diff flys-backend/src/main/java/de/intevation/flys/importer/WstParser.java @ 199:ed38839a6b08

Ported over some WST parsing stuff from desktop flys flys-backend/trunk@1538 c6561f87-3c4e-4783-a992-168aeb5c3f6f
author Sascha L. Teichmann <sascha.teichmann@intevation.de>
date Tue, 22 Mar 2011 15:48:09 +0000
parents c0dcc2357106
children 88048d4f6e4d
line wrap: on
line diff
--- a/flys-backend/src/main/java/de/intevation/flys/importer/WstParser.java	Tue Mar 22 12:15:18 2011 +0000
+++ b/flys-backend/src/main/java/de/intevation/flys/importer/WstParser.java	Tue Mar 22 15:48:09 2011 +0000
@@ -1,15 +1,336 @@
 package de.intevation.flys.importer;
 
+import java.util.ArrayList;
+import java.util.Map;
+import java.util.HashMap;
+
 import java.io.File;
 import java.io.IOException;
+import java.io.LineNumberReader;
+import java.io.InputStreamReader;
+import java.io.FileInputStream;
+
+import java.text.NumberFormat;
+
+import org.apache.log4j.Logger;
+
+import de.intevation.flys.utils.StringUtil;
 
 public class WstParser
 {
+    private static Logger log = Logger.getLogger(WstParser.class);
+
+    public static final String COLUMN_BEZ_TEXT   = "column-bez-text";
+    public static final String COLUMN_BEZ_BREITE = "column-bez-breite";
+    public static final String COLUMN_QUELLE     = "column-quelle";
+    public static final String COLUMN_DATUM      = "column-datum";
+
+    public static final Double UNDEFINED_ZERO = Double.valueOf(0.0);
+
+    public static final String ENCODING = "ISO-8859-1";
+
     public WstParser() {
     }
 
