Extract Tables from PDF - not fixed columns

(0) Share

Report

Posted on by CU01040934-0

I have a PAD flow created to extract values from Tables in a PDF file using the action Extract Tables from PDF. Though it is a PDF table and fixed position, the extraction table columns and roles are not fixed when returning. As a result, it is always a challenge to read the desired value from the data table for further data manipulation. Is there a better way to resolve this?

These are the two tables i extract from the PDF file. The columns and rows in the extraction result table at times varies.

Thanks in advance

P20

Categories:

Power Automate Desktop

I have the same question (0)

All responses (2)

Answers (0)

Suggested answer

VishnuReddy1997 2,656 Super User 2025 Season 2 on at

Like (0)

Report

Hi @CU01040934-0,

If the column names are same try to Retrieve the Required column Data into list as shown in below.

If the column names are also different then you need you either AiBuilder or Azure Document Intelligence.

Regards,

Vishnu Reddy

Was this reply helpful? Yes No
Suggested answer

Riyaz_riz11 3,893 Super User 2025 Season 2 on at

Like (0)

Report
Hi,

Solution 1: Text-Based Extraction with RegEx (Recommended)

Instead of table extraction, use text extraction with pattern matching:

// Extract text from PDF PDFActions.ExtractTextFromPDF File: PDFFile ExtractedText=> ExtractedText // Use RegEx to find specific patterns Text.ParseTextWithRegularExpression.ParseForFirstMatch Text: %ExtractedText% RegularExpressionPattern: '(\d{1,2},\d{3}\s+P)\s+(\d{1,3},\d{3})' IgnoreCase: True Match=> CoasterMatch // Extract specific values using known patterns Text.ParseTextWithRegularExpression.ParseForFirstMatch Text: %ExtractedText% RegularExpressionPattern: '11,000\s+P\s+(\d{1,3},\d{3})\s+(\d{1,3},\d{3})' IgnoreCase: True Match=> ElevenThousandMatch // Access extracted values SET EastValue TO %ElevenThousandMatch.Groups[1].Value% SET WestValue TO %ElevenThousandMatch.Groups[2].Value%

Solution 2: Smart Table Processing with Headers

Process the extracted table by finding headers first:

// After extracting table PDFActions.ExtractTablesFromPDF File: PDFFile ExtractedTables=> ExtractedTables // Find header row containing "East" and "West" LOOP FOREACH CurrentRow IN %ExtractedTables[0]% IF %CurrentRow% CONTAINS 'East' AND %CurrentRow% CONTAINS 'West' THEN SET HeaderRowIndex TO %LoopIndex% // Find column positions LOOP FOREACH CurrentCell IN %CurrentRow% IF %CurrentCell% = 'East' THEN SET EastColumnIndex TO %LoopIndex% END IF IF %CurrentCell% = 'West' THEN SET WestColumnIndex TO %LoopIndex% END IF END LOOP EXIT LOOP END IF END LOOP // Now extract data using found positions SET EastValue TO %ExtractedTables[0][HeaderRowIndex + 1][EastColumnIndex]% SET WestValue TO %ExtractedTables[0][HeaderRowIndex + 1][WestColumnIndex]%

If I have answered your question, please mark it as the preferred solution ✅ . If you like my response, please give it a Thumbs Up 👍.
Regards,
Riyaz

Was this reply helpful? Yes No