2 * Jalview - A Sequence Alignment Editor and Viewer ($$Version-Rel$$)
3 * Copyright (C) $$Year-Rel$$ The Jalview Authors
5 * This file is part of Jalview.
7 * Jalview is free software: you can redistribute it and/or
8 * modify it under the terms of the GNU General Public License
9 * as published by the Free Software Foundation, either version 3
10 * of the License, or (at your option) any later version.
12 * Jalview is distributed in the hope that it will be useful, but
13 * WITHOUT ANY WARRANTY; without even the implied warranty
14 * of MERCHANTABILITY or FITNESS FOR A PARTICULAR
15 * PURPOSE. See the GNU General Public License for more details.
17 * You should have received a copy of the GNU General Public License
18 * along with Jalview. If not, see <http://www.gnu.org/licenses/>.
19 * The Jalview Authors are detailed in the 'AUTHORS' file.
21 package jalview.ext.ensembl;
23 import jalview.io.DataSourceType;
24 import jalview.io.FileParse;
25 import jalview.util.StringUtils;
27 import java.io.BufferedReader;
28 import java.io.DataOutputStream;
29 import java.io.IOException;
30 import java.io.InputStream;
31 import java.io.InputStreamReader;
32 import java.net.HttpURLConnection;
33 import java.net.MalformedURLException;
34 import java.net.ProtocolException;
36 import java.util.HashMap;
37 import java.util.List;
40 import javax.ws.rs.HttpMethod;
42 import org.json.simple.JSONArray;
43 import org.json.simple.JSONObject;
44 import org.json.simple.parser.JSONParser;
47 * Base class for Ensembl REST service clients
51 abstract class EnsemblRestClient extends EnsemblSequenceFetcher
53 private static final int DEFAULT_READ_TIMEOUT = 5 * 60 * 1000; // 5 minutes
55 private static final int CONNECT_TIMEOUT_MS = 10 * 1000; // 10 seconds
57 private static final int MAX_RETRIES = 3;
59 private static final int HTTP_OK = 200;
61 private static final int HTTP_OVERLOAD = 429;
64 * update these constants when Jalview has been checked / updated for
65 * changes to Ensembl REST API (ref JAL-2105)
66 * @see https://github.com/Ensembl/ensembl-rest/wiki/Change-log
67 * @see http://rest.ensembl.org/info/rest?content-type=application/json
69 private static final String LATEST_ENSEMBLGENOMES_REST_VERSION = "6.0";
71 private static final String LATEST_ENSEMBL_REST_VERSION = "6.1";
73 private static final String REST_CHANGE_LOG = "https://github.com/Ensembl/ensembl-rest/wiki/Change-log";
75 private static Map<String, EnsemblData> domainData;
77 // @see https://github.com/Ensembl/ensembl-rest/wiki/Output-formats
78 private static final String PING_URL = "http://rest.ensembl.org/info/ping.json";
80 private final static long AVAILABILITY_RETEST_INTERVAL = 10000L; // 10 seconds
82 private final static long VERSION_RETEST_INTERVAL = 1000L * 3600; // 1 hr
84 protected static final String CONTENT_TYPE_JSON = "?content-type=application/json";
88 domainData = new HashMap<>();
89 domainData.put(DEFAULT_ENSEMBL_BASEURL,
90 new EnsemblData(DEFAULT_ENSEMBL_BASEURL, LATEST_ENSEMBL_REST_VERSION));
91 domainData.put(DEFAULT_ENSEMBL_GENOMES_BASEURL, new EnsemblData(
92 DEFAULT_ENSEMBL_GENOMES_BASEURL, LATEST_ENSEMBLGENOMES_REST_VERSION));
95 protected volatile boolean inProgress = false;
98 * Default constructor to use rest.ensembl.org
100 public EnsemblRestClient()
105 * initialise domain info lazily
107 if (!domainData.containsKey(ensemblDomain))
109 domainData.put(ensemblDomain,
110 new EnsemblData(ensemblDomain, LATEST_ENSEMBL_REST_VERSION));
112 if (!domainData.containsKey(ensemblGenomesDomain))
114 domainData.put(ensemblGenomesDomain, new EnsemblData(
115 ensemblGenomesDomain, LATEST_ENSEMBLGENOMES_REST_VERSION));
120 * Constructor given the target domain to fetch data from
124 public EnsemblRestClient(String d)
130 public boolean queryInProgress()
136 public StringBuffer getRawRecords()
142 * Returns the URL for the client http request
146 * @throws MalformedURLException
148 protected abstract URL getUrl(List<String> ids)
149 throws MalformedURLException;
152 * Returns true if client uses GET method, false if it uses POST
156 protected abstract boolean useGetRequest();
159 * Return the desired value for the Content-Type request header
164 * @see https://github.com/Ensembl/ensembl-rest/wiki/HTTP-Headers
166 protected abstract String getRequestMimeType(boolean multipleIds);
169 * Return the desired value for the Accept request header
172 * @see https://github.com/Ensembl/ensembl-rest/wiki/HTTP-Headers
174 protected abstract String getResponseMimeType();
177 * Checks Ensembl's REST 'ping' endpoint, and returns true if response
