Changeset 4520

Timestamp:

Sep 10, 2013, 4:45:18 PM (10 years ago)

Author:

Fredrik Levander

Message:

Fixes #815. Incorporating msnumpress support.

Location:

trunk/api/core/src

Files:

• ms (added)
• ms/numpress (added)
• ms/numpress/IntDecoder.java (added)
• ms/numpress/MSNumpress.java (added)
• org/proteios/io/Base64Util.java (modified) (16 diffs)
• org/proteios/io/MzMLFileReader.java (modified) (27 diffs)
• org/proteios/io/PeakListFileImpl.java (modified) (1 diff)
• org/proteios/io/mzdata/MzDataImpDataBlock.java (modified) (1 diff)

trunk/api/core/src/org/proteios/io/Base64Util.java

@@ -25 +25 @@
 
 import org.proteios.core.Base64Coder;
+import ms.numpress.MSNumpress;
 import java.nio.ByteBuffer;
 import java.nio.ByteOrder;
@@ -33 +34 @@
 
 /**
- * This class supports conversion to/from Base64-coded data.
- * 
- * Used e.g. to encode the data (peak) part of mass spectra
- * in an mzData file.
+ * This class supports conversion to/from Base64-coded data. Used e.g. to encode
+ * the data (peak) part of mass spectra in an mzData file.
  * 
  * @author Olle
@@ -44 +43 @@
 public class Base64Util
 {
+  
   private static final org.apache.log4j.Logger log = org.apache.log4j.LogManager
     .getLogger("org.proteios.io");
 
-  /**
-   * Decodes Base64 coded data String
-   * and returns a String
-   * containing the extracted data.
-   * 
-   * Uses class org.proteios.core.Base64Coder to decode data.
-   *
+
+  /**
+   * Decodes Base64 coded data String and returns a String containing the
+   * extracted data. Uses class org.proteios.core.Base64Coder to decode data.
+   * 
    * @param dataString String A string with Base64 coded data.
    * @return String A string with decoded data.
@@ -69 +67 @@
 
   /**
-   * Decodes Base64 coded data String
-   * and returns an ArrayList<double>
-   * containing the extracted data.
-   * 
-   * Uses class org.proteios.core.Base64Coder to decode data.
-   *
+   * Decodes Base64 coded data String and returns an ArrayList<double>
+   * containing the extracted data. Uses class org.proteios.core.Base64Coder
+   * to decode data.
+   * 
    * @param doublePrecision boolean flag, true if precision == "64".
    * @param bigEndian boolean flag indicating endian type.
@@ -80 +76 @@
    * @return ArrayList<double> with extracted values.
    */
-  public static List<Double> decode(boolean doublePrecision, boolean bigEndian, String dataString)
+  public static List<Double> decode(boolean doublePrecision,
+      boolean bigEndian, String numpress, String dataString)
   {
     /*
@@ -86 +83 @@
      */
     boolean zLibCompression = false;
-    List<Double> resultArray = decode(doublePrecision, bigEndian, zLibCompression, dataString);
+    List<Double> resultArray = decode(doublePrecision, bigEndian,
+      zLibCompression, numpress, dataString);
     /*
      * Return result as list
@@ -95 +93 @@
 
