X-Git-Url: http://source.jalview.org/gitweb/?a=blobdiff_plain;f=src%2Fjalview%2Fio%2FFileParse.java;h=39d8ad432d6a5f5486d7966ee008472aac53a3ac;hb=33fa5014f3b959dc8d331c95699660861b1f4e3a;hp=dac8fe6f670609849670a919a850fa6afcfe2d4e;hpb=f263182c25e549d39438219b3f5b948f153f233f;p=jalview.git
diff --git a/src/jalview/io/FileParse.java b/src/jalview/io/FileParse.java
index dac8fe6..39d8ad4 100755
--- a/src/jalview/io/FileParse.java
+++ b/src/jalview/io/FileParse.java
@@ -20,6 +20,13 @@
*/
package jalview.io;
+import jalview.api.AlignExportSettingsI;
+import jalview.api.AlignViewportI;
+import jalview.api.AlignmentViewPanel;
+import jalview.api.FeatureSettingsModelI;
+import jalview.util.MessageManager;
+
+import java.io.BufferedInputStream;
import java.io.BufferedReader;
import java.io.ByteArrayInputStream;
import java.io.File;
@@ -30,18 +37,13 @@ import java.io.InputStream;
import java.io.InputStreamReader;
import java.io.Reader;
import java.io.StringReader;
+import java.net.HttpURLConnection;
import java.net.MalformedURLException;
import java.net.URL;
+import java.net.URLConnection;
import java.util.zip.GZIPInputStream;
-import jalview.api.AlignExportSettingsI;
-import jalview.api.AlignViewportI;
-import jalview.api.AlignmentViewPanel;
-import jalview.api.FeatureSettingsModelI;
-import jalview.util.MessageManager;
import jalview.util.Platform;
-import jalview.ws.dbsources.Pfam;
-import jalview.ws.dbsources.Rfam;
/**
* implements a random access wrapper around a particular datasource, for
@@ -53,8 +55,6 @@ public class FileParse
protected static final String TAB = "\t";
- private static final String GZ_EXT = ".gz";
-
/**
* text specifying source of data. usually filename or url.
*/
@@ -195,29 +195,73 @@ public class FileParse
}
if (!error)
{
- if (fileStr.toLowerCase().endsWith(GZ_EXT))
+ try
{
- try
- {
- dataIn = tryAsGzipSource(new FileInputStream(fileStr));
- dataName = fileStr;
- return error;
- } catch (Exception x)
- {
- warningMessage = "Failed to resolve as a GZ stream ("
- + x.getMessage() + ")";
- // x.printStackTrace();
- }
- ;
+ dataIn = checkForGzipStream(new FileInputStream(fileStr));
+ dataName = fileStr;
+ } catch (Exception x)
+ {
+ warningMessage = "Failed to resolve " + fileStr
+ + " as a data source. (" + x.getMessage() + ")";
+ // x.printStackTrace();
+ error = true;
}
-
- dataIn = new BufferedReader(new FileReader(fileStr));
- dataName = fileStr;
+ ;
}
return error;
}
+
+ /**
+ * Recognise the 2-byte magic header for gzip streams
+ *
+ * https://recalll.co/ask/v/topic/java-How-to-check-if-InputStream-is-Gzipped/555aadd62bd27354438b90f6
+ *
+ * @param bytes - at least two bytes
+ * @return
+ */
+ private static boolean isGzipStream(byte[] bytes) {
+ int head = ((int) bytes[0] & 0xff) | ((bytes[1] << 8) & 0xff00);
+ return (GZIPInputStream.GZIP_MAGIC == head);
+ }
- private BufferedReader tryAsGzipSource(InputStream inputStream)
+ /**
+ * Returns a Reader for the given input after wrapping it in a buffered input
+ * stream, and then checking if it needs to be wrapped by a GZipInputStream
+ *
+ * @param input
+ * @return
+ */
+ private BufferedReader checkForGzipStream(InputStream input) throws Exception {
+
+ // NB: stackoverflow https://stackoverflow.com/questions/4818468/how-to-check-if-inputstream-is-gzipped
+ // could use a PushBackInputStream rather than a BufferedInputStream
+
+ BufferedInputStream bufinput;
+ if (!input.markSupported()) {
+ bufinput= new BufferedInputStream(input,16);
+ input = bufinput;
+ }
+ input.mark(4);
+ byte[] bytes=input.readNBytes(2);
+ input.reset();
+ if (bytes.length==2 && isGzipStream(bytes)) {
+ return getGzipReader(input);
+ }
+ // return a buffered reader for the stream.
