRITICAL COMBINING ACUTE ACCENT

With cursor right before the first single é character, it displays | é 0x00E9 - 233'
With cursor right before the é group of two characters, it displays | e 0x0065 - 101'
Moving the cursor rightwards 1 position, “inside” the é group of two characters, it displays | ́ 0x0301 - 769', as expected !

Refer to the complete list of the Combining Diacritical marks, below, for information :

http://www.unicode.org/charts/PDF/U0300.pdf

BR

guy038

PeterJones

@Alan-Kilborn said:

Addicted. Definitely addicted. :)

I am admittedly addicted. But did you see that it helped me use Notepad++ to find a hidden character in someone else’s code? That right there justifies the time I put into it, in my opinion.

Alan Kilborn

I recently found it handy to modify Peter’s script above (the version that runs “on demand”).
Often when dealing with a “weird” character, I want to get rid of it and replace it with something else, sometimes a different character, and sometimes a string.
The following modified version of the script shows the character codes (as before) but then asks you if you want to replace all occurrences of the character, and if you do, prompts you for the replacement.

I called it CharacterCodeShowWithOptionalReplace.py :

# -*- coding: utf-8 -*-

from Npp import editor, notepad

class CCSWOR(object):

    def get_wide_ordinal(self, char):
        # see https://stackoverflow.com/a/7291240/5508606
        if len(char) != 2: return ord(char)
        return 0x10000 + (ord(char[0]) - 0xD800) * 0x400 + (ord(char[1]) - 0xDC00)

    def __init__(self):

        c = editor.getCharAt(editor.getCurrentPos())
        if c < 1 or c > 255:
            p = editor.getCurrentPos()
            q = editor.positionAfter(p)
            s = editor.getTextRange(p,q).decode('utf-8')
            c = self.get_wide_ordinal(s)
        else:
            s = unichr(c)

        try:
            info = "'{1}' = HEX:0x{0:04X} = DEC:{0} ".format(c, s.encode('utf-8') if c not in [13, 10, 0] else 'LINE-ENDING' if c != 0 else 'END-OF-FILE')
        except ValueError:
            info = "HEX:?? DEC:?"
        notepad.messageBox(info, 'Character at caret was...')

        matches = []
        editor.search(s, lambda m: matches.append(m.span(0)))
        user_input = notepad.prompt('Replace the character with something else (all {} occurrences)?\r\n'
            '(Leave box empty to DELETE the character)'.format(len(matches)), 'Replace it?', '')
        if user_input == None: return  # Cancel
        editor.replace(s, user_input)

if __name__ == '__main__': CCSWOR()

Maybe it will be useful to someone else.
Perfect usage: Replace “curly double-quotes” with straight ones. :-)

PeterJones

@Alan-Kilborn said in How to see hex value of character next to cursor?:

Perfect usage: Replace “curly double-quotes” with straight ones. :-)

I wouldn’t call that “perfect”, since [“”] => " does it perfectly well without scripting. :-) Assuming you noticed they were curly in the first place.

Despite that, I do think it’s a nice extension to my script. :-)

Alan Kilborn

@PeterJones

I guess the “curly” thing was a bit contrived… :-)

I’ll tell you my real use case of recent origin:
Getting rid of some invisible LTR marks in a bunch of Skype conversation logs–ugh.

Alan Kilborn

I posted an updated script for this topic, HERE.

Jeff Heath

Can someone clarify how to actually run this script? I’ve followed the installing PythonScript [Guide](link https://notepad-plus-plus.org/community/topic/17256/guide-how-to-install-the-pythonscript-plugin-on-notepad-7-6-3-7-6-4-and-above), but as I search around there are also references to installing the Python Script plug-in, but that’s not available in my version of npp (maybe because it is 64 bit?).

Anyway, if I followed that PythonScript Guide, and have the plugins\PythonScript folder set up, how do I then run the script? I’ve also installed the NppExec plug-in, but don’t immediately see how that could be used.

I seem to be missing something. Can anyone give me a hint?

Ekopalypse

@Jeff-Heath
Once you have installed the PythonScript plugin, you should
see an additional entry in the plugin menu.
In that menu, there is a menu item called Show Console.
Click on it, and if the console opens without reporting a problem, the installation of PS was successful.

To create a new script, use the menu again and select New Script,
But make sure the directory it points to is your ...plugins\Config\PythonScript\scripts\ directory.
Give it a meaningful name and copy the code into it.
Save it.
Now it should appear in the list of the Scripts menu item.
If you want it to be accessible in the PS main menu,
you have to use the PS menu option Configuration…
and add it to the main menu section. That’s it.
Now you can run it by clicking on it or assigning a shortcut, etc.

Jeff Heath

I finally noticed in the Plugins Admin that the Python Script plugin (version 1.3) that I installed manually (with the link above) was listed under the “Incompatible” tab. But I was able to install it on the “Updates” tab to version 2. Now I have an element “Python Script” on the Plugins menu which allows me to create a new script, etc. So I think I can now try to progress and try a few more things.

But why was “Python Script” not available in the Plugins Admin in the first place? All it has is “Python Indent”. I certainly don’t want to have to manually install 1.3 then do the update the next time.

Ekopalypse

@Jeff-Heath

That sounds strange, because normally you have it available.
I’ve never had a problem installing it, and I do it this way with every new npp version. I think I’ve been doing this since 7.9.5, which is how my automated tests run.
The only reason I can think of why it doesn’t show up in the Available tab is if npp thinks it has already found something.