   /**
-   * Decodes Base64 coded data String
-   * and returns an ArrayList<double>
-   * containing the extracted data.
-   * 
-   * Uses class org.proteios.core.Base64Coder to decode data.
-   *
+   * Decodes Base64 coded data String and returns an ArrayList<double>
+   * containing the extracted data. Uses class org.proteios.core.Base64Coder
+   * to decode data.
+   * 
    * @param doublePrecision boolean flag, true if precision == "64".
    * @param bigEndian boolean flag indicating endian type.
@@ -107 +103 @@
    * @return ArrayList<double> with extracted values.
    */
-  public static List<Double> decode(boolean doublePrecision, boolean bigEndian, boolean zLibCompression, String dataString)
+  public static List<Double> decode(boolean doublePrecision,
+      boolean bigEndian, boolean zLibCompression, String numpress,
+      String dataString)
   {
     /*
@@ -118 +116 @@
       Inflater decompresser = new Inflater();
       decompresser.setInput(dataByteArray, 0, dataByteArray.length);
-      byte[] uncompressedByteArray = new byte[3*dataByteArray.length];
+      byte[] uncompressedByteArray = new byte[3 * dataByteArray.length];
       int uncompressedArrayLength = 0;
       try
       {
-        uncompressedArrayLength = decompresser.inflate(uncompressedByteArray);
+        uncompressedArrayLength = decompresser
+          .inflate(uncompressedByteArray);
       }
       catch (DataFormatException e)
@@ -131 +130 @@
       // Use uncompressed byte array
       dataByteArray = new byte[uncompressedArrayLength];
-      for (int i=0; i < uncompressedArrayLength; i++)
+      for (int i = 0; i < uncompressedArrayLength; i++)
       {
         dataByteArray[i] = uncompressedByteArray[i];
       }
     }
-    ByteBuffer dataByteBuffer = ByteBuffer.wrap(dataByteArray);
-    /*
-     * Get size of result list from
-     * byte array size and precision.
-     */
-    int arraySize = dataByteArray.length;
-    int bytesPerValue = 4;
-    if (doublePrecision) {
-      /*
-       * Precision == "64"
-       */
-      bytesPerValue = 8;
-    }
-    int dataLength = arraySize/bytesPerValue;
     /*
      * Create result arrayList
      */
     List<Double> resultArray = new ArrayList<Double>(1);
-    /*
-     * Java works with big endian by default.
-     * If the data is little endian, the
-     * byte order is changed.
-     *
-     * This configuration influences how data
-     * is retrieved using methods like
-     * getFloat() and getDouble(),
-     */
-    if (!bigEndian) {
-      dataByteBuffer.order(ByteOrder.LITTLE_ENDIAN);
-    }
-    /*
-     * Put the data into the result list
-     */
-    if (!doublePrecision) {
-      /*
-       * Precision == "32"
-       */
-      for (int i = 0; i < dataLength; i++) {
-        float value = dataByteBuffer.getFloat();
-        resultArray.add(i, new Float(value).doubleValue());
-      }
-    } else {
-      /*
-       * Precision == "64"
-       */
-      for (int i = 0; i < dataLength; i++) {
-        double value = dataByteBuffer.getDouble();
-        resultArray.add(i, value);
+    if (numpress != null)
+    {
+      int length = dataByteArray.length;
+      log.debug("Numpress decompression:"+numpress+". Array length:"+length);
+      double[] result = MSNumpress.decode(numpress, dataByteArray, length);
+      for (int i=0;i<result.length;i++)
+      {
+        resultArray.add(result[i]);
+      }
+    }
+    else
+    {
+      ByteBuffer dataByteBuffer = ByteBuffer.wrap(dataByteArray);
+      /*
+       * Get size of result list from byte array size and precision.
+       */
+      int arraySize = dataByteArray.length;
+      int bytesPerValue = 4;
+      if (doublePrecision)
+      {
+        /*
+         * Precision == "64"
+         */
+        bytesPerValue = 8;
+      }
+      int dataLength = arraySize / bytesPerValue;
+      /*
+       * Java works with big endian by default. If the data is little
+       * endian, the byte order is changed. This configuration influences
+       * how data is retrieved using methods like getFloat() and
+       * getDouble(),
+       */
+      if (!bigEndian)
+      {
+        dataByteBuffer.order(ByteOrder.LITTLE_ENDIAN);
+      }
+      /*
+       * Put the data into the result list
+       */
+      if (!doublePrecision)
+      {
+        /*
+         * Precision == "32"
+         */
+        for (int i = 0; i < dataLength; i++)
+        {
+          float value = dataByteBuffer.getFloat();
+          resultArray.add(i, new Float(value).doubleValue());
+        }
+      }
+      else
+      {
+        /*
+         * Precision == "64"
+         */
+        for (int i = 0; i < dataLength; i++)
+        {
+          double value = dataByteBuffer.getDouble();
+          resultArray.add(i, value);
+        }
       }
     }
@@ -194 +209 @@
 
   /**
-   * Encodes Base64 coded data String
-   * and returns a String
-   * containing the encoded data.
-   * 
-   * Uses class org.proteios.core.Base64Coder to encode data.
-   *
+   * Encodes Base64 coded data String and returns a String containing the
+   * encoded data. Uses class org.proteios.core.Base64Coder to encode data.
+   * 
    * @param dataString String A string with data to be Base64 encoded.
    * @return String A string with encoded data.
@@ -215 +227 @@
 
