FunctionList Confused

Lycan Thrope

Hello folks,

As I’ve mentioned in another thread, I’m trying to figure out the FunctionList and how to write one for another language. I at first chose C++, since that has a mixed parser as does the language I’m trying to do.

However, since I don’t know C++ that well, I sought other languages FunctionList, found java had one (which also seems to have a mixed parser .xml file) and got a very simple java program to bring into the NPP editor and…no FunctionList activity occurred other than the name of the file showing up at the head of the list. My presumption, of couse, is that the Java language/functionlist ability is working since it’s part of the default languages, yet, only syntax highlighting is working, when selected as the language, and the FunctionList only lists the file name. Not the classes or methods in it. So…as the title states, I’m a bit confused now how to structure my language .xml FunctionList.

Here is the example program I brought in to the editor:

// Class Declaration
 
public class Dog
{
    // Instance Variables
    String name;
    String breed;
    int age;
    String color;
 
    // Constructor Declaration of Class
    public Dog(String name, String breed,
                   int age, String color)
    {
        this.name = name;
        this.breed = breed;
        this.age = age;
        this.color = color;
    }
 
    // method 1
    public String getName()
    {
        return name;
    }
 
    // method 2
    public String getBreed()
    {
        return breed;
    }
 
    // method 3
    public int getAge()
    {
        return age;
    }
 
    // method 4
    public String getColor()
    {
        return color;
    }
 
    @Override
    public String toString()
    {
        return("Hi my name is "+ this.getName()+
               ".\nMy breed,age and color are " +
               this.getBreed()+"," + this.getAge()+
               ","+ this.getColor());
    }
 
    public static void main(String[] args)
    {
        Dog tuffy = new Dog("tuffy","papillon", 5, "white");
        System.out.println(tuffy.toString());
    }
}

When I ran it look in FunctionList, this is what I see:

As I mentioned, I don’t understand most of the C++ structures, but this Java is close to to the way my language does the code…and yet, even with the internal NPP language/functionList files, it’s not working…so you can see why I’m confused. Is there something I’m missing to make it work properly?

Any help in understanding why this isn’t working, and how I can get my functionList project on track would be appreciated.

Lee

Lycan Thrope

@lycan-thrope
As a note, I did find a different java file that the FunctionList did work on, however, I’m curious why it didn’t work on that file. The other file that worked, had a Class Main, which is different from the above file, in that the Main comes later in the file. Not sure if that makes a difference, but my language variant won’t have the Main() function, just the layout that is in the rest of the above file with the exception of that opening/closing braces.

Lee

Lycan Thrope

@lycan-thrope
Two other questions, about the FunctionList .xml file.

If the actual opening of a class is the word “class”, and the closing of it is “endclass”, would the parser find the opening of class in the regex to identify the keyword count as an open and only the keyword “endclass” would be used as: closeSymbole=“endclass”

Or can this work?
openSymbole=“class”
closeSymbole=“endclass”

One of the problems I seem to be having is figuring out how the regex is capturing the needed informtion and where it fits into each category of nameexp, etc. Sadly I can understand what the FAQ presents, but sadly the rest of the explanation, like what actually is happening and how it’s captured in the different sections. Some of the regex in the C++ file, for instance, just looks like a lot of redundant code, which is why I’ve been beating my head trying to decipher it and figure out how it’s working based on the FunctionList FAQ outline.

So it makes some sense, I’ve written a UDL for dBASE Plus which is the OOP version of dBASE, and am trying to write the FunctionList to work with that syntax and language. I’ve successfully been able to use Regex101.com to apparently properly decipher and show the captures of most of the Class syntax including options in the declaration line, so now I’m trying to proceed from that point in baby steps to get this language extension into Notepad++ as it has been requested by our community as the more desired external editor for the language.

So any help along this path would be appreciated.

