- feat: Enhance README with setup instructions, usage, and command documentation

- fix: Update API routes to include DOI URL handling and improve route organization

- chore: Add ORCID preload rule file and ensure proper registration

- docs: Add MIT License to the project for open-source compliance

- feat: Implement command to detect and fix missing dataset cross-references

- feat: Create command for updating DataCite DOI records with detailed logging and error handling

- docs: Add comprehensive documentation for dataset indexing command

- docs: Create detailed documentation for DataCite update command with usage examples and error handling
This commit is contained in:
Kaimbacher 2025-09-19 14:35:23 +02:00
commit c049b22723
11 changed files with 2187 additions and 555 deletions

View file

@ -1,6 +1,3 @@
// import { Client } from 'guzzle';
// import { Log } from '@adonisjs/core/build/standalone';
// import { DoiInterface } from './interfaces/DoiInterface';
import DoiClientContract from '#app/Library/Doi/DoiClientContract';
import DoiClientException from '#app/exceptions/DoiClientException';
import { StatusCodes } from 'http-status-codes';
@ -12,14 +9,14 @@ export class DoiClient implements DoiClientContract {
public username: string;
public password: string;
public serviceUrl: string;
public apiUrl: string;
constructor() {
// const datacite_environment = process.env.DATACITE_ENVIRONMENT || 'debug';
this.username = process.env.DATACITE_USERNAME || '';
this.password = process.env.DATACITE_PASSWORD || '';
this.serviceUrl = process.env.DATACITE_SERVICE_URL || '';
// this.prefix = process.env.DATACITE_PREFIX || '';
// this.base_domain = process.env.BASE_DOMAIN || '';
this.apiUrl = process.env.DATACITE_API_URL || 'https://api.datacite.org';
if (this.username === '' || this.password === '' || this.serviceUrl === '') {
const message = 'issing configuration settings to properly initialize DOI client';
@ -90,4 +87,240 @@ export class DoiClient implements DoiClientContract {
throw new DoiClientException(error.response.status, error.response.data);
}
}
/**
* Retrieves DOI information from DataCite REST API
*
* @param doiValue The DOI identifier e.g. '10.5072/tethys.999'
* @returns Promise with DOI information or null if not found
*/
public async getDoiInfo(doiValue: string): Promise<any | null> {
try {
// Use configurable DataCite REST API URL
const dataciteApiUrl = `${this.apiUrl}/dois/${doiValue}`;
const response = await axios.get(dataciteApiUrl, {
headers: {
Accept: 'application/vnd.api+json',
},
});
if (response.status === 200 && response.data.data) {
return {
created: response.data.data.attributes.created,
registered: response.data.data.attributes.registered,
updated: response.data.data.attributes.updated,
published: response.data.data.attributes.published,
state: response.data.data.attributes.state,
url: response.data.data.attributes.url,
metadata: response.data.data.attributes,
};
}
} catch (error) {
if (error.response?.status === 404) {
logger.debug(`DOI ${doiValue} not found in DataCite`);
return null;
}
logger.debug(`DataCite REST API failed for ${doiValue}: ${error.message}`);
// Fallback to MDS API
return await this.getDoiInfoFromMds(doiValue);
}
return null;
}
/**
* Fallback method to get DOI info from MDS API
*
* @param doiValue The DOI identifier
* @returns Promise with basic DOI information or null
*/
private async getDoiInfoFromMds(doiValue: string): Promise<any | null> {
try {
const auth = {
username: this.username,
password: this.password,
};
// Get DOI URL
const doiResponse = await axios.get(`${this.serviceUrl}/doi/${doiValue}`, { auth });
if (doiResponse.status === 200) {
// Get metadata if available
try {
const metadataResponse = await axios.get(`${this.serviceUrl}/metadata/${doiValue}`, {
auth,
headers: {
Accept: 'application/xml',
},
});
return {
url: doiResponse.data.trim(),
metadata: metadataResponse.data,
created: new Date().toISOString(), // MDS doesn't provide creation dates
registered: new Date().toISOString(), // Use current time as fallback
source: 'mds',
};
} catch (metadataError) {
// Return basic info even if metadata fetch fails
return {
url: doiResponse.data.trim(),
created: new Date().toISOString(),
registered: new Date().toISOString(),
source: 'mds',
};
}
}
} catch (error) {
if (error.response?.status === 404) {
logger.debug(`DOI ${doiValue} not found in DataCite MDS`);
return null;
}
logger.debug(`DataCite MDS API failed for ${doiValue}: ${error.message}`);
