Help on parsing pdf text

Help Studio
, ,
1

Hi all,

See text below. I need to break this string into 3 groups each starting with CHAIN: because each group pertains to a different reporting division and dates. Thoughts? This is in a datatable in Studio.

Column1
CHAIN: ODS390-001 DUMMY COMPANY WIRE DATE: 04/11/23
“04/11 02:32 04/11 POS 1,858,080.72 VISA & MASTERCARD 1,602,745.01”
CHAIN: ODS390-001 DUMMY COMPANY WIRE DATE: 04/11/23
04/10 06:32 04/10 POS 839.62 VISA & MASTERCARD 829.62
“CARD FEES 28,003.59 DB”
BILLING SYSTEM FEES: .00 DB
NET CHARGEBACKS: 336.76 CR
PROCESSING ADJUSTMENTS: .00 DB
CHAIN: ODS390-002 DUMMYCOMPANY WIRE DATE: 04/11/23
“04/10 22:49 04/10 EMD 840,764.18 VISA & MASTERCARD 817,939.83”
CHAIN: 0DS5390-002 DUMMY COMPANY WIRE DATE: 04/11/23
“CARD FEES 14,118.07 DB”
BILLING SYSTEM FEES: .00 DB
“NET CHARGEBACKS: 3,626.40 CR”
PROCESSING ADJUSTMENTS : .00 DB
CHAIN: ODS390-004 DUMMY COMPANY WIRE DATE: 04/11/23
“04/11 00:34 04/11 POS 3,498.27 VISA & MASTERCARD 1,628.69”
“04/11 02:32 04/11 POS 28,216.17 VISA & MASTERCARD 19,174.75”
CHAIN: ODS390-004 DUMMY COMPANY WIRE DATE: 04/11/23
CARD FEES 342.89 DB
BILLING SYSTEM FEES: .00 DB
NET CHARGEBACKS: .00 DB
PROCESSING ADJUSTMENTS: .00 DB
CHAIN: ODS390 DUMMY COMPANY WIRE DATE: 04/11/23

2

have a look here for an initial segmentation:

grafik
grafik756×749 65.5 KB 

(CHAIN\:)[\s\S]+?(?=CHAIN\:|\z)

Some descriptions are unclear

better provide input text as textfile along with expected output sample

3

Thanks my friend. I tried this already but that only gives me the lines starting with CHAIN: I am trying to segment the groups that sit between each CHAIN: key word into different blocks of text. I appreciate the idea though.

4

Hi @Chris_Bolin

Did you try Splitting using “CHAIN:”

That’ll give you text in array which would be splitted using CHAIN:

and then you can apply your other operations.

Thanks,
Aditya

5

if I split on CHAIN: it will give me what is after “CHAIN” but only on the same line. It does not give me the next row of data. What I am attempting to do now is count how many times I find a key word, putting that to an array of Int and then looping through that array of int to split the text into group parts. Since I do not know how many times this new key word will appear in any given document, I created a counter beginning at 1 and in a do while I am splitting, looking for the data I need and then increasing a counter on the split until it reaches the counter for the number of times the word is found. It is just really code heavy which I do not like. If anyone has any other ideas hit me up

6

@Chris_Bolin
we can better help when you give the clear expected output

is too vague

was requested as we can reliable refer to linebreak for the prototypes

7

So isn’t this whole data kept at DT(0)(0)? Instead it is spread accross several rows?

If it is spread accross several rows, you can append all the data in a single string and then apply the split function.

Regards,
Aditya

8

it is in a string - after grouping I will generate a dt and filter that dt for the information I need.

9

the pdf is a highly structured pdf that I had to read with Tessaract OCR to get it into one string. There are potentially over 200 + lines that I ingest into the code. I do not want to write an append for each line. I am looking to group the data between the Key Word CHAIN: which seems to be a break in the data sets. Hope that helps understandng

10

@Chris_Bolin you can try the below regex, it will extract you the only the text without the word chain

(?<=CHAIN:)[\s\S]+?(?=CHAIN:|\z)

image
image641×596 71.6 KB

11

I found a solution myself

12

This topic was automatically closed 3 days after the last reply. New replies are no longer allowed.