boxtec
/
yacy_search_server
mirror of https://github.com/yacy/yacy_search_server
4
1
Fork 0
Code Issues Projects Releases Wiki Activity

Removed remaining unsafe accesses to SimpleDateFormat instances.

Browse Source 
SimpleDateFormat must not be used by concurrent threads without
synchronization for parsing or formating dates as it is not thread-safe
(internally holds a calendar instance that is not synchronized).

Prefer now DateTimeFormatter when possible as it is thread-safe without
concurrent access performance bottleneck (does not internally use
synchronization locks).
pull/186/head
luccioman 6 years ago
parent 6b7883900c
commit bb51555830
5 changed files with 109 additions and 70 deletions
Show Stats Download Patch File Download Diff File

27
source/net/yacy/cora/federate/solr/responsewriter/GSAResponseWriter.java
Unescape View File

@ -23,6 +23,7 @@ package net.yacy.cora.federate.solr.responsewriter;
import java.io.IOException;
import java.io.Writer;
import java.nio.charset.StandardCharsets;
import java.time.DateTimeException;
import java.util.ArrayList;
import java.util.Date;
import java.util.HashMap;
@ -33,13 +34,6 @@ import java.util.Map;
import java.util.Set;
import java.util.regex.Pattern;
import net.yacy.cora.protocol.HeaderFramework;
import net.yacy.cora.util.CommonPattern;
import net.yacy.http.servlets.GSAsearchServlet;
import net.yacy.peers.operation.yacyVersion;
import net.yacy.search.Switchboard;
import net.yacy.search.schema.CollectionSchema;
import org.apache.lucene.document.Document;
import org.apache.lucene.index.IndexableField;
import org.apache.solr.common.params.CommonParams;
@ -54,6 +48,14 @@ import org.apache.solr.search.DocIterator;
import org.apache.solr.search.DocList;
import org.apache.solr.search.SolrIndexSearcher;
import net.yacy.cora.date.ISO8601Formatter;
import net.yacy.cora.protocol.HeaderFramework;
import net.yacy.cora.util.CommonPattern;
import net.yacy.http.servlets.GSAsearchServlet;
import net.yacy.peers.operation.yacyVersion;
import net.yacy.search.Switchboard;
import net.yacy.search.schema.CollectionSchema;
/**
* implementation of a GSA search result.
* example: GET /gsa/searchresult?q=chicken+teriyaki&output=xml&client=test&site=test&sort=date:D:S:d1
@ -359,11 +361,14 @@ public class GSAResponseWriter implements QueryResponseWriter, EmbeddedSolrRespo
* @see ISO8601Formatter
*/
public final String formatGSAFS(final Date date) {
if (date == null) return "";
synchronized (GSAsearchServlet.FORMAT_GSAFS) {
final String s = GSAsearchServlet.FORMAT_GSAFS.format(date);
return s;
if (date == null) {
return "";
}
try {
return GSAsearchServlet.FORMAT_GSAFS.format(date.toInstant());
} catch (final DateTimeException e) {
return "";
}
}
}

2
source/net/yacy/data/wiki/WikiBoard.java
Unescape View File

@ -54,7 +54,7 @@ public class WikiBoard {
private static final String DATE_FORMAT = "yyyyMMddHHmmss";
private static final String ANONYMOUS = "anonymous";
protected static final SimpleDateFormat SimpleFormatter = new SimpleDateFormat(DATE_FORMAT, Locale.US);
private static final SimpleDateFormat SimpleFormatter = new SimpleDateFormat(DATE_FORMAT, Locale.US);
static {
SimpleFormatter.setTimeZone(TimeZone.getTimeZone("GMT"));

