Extract text from global variable

JVDH · August 28, 2025, 9:54am

I have this part of a text in a global variable ‘nasatxt’:

...
 2025-Aug-27 14:49       1.937511   4.060715  131.799177   0.380618
 2025-Aug-27 14:50       1.936684   4.061515  131.790684   0.380606
 2025-Aug-27 14:51       1.935856   4.062316  131.782190   0.380593
 2025-Aug-27 14:52       1.935028   4.063116  131.773697   0.380581
 2025-Aug-27 14:53       1.934201   4.063917  131.765203   0.380568
 2025-Aug-27 14:54       1.933373   4.064717  131.756710   0.380556
 2025-Aug-27 14:55       1.932545   4.065517  131.748216   0.380543
 2025-Aug-27 14:56       1.931717   4.066317  131.739723   0.380531
 2025-Aug-27 14:57       1.930889   4.067117  131.731229   0.380518
 2025-Aug-27 14:58       1.930061   4.067917  131.722736   0.380506
 2025-Aug-27 14:59       1.929233   4.068717  131.714242   0.380493
 2025-Aug-27 15:00       1.928405   4.069516  131.705749   0.380481
...

Codes suggested by ChatGPT and Gemini do not work:

$tc(reg, gv(nasatxt), "2025-Aug-27 14:56\s+(\S+)\s+\S+", "$0")$

$fl(tc(split, gv(nasatxt), "\n"), "i", "if(i ~ """ + "2025-Aug-27 14:56" + """, i, """")")$

Ultimately, I would like to extract the 4 numbers on each line separately, and dynamically with a

 df(yyyy-MMM-dd HH:mm)

Can anyone point me to a correct way of doing this?

Baju_Santiko · August 28, 2025, 10:06am

Hi there,

Could you please tell me what is the 4 number like? Any example from the invite text?

Baju_Santiko · August 28, 2025, 10:32am

@JVDH

Hi James,

If you want to get the 3rd column from your nasatxt global variable, try these steps.

Create a Text global variable to grab all the 3rd column data from the nasatxt global variable. Let’s say its name is: 3rdcol
Type this code snippet into the 3rdcol variable:

$tc(reg,tc(reg,gv(nasatxt),"([0-9]{4}\-[A-Za-z]{3}\-[0-9]{1,2}[\s\t]+[0-9]{1,2}\:[0-9]{1,2})(([\s\t]+[0-9.]+){2})(([\s\t]+)([0-9.]+))(.*[\n\r]*)","$6"),"[\n\r\s\t]+","|")$

Create a new Text object, type this code snippet into it: $tc(split,gv(3rdcol),"|",0)$ ➞ get the first value from the 3rd column data.

TIPS • Table Parser 202508281080×2168 179 KB

I hope this would help.
Did you find the most helpful response to your issue?
Please kindly press the Solution button to mark it as the solution.
This would help others with the similar issue.

JVDH · August 28, 2025, 11:03am

Just to clarify further, the selection of the separate line needs to be done dynamically with

df(yyyy-MMM-dd hh:mm)

Lets say that
df(yyyy-MMM-dd hh:mm) = 2025-Aug-27 14:56
the line would be:

2025-Aug-27 14:56 1.931717 4.066317 131.739723 0.380531

After that I need value 1, 2, 3 and 4 from that line:

1.931717
4.066317
131.739723
0.380531

The numbers change every minute, and I would like use the 4 number from each line and time to make some calculations.

Baju_Santiko · August 28, 2025, 11:06am

Hi James,

Did you mean you need get all the 4 numbers from each row?

JVDH · August 28, 2025, 11:12am

Yes, I need all 4 numbers.
Probably best to put them in 4 different local variables.

Baju_Santiko · August 28, 2025, 11:22am

A question:
Do you need to get just 1 line of data every minute?

JVDH · August 28, 2025, 12:05pm

Yes, I need the values of the 4 numbers that correspond with the actual time. In this case every minute.

JVDH · August 28, 2025, 12:53pm

The frustrating thing is that this regex works on https://regex101.com/ :

2025-Aug-27 14:56\s+(\S+)\s+\S+

and returns Group 1:

1.931717

So I don’t know what is wrong with this code, and why it does not work in KWGT :

$tc(reg, gv(nasatxt), "2025-Aug-27 14:56\s+(\S+)\s+\S+", "$1")$

Baju_Santiko · August 28, 2025, 7:04pm

Hi James,

I have created a Table Parser to parse the nasatxt raw data.

