com.strider.datadefender.discoverer.DatabaseDiscoverer.discover() - Code Metrics - armenak/DataDefender - Measure and Improve Code Quality continuously with Scrutinizer

discover() F
last analyzed 2021-07-07 23:11 UTC

↳ Parent: com.strider.datadefender.discoverer.DatabaseDiscoverer

Complexity

Conditions

Size

Total Lines	124
Code Lines	85

Duplication

Lines	0
Ratio	0 %

Code Coverage

Tests	0
CRAP Score	182

Importance

Changes

Metric	Value
cc	13
eloc	85
c	0
b	0
f	0
dl	0
loc	124
ccs	0
cts	59
cp	0
crap	182
rs	3.4309

How to fix Long Method Complexity

Long Method

Small methods make your code easier to understand, in particular if combined with a good name. Besides, if your method is small, finding a good name is usually much easier.

For example, if you find yourself adding comments to a method's body, this is usually a good sign to extract the commented part to a new method, and use the comment as a starting point when coming up with a good name for this new method.

Commonly applied refactorings include:

If many parameters/temporary variables are present:
- Replace temporary variables with Query
- Introduce parameter object; often combined with preserve whole object
- If the above two are insufficient: Replace method with method object
If you have long conditionals: Decompose Conditional
Otherwise: Extract method

Complexity

Complex classes like com.strider.datadefender.discoverer.DatabaseDiscoverer.discover() often do a lot of different things. To break such a class down, we need to identify a cohesive component within that class. A common approach to find such a component is to look for fields/methods that share the same prefixes, or suffixes.

Once you have determined the fields that belong together, you can apply the Extract Class refactoring. If the component makes sense as a sub-class, Extract Subclass is also a candidate, and is often faster.

/*
 *
 * Copyright 2014-2019, Armenak Grigoryan, and individual contributors as indicated
 * by the @authors tag. See the copyright.txt in the distribution for a
 * full listing of individual contributors.
 *
 * This is free software; you can redistribute it and/or modify it
 * under the terms of the GNU Lesser General Public License as
 * published by the Free Software Foundation; either version 2.1 of
 * the License, or (at your option) any later version.
 *
 * This software is distributed in the hope that it will be useful,
 * but WITHOUT ANY WARRANTY; without even the implied warranty of
 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
 * Lesser General Public License for more details.
 *
 */
package com.strider.datadefender.discoverer;

import com.strider.datadefender.DataDefenderException;
import com.strider.datadefender.ModelDiscoveryConfig;
import com.strider.datadefender.database.IDbFactory;
import com.strider.datadefender.database.metadata.IMetaData;
import com.strider.datadefender.database.metadata.TableMetaData;
import com.strider.datadefender.database.metadata.TableMetaData.ColumnMetaData;
import com.strider.datadefender.database.sqlbuilder.ISqlBuilder;
import com.strider.datadefender.functions.Utils;
import com.strider.datadefender.report.ReportUtil;
import com.strider.datadefender.specialcase.SpecialCase;
import com.strider.datadefender.utils.Score;

import java.io.File;
import java.io.IOException;
import java.io.InputStream;
import java.lang.reflect.InvocationTargetException;
import java.lang.reflect.Method;
import java.nio.charset.StandardCharsets;
import java.sql.Blob;
import java.sql.Clob;
import java.sql.Date;
import java.sql.ResultSet;
import java.sql.SQLException;
import java.sql.Statement;
import java.sql.Time;
import java.sql.Timestamp;
import java.text.DateFormat;
import java.text.ParseException;
import java.text.SimpleDateFormat;
import java.util.ArrayList;
import java.util.List;
import java.util.Locale;
import java.util.Objects;
import java.util.stream.Collectors;

import me.tongfei.progressbar.ProgressBar;

import org.apache.commons.collections4.CollectionUtils;
import org.apache.commons.collections4.ListUtils;
import org.apache.commons.io.IOUtils;
import org.apache.commons.lang3.ClassUtils;
import org.apache.commons.lang3.StringUtils;

import opennlp.tools.util.Span;

import lombok.extern.log4j.Log4j2;