   /**
-   * Encodes Base64 coded data list<double>
-   * and returns a <String>
-   * containing the encoded data.
-   * 
-   * Uses class org.proteios.core.Base64Coder to encode data.
-   *
+   * Encodes Base64 coded data list<double> and returns a <String> containing
+   * the encoded data. Uses class org.proteios.core.Base64Coder to encode
+   * data.
+   * 
    * @param doublePrecision boolean flag, true if precision == "64".
    * @param bigEndian boolean flag indicating endian type.
@@ -226 +236 @@
    * @return <String> with encoded values.
    */
-  public static String encode(boolean doublePrecision, boolean bigEndian, List<? extends Number> inDataList)
-  {
-    /*
-     * Get size of byte array from
-     * input list size and precision.
+  public static String encode(boolean doublePrecision, boolean bigEndian,
+      List<? extends Number> inDataList)
+  {
+    /*
+     * Get size of byte array from input list size and precision.
      */
     int dataLength = inDataList.size();
     int bytesPerValue = 4;
-    if (doublePrecision) {
+    if (doublePrecision)
+    {
       /*
        * Precision == "64"
@@ -240 +251 @@
       bytesPerValue = 8;
     }
-    int arraySize = bytesPerValue*dataLength;
+    int arraySize = bytesPerValue * dataLength;
     /*
      * Create dataByteBuffer ByteBuffer
@@ -247 +258 @@
     ByteBuffer dataByteBuffer = ByteBuffer.wrap(dataByteArray);
     /*
-     * Java works with big endian by default.
-     * If the data is little endian, the
-     * byte order is changed.
-     *
-     * Note! This configuration must be set before
-     * the ByteBuffer is loaded with data using
-     * methods like putFloat() and putDouble(),
-     * as it influences how the data will be stored.
-     */
-    if (!bigEndian) {
+     * Java works with big endian by default. If the data is little endian,
+     * the byte order is changed. Note! This configuration must be set
+     * before the ByteBuffer is loaded with data using methods like
+     * putFloat() and putDouble(), as it influences how the data will be
+     * stored.
+     */
+    if (!bigEndian)
+    {
       dataByteBuffer.order(ByteOrder.LITTLE_ENDIAN);
     }
@@ -262 +271 @@
      * Put the data into the byte buffer
      */
-    if (!doublePrecision) {
+    if (!doublePrecision)
+    {
       /*
        * Precision == "32"
        */
-      for (int i = 0; i < dataLength; i++) {
+      for (int i = 0; i < dataLength; i++)
+      {
         if (inDataList.get(i) instanceof Double)
         {
-          Double doubleVal = (Double)inDataList.get(i);
+          Double doubleVal = (Double) inDataList.get(i);
           dataByteBuffer.putFloat(doubleVal.floatValue());
         }
         else if (inDataList.get(i) instanceof Integer)
         {
-          Integer intVal = (Integer)inDataList.get(i);
+          Integer intVal = (Integer) inDataList.get(i);
           dataByteBuffer.putInt(intVal.intValue());
         }
       }
-    } else {
+    }
+    else
+    {
       /*
        * Precision == "64"
        */
-      for (int i = 0; i < dataLength; i++) {
-        Double doubleVal = (Double)inDataList.get(i);
+      for (int i = 0; i < dataLength; i++)
+      {
+        Double doubleVal = (Double) inDataList.get(i);
         dataByteBuffer.putDouble(doubleVal.doubleValue());
       }

trunk/api/core/src/org/proteios/io/MzMLFileReader.java

@@ -28 +28 @@
 package org.proteios.io;
 
