Getting string from X position until the end of text

[dbm]

Hi again,

I'm searching in forums and google but in the name of $DEITY I don't know how to ask them correctly to get an answer.

I'm trying to get the substring of a given text, begining from a desired position. In other words:

If I've a text like this: "abc def ghi abc dde jkman de" I wanna all text starting from "def"

I don't know if exists in LC a predefined way to do it. By now I could do:

Get the position of the first occurence of "def" using the offset
walk from this position storing each char until the end of text

But How I say in LC "until the end of text" ?

[Dixie]

Using your example string...

Code: Select all

on mouseUp
   put "abc def ghi abc dde jkman de" into theString
   put wordOffset("def",theString) into wordPosition
   put word wordPosition to -1 of theString
end mouseUp

[dbm]

I'm using your example as source, but seems that I'm doing it bad.

I've created this funtion:

Code: Select all

function wordafter letext leword ledelim
   local pos, rtext
   put empty into pos
   put empty into rtext
   put wordoffset(leword,letext) into pos
   if pos is not 0 then
      put word pos to -1 of letext into rtext
      set the itemDelimiter to "="
      --set the itemDelimiter to ledelim
      put word 2 of rtext into rtext
   Else
      put empty into rtext
   end if
   return rtext
end wordafter

This function is called like this:

Code: Select all

put wordafter(ttest,"name=","=") into field cdebug

Even If I force to use the "=" literaly as delimiter, the line

put word 2 of rtext into rtext

returns the word as it is count using space delimiter.

Why uses the space delimiter after I forced to use the "=" delimiter?

Sorry for all this beginner questions

[Dixie]

Why uses the space delimiter after I forced to use the "=" delimiter?

I don't know ?... Why is "=" the delimeter ?... give an example of the string you are trying to parse

[dbm]

Flags: X - disabled, A - active, I - incomplete
0 ;;; comentario de usuario
customer=admin name="user1" actual-profile="test profile"
password="password" location="camp_name" shared-users=1
wireless-psk="" wireless-enc-key="" wireless-enc-algo=none
uptime-used=8m31s download-used=12180371 upload-used=1766641
last-seen=feb/09/2015 22:27:27

[Dixie]

Ok... What are you wanting to return from your function ?

[dbm]

For example, we have the following text :

Flags: X - disabled, A - active, I - incomplete
0 ;;; comentario de usuario
customer=admin name="user1" actual-profile="test profile"
password="password" location="camp_name" shared-users=1
wireless-psk="" wireless-enc-key="" wireless-enc-algo=none
uptime-used=8m31s download-used=12180371 upload-used=1766641
last-seen=feb/09/2015 22:27:27

and in this example we want the value after the "uptime-used", but we can want any other value, the only thing that we can know is the identifier of the value (location, shared-users, wireless-enc-key...)

I want create a generic function that giving a text, a sting to search and a delimiter, returns the word after the string to search.

the call function in this example can be wordafter(ttest,"uptime-used=","=") ( where ttest is a variable containing the example text, "uptime-used=" is the identifier to locate the value and "=" is the delimiter that must be used)

Actually, the code of my function wordafter, does the following:

Code: Select all

function wordafter letext leword ledelim ( letext -> text containing data, leword -> identifier to find, ledelim -> delimiter to be used to split in words)
-- variables to be used as index and to store text
   local pos, rtext
   put empty into pos
   put empty into rtext
-- we get the first position of the identifier inside the text
   put wordoffset(leword,letext) into pos
-- if found 
   if pos is not 0 then
-- copy all text from the identifier until the end of text ( then the identifier becomes the facto the first word in the copied text )
      put word pos to -1 of letext into rtext
-- we force the delimiter to be "=" ( then the value we are searching for will be the 2nd word of the copied text
      set the itemDelimiter to "="
      --set the itemDelimiter to ledelim
-- we get the second word of the copied text ( the value of the identifier )
      put word 2 of rtext into rtext
   Else
      put empty into rtext
   end if
   return rtext
end wordafter

In the example above:

put wordafter(ttest,"uptime-used","=") into field cdebug

It should return into cdebug the text "8m31s" but actually returns "download-used=12180371". In fact, the line put word 2 of rtext into rtext is using space as delimiter, when I forced to use "=", or I think so.

[dbm]

I think that the problem is here:

"By default, items are delimited by commas. You can change the default comma to create your own chunk type by setting the itemDelimiter property to any character. For example:

on mouseUp
set the itemDelimiter to ":"
answer item 7 of field "text"
end mouseUp

An item can contain characters or words, but not lines. Items can be contained in a line, but not in a word or character."