Here is the screencast:
TIPS • Table Parser

NOTES

Add a Text global variable for parsing the raw data, to get the line that match to the current time. Let say its name is: currdata.
Insert this code snippet into the currdata:
$lv(datetime,df("yyyy-MMM-dd hh:mm",gv(datetime))) $$lv(data,tc(reg,tc(reg,gv(nasatxt),"("+#datetime+"[\s\t]*)(([0-9\.]+[\n\r\s\t]*){4})([\n\r\s\t\.\:\-0-9A-Za-z]*)","~$1$2~"),"[\s\t]{2,}","|"))$$if(tc(count,#data,"~")>0,if(tc(split,#data,"~",1)!="",tc(split,#data,"~",1),tc(split,#data,"~",0)))$
Add 5 more Text global variables, to get the data from each columns:
- Date : the date from COLUMN 1
  $if(gv(currdata)!="",if(tc(count,gv(currdata),"|")>=0,tc(split,gv(currdata),"|",0)))$
- Data1 : the number from COLUMN 2
  $if(gv(currdata)!="",if(tc(count,gv(currdata),"|")>0,tc(split,gv(currdata),"|",1)))$
- Data2 : the number from COLUMN 3
  $if(gv(currdata)!="",if(tc(count,gv(currdata),"|")>1,tc(split,gv(currdata),"|",2)))$
- Data3 : the number from COLUMN 4
  $if(gv(currdata)!="",if(tc(count,gv(currdata),"|")>2,tc(split,gv(currdata),"|",3)))$
- Data4 : the number from COLUMN 5
  $if(gv(currdata)!="",if(tc(count,gv(currdata),"|")>3,tc(split,gv(currdata),"|",4)))$

JVDH · August 28, 2025, 7:28pm

Impressive!
Do you mind sharing how you achieved this?

Baju_Santiko · August 28, 2025, 11:58pm

Hi James,

I updated my reply above for additional details, please take a look.

I hope this would help.
Did you find the most helpful response to your issue?
Please kindly press the Solution button to mark it as the solution.
This would help others with the similar issue.

RonC · August 29, 2025, 9:04am

Holy ****, dude, you are so good with these things and so fast!
Hats off to you!

Baju_Santiko · August 29, 2025, 9:08am

Hi Ron,

I just tried to help as far as I know the solution.

JVDH · August 29, 2025, 9:40am

I totally agree with RonC!

Unfortunately, the text preceeding the line(s) I need trips up the regex. My result is:

2013|

I’m pulling the full version of ‘nasatxt’ from here:

https://ssd.jpl.nasa.gov/api/horizons.api?format=text&COMMAND=%27301%27&OBJ_DATA=%27YES%27&MAKE_EPHEM=%27YES%27&EPHEM_TYPE=%27OBSERVER%27&CENTER=%27500@399%27&START_TIME=%272025-08-29%27&STOP_TIME=%272025-08-30%27&STEP_SIZE=%271%20m%27&QUANTITIES=%2714,15%27

The flow in KWGT, to update the global variable ‘nasatxt’, runs once a day with:

https://ssd.jpl.nasa.gov/api/horizons.api?format=text&COMMAND=%27301%27&OBJ_DATA=%27YES%27&MAKE_EPHEM=%27YES%27&EPHEM_TYPE=%27OBSERVER%27&CENTER=%27500@399%27&START_TIME=%27$df(yyyy-MM-dd)$%27&STOP_TIME=%27$df(yyyy-MM-dd, a1d)$%27&STEP_SIZE=%271%20m%27&QUANTITIES=%2714,15%27

I tried CrapGPT to update your code, but it screwed up again…

Baju_Santiko · August 29, 2025, 9:54am

Hi James,

So, the raw data is updated on daily basis and not hourly basis?

Hmm…

JVDH · August 29, 2025, 10:05am

The START_TIME to STOP_TIME of the API pull is a full day, and the STEP_SIZE is 1 minute. That could be changed to whatever is needed. But it will not make the junk info at the start of the text go away.

Baju_Santiko · August 29, 2025, 11:42am

Hi James,

Since now I see the API’s URL, and I can see the whole/complete raw data, this makes the previous logics slightly changed, specifically the data parsing logics.

Now, here’s the update I have made.

TIPS • Table Parser 002
I intentionally display the second numbers, to show the automatic changes in between minutes.