+import ms.numpress.MSNumpress;
 import org.proteios.core.InvalidDataException;
 import org.xml.sax.SAXException;
@@ -79 +80 @@
   private final String ACC_CHARGE_STATE = "MS:1000041";
   private final String ACC_INTENSITY = "MS:1000042";
+
   private final HashMap<String, String> accInstrumentModelHashMap = new HashMap<String, String>();
   /*
@@ -98 +100 @@
   private String precision = new String("");
   private String compression = new String("");
+  private String numpress = null;
   private String endian = new String("");
   private int dataLength = 0;
@@ -1205 +1208 @@
                 getSpectrumInstrumentData()
                   .setInstrumentSerialNo(instrumentSerialNo);
-                log
-                  .debug("instrumentSerialNo: \"" + instrumentSerialNo + "\"");
+                log.debug("instrumentSerialNo: \"" + instrumentSerialNo + "\"");
               }
               // Check if instrument model set in referenceable
@@ -1220 +1222 @@
                 getSpectrumInstrumentData().getAdditional()
                   .add(stringPair);
-                log
-                  .debug("instrumentModel: \"" + instrumentModelStr + "\"");
+                log.debug("instrumentModel: \"" + instrumentModelStr + "\"");
               }
             }
@@ -1255 +1256 @@
               {
                 inIntensityTargetBinaryDataArrayBlock = true;
-                log
-                  .debug("intensity binaryDataArray block found");
+                log.debug("intensity binaryDataArray block found");
               }
               /*
@@ -1269 +1269 @@
               {
                 setCompression(compression);
-                log
-                  .debug("compression: \"" + compression + "\"");
+                log.debug("compression: \"" + compression + "\"");
               }
             }
@@ -1322 +1321 @@
             SpectrumImpl currentSpectrum = new SpectrumImpl();
             setSpectrum(currentSpectrum);
-            log
-              .debug("SpectrumId (" + spectrumIndex + ") = \"" + getCurrentSpectrumId() + "\" found.");
+            log.debug("SpectrumId (" + spectrumIndex + ") = \"" + getCurrentSpectrumId() + "\" found.");
           }
         }
@@ -1492 +1490 @@
             {
               // Add analyzer data to last list entry
-              analyzerList.get(listSize - 1).getAnalyzer().add(
-                stringPair);
+              analyzerList.get(listSize - 1).getAnalyzer()
+                .add(stringPair);
             }
           }
@@ -1592 +1590 @@
             compression = new String("zlib");
           }
+          else if (currentCvParamAccessionStr
+            .equals(MSNumpress.ACC_NUMPRESS_LINEAR) || currentCvParamAccessionStr
+            .equals(MSNumpress.ACC_NUMPRESS_SLOF) || currentCvParamAccessionStr
+            .equals(MSNumpress.ACC_NUMPRESS_PIC))
+          {
+            numpress = currentCvParamAccessionStr;
+            log.debug ("Numpress compression detected:"+numpress);
+          }
+
           else if (currentCvParamAccessionStr.equals(ACC_MZ_ARRAY))
           {
@@ -1681 +1688 @@
             .getRetentionTimeInMinutes() == null)
           {
-            log
-              .debug("Accession value when looking for scan time = \"" + currentCvParamAccessionStr + "\"");
+            log.debug("Accession value when looking for scan time = \"" + currentCvParamAccessionStr + "\"");
             cvParamProcessed = true;
             // Check unit
@@ -1688 +1694 @@
             String currentCvParamUnitAccessionStr = XMLImportUtil
               .seekAttribute("unitAccession", parser);
-            log
-              .debug("currentCvParamUnitAccessionStr = \"" + currentCvParamUnitAccessionStr + "\"");
+            log.debug("currentCvParamUnitAccessionStr = \"" + currentCvParamUnitAccessionStr + "\"");
             if (currentCvParamUnitAccessionStr != null)
             {
@@ -1717 +1722 @@
             retentionTimeStr = XMLImportUtil.seekAttribute("value",
               parser);
-            log
-              .debug("retentionTimeStr =\"" + retentionTimeStr + "\"");
+            log.debug("retentionTimeStr =\"" + retentionTimeStr + "\"");
             if (retentionTimeStr != null && !retentionTimeStr
               .equals(""))
@@ -1737 +1741 @@
                 getSpectrum().setRetentionTimeInMinutes(
                   retentionTime);
-                log
-                  .debug("cvParam block with retentionTimeInMinutes found, time = " + retentionTime);
+                log.debug("cvParam block with retentionTimeInMinutes found, time = " + retentionTime);
               }
             }
@@ -1745 +1748 @@
             .equals(ACC_SELECTED_MASS_TO_CHARGE_RATIO))
           {
-            log
-              .debug("Accession value when looking for precursor m/z = \"" + currentCvParamAccessionStr + "\"");
+            log.debug("Accession value when looking for precursor m/z = \"" + currentCvParamAccessionStr + "\"");
             cvParamProcessed = true;
             if (inTargetPrecursorBlock)
@@ -1763 +1765 @@
             .equals(ACC_CHARGE_STATE))
           {
-            log
-              .debug("Accession value when looking for precursor charge state = \"" + currentCvParamAccessionStr + "\"");
+            log.debug("Accession value when looking for precursor charge state = \"" + currentCvParamAccessionStr + "\"");
             cvParamProcessed = true;
             if (inTargetPrecursorBlock)
@@ -1780 +1781 @@
           else if (currentCvParamAccessionStr.equals(ACC_INTENSITY))
           {
-            log
-              .debug("Accession value when looking for precursor intensity = \"" + currentCvParamAccessionStr + "\"");
+            log.debug("Accession value when looking for precursor intensity = \"" + currentCvParamAccessionStr + "\"");
             cvParamProcessed = true;
             if (inTargetPrecursorBlock)
@@ -1792 +1792 @@
                   Double.valueOf(intensity));
               }
-              log
-                .debug("precursor intensity =\"" + intensity + "\"");
+              log.debug("precursor intensity =\"" + intensity + "\"");
             }
           }
@@ -1872 +1871 @@
             {
               // Add analyzer data to last list entry
-              analyzerList.get(listSize - 1).getAnalyzer().add(
-                stringPair);
+              analyzerList.get(listSize - 1).getAnalyzer()
+                .add(stringPair);
             }
           }
@@ -1913 +1912 @@
         setEndian("little");
         setDataLength(dataLength);
+        numpress = null;
         log.debug("dataLength: " + dataLength);
-        log
-          .debug("endian: \"" + endian + "\" dataLength: " + dataLength);
+        log.debug("endian: \"" + endian + "\" dataLength: " + dataLength);
       }
     }
@@ -1925 +1924 @@
         log.debug("binary block found");
 