/**
 *
 * @author Armenak Grigoryan
 */
@Log4j2
public class DatabaseDiscoverer extends Discoverer {

    protected final IDbFactory factory;

    public DatabaseDiscoverer(ModelDiscoveryConfig config, IDbFactory factory) throws IOException {
        super(config);
        this.factory = factory;
    }

    /**
     * Calls a function defined as an extension
     * @param function
     * @param data
     * @param text
     * @return
     * @throws SQLException
     * @throws NoSuchMethodException
     * @throws SecurityException
     * @throws IllegalAccessException
     * @throws IllegalArgumentException
     * @throws InvocationTargetException
     */
    private Object callExtension(final String function, final ColumnMetaData data, final String text)
            throws SQLException, NoSuchMethodException, SecurityException, IllegalAccessException,
                   IllegalArgumentException, InvocationTargetException {

        if (StringUtils.isBlank(function)) {
            return null;
        }

        Object value = null;

        try {
            final String className = Utils.getClassName(function);
            final String methodName = Utils.getMethodName(function);
            final Method method = Class.forName(className).getDeclaredMethod(
                methodName,
                new Class[] { ColumnMetaData.class, String.class }
            );
            final SpecialCase instance = (SpecialCase) Class.forName(className).getConstructor().newInstance();
            value = method.invoke(instance, data, text);

        } catch (InstantiationException | ClassNotFoundException ex) {
            log.error(ex.toString());
            log.debug(ex.toString(), ex);
        }

        return value;
    }

    @SuppressWarnings("unchecked")
    public List<ColumnMetaData> discover()
        throws ParseException,
        DataDefenderException,
        IOException,
        SQLException {

        List<ColumnMatch> finalList = new ArrayList<>();

        try (ProgressBar pb = new ProgressBar(
            "Discovering by model...",
            CollectionUtils.size(config.getModels()) + CollectionUtils.size(config.getFileModels())
        )) {
            for (final String sm : CollectionUtils.emptyIfNull(config.getModels())) {
                log.info("********************************");
                log.info("Processing model " + sm);
                log.info("********************************");
                pb.setExtraMessage("Model: " + sm);

                final Model model = createModel(sm);
                matches = discoverAgainstSingleModel(model);
                finalList = ListUtils.union(finalList, matches);
                pb.step();
            }
            for (final File fm : CollectionUtils.emptyIfNull(config.getFileModels())) {
                log.info("********************************");
                log.info("Processing model " + fm);
                log.info("********************************");
                pb.setExtraMessage("Model: " + fm.getName());

                final Model model = createModel(fm);
                matches = discoverAgainstSingleModel(model);
                finalList = ListUtils.union(finalList, matches);
                pb.step();
            }
        }

        log.info("List of suspects:");

        final Score score           = new Score();
        int         highRiskColumns = 0;
        int         rowCount        = 0;

        for (final ColumnMatch match : finalList) {

            ColumnMetaData column = match.getColumn();
            // Row count
            if (config.getCalculateScore()) {
                log.debug("Counting number of rows ...");
                rowCount = ReportUtil.rowCount(factory, 
                               column.getTable().getTableName());
            } else {
                log.debug("Skipping counting number of rows ...");
            }

            // Getting 5 sample values
            final List<String> sampleDataList = ReportUtil.sampleData(factory, column);
            // Output
            log.info("Column                      : " + column.toString());
            log.info(StringUtils.repeat('=', column.toString().length() + 30));
            log.info("Model                       : " + match.getModel());
            log.info("Number of rows in the table : " + rowCount);

            if (config.getCalculateScore()) {
                log.info("Score                       : " + score.columnScore(rowCount));
            } else {
                log.info("Score                       : N/A");
            }

            log.info("Sample data");
            log.info(StringUtils.repeat('-', 11));
            
            sampleDataList.forEach((sampleData) -> {
                log.info(sampleData);
            });

            log.info("");

