Struct PmcCloudClient 

pub struct PmcCloudClient { /* private fields */ }

Download client for PMC Open Access articles via the PMC OA Cloud (AWS S3).

Fetches an article’s full-text XML, media, and supplementary files as individual per-article objects from the pmc-oa-opendata S3 bucket. This replaces the retired PMC FTP service and its legacy oa_package tar.gz bundles (removed by NCBI in August 2026).

Implementations

impl PmcCloudClient

pub fn new(config: ClientConfig) -> Self

Create a new PMC OA Cloud client with configuration

pub async fn download_files<P: AsRef<Path>>( &self, pmcid: &str, output_dir: P, ) -> Result<Vec<String>>

Download a PMC article’s files from the PMC OA Cloud (AWS S3) service.

NCBI retired the PMC FTP service and the legacy oa_package tar.gz bundles (August 2026). This downloads each of the article’s files (full-text XML, media, supplementary materials, PDF, etc.) individually from the pmc-oa-opendata S3 bucket into output_dir.

Arguments

• pmcid - PMC ID (with or without “PMC” prefix)
• output_dir - Directory to download the article’s files into

Returns

Returns a Result<Vec<String>> containing the list of downloaded file paths

Errors

• ParseError::InvalidPmid - If the PMCID format is invalid
• PubMedError::RequestError - If the HTTP request fails
• ParseError::IoError - If file operations fail
• ParseError::PmcNotAvailable - If the article is not available in the OA Cloud

Example

use pubmed_client::pmc::cloud::PmcCloudClient;
use pubmed_client::ClientConfig;
use std::path::Path;

#[tokio::main]
async fn main() -> Result<(), Box<dyn std::error::Error>> {
    let config = ClientConfig::new();
    let client = PmcCloudClient::new(config);
    let output_dir = Path::new("./extracted_articles");
    let files = client.download_files("PMC7906746", output_dir).await?;

    for file in files {
        println!("Downloaded: {}", file);
    }
    Ok(())
}

pub async fn extract_figures_with_captions<P: AsRef<Path>>( &self, pmcid: &str, output_dir: P, ) -> Result<Vec<ExtractedFigure>>

Download the article’s files and match figures with their captions from XML

Arguments

• pmcid - PMC ID (with or without “PMC” prefix)
• output_dir - Directory to download the article’s files into

Returns

Returns a Result<Vec<ExtractedFigure>> containing figures with both XML metadata and file paths

Errors

• ParseError::InvalidPmid - If the PMCID format is invalid
• PubMedError::RequestError - If the HTTP request fails
• ParseError::IoError - If file operations fail
• ParseError::PmcNotAvailable - If the article is not available in OA

Example

use pubmed_client::pmc::cloud::PmcCloudClient;
use pubmed_client::ClientConfig;
use std::path::Path;

#[tokio::main]
async fn main() -> Result<(), Box<dyn std::error::Error>> {
    let config = ClientConfig::new();
    let client = PmcCloudClient::new(config);
    let output_dir = Path::new("./extracted_articles");
    let figures = client.extract_figures_with_captions("PMC7906746", output_dir).await?;

    for figure in figures {
        println!("Figure {}: {:?}", figure.figure.id, figure.figure.caption);
        println!("File: {}", figure.extracted_file_path);
    }
    Ok(())
}

pub fn find_matching_file( figure: &Figure, extracted_files: &[String], image_extensions: &[&str], ) -> Option<String>

Find a matching file for a figure based on ID, label, or filename patterns.

Three rules are tried in order, returning the first extracted file that matches:

1. the explicit graphic_href (case-sensitive substring of the file name, any extension);
2. the figure id (case-insensitive substring) with an image extension;
3. the figure label with whitespace/dots stripped (case-insensitive) with an image extension.

Trait Implementations

impl Clone for PmcCloudClient

fn clone(&self) -> PmcCloudClient

Returns a duplicate of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.