borb-pdf-corpus

This repository contains a curated corpus of PDF documents and their extracted content, organized to support document analysis, processing, and duplication detection workflows. Each PDF is accompanied by its full text (txt/), a first-page extract (first-page-pdf/ and first-page-txt/), and a corresponding SHA-256 digest (digest/) for efficient duplication checks.

---
config:
theme: default
---
graph TD
pdf
pdf --> txt
pdf --> digest
pdf --> first-page
first-page --> first-page-pdf
first-page --> first-page-txt

%% Define classes
classDef gray fill:#ccc,stroke:#999,stroke-width:1px;
classDef highlight fill:#F1CD2E,stroke:#999,stroke-width:2px;

%% Assign classes
class pdf highlight;
class txt,digest,first-page,first-page-pdf,first-page-txt gray;

Loading

The repository also includes automated metrics to help understand the overall structure, size, and temporal distribution of the documents.

1. File Size

Property	Value
Smallest PDF	2.00 KB
Average PDF	1.48 MB
Largest PDF	55.19 MB

2. Creation Year

Property	Value
Youngest PDF	2025
Average PDF	2015
Oldest PDF	1999

3. Word Count

Property	Value
Largest PDF	346574
Average PDF	6682
Smallest PDF	14