Does the UDL miss the ablity to recognize methods and classes?

Andreas Wiese

Regarding syntax highlighting, I’m able to define the styles of operators, keywords and comments. What about identifiers for methods (method names) and classes (class names). They aren’t recognized so I’m not able to define their style. Don’t I just see a feature of the “User Defined Language”?

Andreas Wiese

Maybe I’ve not been precisely enough. To clarify what I mean, here’s a screenshot: [http://fs1.directupload.net/images/180808/82g2vbd7.png]

The highlighting algorithm recognizes that there is a class/type to the left of the point and a method to the right of the point and displays both elements in the corresponding color.

I’m looking for something similar for Notepad++.

Claudia Frank

@Andreas-Wiese

no, the current UDL version does not support it directly.
A hack to get the method colored might be to use the dot as an opening delimiter
and the open-round-bracket as a closing delimiter but the class is still open.

Another, possible, approach is discussed here.

Cheers
Claudia

Andreas Wiese

Thank you for your reply. Well, at least in Java classes are recognizable by their capital letter at the front, while identifiers (method names/variable names) usually start with a lower case letter. Does your proposed thread say anything about regular expressions and how to use them with Notepad++? I’m not sure. Maybe I should go over it once more.

Claudia Frank

@Andreas-Wiese

if you are asking whether udl supports regex then the answer is no, not currently
but if the question is if the python script could be used with regex then yes, it can and
it is used in this example

As you see, only the keywords true and null have been defined, as well as
two delimiters, one using ( and ) and the other for strings using ’ and ’

A script to accomplish this could look like this

from Npp import editor, notepad, NOTIFICATION, SCINTILLANOTIFICATION, INDICATORSTYLE

# to use this for a different language, three steps need to be done

# 1. replace the EnhancedUDLLexer string with a different but unique name,
# Note, checkout the string_needs_to_be_replace_here marks
try:
    # string_needs_to_be_replace_here
    EnhancedUDLLexer().main()
except NameError:

    # string_needs_to_be_replace_here
    class SingletonEnhancedUDLLexer(type):
        _instances = {}
        def __call__(cls, *args, **kwargs):
            if cls not in cls._instances:
                # string_needs_to_be_replace_here
                cls._instances[cls] = super(SingletonEnhancedUDLLexer, cls).__call__(*args, **kwargs)
            return cls._instances[cls]

    # string_needs_to_be_replace_here
    class EnhancedUDLLexer(object):

        # string_needs_to_be_replace_here
        __metaclass__ = SingletonEnhancedUDLLexer

        @staticmethod
        def set_indicator_attributes(indicator_number, fore=(0,0,0)):
            editor.indicSetFore(indicator_number, fore)
            editor.indicSetStyle(indicator_number, INDICATORSTYLE.TEXTFORE)
            return indicator_number


        @staticmethod
        def paint_it(indicator, pos, length):
            editor.setIndicatorCurrent(indicator)
            editor.indicatorFillRange(pos,length)


        def do_regex(self, regex, indicator, start_position, end_position):
            editor.setIndicatorCurrent(indicator)
            editor.indicatorClearRange(start_position, end_position-start_position)
            editor.research(regex,
                            lambda m: self.paint_it(indicator, m.span(0)[0], m.span(0)[1] - m.span(0)[0]),
                            0,
                            start_position,
                            end_position)


        def style(self):
            line_number = editor.getFirstVisibleLine()
            start_position = editor.positionFromLine(line_number)
            end_position = editor.getLineEndPosition(line_number + editor.linesOnScreen())
            for indicator, regex in self.regex_dict.items():
                self.do_regex(regex, indicator, start_position, end_position)


        def __init__(self):
            editor.callbackSync(self.on_updateui, [SCINTILLANOTIFICATION.UPDATEUI])
            notepad.callback(self.on_langchanged, [NOTIFICATION.LANGCHANGED])
            notepad.callback(self.on_bufferactivated, [NOTIFICATION.BUFFERACTIVATED])

            self.__is_lexer_doc = False
            self.get_lexer_name = lambda: notepad.getLanguageName(notepad.getLangType())


        def set_lexer_doc(self,bool_value):
            editor.setProperty(self.__class__.__name__, 1 if bool_value is True else -1)
            self.__is_lexer_doc = bool_value


        def on_bufferactivated(self,args):
            if self.get_lexer_name() == self.lexer_name and editor.getPropertyInt(self.__class__.__name__) != -1:
                self.__is_lexer_doc = True
            else:
                self.__is_lexer_doc = False


        def on_updateui(self,args):
            if self.__is_lexer_doc:
                self.style()


        def on_langchanged(self,args):
            self.set_lexer_doc(True if self.get_lexer_name() == self.lexer_name else False)


        def main(self):
            # 2. define the lexer language, normally like what is returned by notepad.getLanguageName
            self.lexer_name = 'udf - test'

            # 3. define needed indicator_ids, indicator_fore_colors and regexes
            self.regex_dict = {
                                self.set_indicator_attributes(18, fore=(131,255,245)) : u'([A-Za-z][\w+]*)(?=\()',
                                self.set_indicator_attributes(19, fore=(255,137,115)) : u'([A-Za-z][\w+]*)(?=\.)',
                              }

            self.set_lexer_doc(True)
            self.style()

    # string_needs_to_be_replace_here
    EnhancedUDLLexer().main()

Note, focus was on portability not on performance and
you need to call it for each of the two views npp supports if you want
to have it automatically coloring udl docs in both views.

