Document in (Un)Perplexed Spready — Process Documents Without Leaving the Application

Introduction

(Un)Perplexed Spready is a spreadsheet application that combines the power of traditional spreadsheets with advanced artificial intelligence capabilities. Version 1.2.X represents the biggest leap in the application's history, bringing a range of features that elevate data work to an entirely new level.

This version introduces four major improvement areas: Web Search, Vision & Document Processing, SQL Query, and Advanced Statistics.
Best of all, SQL Query and Advanced Statistics functions are completely free! If you are looking for a free statistical software, you are on the right page!

In a business environment, data often comes in various formats — PDF reports, Word documents, Excel spreadsheets, presentations, web pages... Traditionally, each format requires a different tool for processing.

(Un)Perplexed Spready v1.2.X introduces Document functionality — the ability to load various document formats directly into AI context, without leaving the application.

How It Works

Document functionality allows you to specify the path to a document (local or URL) in a formula, and the AI processes the content and returns the result to the cell.

Basic syntax:

=ASK_LOCAL1("/path/to/document.pdf", "instruction")
=ASK_LOCAL1("https://example.com/document.docx", "instruction")

Supported Formats

Text Formats

Format	Extensions	Notes
Text	.txt	Plain text
Markdown	.md	Formatted text
CSV	.csv, .tsv	Data separated by delimiter
SQL	.sql	SQL scripts
YAML	.yaml, .yml	Configuration files
Code	.py, .js, .pas, .sh, etc.	Source code

Web Formats

Format	Extensions
HTML	.html, .htm, .xhtml

Documents

Format	Extensions	Notes
PDF	.pdf	Universal document format
Word	.docx, .doc, .rtf	Microsoft Word documents
OpenDocument	.odt	LibreOffice/OpenOffice documents
EPUB	.epub	E-books

Spreadsheets

Format	Extensions	Notes
Excel	.xlsx, .xlsm, .xls	Microsoft Excel
OpenDocument	.ods	LibreOffice/OpenOffice Calc

Presentations

Format	Extensions
PowerPoint	.pptx, .ppt

Data Formats

Format	Extensions
XML	.xml
JSON	.json, .geojson

Usage Methods

1. Local Documents

Documents stored on your computer:

=ASK_LOCAL1("/home/user/Documents/report.pdf", "Summarize main points")
=ASK_LOCAL1("C:\Users\User\Documents\contract.docx", "Extract key provisions")
=ASK_LOCAL1("./data/data.xlsx", "Analyze data structure")

2. Online Documents (HTTP/HTTPS URLs)

Documents from the internet:

=ASK_LOCAL1("https://example.com/report.pdf", "Summarize this report")
=ASK_LOCAL1("https://example.com/data.json", "Describe the structure of this data")

3. Combining with Other Data

=ASK_LOCAL2(A2, "/path/to/specs.pdf", "Check if product in A2 complies with specifications")
=ASK_LOCAL3(A2, B2, "https://example.com/regulations.pdf", "Check compliance with regulations")

Practical Applications

PDF Report Analysis

=ASK_LOCAL1("/reports/annual_report_2025.pdf", "Summarize financial results")
=ASK_LOCAL1("/reports/market_analysis.pdf", "Extract key trends and figures")

Contract Data Extraction

=ASK_LOCAL1("/contracts/contract_001.docx", "Extract: parties, date, subject, amount")
=ASK_LOCAL1("/contracts/NDA.pdf", "List confidentiality clauses")

Word Document Analysis

=ASK_LOCAL1("/proposals/proposal.docx", "Summarize main points of the proposal")
=ASK_LOCAL1("/policies/company_policy.docx", "Extract key guidelines")

Excel Spreadsheet Reading

=ASK_LOCAL1("/data/sales.xlsx", "Describe structure of this table: sheets, columns, row count")
=ASK_LOCAL1("/data/budget.xlsx", "Summarize financial data")

JSON/XML Data Analysis

=ASK_LOCAL1("/api/response.json", "Describe structure of this JSON response")
=ASK_LOCAL1("/config/settings.xml", "Extract configuration values")

Web Page Reading

=ASK_LOCAL1("https://example.com/article.html", "Summarize this article")
=ASK_LOCAL1("https://docs.example.com/guide.html", "Extract main instructions")

Cross-Sheet Range with Documents

You can combine documents with range data from other sheets:

=ASK_LOCAL2("/products/specs.pdf", "RANGE:Products!A2:A50", "Check if all products comply with specifications")

Limitations

Document Size

Maximum length: 50,000 characters
Larger documents are truncated (first 50,000 characters)
For very large documents, consider splitting into smaller parts

PDF Specifics

Text-based PDF: Fully supported
Scanned PDF (images): May require OCR; better to use Vision functionality

Online Documents

Requires internet connection
Possible timeout for large files

Configuration

Required Models

Document functionality works with standard LLM models (does not require special vision models).

Recommended models:

deepseek-v3 — Excellent for document analysis
qwen2.5 — Good balance of speed and quality
llama3.1 — Fast and efficient

Installing models (Ollama):

ollama pull deepseek-v3
ollama pull qwen2.5

Real-World Examples

Legal Documentation

You have a folder with contracts:

Cell A1: =ASK_LOCAL1("/contracts/001.pdf", "Parties")
Cell B1: =ASK_LOCAL1("/contracts/001.pdf", "Date")
Cell C1: =ASK_LOCAL1("/contracts/001.pdf", "Amount")

Regulatory Compliance

=ASK_LOCAL2("/regulations/GDPR.pdf", "RANGE:Processes!A2:A20", "Check compliance of processes with GDPR regulations")

Data Preprocessing

=ASK_LOCAL1("/raw_data.json", "Convert this JSON to tabular format: columns, values")
=ASK_LOCAL1("/export.csv", "Analyze structure and suggest data cleaning")

Tips for Best Results

Be specific — Specify exactly what you want to extract from the document
Control the format — Request JSON, CSV, or tabular output format
Use portions — For large documents, focus on specific sections
Validate results — Always check AI extraction against original document
Choose the right model — deepseek-v3 is excellent for complex documents

Error Messages

Error	Meaning
FILE NOT FOUND	Document does not exist at the specified path
UNSUPPORTED FORMAT	Document format is not supported
APIERR	Model not available
DOCUMENT TOO LARGE	Document exceeds 50,000 characters

Privacy and Security

Local models: Documents stay on your computer
Remote models: Documents are sent to remote server
Temporary: Document content is not permanently stored

Conclusion

Document functionality in (Un)Perplexed Spready simplifies working with various document formats. Instead of opening multiple applications, you can process PDFs, Word documents, Excel spreadsheets, and other formats directly from your spreadsheet.

Whether you're analyzing reports, extracting data from contracts, or processing data files — Document formulas give you the power of AI analysis at your fingertips.

Document is a PREMIUM feature. Requires LLM model (local or cloud).

Home

(Un)Perplexed Spready v1.2.X — Process Documents in Spreadsheets