178 * indicates available, else false
180 * @see http://rest.ensembl.org/documentation/info/ping
183 boolean checkEnsembl()
185 BufferedReader br = null;
186 String pingUrl = getDomain() + "/info/ping" + CONTENT_TYPE_JSON;
189 // note this format works for both ensembl and ensemblgenomes
190 // info/ping.json works for ensembl only (March 2016)
191 URL ping = new URL(pingUrl);
194 * expect {"ping":1} if ok
195 * if ping takes more than 2 seconds to respond, treat as if unavailable
197 br = getHttpResponse(ping, null, 2 * 1000);
200 // error reponse status
203 JSONParser jp = new JSONParser();
204 JSONObject val = (JSONObject) jp.parse(br);
205 String pingString = val.get("ping").toString();
206 return pingString != null;
207 } catch (Throwable t)
210 "Error connecting to " + pingUrl + ": " + t.getMessage());
218 } catch (IOException e)
228 * returns a reader to a Fasta response from the Ensembl sequence endpoint
232 * @throws IOException
234 protected FileParse getSequenceReader(List<String> ids) throws IOException
236 URL url = getUrl(ids);
238 BufferedReader reader = getHttpResponse(url, ids);
244 FileParse fp = new FileParse(reader, url.toString(),
250 * Gets a reader to the HTTP response, using the default read timeout of 5
256 * @throws IOException
258 protected BufferedReader getHttpResponse(URL url, List<String> ids)
261 return getHttpResponse(url, ids, DEFAULT_READ_TIMEOUT);
265 * Sends the HTTP request and gets the response as a reader
269 * written as Json POST body if more than one
273 * @throws IOException
274 * if response code was not 200, or other I/O error
276 protected BufferedReader getHttpResponse(URL url, List<String> ids,
277 int readTimeout) throws IOException
279 int retriesLeft = MAX_RETRIES;
280 HttpURLConnection connection = null;
281 int responseCode = 0;
283 while (retriesLeft > 0)
285 connection = tryConnection(url, ids, readTimeout);
286 responseCode = connection.getResponseCode();
287 if (responseCode == HTTP_OVERLOAD) // 429
290 checkRetryAfter(connection);
297 if (responseCode != HTTP_OK) // 200
300 * note: a GET request for an invalid id returns an error code e.g. 415
301 * but POST request returns 200 and an empty Fasta response
303 System.err.println("Response code " + responseCode + " for " + url);
307 InputStream response = connection.getInputStream();
309 // System.out.println(getClass().getName() + " took "
310 // + (System.currentTimeMillis() - now) + "ms to fetch");
312 BufferedReader reader = null;
313 reader = new BufferedReader(new InputStreamReader(response, "UTF-8"));
322 * @throws IOException
323 * @throws ProtocolException
325 protected HttpURLConnection tryConnection(URL url, List<String> ids,
326 int readTimeout) throws IOException, ProtocolException
328 // System.out.println(System.currentTimeMillis() + " " + url);
329 HttpURLConnection connection = (HttpURLConnection) url.openConnection();
332 * POST method allows multiple queries in one request; it is supported for
333 * sequence queries, but not for overlap
335 boolean multipleIds = ids != null && ids.size() > 1;
336 connection.setRequestMethod(
337 multipleIds ? HttpMethod.POST : HttpMethod.GET);
338 connection.setRequestProperty("Content-Type",
339 getRequestMimeType(multipleIds));
340 connection.setRequestProperty("Accept", getResponseMimeType());
342 connection.setUseCaches(false);
343 connection.setDoInput(true);
344 connection.setDoOutput(multipleIds);
346 connection.setConnectTimeout(CONNECT_TIMEOUT_MS);
347 connection.setReadTimeout(readTimeout);
351 writePostBody(connection, ids);
357 * Inspects response headers for a 'retry-after' directive, and waits for the
358 * directed period (if less than 10 seconds)
360 * @see https://github.com/Ensembl/ensembl-rest/wiki/Rate-Limits
363 void checkRetryAfter(HttpURLConnection connection)
365 String retryDelay = connection.getHeaderField("Retry-After");
370 if (retryDelay != null)
374 int retrySecs = Integer.valueOf(retryDelay);
375 if (retrySecs > 0 && retrySecs < 10)
378 .println("Ensembl REST service rate limit exceeded, waiting "
379 + retryDelay + " seconds before retrying");
380 Thread.sleep(1000 * retrySecs);
382 } catch (NumberFormatException | InterruptedException e)
384 System.err.println("Error handling Retry-After: " + e.getMessage());
390 * Rechecks if Ensembl is responding, unless the last check was successful and
391 * the retest interval has not yet elapsed. Returns true if Ensembl is up,
392 * else false. Also retrieves and saves the current version of Ensembl data
393 * and REST services at intervals.