Lee

Lycan Thrope

@lycan-thrope
Here’s the other java program that works in the FunctionList:

// Create a Main class
public class Main {
 
  // Create a fullThrottle() method
  public void fullThrottle() {
    System.out.println("The car is going as fast as it can!");
  }

  // Create a speed() method and add a parameter
  public void speed(int maxSpeed) {
    System.out.println("Max speed is: " + maxSpeed);
  }

  // Inside main, call the methods on the myCar object
  public static void main(String[] args) {
    Main myCar = new Main();     // Create a myCar object
    myCar.fullThrottle();      // Call the fullThrottle() method
    myCar.speed(200);          // Call the speed() method
  }
}

And this is the screenshot of what it produces:

So my language isssues aside, this is an interesting look at what may not work for Java programs also. :) Maybe a bug?

Lee

MAPJe71

Make sure there is extra white space (e.g. newline) after the last closing brace.

Lycan Thrope

@mapje71
Thanks mapje71.

That did the trick…I obviously cut and paste too few space on the copy.

Now if I can just figure out what regex, goes where…etc. :) Thanks again, and yes, I did figure out that the Java FunctionList xml isn’t a mixed parser, just a classrange parser that can find the methods in the class. At least, most of the regex is readable in that file for me to try and get a handle on it.

Lee

Lycan Thrope

@lycan-thrope said in FunctionList Confused:

openSymbole=“class”
closeSymbole=“endclass”

Still need to figure out if this will work, but in the meantime, I think I had an epiphany about how the regex fits into the FunctionList, and would appreciate an up or down on if the understanding is correct, bearing in mind, I’m just picking up regex, the boost flavor, and learning the NPP FunctionList functionality. :) No pressure.

So the “mainexpr” gathers all the possible text that can be gathered in a file, just to read it, not necessarily to parse it, so capturing or non-capturing shouldn’t matter (or does it?) and the following breakdowns start reading through the same code, most of it non-capturing until it gets to the desired area (Class or Function name in this case) and then captures that, so the next step in the breakdown is left only to look at the captured text from the previous breakdown…and is therefore used by the parser to identify and place that Class, Class Function(method), or separate Functions outside that body into their proper place for the FunctionList. Would that be a proper understanding of the process?

In my testing, I’ve been using all capture groups to isolate and identify syntax, and I think that’s why I didn’t get any kind of response from my FunctionList testing for class…besides not having the Classrange,Classname function working…but like I said, I’m confused and unsure about this process…but am plugging away at trying to understand what I need to do to get it to function properly. This part of the process isn’t that well documented, which is why I’m asking if this is the proper analogy while I continue to beat my head against the wall. :)

Lee

Michael Vincent

@lycan-thrope said in FunctionList Confused:

I think I had an epiphany about how the regex fits into the FunctionList, and would appreciate an up or down on if the understanding is correct, bearing in mind, I’m just picking up regex, the boost flavor, and learning the NPP FunctionList functionality.

Have you read the user manual that explains basic Function List behavior?

Lycan Thrope

@michael-vincent
Yes I have, but like I said, I’m new to all these tools, so I’m trying to understand what I see there and “how” to implement it. Understanding what you’re trying to do makes it a bit easier than trial and error…that’s all. So does my description make sense of what I’m seeing? That’s what I’m asking…my understanding, versus the mainstream understanding. Like people that learn by reaading, others learn by seeing, and yet others learn by trial and error, all good, but the important thing is do they understand…and that’s what makes clarity of a process…I’m at that point, of trying to understand what I’m trying to do.

Does that make sense?

Lee

Lycan Thrope

@michael-vincent
Maybe this will clarify what I mean. I have taken these variations of the syntax for the Class declaration starting line. (Watch for wrap):

class TruckNotebookForm of TBASE from :Truck:Truckbase.cfm
class PlainObjectListForm of FORM
class FrameCtrl(frameObj) of dBCWndCtrl custom
class dBCWndCtrl
class FrameAppCtrl of FrameCtrl custom
class ToolButtonFx(oParent) of Toolbutton(oParent) custom
class MenuFx(oParent,cName)
class dContainersForm of DFORM from “dForm.cfm”
class LGCENTRYFIELD(parentObj, name) of ENTRYFIELD(parentObj, name) custom