            // Score calculation is evaluated with score_calculation parameter
            if (config.getCalculateScore() && score.columnScore(rowCount).equals("High")) {
                highRiskColumns++;
            }
        }

        // Only applicable when parameter table_rowcount=yes otherwise score calculation should not be done
        if (config.getCalculateScore()) {
            log.info("Overall score: " + score.dataStoreScore());
            log.info("");

            if ((finalList != null) && (finalList.size() > 0)) {
                log.info("============================================");

                if (finalList.size() > config.getThresholdCount()) {
                    log.info(
                        "Number of PI [{}] columns is higher than defined threashold [{}]",
                        finalList.size(),
                        config.getThresholdCount()
                    );
                } else {
                    log.info(
                        "Number of PI [{}] columns is lower than or equal to defined threashold [{}]",
                        finalList.size(),
                        config.getThresholdCount()
                    );
                }
                if (highRiskColumns > config.getThresholdHighRisk()) {
                    log.info(
                        "Number of High risk PI [{}] columns is higher than defined threashold [{}]",
                        highRiskColumns,
                        config.getThresholdHighRisk()
                    );
                } else {
                    log.info(
                        "Number of High risk PI [{}] columns is lower than or equal to defined threashold [{}]",
                        highRiskColumns,
                        config.getThresholdHighRisk()
                    );
                }
            }
        } else {
            log.info("Overall score: N/A");
        }

        return matches.stream().map((c) -> c.getColumn()).collect(Collectors.toList());
    }

    private List<ColumnMatch> discoverAgainstSingleModel(final Model model)
        throws ParseException,
        DataDefenderException,
        IOException,
        SQLException {

        final IMetaData           metaData = factory.fetchMetaData();
        final List<TableMetaData> map      = metaData.getMetaData();

        // Start running NLP algorithms for each column and collect percentage
        matches = new ArrayList<>();

        ColumnMatch             specialCaseData;
        final List<ColumnMatch> specialCaseDataList  = new ArrayList();
        List<String>            specialCaseFunctions = config.getExtensions();
        boolean                 specialCase          = CollectionUtils.isNotEmpty(specialCaseFunctions);

        log.info("Extension list: {}", specialCaseFunctions);

        final ISqlBuilder sqlBuilder = factory.createSQLBuilder();
        List<Probability> probabilityList;

        for (final TableMetaData table : map) {

            final String tableName  = table.getTableName();
            final String prefixed = sqlBuilder.prefixSchema(tableName);
            final String cntQuery = "SELECT COUNT(*) FROM " + prefixed;

            int numRows = config.getLimit();
            try (
                Statement stmt = factory.getConnection().createStatement();
                ResultSet rs = stmt.executeQuery(cntQuery)
            ) {
                rs.next();
                numRows = Math.min(numRows, rs.getInt(1));
            } catch (SQLException e) {
            }

            List<ColumnMetaData> cols = table.getColumns().stream()
                .filter((c) -> !c.isForeignKey() && !c.isPrimaryKey())
                .collect(Collectors.toList());

            int numSteps = numRows * cols.size();
            try (ProgressBar pb = new ProgressBar(model.getName() + " in " + tableName, numSteps)) {
                for (final ColumnMetaData data : cols) {
                    
                    final String columnName = data.getColumnName();
                    pb.setExtraMessage(columnName);

                    log.debug("Column type: [" + data.getColumnType() + "]");
                    probabilityList = new ArrayList<>();
                    log.info("Analyzing column [" + tableName + "].[" + columnName + "]");

                    final String query = sqlBuilder.buildSelectWithLimit(
                        "SELECT " + columnName + " FROM "  + prefixed + " WHERE "
                            + columnName + " IS NOT NULL",
                        config.getLimit()
                    );

                    log.debug("Executing query against database: " + query);
                    try (
                        Statement stmt = factory.getConnection().createStatement();
                        ResultSet resultSet = stmt.executeQuery(query)
                    ) {
                        while (resultSet.next()) {
                            pb.step();
                            if (Objects.equals(Blob.class, data.getColumnType())) {
                                continue;
                            }
                            if (model.getName().equals("location") && ClassUtils.isAssignable(data.getColumnType(), Number.class)) {
                                continue;
                            }

