Parse specific numbers on a text

[Sambuddy]

This is a sample of a *text I have. I am trying to select a bunch of Single and Multiline texts and then change their values from example this 311/312/313/314 to this 311/312/313/314 or from this A10/A11/A16/A18 to this A20/A21/A26/A28. In Summary the ones to twos or threes. What is consistant is the number 1 either in 3(1)1 or A(1)0 or E(1) in a set of numbers that are separated by / that I would like to change to 2s or 3s. Can someone help me either improve what I have so far or get me there please?

(defun c:sector ( / ss newValue ent entData text segments newSegments newText )
  (prompt "\nSelect Mtext objects to modify: ")
  (setq ss (ssget '((0 . "MTEXT")))) ; Select Mtext objects
  (if ss
    (progn
      (setq newValue (getint "\nEnter the new middle value (1, 2, or 3): "))
      (if (member newValue '(1 2 3))
        (progn
          (setq newValue (itoa newValue))
          (repeat (sslength ss)
            (setq ent (ssname ss (setq count (1- (sslength ss)))))
            (setq entData (entget ent))
            (setq text (cdr (assoc 1 entData)))
            (if (and text (> (strlen text) 0))
              (progn
                ;; Split the text into segments
                (setq segments (split-string text "/"))
                ;; Update each segment
                (setq newSegments
                      (mapcar
                        (lambda (seg)
                          (if (>= (strlen seg) 3)
                            ;; Replace the middle digit
                            (strcat (substr seg 1 1)
                                    newValue
                                    (substr seg 3))
                            seg
                          )
                        )
                        segments
                      )
                )
                ;; Join the segments back together
                (setq newText (apply 'strcat (mapcar (lambda (s) (strcat s "/")) newSegments)))
                ;; Remove the trailing "/"
                (setq newText (substr newText 1 (1- (strlen newText))))
                ;; Update the Mtext object
                (entmod (subst (cons 1 newText) (assoc 1 entData) entData))
                (entupd ent)
              )
              (prompt "\nSkipped: Mtext is empty.")
            )
          )
          (princ "\nMtext objects updated.")
        )
        (prompt "\nInvalid value. Please enter 1, 2, or 3.")
      )
    )
    (prompt "\nNo Mtext objects selected.")
  )
  (princ)
)

(defun split-string (str sep / pos start result)
  (setq result '() start 1)
  (while (setq pos (vl-string-search sep str start))
    (setq result (append result (list (substr str start (- pos start)))))
    (setq start (+ pos (strlen sep))))
  (append result (list (substr str start)))
)

 

[mhupp]

getkword will only allow user to input 1 2 or 3 before continuing.

Changed the lambda to work with either 3 or 2 string lengths.

UMV.lsp

Edited June 20 by mhupp

[Sambuddy]

10 hours ago, mhupp said:

getkword will only allow user to input 1 2 or 3 before continuing.

Changed the lambda to work with either 3 or 2 string lengths.

UMV.lsp 1.9 kB · 1 download

I could not get it to work. it tells me that the Mtexts are updated, but no changes for values.

[CyberAngel]

As usual, Lee Mac is way ahead of us. On this page you'll find a regular expression find/replace. Or you may only need to work the wcmatch command into your code.

[BIGAL]

Over at "Theswamp" is a section called Challenges it is for this type of problem, a generic answer to list manipulation. I have asked there and is a very helpful section. Give it a try.

 

http://www.theswamp.org/index.php?board=77.0