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, EnsemblInfo> domainData = new HashMap<>();
77 private final static long AVAILABILITY_RETEST_INTERVAL = 10000L; // 10 seconds
79 private final static long VERSION_RETEST_INTERVAL = 1000L * 3600; // 1 hr
81 protected static final String CONTENT_TYPE_JSON = "?content-type=application/json";
85 domainData.put(DEFAULT_ENSEMBL_DOMAIN,
86 new EnsemblInfo(DEFAULT_ENSEMBL_DOMAIN, LATEST_ENSEMBL_REST_VERSION));
87 domainData.put(DEFAULT_ENSEMBL_GENOMES_DOMAIN, new EnsemblInfo(
88 DEFAULT_ENSEMBL_GENOMES_DOMAIN, LATEST_ENSEMBLGENOMES_REST_VERSION));
91 protected volatile boolean inProgress = false;
94 * Default constructor to use rest.ensembl.org
96 public EnsemblRestClient()
101 * initialise domain info lazily
103 if (!domainData.containsKey(ensemblDomain))
105 domainData.put(ensemblDomain,
106 new EnsemblInfo(ensemblDomain, LATEST_ENSEMBL_REST_VERSION));
108 if (!domainData.containsKey(ensemblGenomesDomain))
110 domainData.put(ensemblGenomesDomain, new EnsemblInfo(
111 ensemblGenomesDomain, LATEST_ENSEMBLGENOMES_REST_VERSION));
116 * Constructor given the target domain to fetch data from
120 public EnsemblRestClient(String d)
126 public boolean queryInProgress()
132 public StringBuffer getRawRecords()
138 * Returns the URL for the client http request
142 * @throws MalformedURLException
144 protected abstract URL getUrl(List<String> ids)
145 throws MalformedURLException;
148 * Returns true if client uses GET method, false if it uses POST
152 protected abstract boolean useGetRequest();
155 * Return the desired value for the Content-Type request header
160 * @see https://github.com/Ensembl/ensembl-rest/wiki/HTTP-Headers
162 protected abstract String getRequestMimeType(boolean multipleIds);
165 * Return the desired value for the Accept request header
168 * @see https://github.com/Ensembl/ensembl-rest/wiki/HTTP-Headers
170 protected abstract String getResponseMimeType();
173 * Checks Ensembl's REST 'ping' endpoint, and returns true if response
174 * indicates available, else false
176 * @see http://rest.ensembl.org/documentation/info/ping
179 boolean checkEnsembl()
181 BufferedReader br = null;
182 String pingUrl = getDomain() + "/info/ping" + CONTENT_TYPE_JSON;
185 // note this format works for both ensembl and ensemblgenomes
186 // info/ping.json works for ensembl only (March 2016)
187 URL ping = new URL(pingUrl);
190 * expect {"ping":1} if ok
191 * if ping takes more than 2 seconds to respond, treat as if unavailable
193 br = getHttpResponse(ping, null, 2 * 1000);
196 // error reponse status
199 JSONParser jp = new JSONParser();
200 JSONObject val = (JSONObject) jp.parse(br);
201 String pingString = val.get("ping").toString();
202 return pingString != null;
203 } catch (Throwable t)
206 "Error connecting to " + pingUrl + ": " + t.getMessage());
214 } catch (IOException e)
224 * returns a reader to a Fasta response from the Ensembl sequence endpoint
228 * @throws IOException
230 protected FileParse getSequenceReader(List<String> ids) throws IOException
232 URL url = getUrl(ids);
234 BufferedReader reader = getHttpResponse(url, ids);
240 FileParse fp = new FileParse(reader, url.toString(),
246 * Gets a reader to the HTTP response, using the default read timeout of 5
252 * @throws IOException
254 protected BufferedReader getHttpResponse(URL url, List<String> ids)
257 return getHttpResponse(url, ids, DEFAULT_READ_TIMEOUT);
261 * Sends the HTTP request and gets the response as a reader
265 * written as Json POST body if more than one
269 * @throws IOException
270 * if response code was not 200, or other I/O error
272 protected BufferedReader getHttpResponse(URL url, List<String> ids,
273 int readTimeout) throws IOException
275 int retriesLeft = MAX_RETRIES;
276 HttpURLConnection connection = null;
277 int responseCode = 0;
279 while (retriesLeft > 0)
281 connection = tryConnection(url, ids, readTimeout);
282 responseCode = connection.getResponseCode();
283 if (responseCode == HTTP_OVERLOAD) // 429
286 checkRetryAfter(connection);
293 if (responseCode != HTTP_OK) // 200
296 * note: a GET request for an invalid id returns an error code e.g. 415
297 * but POST request returns 200 and an empty Fasta response
299 System.err.println("Response code " + responseCode + " for " + url);
303 InputStream response = connection.getInputStream();
305 // System.out.println(getClass().getName() + " took "
306 // + (System.currentTimeMillis() - now) + "ms to fetch");
308 BufferedReader reader = null;
309 reader = new BufferedReader(new InputStreamReader(response, "UTF-8"));
318 * @throws IOException
319 * @throws ProtocolException
321 protected HttpURLConnection tryConnection(URL url, List<String> ids,
322 int readTimeout) throws IOException, ProtocolException
324 // System.out.println(System.currentTimeMillis() + " " + url);
325 HttpURLConnection connection = (HttpURLConnection) url.openConnection();
328 * POST method allows multiple queries in one request; it is supported for
329 * sequence queries, but not for overlap
331 boolean multipleIds = ids != null && ids.size() > 1;
332 connection.setRequestMethod(
333 multipleIds ? HttpMethod.POST : HttpMethod.GET);
334 connection.setRequestProperty("Content-Type",
335 getRequestMimeType(multipleIds));
336 connection.setRequestProperty("Accept", getResponseMimeType());
338 connection.setUseCaches(false);
339 connection.setDoInput(true);
340 connection.setDoOutput(multipleIds);
342 connection.setConnectTimeout(CONNECT_TIMEOUT_MS);
343 connection.setReadTimeout(readTimeout);
347 writePostBody(connection, ids);
353 * Inspects response headers for a 'retry-after' directive, and waits for the
354 * directed period (if less than 10 seconds)
356 * @see https://github.com/Ensembl/ensembl-rest/wiki/Rate-Limits
359 void checkRetryAfter(HttpURLConnection connection)
361 String retryDelay = connection.getHeaderField("Retry-After");
366 if (retryDelay != null)
370 int retrySecs = Integer.valueOf(retryDelay);
371 if (retrySecs > 0 && retrySecs < 10)
374 .println("Ensembl REST service rate limit exceeded, waiting "
375 + retryDelay + " seconds before retrying");
376 Thread.sleep(1000 * retrySecs);
378 } catch (NumberFormatException | InterruptedException e)
380 System.err.println("Error handling Retry-After: " + e.getMessage());
386 * Rechecks if Ensembl is responding, unless the last check was successful and
387 * the retest interval has not yet elapsed. Returns true if Ensembl is up,
388 * else false. Also retrieves and saves the current version of Ensembl data
389 * and REST services at intervals.