                            String sentence = "";
                            if (Objects.equals(Clob.class, data.getColumnType())) {
                                Clob clob = resultSet.getClob(1);
                                InputStream is = clob.getAsciiStream();
                                sentence = IOUtils.toString(is, StandardCharsets.UTF_8.name());
                            } else {
                                sentence = resultSet.getString(1);
                            }
                            log.debug(sentence);
                            if (specialCaseFunctions != null && specialCase) {
                                try {
                                    for (String specialCaseFunction : specialCaseFunctions) {
                                        if ((sentence != null) && !sentence.isEmpty()) {
                                            log.debug("sentence: " + sentence);
                                            log.debug("data: " + data);
                                            specialCaseData = (ColumnMatch) callExtension(specialCaseFunction, data, sentence);
                                            if (specialCaseData != null) {
                                                if (!specialCaseDataList.contains(specialCaseData)) {
                                                    log.debug("Adding new special case data: " + specialCaseData.toString());
                                                    specialCaseDataList.add(specialCaseData);
                                                }
                                            } else {
                                                log.debug("No special case data found");
                                            }
                                        }
                                    }
                                } catch (NoSuchMethodException | IllegalAccessException | InvocationTargetException e) {
                                    log.error(e.toString());
                                }
                            }

                            if ((sentence != null) &&!sentence.isEmpty()) {
                                String processingValue;

                                if (Objects.equals(Date.class, data.getColumnType())
                                    || Objects.equals(Timestamp.class, data.getColumnType())
                                    || Objects.equals(Time.class, data.getColumnType())) {

                                    final DateFormat     originalFormat = new SimpleDateFormat(sentence, Locale.ENGLISH);
                                    final DateFormat     targetFormat   = new SimpleDateFormat("MMM d, yy", Locale.ENGLISH);
                                    final java.util.Date date           = originalFormat.parse(sentence);

                                    processingValue = targetFormat.format(date);
                                } else {
                                    processingValue = sentence;
                                }

                                // LOG.debug(sentence);
                                // Convert sentence into tokens
                                final String tokens[] = model.getTokenizer().tokenize(processingValue);

                                // Find names
                                final Span nameSpans[] = model.getNameFinder().find(tokens);

                                // find probabilities for names
                                final double[] spanProbs = model.getNameFinder().probs(nameSpans);

                                // Collect top X tokens with highest probability
                                // display names
                                for (int i = 0; i < nameSpans.length; i++) {
                                    final String span = nameSpans[i].toString();

                                    if (span.length() > 2) {
                                        log.debug("Span: " + span);
                                        log.debug("Covered text is: " + tokens[nameSpans[i].getStart()]);
                                        log.debug("Probability is: " + spanProbs[i]);
                                        probabilityList.add(new Probability(tokens[nameSpans[i].getStart()], spanProbs[i]));
                                    }
                                }

                                // From OpenNLP documentation:
                                // After every document clearAdaptiveData must be called to clear the adaptive data in the feature generators.
                                // Not calling clearAdaptiveData can lead to a sharp drop in the detection rate after a few documents.
                                model.getNameFinder().clearAdaptiveData();
                            }
                        }
                    } catch (SQLException sqle) {
                        log.error(sqle.toString());
                    }

                    final double averageProbability = calculateAverage(probabilityList);

                    if (averageProbability >= config.getProbabilityThreshold()) {
                        matches.add(new ColumnMatch(
                            data,
                            averageProbability,
                            model.getName(),
                            probabilityList)
                        );
                    }
                }
                pb.stepTo(numSteps);
            }
        }

