medical_parser

This Rust library provides a simple XML-like parser for medical patient records. It extracts structured information about patients, their personal data, and visit histories, using the pest parsing framework.

Brief Description

• Parses a tag-based XML dialect containing <patient>, <name>, <age>, and optional <visit> sections.
• Produces strongly typed Patient and Visit records that can be consumed programmatically.
• Provides CLI summaries reporting patient counts, visit diagnoses, temperatures, and notes.

Parsing Overview

A XML document consists of one or more <patient> elements. Each patient must define a <name> and <age> tag; visits are optional and may appear multiple times. Every visit is built from optional <diagnosis>, <temperature>, and <notes> tags.

Step-by-Step Flow

XML file ──► parse medical document ──► Patient & Visit structures ──► CLI summary & analytics helpers

1. The CLI reads input from a file path.
2. pest parses the XML grammar described below.
3. Parsed pairs are transformed into Patient and Visit structures.
4. Library consumers operate on the typed structures; the CLI prints neatly formatted summaries.

Full Grammar (src/grammar.pest)

file = { SOI ~ patient+ ~ EOI }
patient = { "<patient>" ~ name ~ age ~ visit* ~ "</patient>" }
visit = { "<visit>" ~ (diagnosis | temperature | notes)* ~ "</visit>" }
name = { "<name>" ~ text ~ "</name>" }
age = { "<age>" ~ number ~ "</age>" }
diagnosis = { "<diagnosis>" ~ text ~ "</diagnosis>" }
temperature = { "<temperature>" ~ number ~ "</temperature>" }
notes = { "<notes>" ~ text ~ "</notes>" }
text = @{ (!"<" ~ ANY)+ }
number = @{ "-"? ~ ASCII_DIGIT+ ~ ("." ~ ASCII_DIGIT+)? }
WHITESPACE = _{ " " | "\t" | "\r" | "\n" }

Library API Highlights

• parse_medical_document converts raw XML into typed data.
• Patient and Visit structs expose parsed fields for further analytics.
• Helper functions (e.g., print_patients, count_patients) support quick reporting.
• Errors use thiserror for library code and anyhow in tests to keep diagnostics ergonomic.

Credits

Author: Anna Rechkalova (NaUKMA SE-4) with the support of the Ukrainian Rust community.