-        log
-          .debug("precision: \"" + precision + "\" endian: \"" + endian + "\" dataLength: " + dataLength);
+        log.debug("precision: \"" + precision + "\" endian: \"" + endian + "\" dataLength: " + dataLength);
       }
     }
@@ -2028 +2026 @@
             spectrumArray[i] = getSpectrum();
             spectrumTagsFound++;
-            log
-              .debug("SpectrumId (" + i + ") = \"" + getCurrentSpectrumId() + "\" processed, spectrumTagsFound = " + spectrumTagsFound);
-            log
-              .debug("-----------------------------------------------");
+            log.debug("SpectrumId (" + i + ") = \"" + getCurrentSpectrumId() + "\" processed, spectrumTagsFound = " + spectrumTagsFound);
+            log.debug("-----------------------------------------------");
           }
         }
@@ -2079 +2075 @@
          */
         decodedBase64List = dataItem(doublePrecision, bigEndian,
-          zLibCompression, dataBase64);
+          zLibCompression, numpress, dataBase64);
 
         /*
@@ -2154 +2150 @@
    */
   private List<Double> dataItem(boolean doublePrecision, boolean bigEndian,
-      boolean zLibCompression, String dataBase64Raw)
+      boolean zLibCompression, String numpress, String dataBase64Raw)
   {
     /*
@@ -2197 +2193 @@
     }
     String dataBase64 = dataBase64StrBuf.toString();
-    log
-      .debug("nLines = " + nLines + " lineLength = " + lineLength + " nChars = " + nChars + " dataBase64Raw.length() = " + dataBase64Raw
-        .length());
+    log.debug("nLines = " + nLines + " lineLength = " + lineLength + " nChars = " + nChars + " dataBase64Raw.length() = " + dataBase64Raw
+      .length());
 
     /*
@@ -2211 +2206 @@
       {
         decodedBase64 = Base64Util.decode(doublePrecision, bigEndian,
-          zLibCompression, dataBase64);
+          zLibCompression, numpress, dataBase64);
       }
     }
@@ -2264 +2259 @@
     List<SpectrumInterface> spectrumList = new ArrayList<SpectrumInterface>();
     spectrumList = fetchSpectrum(spectrumIds);
-    log
-      .debug("spectrumIds.size() = " + spectrumIds.size() + ", spectrumList.size() = " + spectrumList
-        .size());
+    log.debug("spectrumIds.size() = " + spectrumIds.size() + ", spectrumList.size() = " + spectrumList
+      .size());
 
     /*

trunk/api/core/src/org/proteios/io/PeakListFileImpl.java

@@ -1540 +1540 @@
       if (dataBase64.length() > 0)
       {
-        decodedBase64 = Base64Util.decode(doublePrecision, bigEndian,
+        decodedBase64 = Base64Util.decode(doublePrecision, bigEndian, null,
           dataBase64);
       }

trunk/api/core/src/org/proteios/io/mzdata/MzDataImpDataBlock.java

@@ -345 +345 @@
     if (dataBase64 != null) {
       if (dataBase64.length() > 0) {
-        decodedBase64 = Base64Util.decode(doublePrecision, bigEndian, dataBase64);
+        decodedBase64 = Base64Util.decode(doublePrecision, bigEndian, null, dataBase64);
       }
     }