        // Special processing
        if (!specialCaseDataList.isEmpty()) {
            log.debug("Special case data is processed :" + specialCaseDataList.toString());

            specialCaseDataList.forEach((specialData) -> {
                matches.add(specialData);
            });
        }

        return matches;
    }
}


1			/*
2			*
3			* Copyright 2014-2019, Armenak Grigoryan, and individual contributors as indicated
4			* by the @authors tag. See the copyright.txt in the distribution for a
5			* full listing of individual contributors.
6			*
7			* This is free software; you can redistribute it and/or modify it
8			* under the terms of the GNU Lesser General Public License as
9			* published by the Free Software Foundation; either version 2.1 of
10			* the License, or (at your option) any later version.
11			*
12			* This software is distributed in the hope that it will be useful,
13			* but WITHOUT ANY WARRANTY; without even the implied warranty of
14			* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
15			* Lesser General Public License for more details.
16			*
17			*/
18			package com.strider.datadefender.discoverer;
19
20			import com.strider.datadefender.DataDefenderException;
21			import com.strider.datadefender.ModelDiscoveryConfig;
22			import com.strider.datadefender.database.IDbFactory;
23			import com.strider.datadefender.database.metadata.IMetaData;
24			import com.strider.datadefender.database.metadata.TableMetaData;
25			import com.strider.datadefender.database.metadata.TableMetaData.ColumnMetaData;
26			import com.strider.datadefender.database.sqlbuilder.ISqlBuilder;
27			import com.strider.datadefender.functions.Utils;
28			import com.strider.datadefender.report.ReportUtil;
29			import com.strider.datadefender.specialcase.SpecialCase;
30			import com.strider.datadefender.utils.Score;
31
32			import java.io.File;
33			import java.io.IOException;
34			import java.io.InputStream;
35			import java.lang.reflect.InvocationTargetException;
36			import java.lang.reflect.Method;
37			import java.nio.charset.StandardCharsets;
38			import java.sql.Blob;
39			import java.sql.Clob;
40			import java.sql.Date;
41			import java.sql.ResultSet;
42			import java.sql.SQLException;
43			import java.sql.Statement;
44			import java.sql.Time;
45			import java.sql.Timestamp;
46			import java.text.DateFormat;
47			import java.text.ParseException;
48			import java.text.SimpleDateFormat;
49			import java.util.ArrayList;
50			import java.util.List;
51			import java.util.Locale;
52			import java.util.Objects;
53			import java.util.stream.Collectors;
54
55			import me.tongfei.progressbar.ProgressBar;
56
57			import org.apache.commons.collections4.CollectionUtils;
58			import org.apache.commons.collections4.ListUtils;
59			import org.apache.commons.io.IOUtils;
60			import org.apache.commons.lang3.ClassUtils;
61			import org.apache.commons.lang3.StringUtils;
62
63			import opennlp.tools.util.Span;
64
65			import lombok.extern.log4j.Log4j2;
66
67			/**
68			*
69			* @author Armenak Grigoryan
70			*/
71			@Log4j2
72			public class DatabaseDiscoverer extends Discoverer {
73
74			protected final IDbFactory factory;
75
76			public DatabaseDiscoverer(ModelDiscoveryConfig config, IDbFactory factory) throws IOException {
77			super(config);
78			this.factory = factory;
79			}
80
81			/**
82			* Calls a function defined as an extension
83			* @param function
84			* @param data
85			* @param text
86			* @return
87			* @throws SQLException
88			* @throws NoSuchMethodException
89			* @throws SecurityException
90			* @throws IllegalAccessException
91			* @throws IllegalArgumentException
92			* @throws InvocationTargetException
93			*/
94			private Object callExtension(final String function, final ColumnMetaData data, final String text)
95			throws SQLException, NoSuchMethodException, SecurityException, IllegalAccessException,