85
source/net/yacy/document/DateDetection.java
Unescape View File

@ -20,12 +20,12 @@
package net.yacy.document;
import java.text.ParseException;
import java.text.SimpleDateFormat;
import java.time.DayOfWeek;
import java.time.LocalDate;
import java.time.LocalTime;
import java.time.ZoneOffset;
import java.time.ZonedDateTime;
import java.time.format.DateTimeFormatter;
import java.time.temporal.TemporalAdjuster;
import java.time.temporal.TemporalAdjusters;
import java.util.ArrayList;
@ -63,8 +63,10 @@ import net.yacy.cora.date.GenericFormatter;
public class DateDetection {
private static final TimeZone UTC_TIMEZONE = TimeZone.getTimeZone("UTC");
private static final String CONPATT = "yyyy/MM/dd";
private static final SimpleDateFormat CONFORM = new SimpleDateFormat(CONPATT, Locale.US);
private static final String CONPATT = "uuuu/MM/dd";
private static final DateTimeFormatter CONFORM = DateTimeFormatter.ofPattern(CONPATT).withLocale(Locale.US)
.withZone(ZoneOffset.UTC);
private static final LinkedHashMap<Language, String[]> Weekdays = new LinkedHashMap<>();
private static final LinkedHashMap<Language, String[]> Months = new LinkedHashMap<>();
private static final int[] MaxDaysInMonth = new int[]{31,29,31,30,31,30,31,31,30,31,30,31};
@ -75,7 +77,6 @@ public class DateDetection {
}
static {
CONFORM.setTimeZone(UTC_TIMEZONE);
// all names must be lowercase because compared strings are made to lowercase as well
Weekdays.put(Language.GERMAN, new String[]{"montag", "dienstag", "mittwoch", "donnerstag", "freitag", "samstag" /*oder: "sonnabend"*/, "sonntag"});
Weekdays.put(Language.ENGLISH, new String[]{"monday", "tuesday", "wednesday", "thursday", "friday", "saturday", "sunday"});
@ -133,8 +134,7 @@ public class DateDetection {
}
}
private final static Date TODAY = new Date();
private final static int CURRENT_YEAR = Integer.parseInt(CONFORM.format(TODAY).substring(0, 4)); // we need that to parse dates without given years, see the ShortStyle class
private final static int CURRENT_YEAR = LocalDate.now().getYear(); // we need that to parse dates without given years, see the ShortStyle class
private final static String BODNCG = "(?:\\s|^)"; // begin of date non-capturing group
private final static String EODNCG = "(?:[).:;! ]|$)"; // end of date non-capturing group
@ -312,7 +312,7 @@ public class DateDetection {
*/
private static Date[] sameDayEveryYear(final int month, final int day, final int currentYear) {
final Date[] r = new Date[4];
final Calendar cal = CONFORM.getCalendar();
final Calendar cal = new GregorianCalendar(UTC_TIMEZONE);
cal.clear();
cal.set(currentYear - 1, month, day); // set start in previous year
r[0] = cal.getTime();
@ -336,7 +336,7 @@ public class DateDetection {
january1Calendar.clear();
/* Calendar using UTC time zone to produce date results */
final Calendar utcCalendar = CONFORM.getCalendar();
final Calendar utcCalendar = new GregorianCalendar(UTC_TIMEZONE);
/* Calendar using the same time zone as in the holidayrule to extract year,month, and day fields */
final Calendar ruleCalendar = new GregorianCalendar(ruleTimeZone);
@ -552,11 +552,11 @@ public class DateDetection {
int month = this.firstEntity == EntityType.MONTH ? i1 : this.secondEntity == EntityType.MONTH ? i2 : i3;
if (day > MaxDaysInMonth[month - 1]) continue; // validity check of the day number
int year = this.firstEntity == EntityType.YEAR ? i1 : this.secondEntity == EntityType.YEAR ? i2 : i3;
synchronized (CONFORM) {try {
dates.add(CONFORM.parse(year + "/" + (month < 10 ? "0" : "") + month + "/" + (day < 10 ? "0" : "") + day));
} catch (ParseException e) {
continue;
}}
final Date parsed = parseDateSafely(
year + "/" + (month < 10 ? "0" : "") + month + "/" + (day < 10 ? "0" : "") + day, CONFORM);
if(parsed != null) {
dates.add(parsed);
}
if (dates.size() > 100) {dates.clear(); break;} // that does not make sense
}
return dates;
@ -564,6 +564,30 @@ public class DateDetection {
}
/**
* Safely parse the given string to an instant using the given formatter. Return
* null when the format can not be applied to the given string or when any
* parsing error occurred.
*
* @param str
* the string to parse
* @param formatter
* the formatter to use
* @return an Instant instance or null
*/
protected static Date parseDateSafely(final String str, final DateTimeFormatter formatter) {
Date res = null;
if (str != null && !str.isEmpty()) {
try {
if (formatter != null) {
res = Date.from(LocalDate.parse(str, formatter).atStartOfDay().toInstant(ZoneOffset.UTC));
}
} catch (final RuntimeException ignored) {
}
}
return res;
}
public static enum ShortStyle implements StyleParser {
MD_ENGLISH(EntityType.MONTH, EntityType.DAY, // Big-endian (month, day), e.g. "from october 1st to september 13th"
ENGLISH_LANGUAGE,
@ -619,16 +643,18 @@ public class DateDetection {
if (day > MaxDaysInMonth[month - 1]) continue; // validity check of the day number
int thisyear = CURRENT_YEAR;
int nextyear = CURRENT_YEAR + 1;
synchronized (CONFORM) {try {
String datestub = "/" + (month < 10 ? "0" : "") + month + "/" + (day < 10 ? "0" : "") + day;
Date atThisYear = CONFORM.parse(thisyear + datestub);
Date atNextYear = CONFORM.parse(nextyear + datestub);
dates.add(atThisYear);
dates.add(atNextYear);
//dates.add(atThisYear.after(TODAY) ? atThisYear : atNextYear); // we consider these kind of dates as given for the future
} catch (ParseException e) {
continue;
}}
String datestub = "/" + (month < 10 ? "0" : "") + month + "/" + (day < 10 ? "0" : "") + day;
final Date atThisYear = parseDateSafely(thisyear + datestub, CONFORM);
if(atThisYear != null) {
dates.add(atThisYear);
}
final Date atNextYear = parseDateSafely(nextyear + datestub, CONFORM);
if(atNextYear != null) {
dates.add(atNextYear);
}
//dates.add(atThisYear.after(TODAY) ? atThisYear : atNextYear); // we consider these kind of dates as given for the future
if (dates.size() > 100) {dates.clear(); break;} // that does not make sense
}
return dates;
@ -670,12 +696,15 @@ public class DateDetection {
* @return determined date or null
*/
public static Date parseLine(final String text, final int timezoneOffset) {
Date d = null;
// check standard date formats
try {d = CONFORM.parse(text);} catch (ParseException e) {}
Date d = parseDateSafely(text, CONFORM);
//if (d == null) try {d = GenericFormatter.FORMAT_SHORT_DAY.parse(text);} catch (ParseException e) {} // did not work well and fired for wrong formats; do not use
if (d == null) try {d = GenericFormatter.newRfc1123ShortFormat().parse(text);} catch (ParseException e) {}
if (d == null) try {d = GenericFormatter.newAnsicFormat().parse(text);} catch (ParseException e) {}
if (d == null) {
d = parseDateSafely(text, GenericFormatter.FORMAT_RFC1123_SHORT);
}
if (d == null) {
d = parseDateSafely(text, GenericFormatter.FORMAT_ANSIC);
}
if (d == null) {
// check other date formats

