Skip to content

README.md

Latest commit

 

History

History
85 lines (57 loc) · 2.11 KB

README.md

File metadata and controls

85 lines (57 loc) · 2.11 KB
Raw

Scientific Paper Generator (SPG)

Scientific Paper Generator is a small command-line helper that turns a guided questionnaire into a standard LaTeX draft of a scientific paper. It asks for the core pieces of a manuscript (title, authors, abstract, introduction, methods, results, discussion, conclusions, and acknowledgments), places the answers in the right sections, and optionally compiles the result to PDF.

The goal is not to replace the author. SPG creates a structured first draft so that roughly three quarters of the paper is already organized; the remaining work is to connect the text naturally, add figures/tables/references, and check every scientific detail.

Features

  • Interactive questionnaire for a complete draft-paper structure.
  • draft paper format that writes output/draft.tex.
  • Optional PDF generation when pdflatex is installed.
  • JSON answer-file support for repeatable, non-interactive generation.
  • Sample answer generation for testing or as a starting point.

Dependencies

Required:

  • Python 3.10 or newer.

Optional, for PDF output:

  • A LaTeX distribution that provides pdflatex (for example TeX Live).

Usage

Create a paper interactively:

make draft

or run the script directly:

python3 spg.py draft

The generated LaTeX file is saved as:

output/draft.tex

If pdflatex is available, SPG also creates:

output/draft.pdf

Non-interactive usage

Create a sample answers file:

python3 spg.py --sample-answers output/sample_answers.json

Generate a draft from a JSON answers file without compiling a PDF:

python3 spg.py draft --answers output/sample_answers.json --tex-only

Generate a draft from a JSON answers file and compile the PDF if pdflatex is installed:

python3 spg.py draft --answers output/sample_answers.json

Development status

SPG currently supports one format: draft. Future improvements can add journal- specific templates, figure/reference management, richer validation, and optional Natural Language Processing support to improve the generated prose.

Suggestions and improvements are welcome.