96			IllegalArgumentException, InvocationTargetException {
97
98			if (StringUtils.isBlank(function)) {
99			return null;
100			}
101
102			Object value = null;
103
104			try {
105			final String className = Utils.getClassName(function);
106			final String methodName = Utils.getMethodName(function);
107			final Method method = Class.forName(className).getDeclaredMethod(
108			methodName,
109			new Class[] { ColumnMetaData.class, String.class }
110			);
111			final SpecialCase instance = (SpecialCase) Class.forName(className).getConstructor().newInstance();
112			value = method.invoke(instance, data, text);
113
114			} catch (InstantiationException \| ClassNotFoundException ex) {
115			log.error(ex.toString());
116			log.debug(ex.toString(), ex);
117			}
118
119			return value;
120			}
121
122			@SuppressWarnings("unchecked")
123			public List<ColumnMetaData> discover()
124			throws ParseException,
125			DataDefenderException,
126			IOException,
127			SQLException {
128
129			List<ColumnMatch> finalList = new ArrayList<>();
130
131			try (ProgressBar pb = new ProgressBar(
132			"Discovering by model...",
133			CollectionUtils.size(config.getModels()) + CollectionUtils.size(config.getFileModels())
134			)) {
135			for (final String sm : CollectionUtils.emptyIfNull(config.getModels())) {
136			log.info("********************************");
137			log.info("Processing model " + sm);
138			log.info("********************************");
139			pb.setExtraMessage("Model: " + sm);
140
141			final Model model = createModel(sm);
142			matches = discoverAgainstSingleModel(model);
143			finalList = ListUtils.union(finalList, matches);
144			pb.step();
145			}
146			for (final File fm : CollectionUtils.emptyIfNull(config.getFileModels())) {
147			log.info("********************************");
148			log.info("Processing model " + fm);
149			log.info("********************************");
150			pb.setExtraMessage("Model: " + fm.getName());
151
152			final Model model = createModel(fm);
153			matches = discoverAgainstSingleModel(model);
154			finalList = ListUtils.union(finalList, matches);
155			pb.step();
156			}
157			}
158
159			log.info("List of suspects:");
160
161			final Score score = new Score();
162			int highRiskColumns = 0;
163			int rowCount = 0;
164
165			for (final ColumnMatch match : finalList) {
166
167			ColumnMetaData column = match.getColumn();
168			// Row count
169			if (config.getCalculateScore()) {
170			log.debug("Counting number of rows ...");
171			rowCount = ReportUtil.rowCount(factory,
172			column.getTable().getTableName());
173			} else {
174			log.debug("Skipping counting number of rows ...");
175			}
176
177			// Getting 5 sample values
178			final List<String> sampleDataList = ReportUtil.sampleData(factory, column);
179			// Output
180			log.info("Column : " + column.toString());
181			log.info(StringUtils.repeat('=', column.toString().length() + 30));
182			log.info("Model : " + match.getModel());
183			log.info("Number of rows in the table : " + rowCount);
184
185			if (config.getCalculateScore()) {
186			log.info("Score : " + score.columnScore(rowCount));
187			} else {
188			log.info("Score : N/A");
189			}
190
191			log.info("Sample data");
192			log.info(StringUtils.repeat('-', 11));
193
194			sampleDataList.forEach((sampleData) -> {
195			log.info(sampleData);
196			});
197
198			log.info("");
199
200			// Score calculation is evaluated with score_calculation parameter
201			if (config.getCalculateScore() && score.columnScore(rowCount).equals("High")) {
202			highRiskColumns++;
203			}
204			}
205
206			// Only applicable when parameter table_rowcount=yes otherwise score calculation should not be done
207			if (config.getCalculateScore()) {
208			log.info("Overall score: " + score.dataStoreScore());
209			log.info("");
210
211			if ((finalList != null) && (finalList.size() > 0)) {