}
return null;
}
/**
* Checks if a DOI exists in DataCite
*
* @param doiValue The DOI identifier
* @returns Promise<boolean> True if DOI exists
*/
public async doiExists(doiValue: string): Promise<boolean> {
const doiInfo = await this.getDoiInfo(doiValue);
return doiInfo !== null;
}
/**
* Gets the last modification date of a DOI
*
* @param doiValue The DOI identifier
* @returns Promise<Date | null> Last modification date or creation date if never updated, null if not found
*/
public async getDoiLastModified(doiValue: string): Promise<Date | null> {
const doiInfo = await this.getDoiInfo(doiValue);
if (doiInfo) {
// Use updated date if available, otherwise fall back to created/registered date
const dateToUse = doiInfo.updated || doiInfo.registered || doiInfo.created;
if (dateToUse) {
logger.debug(
`DOI ${doiValue}: Using ${doiInfo.updated ? 'updated' : doiInfo.registered ? 'registered' : 'created'} date: ${dateToUse}`,
);
return new Date(dateToUse);
}
}
return null;
}
/**
* Makes a DOI unfindable (registered but not discoverable)
* Note: DOIs cannot be deleted, only made unfindable
* await doiClient.makeDoiUnfindable('10.21388/tethys.231');
*
* @param doiValue The DOI identifier e.g. '10.5072/tethys.999'
* @returns Promise<AxiosResponse<any>> The http response
*/
public async makeDoiUnfindable(doiValue: string): Promise<AxiosResponse<any>> {
const auth = {
username: this.username,
password: this.password,
};
try {
// First, check if DOI exists
const exists = await this.doiExists(doiValue);
if (!exists) {
throw new DoiClientException(404, `DOI ${doiValue} not found`);
}
// Delete the DOI URL mapping to make it unfindable
// This removes the URL but keeps the metadata registered
const response = await axios.delete(`${this.serviceUrl}/doi/${doiValue}`, { auth });
// Response Codes for DELETE /doi/{doi}
// 200 OK: operation successful
// 401 Unauthorized: no login
// 403 Forbidden: login problem, quota exceeded
// 404 Not Found: DOI does not exist
if (response.status !== 200) {
const message = `Unexpected DataCite MDS response code ${response.status}`;
logger.error(message);
throw new DoiClientException(response.status, message);
}
logger.info(`DOI ${doiValue} successfully made unfindable`);
return response;
} catch (error) {
logger.error(`Failed to make DOI ${doiValue} unfindable: ${error.message}`);
if (error instanceof DoiClientException) {
throw error;
}
throw new DoiClientException(error.response?.status || 500, error.response?.data || error.message);
}
}
/**
* Makes a DOI findable again by re-registering the URL
* await doiClient.makeDoiFindable(
* '10.21388/tethys.231',
* 'https://doi.dev.tethys.at/10.21388/tethys.231'
* );
*
* @param doiValue The DOI identifier e.g. '10.5072/tethys.999'
* @param landingPageUrl The landing page URL
* @returns Promise<AxiosResponse<any>> The http response
*/
public async makeDoiFindable(doiValue: string, landingPageUrl: string): Promise<AxiosResponse<any>> {
const auth = {
username: this.username,
password: this.password,
};
try {
// Re-register the DOI with its URL to make it findable again
const response = await axios.put(`${this.serviceUrl}/doi/${doiValue}`, `doi=${doiValue}\nurl=${landingPageUrl}`, { auth });
// Response Codes for PUT /doi/{doi}
// 201 Created: operation successful
// 400 Bad Request: request body must be exactly two lines: DOI and URL
// 401 Unauthorized: no login
// 403 Forbidden: login problem, quota exceeded
// 412 Precondition failed: metadata must be uploaded first
if (response.status !== 201) {
const message = `Unexpected DataCite MDS response code ${response.status}`;
logger.error(message);
throw new DoiClientException(response.status, message);
}
logger.info(`DOI ${doiValue} successfully made findable again`);
return response;
} catch (error) {
logger.error(`Failed to make DOI ${doiValue} findable: ${error.message}`);
if (error instanceof DoiClientException) {
throw error;
}
throw new DoiClientException(error.response?.status || 500, error.response?.data || error.message);
}
}
/**
* Gets the current state of a DOI (draft, registered, findable)
* const state = await doiClient.getDoiState('10.21388/tethys.231');
* console.log(`Current state: ${state}`); // 'findable'
*
* @param doiValue The DOI identifier
* @returns Promise<string | null> The DOI state or null if not found
*/
public async getDoiState(doiValue: string): Promise<string | null> {
const doiInfo = await this.getDoiInfo(doiValue);
return doiInfo?.state || null;
}
}