397 protected boolean isEnsemblAvailable()
399 EnsemblData info = domainData.get(getDomain());
401 long now = System.currentTimeMillis();
404 * recheck if Ensembl is up if it was down, or the recheck period has elapsed
406 boolean retestAvailability = (now
407 - info.lastAvailableCheckTime) > AVAILABILITY_RETEST_INTERVAL;
408 if (!info.restAvailable || retestAvailability)
410 info.restAvailable = checkEnsembl();
411 info.lastAvailableCheckTime = now;
415 * refetch Ensembl versions if the recheck period has elapsed
417 boolean refetchVersion = (now
418 - info.lastVersionCheckTime) > VERSION_RETEST_INTERVAL;
421 checkEnsemblRestVersion();
422 checkEnsemblDataVersion();
423 info.lastVersionCheckTime = now;
426 return info.restAvailable;
430 * Constructs, writes and flushes the POST body of the request, containing the
431 * query ids in JSON format
435 * @throws IOException
437 protected void writePostBody(HttpURLConnection connection,
438 List<String> ids) throws IOException
441 StringBuilder postBody = new StringBuilder(64);
442 postBody.append("{\"ids\":[");
444 for (String id : ids)
448 postBody.append(",");
451 postBody.append("\"");
452 postBody.append(id.trim());
453 postBody.append("\"");
455 postBody.append("]}");
456 byte[] thepostbody = postBody.toString().getBytes();
457 connection.setRequestProperty("Content-Length",
458 Integer.toString(thepostbody.length));
459 DataOutputStream wr = new DataOutputStream(
460 connection.getOutputStream());
461 wr.write(thepostbody);
467 * Fetches and checks Ensembl's REST version number
471 private void checkEnsemblRestVersion()
473 EnsemblData info = domainData.get(getDomain());
475 JSONParser jp = new JSONParser();
479 url = new URL(getDomain() + "/info/rest" + CONTENT_TYPE_JSON);
480 BufferedReader br = getHttpResponse(url, null);
485 JSONObject val = (JSONObject) jp.parse(br);
486 String version = val.get("release").toString();
487 String majorVersion = version.substring(0, version.indexOf("."));
488 String expected = info.expectedRestVersion;
489 String expectedMajorVersion = expected.substring(0,
490 expected.indexOf("."));
491 info.restMajorVersionMismatch = false;
495 * if actual REST major version is ahead of what we expect,
496 * record this in case we want to warn the user
498 if (Float.valueOf(majorVersion) > Float
499 .valueOf(expectedMajorVersion))
501 info.restMajorVersionMismatch = true;
503 } catch (NumberFormatException e)
505 System.err.println("Error in REST version: " + e.toString());
509 * check if REST version is later than what Jalview has tested against,
510 * if so warn; we don't worry if it is earlier (this indicates Jalview has
511 * been tested in advance against the next pending REST version)
513 boolean laterVersion = StringUtils.compareVersions(version,
517 System.err.println(String.format(
518 "EnsemblRestClient expected %s REST version %s but found %s, see %s",
519 getDbSource(), expected, version, REST_CHANGE_LOG));
521 info.restVersion = version;
522 } catch (Throwable t)
525 "Error checking Ensembl REST version: " + t.getMessage());
529 public boolean isRestMajorVersionMismatch()
531 return domainData.get(getDomain()).restMajorVersionMismatch;
535 * Fetches and checks Ensembl's data version number
539 private void checkEnsemblDataVersion()
541 JSONParser jp = new JSONParser();
543 BufferedReader br = null;
547 url = new URL(getDomain() + "/info/data" + CONTENT_TYPE_JSON);
548 br = getHttpResponse(url, null);
551 JSONObject val = (JSONObject) jp.parse(br);
552 JSONArray versions = (JSONArray) val.get("releases");
553 domainData.get(getDomain()).dataVersion = versions.get(0)
556 } catch (Throwable t)
559 "Error checking Ensembl data version: " + t.getMessage());
567 } catch (IOException e)
575 public String getEnsemblDataVersion()
577 return domainData.get(getDomain()).dataVersion;
581 public String getDbVersion()
583 return getEnsemblDataVersion();