In my text, from the point of view of LC, the 2 itmes are a word and setdelimiter is not applied, then?

If I convert the text before, to have an space before and after each "=" then It should work, but I don't like very much this solution 'cause it will have problems returning data from a pair of fields like
identifier="text value"

By now I changed my function and it returns de desired value when the pair of fields are in the form identifier=value

Code: Select all

function wordafter letext leword ledelim
   local pos, rtext
   put empty into pos
   put empty into rtext
   put wordoffset(leword,letext) into pos
   if pos is not 0 then
      put word pos to -1 of letext into rtext
      replace ledelim with space & ledelim & space in rtext
      put word 3 of rtext into rtext
   Else
      put empty into rtext
   end if
   return rtext
end wordafter

[Dixie]

The function below will work, using the text that you supplied... as long as the returned value is one whole string, which it is in all cases except :-
actual-profile="test profile" ...

Code: Select all

on mouseUp
   put fld 1 into theString
   put "download-used" into theValue
   put "=" into theDelim
   
   put wordAfter( theString,theValue,theDelim)
end mouseUp

function wordAfter theString, theValue theDelim
   set the itemDel to theDelim
   put itemOffset(theValue,theString) into theItemNo
   put item theItemNo + 1 of theString into temp
   set the itemDel to space
   return item 1 of temp of theString
end wordAfter

[dbm]

You're right Dixie, is what I say in my last post. I'm working to find a way to be able to process all possibilities identifier=value and identifier="text value"


Thanks for the optimisation. I learn a lot.

[Thierry]

Flags: X - disabled, A - active, I - incomplete
0 ;;; comentario de usuario
customer=admin name="user1" actual-profile="test profile"
password="password" location="camp_name" shared-users=1
wireless-psk="" wireless-enc-key="" wireless-enc-algo=none
uptime-used=8m31s download-used=12180371 upload-used=1766641
last-seen=feb/09/2015 22:27:27

Hello dbm,

I've made a two-line-script which extract nicely the key and associated value.

However, it appears that there is one inconsistency in this set of data.
For the key "last-seen", its value ( feb/09/2015 22:27:27) should be quoted.

Can you modify your data?
Or can you confirm that it will *always* be the last item of your dataSet?
Both cases will set the complexity to none.

Regards,

Thierry

[dbm]

Merci monsieur Duez!

Thanks for your answer.

The data is chaotic and effectively in a coherent world the value of "last-seen" should be quoted 'cause is a value composed from 2 alphanumeric strings, but in this case, the data returned by the router for the identifier "last-seen" comes in the format date time without quotes.

In this case yes, the "last-seen" is the last value pair in the string, but is not the same case always (see example from another answer from router):

0 customer=admin user="user1" nas-port=2147483656 nas-port-type=ethernet nas-port-id="ether1" calling-station-id="00:1D:72:8E:C0:35" user-ip=10.0.1.194 host-ip=127.0.0.1
status=start,stop,interim from-time=feb/09/2015 22:11:07 till-time=feb/09/2015 22:13:49 terminate-cause=user-request uptime=2m42s download=71441 upload=27396

Could you kindly share this two line script? I'm very curious!

[jiml]

You might try capturing the key value pairs in an array.

Code: Select all

function getVariables theText
   delete line 1 to lineoffset("=",theText) of theText
   repeat with x = the number of words of theText down to 1
      if word x of theText contains "=" then put cr before word x of theText
   end repeat
   filter theText without ""
   replace quote with empty in theText
   split theText with cr and "="
   return theText
end getVariables

If you pass the text you shared:

Flags: X - disabled, A - active, I - incomplete
0 ;;; comentario de usuario
customer=admin name="user1" actual-profile="test profile"
password="password" location="camp_name" shared-users=1
wireless-psk="" wireless-enc-key="" wireless-enc-algo=none
uptime-used=8m31s download-used=12180371 upload-used=1766641
last-seen=feb/09/2015 22:27:27

you'll end up with an array with these keys:

wireless-enc-algo
uptime-used
password
last-seen
download-used
wireless-psk
location
upload-used
shared-users
wireless-enc-key

each key having its value.

Jim Lambert

[jiml]

Slight correction based on your other text examples

Code: Select all

function getVariables theText
   if "=" is not in line 1 of theText then delete line 1 to lineoffset("=",theText) of theText

[jiml]

better

Code: Select all

function getVariables theText
   repeat with x = the number of words of theText down to 1
      if word x of theText contains "=" then 
         put cr before word x of theText
      else
         delete word x of theText
      end if
   end repeat
   filter theText without ""
   replace quote with empty in theText
   split theText with cr and "="
   return theText
end getVariables