I have come up with variations of the regex to match all of these, the last ones working the best to cover all variations. (Watch wrap):

1: ^(class)\s([a-zA-Z0-9]+)\s([a-zA-Z0-9+]+)\s([a-zA-Z0-9+]+)
2: ^(class)\s(\w*)\s(\w*)\s(\w*)
3: (?x)^[\t\x20](class)[\t\x20]+((\w+)|(\w+)[\x20]+(((\w+)(,\w+)?))?)?((\w+|[^\r\n])[\t\x20]+)(custom)? <—best
4: (class)[\t\x20](\w+)([\t\x20]?(\w(,[\t\x20]?\w*)?))?([\t\x20]of[\t\x20]\w((\w*(,[\t\x20]?\w*)?))?([\t\x20]custom)?([\t\x20]from\t\x20)?)?
5: (class)[\t\x20](\w+)([\t\x20]?(\w(,[\t\x20]?\w*)?))?(([\t(\x20]of)[\t\x20]\w((\w*(,[\t\x20]?\w*)?))?([\t\x20]custom)?([\t\x20]from\t\x20)?)?
6: (class)[\t\x20](\w+)([\t\x20]?(\w(,[\t\x20]?\w*)?))?([\t(\x20]of\t\x20)?((\w(,[\t\x20]?\w*)?))?([\t\x20]custom)?([\t\x20]from\t\x20)?
7: (?:class)[\t\x20](?‘Classname’\w+)([\t\x20]?(\w(,[\t\x20]?\w*)?))?(([\t(\x20]of)\t\x20((\w(,[\t\x20]?\w*)*?))?(\t\x20)?([\t\x20]from\t\x20)?)?

Now having done that, what’s the next step, or did I do it right, or have I included enough information to get what I need in “mainexpr”, and where do I get “nameexpr” to extrapolate from that?..How do I seperate the wheat from the chaff, so to speak? If you run this through regex101.com, it works like a charm. Trying to put it in the .xml file, hasn’t been as successful. ::shrug::

Lee

Lycan Thrope

@lycan-thrope
Yikes, it looks like the forum editor did some editing of my strings in the regex…argh…but I hope you get the point. I’m knowing just enough to be dangerous. :)

Lee

PeterJones

@lycan-thrope said in FunctionList Confused:

Yikes, it looks like the forum editor did some editing of my strings in the regex

You should read the FAQ: Formatting Forum Posts

Lycan Thrope

@peterjones
Apparently, I need to re-read it…thanks. :)

Lee

Lycan Thrope

@lycan-thrope Test if this works? Here’s hoping. :)
1: ‘^(class)\s([a-zA-Z0-9]+)\s([a-zA-Z0-9+]+)\s([a-zA-Z0-9+]+)’
2: ‘^(class)\s(\w*)\s(\w*)\s(\w*)’
3: ‘(?x)^[\t\x20](class)[\t\x20]+((\w+)|(\w+)[\x20]+(((\w+)(,\w+)?))?)?((\w+|[^\r\n])[\t\x20]+)(custom)?’
4: '(class)[\t\x20](\w+)([\t\x20]?(\w(,[\t\x20]?\w*)?))?([\t\x20]of[\t\x20]\w((\w*(,[\t\x20]?\w*)?))?([\t\x20]custom)?([\t\x20]from’[‘\t\x20’]'(:\w:\w*.\w*|“\w*.\w*”))?)?’
5: ‘(class)[\t\x20](\w+)([\t\x20]?(\w*(,[\t\x20]?\w*)?))?(([\t(\x20]of)[\t\x20]\w((\w*(,[\t\x20]?\w*)?))?([\t\x20]custom)?([\t\x20]from’[‘\t\x20’]'(:\w:\w*.\w*|“\w*.\w*”))?)?’
6: ‘(class)[\t\x20](\w+)([\t\x20]?(\w*(,[\t\x20]?\w*)?))?([\t(\x20]of’[‘\t\x20’]'(\w))?((\w*(,[\t\x20]?\w*)?))?([\t\x20]custom)?([\t\x20]from’[‘\t\x20’]'((:\w:\w*.\w*)|(“\w*.\w*”)))?’
7: ‘(?:class)[\t\x20](?‘Classname’\w+)([\t\x20]?(\w*(,[\t\x20]?\w*)?))?(([\t(\x20]of)‘[’\t\x20’]'(?‘Superclass’\w)((\w*(,[\t\x20]?\w*)?))?(‘[’\t\x20’]‘(?‘Custom’custom))?(’[’\t\x20’]‘from’[‘\t\x20’]'(:\w:\w*.\w*|“\w*.\w*”))?)?’