212			log.info("============================================");
213
214			if (finalList.size() > config.getThresholdCount()) {
215			log.info(
216			"Number of PI [{}] columns is higher than defined threashold [{}]",
217			finalList.size(),
218			config.getThresholdCount()
219			);
220			} else {
221			log.info(
222			"Number of PI [{}] columns is lower than or equal to defined threashold [{}]",
223			finalList.size(),
224			config.getThresholdCount()
225			);
226			}
227			if (highRiskColumns > config.getThresholdHighRisk()) {
228			log.info(
229			"Number of High risk PI [{}] columns is higher than defined threashold [{}]",
230			highRiskColumns,
231			config.getThresholdHighRisk()
232			);
233			} else {
234			log.info(
235			"Number of High risk PI [{}] columns is lower than or equal to defined threashold [{}]",
236			highRiskColumns,
237			config.getThresholdHighRisk()
238			);
239			}
240			}
241			} else {
242			log.info("Overall score: N/A");
243			}
244
245			return matches.stream().map((c) -> c.getColumn()).collect(Collectors.toList());
246			}
247
248			private List<ColumnMatch> discoverAgainstSingleModel(final Model model)
249			throws ParseException,
250			DataDefenderException,
251			IOException,
252			SQLException {
253
254			final IMetaData metaData = factory.fetchMetaData();
255			final List<TableMetaData> map = metaData.getMetaData();
256
257			// Start running NLP algorithms for each column and collect percentage
258			matches = new ArrayList<>();
259
260			ColumnMatch specialCaseData;
261			final List<ColumnMatch> specialCaseDataList = new ArrayList();
262			List<String> specialCaseFunctions = config.getExtensions();
263			boolean specialCase = CollectionUtils.isNotEmpty(specialCaseFunctions);
264
265			log.info("Extension list: {}", specialCaseFunctions);
266
267			final ISqlBuilder sqlBuilder = factory.createSQLBuilder();
268			List<Probability> probabilityList;
269
270			for (final TableMetaData table : map) {
271
272			final String tableName = table.getTableName();
273			final String prefixed = sqlBuilder.prefixSchema(tableName);
274			final String cntQuery = "SELECT COUNT(*) FROM " + prefixed;
275
276			int numRows = config.getLimit();
277			try (
278			Statement stmt = factory.getConnection().createStatement();
279			ResultSet rs = stmt.executeQuery(cntQuery)
280			) {
281			rs.next();
282			numRows = Math.min(numRows, rs.getInt(1));
283			} catch (SQLException e) {
284			}
285
286			List<ColumnMetaData> cols = table.getColumns().stream()
287			.filter((c) -> !c.isForeignKey() && !c.isPrimaryKey())
288			.collect(Collectors.toList());
289
290			int numSteps = numRows * cols.size();
291			try (ProgressBar pb = new ProgressBar(model.getName() + " in " + tableName, numSteps)) {
292			for (final ColumnMetaData data : cols) {
293
294			final String columnName = data.getColumnName();
295			pb.setExtraMessage(columnName);
296
297			log.debug("Column type: [" + data.getColumnType() + "]");
298			probabilityList = new ArrayList<>();
299			log.info("Analyzing column [" + tableName + "].[" + columnName + "]");
300
301			final String query = sqlBuilder.buildSelectWithLimit(
302			"SELECT " + columnName + " FROM " + prefixed + " WHERE "
303			+ columnName + " IS NOT NULL",
304			config.getLimit()
305			);
306
307			log.debug("Executing query against database: " + query);
308			try (
309			Statement stmt = factory.getConnection().createStatement();
310			ResultSet resultSet = stmt.executeQuery(query)
311			) {
312			while (resultSet.next()) {
313			pb.step();
314			if (Objects.equals(Blob.class, data.getColumnType())) {
315			continue;
316			}
317			if (model.getName().equals("location") && ClassUtils.isAssignable(data.getColumnType(), Number.class)) {
318			continue;
319			}
320
321			String sentence = "";
322			if (Objects.equals(Clob.class, data.getColumnType())) {