     public void parse(File file) throws IOException {
-        // TODO: Implement me!
+
+        log.info("Parsing WST file '" + file + "'");
+
+        LineNumberReader in = null;
+        try {
+            in =
+                new LineNumberReader(
+                new InputStreamReader(
+                new FileInputStream(file), ENCODING));
+
+            String input;
+            boolean first = true;
+            int columnCount = 0;
+
+            String [] lsHeader       = null;
+            String [] lsBezeichner   = null;
+            String [] langBezeichner = null;
+            int    [] colNaWidths    = null;
+            String [] quellen        = null;
+            String [] daten          = null;
+            double [] aktAbfluesse   = null;
+            double [] firstAbfluesse = null;
+
+            double minKm =  Double.MAX_VALUE;
+            double maxKm = -Double.MAX_VALUE;
+
+            boolean bFirstComment = true;
+            boolean columnHeaderChecked = false;
+
+            double lastKm = Double.MAX_VALUE;
+
+            String einheit = "Wassserstand [NN + m]";
+
+            HashMap<String, Double> oldEscapeLine = null;
+
+            while ((input = in.readLine()) != null) {
+                String line = input;
+                if (first) { // fetch number of columns
+                    if ((line = line.trim()).length() == 0) {
+                        continue;
+                    }
+                    try {
+                        columnCount = Integer.parseInt(line);
+                        if (columnCount <= 0) {
+                            throw new NumberFormatException(
+                                "number columns <= 0");
+                        }
+                        log.debug("Number of columns: " + columnCount);
+                        lsBezeichner = new String[columnCount];
+                        lsHeader     = new String[columnCount];
+                        aktAbfluesse = new double[columnCount];
+                    }
+                    catch (NumberFormatException nfe) {
+                        log.warn(nfe);
+                        continue;
+                    }
+                    first = false;
+                    continue;
+                }
+
+                line = line.replace(',', '.');
+
+                if (line.startsWith("*\u001f")) {
+                    Double [] data =
+                        parseLineAsDouble(line, columnCount, false, true);
+
+                    if (oldEscapeLine != null) {
+                        addInterval(minKm, maxKm, oldEscapeLine);
+                        minKm =  Double.MAX_VALUE;
+                        maxKm = -Double.MAX_VALUE;
+                    }
+
+                    oldEscapeLine = new HashMap<String, Double>();
+                    for (int i = 0; i < columnCount; ++i) {
+                        if (lsHeader[i] != null) {
+                            oldEscapeLine.put(lsHeader[i], data[i]);
+                        }
+                    }
+
+                    for (int i = Math.min(data.length, aktAbfluesse.length)-1;
+                         i >= 0; --i) {
+                        aktAbfluesse[i] = data[i].doubleValue();
+                    }
+
+                    if (firstAbfluesse == null) {
+                        firstAbfluesse = (double [])aktAbfluesse.clone();
+                    }
+                    continue;
+                }
+
+                if (line.startsWith("*!")) {
+                    String spezial = line.substring(2).trim();
+
+                    if (spezial.length() == 0) {
+                        continue;
+                    }
+
+                    if (spezial.startsWith(COLUMN_BEZ_TEXT)) {
+                        spezial = spezial.substring(COLUMN_BEZ_TEXT.length()).trim();
+                        if (spezial.length() == 0) {
+                            continue;
+                        }
+                        langBezeichner = StringUtil.splitQuoted(spezial, '"');
+                    }
+                    else if (spezial.startsWith(COLUMN_BEZ_BREITE)) {
+                        spezial = spezial.substring(COLUMN_BEZ_BREITE.length()).trim();
+
+                        if (spezial.length() == 0) {
+                            continue;
+                        }
+
+                        String[] split = spezial.split("\\s+");
+
+                        colNaWidths = new int[split.length];
+                        for (int i=0; i < split.length; i++) {
+                            colNaWidths[i] = Integer.parseInt(split[i]);
+                        }
+                    }
+                    else if (spezial.startsWith(COLUMN_QUELLE)) {
+                        if (spezial.length() == 0) {
+                            continue;
+                        }
+                        quellen = StringUtil.splitQuoted(spezial, '"');
+                    }
+                    else if (spezial.startsWith(COLUMN_DATUM)) {
+                        spezial = spezial.substring(COLUMN_DATUM.length()).trim();
+                        if (spezial.length() == 0) {
+                            continue;
+                        }
+                        daten = StringUtil.splitQuoted(spezial, '"');
+                    }
+                    continue;
+                }
+
+                if (line.startsWith("*")) {
+                    if (bFirstComment && line.length() >= 11) {
+                        String yAxis = line.substring(10).trim();
+                        if (yAxis.length() > 0) {
+                            einheit = yAxis;
+                        }
+                        bFirstComment = false;
+                    }
+                    continue;
+                }
+
+                if (firstAbfluesse != null) {
+                    if (!columnHeaderChecked) {
+                        int unknownCount = 0;
+                        for (int i = 0; i < lsHeader.length; ++i) {
+                            if (lsBezeichner[i] == null
+                            || lsBezeichner[i].length() == 0) {
+                                double q = firstAbfluesse[i];
+                                if (q < 0.001) {
+                                    lsBezeichner[i] =
+                                        "<unbekannt#" + unknownCount + ">";
+                                    ++unknownCount;
+                                }
+                                else {
+                                    lsBezeichner[i] = "Q="+format(q);
+                                }
+                            }
+                            lsHeader[i] = lsBezeichner[i] + " " + einheit;
+                        }
+                        columnHeaderChecked = true;
+                    }
+
+                    Double [] data =
+                        parseLineAsDouble(line, columnCount, true, false);
+
+                    double kaem = data[0];
+
+                    if (kaem < minKm) {
+                        minKm = kaem;
+                    }
+                    if (kaem > maxKm) {
+                        maxKm = kaem;
+                    }
+
+                    lastKm = kaem;
+
+                    // extract values
+                    for (int i = 0; i < columnCount; ++i) {
+                        addValue(kaem, data[i].doubleValue(), lsBezeichner[i]);
+                    }
+
+                }
+                else { // firstAbfluesse == null
+                    if (langBezeichner != null) {
+                        lsBezeichner = StringUtil.fitArray(
+                            langBezeichner, lsBezeichner);
+                    }
+                    else if (colNaWidths != null) {
+                        for (int j = 0, i = 0, N = input.length();
+                             j < colNaWidths.length && i < N;
+                             i += colNaWidths[j++]
+                        ) {
+                            lsBezeichner[j] = input.substring(
+                                i, i+colNaWidths[j]).trim();
+                        }
+                    }
+                    else {
+                        // first column begins at position 8 in line
+                        for (int i = 8, col = 0; i < input.length(); i += 9) {
+                            if ((i + 9) > input.length()) {
+                                i = input.length() - 10;
+                            }
+                            // one column header is 9 chars wide
+                            lsBezeichner[col++] =
+                                input.substring(i, i + 9).trim();
+
+                            if (col == lsBezeichner.length) {
+                                break;
+                            }
+                        }
+                    }
+                }
+
+            }
+            addInterval(minKm, maxKm, oldEscapeLine);
+        }
+        finally {
+            if (in != null) {
+                in.close();
+            }
+        }
+    }
+
+    protected void addValue(double km, double w, String columnName) {
+        // TODO: store me!
+    }
+
+    protected static String format(double value) {
+        NumberFormat nf = NumberFormat.getInstance();
+        nf.setMinimumFractionDigits(2);
+        nf.setMaximumFractionDigits(2);
+        return nf.format(value);
+    }
+
+    protected void addInterval(
+        double              from,
+        double              to, 
+        Map<String, Double> values
+    ) {
+        log.debug("addInterval: " + from + " " + to);
+        if (values == null) {
+            return;
+        }
+    }
+
+    private static final Double [] parseLineAsDouble(
+        String  line, 
+        int     count, 
+        boolean bStation,
+        boolean bParseEmptyAsZero
+    ) {
+        String [] tokens = parseLine(line, count, bStation);
+
+        Double [] doubles = new Double[tokens.length];
+
+        for (int i = 0; i < doubles.length; ++i) {
+            String token = tokens[i].trim();
+            if (token.length() != 0) {
+                doubles[i] = Double.valueOf(token);
+            }
+            else if (bParseEmptyAsZero) {
+                doubles[i] = UNDEFINED_ZERO;
+            }
+        }
+
+        return doubles;
+    }
+
+    private static String [] parseLine(
+        String  line, 
+        int     tokenCount,
+        boolean bParseStation
+    ) {
+        ArrayList<String> strings = new ArrayList<String>();
+
+        if (bParseStation) {
+            if (line.length() < 8) {
+                throw new IllegalArgumentException("station too short");
+            }
+            strings.add(line.substring(0, 8));
+        }
+
+        int pos = 9;
+        for (int i = 0; i < tokenCount; ++i) {
+            if (line.length() >= pos + 8) {
+                strings.add(line.substring(pos, pos + 8));
+            }
+            else {
+                strings.add("");
+            }
+            pos += 9;
+        }
+
+        return strings.toArray(new String[strings.size()]);
     }
 }
 // vim:set ts=4 sw=4 si et sta sts=4 fenc=utf8 :

http://dive4elements.wald.intevation.org