Lycan Thrope

@lycan-thrope
Unfortunately, the backticks to make the multiline brackets work, will have to be manually removed to run the regex’s…and the backticks around the whole regex did not turn it red typewriter text. :(

Lee

Lycan Thrope

@lycan-thrope Hmmmm Let me try this. Test
'red typewriter text’1: ^(class)\s([a-zA-Z0-9]+)\s([a-zA-Z0-9+]+)\s([a-zA-Z0-9+]+)

Michael Vincent

@Lycan-Thrope
@michael-vincent said in FunctionList Confused:

I think I had an epiphany about how the regex fits into the FunctionList, and would appreciate an up or down on if the understanding is correct, bearing in mind, I’m just picking up regex, the boost flavor, and learning the NPP FunctionList functionality.

Have you read the user manual that explains basic Function List behavior?

Maybe this community post is a little more detailed than the user manual?

Cheers.

PeterJones

@lycan-thrope said in FunctionList Confused:

will have to be manually removed to run the regex’s

You have read the FAQ wrong.

the syntax

1. `^(class)\s([a-zA-Z0-9]+)\s([a-zA-Z0-9+]+)\s([a-zA-Z0-9+]+)`

renders as

^(class)\s([a-zA-Z0-9]+)\s([a-zA-Z0-9+]+)\s([a-zA-Z0-9+]+)

which you can then just copy/paste the red text.

Please note that what you had,

1: ‘^(class)\s([a-zA-Z0-9]+)\s([a-zA-Z0-9+]+)\s([a-zA-Z0-9+]+)’

is not the same as

1: `^(class)\s([a-zA-Z0-9]+)\s([a-zA-Z0-9+]+)\s([a-zA-Z0-9+]+)`

which renders correctly as
1: ^(class)\s([a-zA-Z0-9]+)\s([a-zA-Z0-9+]+)\s([a-zA-Z0-9+]+)

The ` key (often called “backtick” or “grave”) is the one that on a standard US keyboard is the same key as ~, but unshifted . You are typing the single-quote / apostrophe key ' (which on a US keyboard, is the unshifted version of ", the double quote). The single-quote in the forum is then being rendered as ‘smart single quotes’ ‘...’ in your post.

The Formatting Forum Posts gives you plenty of copies of the ` character, which you can copy/paste if you are unable to type the character on your keyboard.

Lycan Thrope

@peterjones said in FunctionList Confused:

`

Peter, thanks for clarifying that for me. I just want to show you why I used that character, besides not realizing where the backtick key was. In this screenshot of your message to me, as well as in the documentation FAQ for posting, I highlighted the character to see if these old eyes could make out what it was, and as you see, the little pop up that “says” what it is, is mistaken. :) Thanks for fixing that for me. :) Screenshot of what I saw:

So really, thanks for clearing that up for me. :)

Lee

Lycan Thrope

@lycan-thrope
So…let’s try this again, so I can see if it work right, with just the backtick.
7: (?:class)[\t\x20]*(?'Classname'\w+)([\t\x20]*?\(\w*(,[\t\x20]?\w*)*?\))?(([\t(\x20]of)[\t\x20](?'Superclass'\w*)(\(\w*(,[\t\x20]?\w*)*?\))?([\t\x20](?'Custom'custom))?([\t\x20]from[\t\x20](:\w*:\w*\.\w*|"\w*\.\w*"))?)?

Woohoo…the preview shows it working, but as per the FAQ let’s see if it does it without all the backticking in side the regex on the multiline brackets. :)

Lee