+ InputStreamReader isReader= new InputStreamReader(input);
+ BufferedReader toReadFrom=new BufferedReader(isReader);
+ return toReadFrom;
+ }
+ /**
+ * Returns a {@code BufferedReader} which wraps the input stream with a
+ * GZIPInputStream. Throws a {@code ZipException} if a GZIP format error
+ * occurs or the compression method used is unsupported.
+ *
+ * @param inputStream
+ * @return
+ * @throws Exception
+ */
+ private BufferedReader getGzipReader(InputStream inputStream)
throws Exception
{
BufferedReader inData = new BufferedReader(
@@ -228,65 +272,74 @@ public class FileParse
return inData;
}
- private boolean checkURLSource(String fileStr)
+ /**
+ * Tries to read from the given URL. If successful, saves a reader to the
+ * response in field {@code dataIn}, otherwise (on exception, or HTTP response
+ * status not 200), throws an exception.
+ *
+ * If the response status includes
+ *
+ *
+ * Content-Type : application/x-gzip
+ *
+ *
+ * then tries to read as gzipped content.
+ *
+ * @param urlStr
+ * @throws IOException
+ * @throws MalformedURLException
+ */
+ private void checkURLSource(String urlStr)
throws IOException, MalformedURLException
{
errormessage = "URL NOT FOUND";
- URL url = new URL(fileStr);
- //
- // GZIPInputStream code borrowed from Aquaria (soon to be open sourced) via
- // Kenny Sabir
- Exception e = null;
- if (isGzipped(fileStr))
+ URL url = new URL(urlStr);
+ URLConnection _conn = url.openConnection();
+ if (_conn instanceof HttpURLConnection)
{
- try
+ HttpURLConnection conn = (HttpURLConnection) _conn;
+ int rc = conn.getResponseCode();
+ if (rc != HttpURLConnection.HTTP_OK)
{
- InputStream inputStream = url.openStream();
- dataIn = tryAsGzipSource(inputStream);
- dataName = fileStr;
- return false;
+ throw new IOException(
+ "Response status from " + urlStr + " was " + rc);
+ }
+ } else {
+ try {
+ dataIn = checkForGzipStream(_conn.getInputStream());
+ dataName=urlStr;
+ } catch (IOException ex)
+ {
+ throw new IOException("Failed to handle non-HTTP URI stream",ex);
} catch (Exception ex)
{
- e = ex;
+ throw new IOException("Failed to determine type of input stream for given URI",ex);
}
+ return;
}
-
- try
- {
- dataIn = new BufferedReader(new InputStreamReader(url.openStream()));
- } catch (IOException q)
+ String encoding = _conn.getContentEncoding();
+ String contentType = _conn.getContentType();
+ boolean isgzipped = "application/x-gzip".equalsIgnoreCase(contentType)
+ || "gzip".equals(encoding);
+ Exception e = null;
+ InputStream inputStream = _conn.getInputStream();
+ if (isgzipped)
{
- if (e != null)
+ try
+ {
+ dataIn = getGzipReader(inputStream);
+ dataName = urlStr;
+ } catch (Exception e1)
{
throw new IOException(MessageManager
.getString("exception.failed_to_resolve_gzip_stream"), e);
}
- throw q;
+ return;
}
- // record URL as name of datasource.
- dataName = fileStr;
- return false;
- }
- /**
- * Answers true if the filename (or URL) has a format which Jalview recognises
- * as denoting gzipped content.
- *
- * Currently this means having a ".gz" extension, or ending in "/gzipped" or
- * "?gz=1" (used to retrieve gzipped from Pfam and Rfam respectively).
- *
- * @param filename
- * @return
- */
- protected static boolean isGzipped(String filename)
- {
- if (filename == null)
- {
- return false;
- }
- String lower = filename.toLowerCase();
- return lower.endsWith(GZ_EXT) || lower.endsWith(Pfam.GZIPPED)
- || lower.endsWith(Rfam.GZIPPED);
+ dataIn = new BufferedReader(new InputStreamReader(inputStream));
+ dataName = urlStr;
+ return;
}
/**