Macro Complex instructions.

Annas Taher · Dec 19, 2017, 3:35 PM

Well. I don’t know how to explain it so I will give the example right away:
I have 2 Files
File1 is a list:
aaa
bbb
ccc
ddd
etc…
And File2: all X are the same value
Blue Box xxx
It contains grapes and xxx
Red Box xxx
It contains tomatoes and xxx
Green Box xxx
It contains oranges and xxx
Yellow Box xxx
It contains lemons and xxx
etc.

I want is so macro can take from File1 and replace the first two xxx with aaa the 2nd two xxx with bbb and so on .

How do I do that?

PeterJones · Dec 19, 2017, 2:44 PM

You could use a scripting language, like PythonScript . The general procedure would be akin to

Open both File1 and File2.
RClick on the File2 tab, and Move to Other View

Run a PythonScript that does something like the following (not valid PythonScript… this just gives an outline in pseudocode.)

 // editor1 is the PythonScript object instance for the first (left or top) document window in NPP
 // editor2 is the instance for the second (right or bottom) document window
 for( lineNumber = 1 to editor1.getLineCount() )
     // grab the nth line from File1
     editor1.gotoLine(lineNumber)
     newValue = editor1.getCurLine()
     for ( time = 1 to 2 )
         // look for the first occurrence of 'xxx'
         position = editor2.findText( flags, 0, end, "xxx")      // Look in the PythonScript documentation
         //      for Editor Object > Scintilla Methods:
         //      I would need to study .findText() and the associated SCI_FINDTEXT more
         //      to figure out the right values for flags, and then search thru to find
         //      the right command for finding the end-location of editor2
         editor2.????(position, position+3)                      // select the text from "position" to "position+3" or whatever length of "xxx" is
         editor2.replaceSel(newValue)                            // replace the selected "xxx" with the newValue

Sorry, I’m not a Python/PythonScript expert, and didn’t have much time this morning, otherwise I could have researched a bit more and tried to get valid/running code. Hopefully, this outline will spark a PythonScript guru (@Claudia-Frank , hint, hint) could turn this pseudocode into a valid PythonScript in a few minutes.

(edit: added the for time=1 to 2 loop, because I forgot you wanted to replace two instances of each “xxx” with a given newValue)

Claudia Frank · Dec 19, 2017, 3:35 PM

@Annas-Taher

Thx to Peter :-), I already have the logic to follow - should be straightforward.
When back home I will build something but in the meantime you need to install
the python script plugin (preferable use the msi ) to make it work. And you have to use the 32bit version of notepad++
as there is no python script plugin for 64 bit available yet.

Cheers
Claudia

PeterJones · Dec 19, 2017, 3:44 PM

Actually, it was pestering me so much I couldn’t think about anything else…

This appears to do what @Annas-Taher described:

# editor1 is the PythonScript object instance for the first (left or top) document window in NPP
# editor2 is the instance for the second (right or bottom) document window

for lineNumber in range(0, editor1.getLineCount() ):
    # grab the nth line from File1 (exclude newline)
    editor1.gotoLine(lineNumber)
    newValue = editor1.getCurLine().rstrip()
    console.write("editor1: #" + str(lineNumber) + " = \"" + newValue + "\"\n")

    editor2.documentEnd()               # go to the last position
    end2 = editor2.getCurrentPos()      # record the position
    console.write("editor2.end = " + str(end2)+"\n")
    editor2.documentStart()             # back to the beginning
    start2 = editor2.getCurrentPos()    # record the position
    console.write("editor2.start = " + str(start2)+"\n")

    # want to replace two "xxx" entries in File2 with each line from File1
    for time in range(1, 2):

        # look for the first occurrence of 'xxx', starting at start2 and ending at end2
        #  position is a tuple with the start and end locations of the match
        position = editor2.findText( FINDOPTION.MATCHCASE, start2, end2, "xxx")
        console.write("editor2: findText @ " + str(position[0]) + ":" + str(position[1]) + "\n")

        # select the "xxx"
        editor2.setSelectionStart(position[0])
        editor2.setSelectionEnd(position[1])

        # replace the selection with newValue
        editor2.replaceSel(newValue)

        # the cursor is now at the end of the replaced value, and we want to
        start2 = editor2.getCurrentPos()
        console.write("Start2: " + str(start2)+"\n")

Claudia Frank · Dec 19, 2017, 3:50 PM

@PeterJones

So you gave me a job and fired me instantly :-D
To quote Meat Loaf: “Is this the way to treat an employee (well he said “an expensive guitar”)” :-D

Cheers
Claudia

PeterJones · Dec 19, 2017, 3:53 PM

Well, there is a bug in the code: The nested loop isn’t working, and it’s just replacing one of each, rather than two xxx with each line from File1. So whichever of us finds the Round Tuit to debug the code first will get credit for final solution. :-)

Claudia Frank · Dec 19, 2017, 4:54 PM

@Annas-Taher
@PeterJones

from your description I would solve it like this

editor1_list = editor1.getText().split()
string_to_be_replaced = 'xxx'
line_counter_editor2 = 0

editor.beginUndoAction()
for i in editor1_list:
    editor2.replaceLine(line_counter_editor2, editor2.getLine(line_counter_editor2).replace(string_to_be_replaced, i).rstrip())
    editor2.replaceLine(line_counter_editor2+1, editor2.getLine(line_counter_editor2+1).replace(string_to_be_replaced, i).rstrip())
    line_counter_editor2 += 2
editor.endUndoAction()

If the code doesn’t explain itself, let me know.