Global variables

Add a Text global variable for parsing the raw data, to get the line that match to the current time. Let say its name is: currdata.
Insert this code snippet into the currdata:
$lv(datetime,df("yyyy-MMM-dd hh:mm",gv(datetime)))$$lv(data,tc(reg,tc(reg,gv(nasatxt),"("+#datetime+"[\s\t]*)"+"(([0-9\.]+[\s\t]*){4})"+"([\s\t\.\:\-0-9A-Za-z]+)","~$1$2~"), "[\s\t]+", "|"))$$if(tc(count, #data, "~")>0,if(tc(split, #data, "~", 1)!="",tc(split, #data, "~", 1),tc(split, #data, "~", 0)))$
Add 5 more Text global variables, to get the data from each columns:
- Date : the date from COLUMN 1
  $if(gv(currdata)!="",if(tc(count,gv(currdata),"|")>0,tc(split,gv(currdata),"|",0)+" "+tc(split,gv(currdata),"|",1)))$
- Data1 : the number from COLUMN 2
  $if(gv(currdata)!="",if(tc(count,gv(currdata),"|")>2,tc(split,gv(currdata),"|",2)))$
- Data2 : the number from COLUMN 3
  $if(gv(currdata)!="",if(tc(count,gv(currdata),"|")>3,tc(split,gv(currdata),"|",3)))$
- Data3 : the number from COLUMN 4
  $if(gv(currdata)!="",if(tc(count,gv(currdata),"|")>4,tc(split,gv(currdata),"|",4)))$
- Data4 : the number from COLUMN 5
  $if(gv(currdata)!="",if(tc(count,gv(currdata),"|")>5,tc(split,gv(currdata),"|",5)))$

The flow

The flow is used to retrieve the raw data from an external text file, in this example: Horizons.txt

Currently, the flow trigger is only Manual. You can add a Cron trigger to periodically retrieve the raw data from the updated text file.

Here is a screenshot that show the list of global variables that is used.

I hope this would help.

Baju_Santiko · August 29, 2025, 12:26pm

Hi James (@JVDH ),

I made a small update, to make the raw data updated directly from the API URL. So, there is no external text file needed for the raw data.

Add a global variable for the API URL, let say its name is: apiurl.

The value of this apiurl variable is:

https://ssd.jpl.nasa.gov/api/horizons.api?format=text&COMMAND=%27301%27&OBJ_DATA=%27YES%27&MAKE_EPHEM=%27YES%27&EPHEM_TYPE=%27OBSERVER%27&CENTER=%27500@399%27&START_TIME=%27$df("yyyy-MM-dd",gv(datetime))$%27&STOP_TIME=%27$df("yyyy-MM-dd",gv(datetime)+"a1d")$%27&STEP_SIZE=%271%20m%27&QUANTITIES=%2714,15%27

The flow

I hope this would help.

JVDH · August 29, 2025, 9:44pm

Oh Boy!
Never thought something seemingly so trivial and straightforward would turn into this!

I had to do some tweaks to make it work on my end.
This is the complete workflow I used:

Create a gv(apiurl) containing the API site and correct dates to call:

https://ssd.jpl.nasa.gov/api/horizons.api?format=text&COMMAND=%27301%27&OBJ_DATA=%27YES%27&MAKE_EPHEM=%27YES%27&EPHEM_TYPE=%27OBSERVER%27&CENTER=%27500@399%27&START_TIME=%27$df(yyyy-MM-dd)$%27&STOP_TIME=%27$df(yyyy-MM-dd, a1d)$%27&STEP_SIZE=%271%20m%27&QUANTITIES=%2714,15%27

Create a Flow (CleanNasa) to fetch the content of the gv(apiurl):
Formula

$wg(gv(apiurl), txt)$

strip everything after ‘$$EOE’ (included):
Formula

$tc(split,#last,"$$ЕОЕ", 0)$

strip everything before ‘$$SOE’ (included):
Formula

$tc(split,#last,"$$SOE", 1)$

and put what’s left in gv(nasatxt):
Set Global Var

Global: nasatxt, store mode: AUTO

Create a gv(currdata) containing the actual time and corresponding 4 numbers extracted from gv(nasatxt):

$lv(datetime,df(yyyy-MMM-dd hh:mm))$$lv(data,tc(reg,tc(reg,gv(nasatxt),"("+#datetime+"[\s\t]*)"+"(([0-9\.]+[\s\t]*){4})"+"([\s\t\.\:\-0-9A-Za-z]+)","~$1$2~"), "[\s\t]+", "|"))$$if(tc(count, #data, "~")>0,if(tc(split, #data, "~", 1)!="",tc(split, #data, "~", 1),tc(split, #data, "~", 0)))$

Create four additional global variables to capture the 4 numbers:

gv(MSubLon)

$if(gv(currdata)!="",if(tc(count,gv(currdata),"|")>2,tc(split,gv(currdata),"|",2)))$

gv(MSubLat)

$if(gv(currdata)!="",if(tc(count,gv(currdata),"|")>3,tc(split,gv(currdata),"|",3)))$

gv(SSubLon)

$if(gv(currdata)!="",if(tc(count,gv(currdata),"|")>4,tc(split,gv(currdata),"|",4)))$

gv(SSubLon)

$if(gv(currdata)!="",if(tc(count,gv(currdata),"|")>5,tc(split,gv(currdata),"|",5)))$

Now, by no means I claim this to be my solution…

Thank you so much @Baju_Santiko for the effort and time you put in! This would have never worked without you!
Let me know how I can get you some beers