323			Clob clob = resultSet.getClob(1);
324			InputStream is = clob.getAsciiStream();
325			sentence = IOUtils.toString(is, StandardCharsets.UTF_8.name());
326			} else {
327			sentence = resultSet.getString(1);
328			}
329			log.debug(sentence);
330			if (specialCaseFunctions != null && specialCase) {
331			try {
332			for (String specialCaseFunction : specialCaseFunctions) {
333			if ((sentence != null) && !sentence.isEmpty()) {
334			log.debug("sentence: " + sentence);
335			log.debug("data: " + data);
336			specialCaseData = (ColumnMatch) callExtension(specialCaseFunction, data, sentence);
337			if (specialCaseData != null) {
338			if (!specialCaseDataList.contains(specialCaseData)) {
339			log.debug("Adding new special case data: " + specialCaseData.toString());
340			specialCaseDataList.add(specialCaseData);
341			}
342			} else {
343			log.debug("No special case data found");
344			}
345			}
346			}
347			} catch (NoSuchMethodException \| IllegalAccessException \| InvocationTargetException e) {
348			log.error(e.toString());
349			}
350			}
351
352			if ((sentence != null) &&!sentence.isEmpty()) {
353			String processingValue;
354
355			if (Objects.equals(Date.class, data.getColumnType())
356			\|\| Objects.equals(Timestamp.class, data.getColumnType())
357			\|\| Objects.equals(Time.class, data.getColumnType())) {
358
359			final DateFormat originalFormat = new SimpleDateFormat(sentence, Locale.ENGLISH);
360			final DateFormat targetFormat = new SimpleDateFormat("MMM d, yy", Locale.ENGLISH);
361			final java.util.Date date = originalFormat.parse(sentence);
362
363			processingValue = targetFormat.format(date);
364			} else {
365			processingValue = sentence;
366			}
367
368			// LOG.debug(sentence);
369			// Convert sentence into tokens
370			final String tokens[] = model.getTokenizer().tokenize(processingValue);
371
372			// Find names
373			final Span nameSpans[] = model.getNameFinder().find(tokens);
374
375			// find probabilities for names
376			final double[] spanProbs = model.getNameFinder().probs(nameSpans);
377
378			// Collect top X tokens with highest probability
379			// display names
380			for (int i = 0; i < nameSpans.length; i++) {
381			final String span = nameSpans[i].toString();
382
383			if (span.length() > 2) {
384			log.debug("Span: " + span);
385			log.debug("Covered text is: " + tokens[nameSpans[i].getStart()]);
386			log.debug("Probability is: " + spanProbs[i]);
387			probabilityList.add(new Probability(tokens[nameSpans[i].getStart()], spanProbs[i]));
388			}
389			}
390
391			// From OpenNLP documentation:
392			// After every document clearAdaptiveData must be called to clear the adaptive data in the feature generators.
393			// Not calling clearAdaptiveData can lead to a sharp drop in the detection rate after a few documents.
394			model.getNameFinder().clearAdaptiveData();
395			}
396			}
397			} catch (SQLException sqle) {
398			log.error(sqle.toString());
399			}
400
401			final double averageProbability = calculateAverage(probabilityList);
402
403			if (averageProbability >= config.getProbabilityThreshold()) {
404			matches.add(new ColumnMatch(
405			data,
406			averageProbability,
407			model.getName(),
408			probabilityList)
409			);
410			}
411			}
412			pb.stepTo(numSteps);
413			}
414			}
415
416			// Special processing
417			if (!specialCaseDataList.isEmpty()) {
418			log.debug("Special case data is processed :" + specialCaseDataList.toString());
419
420			specialCaseDataList.forEach((specialData) -> {
421			matches.add(specialData);
422			});
423			}
424
425			return matches;
426			}
427			}
428

armenak / DataDefender

discover() F last analyzed 2021-07-07 23:11 UTC

Complexity

Size

Duplication

Code Coverage

Importance

How to fix Long Method Complexity

Long Method

Complexity

Duplication Side-by-Side

Filter issues like

discover() F
last analyzed 2021-07-07 23:11 UTC