393 protected boolean isEnsemblAvailable()
395 EnsemblInfo info = domainData.get(getDomain());
397 long now = System.currentTimeMillis();
400 * recheck if Ensembl is up if it was down, or the recheck period has elapsed
402 boolean retestAvailability = (now
403 - info.lastAvailableCheckTime) > AVAILABILITY_RETEST_INTERVAL;
404 if (!info.restAvailable || retestAvailability)
406 info.restAvailable = checkEnsembl();
407 info.lastAvailableCheckTime = now;
411 * refetch Ensembl versions if the recheck period has elapsed
413 boolean refetchVersion = (now
414 - info.lastVersionCheckTime) > VERSION_RETEST_INTERVAL;
417 checkEnsemblRestVersion();
418 checkEnsemblDataVersion();
419 info.lastVersionCheckTime = now;
422 return info.restAvailable;
426 * Constructs, writes and flushes the POST body of the request, containing the
427 * query ids in JSON format
431 * @throws IOException
433 protected void writePostBody(HttpURLConnection connection,
434 List<String> ids) throws IOException
437 StringBuilder postBody = new StringBuilder(64);
438 postBody.append("{\"ids\":[");
440 for (String id : ids)
444 postBody.append(",");
447 postBody.append("\"");
448 postBody.append(id.trim());
449 postBody.append("\"");
451 postBody.append("]}");
452 byte[] thepostbody = postBody.toString().getBytes();
453 connection.setRequestProperty("Content-Length",
454 Integer.toString(thepostbody.length));
455 DataOutputStream wr = new DataOutputStream(
456 connection.getOutputStream());
457 wr.write(thepostbody);
463 * Fetches and checks Ensembl's REST version number
467 private void checkEnsemblRestVersion()
469 EnsemblInfo info = domainData.get(getDomain());
471 JSONParser jp = new JSONParser();
475 url = new URL(getDomain() + "/info/rest" + CONTENT_TYPE_JSON);
476 BufferedReader br = getHttpResponse(url, null);
481 JSONObject val = (JSONObject) jp.parse(br);
482 String version = val.get("release").toString();
483 String majorVersion = version.substring(0, version.indexOf("."));
484 String expected = info.expectedRestVersion;
485 String expectedMajorVersion = expected.substring(0,
486 expected.indexOf("."));
487 info.restMajorVersionMismatch = false;
491 * if actual REST major version is ahead of what we expect,
492 * record this in case we want to warn the user
494 if (Float.valueOf(majorVersion) > Float
495 .valueOf(expectedMajorVersion))
497 info.restMajorVersionMismatch = true;
499 } catch (NumberFormatException e)
501 System.err.println("Error in REST version: " + e.toString());
505 * check if REST version is later than what Jalview has tested against,
506 * if so warn; we don't worry if it is earlier (this indicates Jalview has
507 * been tested in advance against the next pending REST version)
509 boolean laterVersion = StringUtils.compareVersions(version,
513 System.err.println(String.format(
514 "EnsemblRestClient expected %s REST version %s but found %s, see %s",
515 getDbSource(), expected, version, REST_CHANGE_LOG));
517 info.restVersion = version;
518 } catch (Throwable t)
521 "Error checking Ensembl REST version: " + t.getMessage());
525 public boolean isRestMajorVersionMismatch()
527 return domainData.get(getDomain()).restMajorVersionMismatch;
531 * Fetches and checks Ensembl's data version number
535 private void checkEnsemblDataVersion()
537 JSONParser jp = new JSONParser();
539 BufferedReader br = null;
543 url = new URL(getDomain() + "/info/data" + CONTENT_TYPE_JSON);
544 br = getHttpResponse(url, null);
547 JSONObject val = (JSONObject) jp.parse(br);
548 JSONArray versions = (JSONArray) val.get("releases");
549 domainData.get(getDomain()).dataVersion = versions.get(0)
552 } catch (Throwable t)
555 "Error checking Ensembl data version: " + t.getMessage());
563 } catch (IOException e)
571 public String getEnsemblDataVersion()
573 return domainData.get(getDomain()).dataVersion;
577 public String getDbVersion()
579 return getEnsemblDataVersion();