Cheers
Claudia

Scott Sumner

@Claudia-Frank

I’m looking forward to seeing steps 2 and 3! :-D

Claudia Frank

@Scott-Sumner

well, might happen when you start scrolling :-D

Cheers
Claudia

Scott Sumner

@Claudia-Frank

Got it! (I guess I expected them to be closer together, like is shown here)

:-D

Claudia Frank

@Scott-Sumner

well sometimes you have to have changes in your life to keep it exciting :-)
Seems we are back to future again - I answered your question 1 minute before you
were asking it and you did it even 3 minutes before I was answering it :-D

Cheers
Claudia

Scott Sumner

@Claudia-Frank

back to the future again

Yea, I’d much rather see 10-Aug-2018 7:47a than 2 minutes ago or six months ago anyway, but geez, if they’re going to do it that way (relatively), can’t they at least get it right? This will be confusing to any future readers, so here is the progression of the last several posting timestamps, as of RIGHT NOW:

Imgur

It would be really cool to go to an absolute timestamp here, how can that be made to happen?

Claudia Frank

@Scott-Sumner

Now it is back on track - thx for your image otherwise future readers must think we are insane :-)

Cheers
Claudia

PeterJones

There have been times that immediately after hitting SUBMIT, I see “PeterJones posted 7 minutes ago”…

I almost wonder if the timestamp issue here is similar to the one I described in the browser refresh thread: if there’s more than one server here, with different times – so that some people get one timestamp and some get the other.

The other possibility I can think of: NodeBB might assign the timestamp when you hit REPLY and get the compose window, rather than when you hit SUBMIT. So if you take longer to author and post your node, your post goes farther into the past. (Though if you’re replying to an existing node, I don’t see how that would explain your post showing with an earlier timestamp than the one you’re replying to. That goes back to my multiple-server-times hypothesis.)

update: appropriately, this one showed up as exactly “7 minutes ago” like I said above. :-)

Claudia Frank

@PeterJones

what is strange is that if you hover over the timestamp then it does show the correct time.
Although I don’t know the server setup I would say that nowadays no server-ops should
have troubles to install “clustered” servers with time sync - most OSes which
support clustering do have it anyway and rely on it to work correctly.
It’s really strange …

Cheers
Claudia

Update: just said and proven to be incorrect. Just hovered, immediately after submitting the post,
over the timestamp and … it is wrong.

Claudia Frank

I see - once the post passes the “lead time” it is correct again.

Cheers
Claudia

Χρήστος Κοτζιάς

@Claudia-Frank

*AUTO
DEMODISK = 0
ORG = $5400
TR ON
LST OFF

***PRINCE OF PERSIA
***COPYRIGHT 1989 JORDAN MECHNER

ORG ORG

JMP AUTOCTRL
JMP CHECKSTRIKE
JMP CHECKSTAB
JMP AUTOPLAYBACK
JMP CUTCHECK

JMP CUTGUARD
JMP ADDGUARD
JMP CUT

ADDAMOB LDX $0500 ;THIS IS A COMMENT, THE ABOVE LINE ISN’T ACTUALCODE

(LABEL MNEMONIC(LDX) OPERAND($0500) ;COMMENT)

CPX #MAXMOB
BEQ :RTS

INX
STX NUMMOB

JMP SAVEMOB

.
.
.
.

This is a 6502 assembly source code made with the Merlin assembler for the Prince of Persia game (copyright to Jordan Mechner)

Somewhere in the middle of that code you will see the typical syntax of the assembler. It goes like --> a) LABEL b) MENMONIC c) OPERAND and d) COMMENTS

4 columns all the time, some of them are optional (LABELS and COMMENTS and rarely OPERANDS)

Mnemonics are fixed keywords. I can highlight them. :)
Operands are sometimes fixed and sometimes not, but I don’t want to highlight the non-fixed since I can highlight the fixed ones.
Comments start with “;” or “*” and I can highlight them.
Labels are not fixed keywords and I want to use a different color for them.
LABELS -always- start in the beginning of a new line. NO space precedes them.
Is it possible to color every word and ONLY THAT WORD who is the first ONE of each line of code???

Thanks in advance :)

Claudia Frank

@Χρήστος-Κοτζιάς

Yes, can be done when you are willing to install python script plugin.

From my understanding of 6502 assembler and your description the label needs to be
followed by known mnemonic fields. Keeping this in mind and using a slightly
modified version of the script I posted above you would get something like

The part of the script which needs to be modified is within the main function

    def main(self):
        # 2. define the lexer language, normally like what is returned by notepad.getLanguageName
        self.lexer_name = 'Assembly'

        # 3. define needed indicator_ids, indicator_fore_colors and regexes
        self.regex_dict = {
                            self.set_indicator_attributes(18, fore=(131,255,245)) : u'^\w+?(?= LDA| LDX| LDY)',
                          }

        self.set_lexer_doc(True)
        self.style()

First you need to change the self.lexer_name with your udl name.
Run notepad.getLanguageName(notepad.getLangType()) in the python script console
after assigning your udl language to a document. I chose the assembly lexer for
demonstration purpose.

Second you need to extend the list of know mnemonic words in the regex

u'^\w+?(?= LDA| LDX| LDY)',

Cheers
Claudia

Χρήστος Κοτζιάς

This post is deleted!

Χρήστος Κοτζιάς

@Claudia-Frank

http://tinypic.com/r/4r4904/9

I have uploaded a screenshot from the built-in editor of the Merlin Pro assembler (Applewin emulator, Enhanced Apple IIe system)

Can you see it?

Claudia Frank

@Χρήστος-Κοτζιάς

Yes, I’m able to see your uploaded image.
What is your question or what did I miss?

Cheers
Claudia