Cheers
Claudia

guy038 · Dec 20, 2017, 9:35 PM

Hello @annas-taher, and All,

Surely, Claudia and Peter, your Python scripts should give nice results. I just imagined … a regex S/R equivalent method :-))

Annas, my method, below, will work, assuming two facts :

The expression xxx ends all the lines of File2
The number of lines of the list, in File1, is, exactly, half the number of lines of File2

If so, first, merge the File1 contents, right after the File2 contents, as below, and save these contents as File3 :

Blue Box xxx
It contains grapes and xxx
Red Box xxx
It contains tomatoes and xxx
Green Box xxx
It contains oranges and xxx
Yellow Box xxx
It contains lemons and xxx
aaa
bbb
ccc
ddd

Then, in File3, with a first regex S/R, we join two consecutive lines, ending with the xxx expression

SEARCH (?-s)^(.+xxx)\R(.+xxx)

REPLACE \1\2

OPTIONS Wrap Around and Regular expression checked

ACTION Click on the Replace All button

You should obtain the text, below :

Blue Box xxxIt contains grapes and xxx
Red Box xxxIt contains tomatoes and xxx
Green Box xxxIt contains oranges and xxx
Yellow Box xxxIt contains lemons and xxx
aaa
bbb
ccc
ddd

Notes :

As usual, the (?-s) modifier, ensures that the dot special character will match standard characters, only
The ^ assertion represents the position beginning of line, which must be verified
The two syntaxes (.+xxx) catch the contents of two consecutive lines, ending with xxx, and stored, successively, in group1 and group 2, due to each pair of parentheses
The \R form represents the line-break characters between these two lines
In replacement, we rewrite the contents of the two lines, \1\2, except for the line break

Well. now, note that after the first line, you have to cross 3 complete lines to get the aaa first line of your list

Therefore, we’ll use the second and last regex S/R, below :

SEARCH (?-s)^(.+)xxx(.+)xxx\R(?=(?:.+\R){3}(.+\R))

REPLACE \1\3\2\3

OPTIONS Wrap Around and Regular expression checked

ACTION Click on the Replace All button

=> Your File3 contents should be as below :

Blue Box aaa
It contains grapes and aaa
Red Box bbb
It contains tomatoes and bbb
Green Box ccc
It contains oranges and ccc
Yellow Box ddd
It contains lemons and ddd
aaa
bbb
ccc
ddd

Notes :

This time, group 1 and group 2, (.+), contain the totality of the two consecutive lines, apart from the xxx expression
So, the part ^(.+)xxx(.+)xxx\R matches each complete line, with its line-break
But, ONLY IF, further on, the positive look-ahead (?=(?:.+\R){3}(.+\R)), ( where it looks for 3 complete lines, with its line-break character(s), in a non-capturing group, (?:.+\R){3}, followed with an other complete line (.+\R) ) , is verified
Note that this condition is always true, but, thanks to this syntax, we can store any line of the list aaa, bbb,… as group 3
In replacement, we simply, rewrite groups 1 and 2, replacing the old expression xxx by \3 which represents any value of the list aaa, bbb and so on…, with its line breaks, to get the initial form of the text

Finally, just get rid of the list aaa......ddd, at the end of File3 ( Ctrl + Shift + End and Del ! )

Of course, for n lines, between the first one and the aaa line, simply change the search regex as :

(?-s)^(.+?)xxx(.+?)xxx\R(?=(?:.+\R){n}(.+\R)) , with the appropriate integer n

Best Regards,

guy038

PeterJones · Dec 19, 2017, 10:10 PM

For completeness, I eventually found the bug in mine: I was misusing range(start, end) – I thought it included end when iterating over the range, but it only uses the values less than end.

The following is my code, with the bug fixed (and some of my debug prints left in) that I think now does what was requested… though @Claudia-Frank did it much more succinctly… and the magic of @guy038 regex ability is virtually limitless. :-)

# editor1 is the PythonScript object instance for the first (left or top) document window in NPP
# editor2 is the instance for the second (right or bottom) document window

# Python's range(start, stop) operator will keep going while the index variable is LESS THAN stop, so it will never use stop: https://docs.python.org/2/library/functions.html#range
for lineNumber in range(0, editor1.getLineCount() ):
    # grab the nth line from File1 (exclude newline)
    editor1.gotoLine(lineNumber)
    newValue = editor1.getCurLine().rstrip()
    console.write("editor1: #" + str(lineNumber) + " = \"" + newValue + "\"\n")

    editor2.documentEnd()               # go to the last position
    end2 = editor2.getCurrentPos()      # record the position
    console.write("editor2.end = " + str(end2)+"\n")
    editor2.documentStart()             # back to the beginning
    start2 = editor2.getCurrentPos()    # record the position
    console.write("editor2.start = " + str(start2)+"\n")

    # want to replace two "xxx" entries in File2 with each line from File1
    for time in range(0, 2):
        console.write("time# " + str(time) + "!!!\n")

        # look for the first occurrence of 'xxx', starting at start2 and ending at end2
        #  position is a tuple with the start and end locations of the match
        position = editor2.findText( FINDOPTION.MATCHCASE, start2, end2, "xxx")
        if position is None:
            console.write("editor2.position is NONE, so skipping...\n")
            break                   # don't try to replace

        console.write("editor2: findText @ " + str(position[0]) + ":" + str(position[1]) + "\n")

        # select the "xxx"
        editor2.setSelectionStart(position[0])
        editor2.setSelectionEnd(position[1])

        # replace the selection with newValue
        editor2.replaceSel(newValue)

        # the cursor is now at the end of the replaced value, and we want to
        start2 = editor2.getCurrentPos()
        console.write("Start2: " + str(start2)+"\n")

    console.write("next source line...\n")