46
source/net/yacy/http/servlets/GSAsearchServlet.java
Unescape View File

@ -24,8 +24,9 @@ import java.io.OutputStream;
import java.io.OutputStreamWriter;
import java.io.Writer;
import java.nio.charset.StandardCharsets;
import java.text.ParseException;
import java.text.SimpleDateFormat;
import java.time.LocalDate;
import java.time.ZoneId;
import java.time.format.DateTimeFormatter;
import java.util.Date;
import java.util.Iterator;
import java.util.List;
@ -37,6 +38,17 @@ import javax.servlet.http.HttpServlet;
import javax.servlet.http.HttpServletRequest;
import javax.servlet.http.HttpServletResponse;
import org.apache.solr.common.SolrDocumentList;
import org.apache.solr.common.SolrException;
import org.apache.solr.common.params.CommonParams;
import org.apache.solr.common.params.DisMaxParams;
import org.apache.solr.request.SolrQueryRequest;
import org.apache.solr.request.SolrRequestInfo;
import org.apache.solr.response.QueryResponseWriter;
import org.apache.solr.response.ResultContext;
import org.apache.solr.response.SolrQueryResponse;
import org.apache.solr.util.FastWriter;
import net.yacy.cora.date.ISO8601Formatter;
import net.yacy.cora.federate.solr.Ranking;
import net.yacy.cora.federate.solr.connector.EmbeddedSolrConnector;
@ -53,17 +65,6 @@ import net.yacy.search.query.SearchEvent;
import net.yacy.search.schema.CollectionSchema;
import net.yacy.server.serverObjects;
import org.apache.solr.common.SolrDocumentList;
import org.apache.solr.common.SolrException;
import org.apache.solr.common.params.CommonParams;
import org.apache.solr.common.params.DisMaxParams;
import org.apache.solr.request.SolrQueryRequest;
import org.apache.solr.request.SolrRequestInfo;
import org.apache.solr.response.QueryResponseWriter;
import org.apache.solr.response.ResultContext;
import org.apache.solr.response.SolrQueryResponse;
import org.apache.solr.util.FastWriter;
/**
* This is a gsa result formatter for solr search results.
@ -74,9 +75,11 @@ public class GSAsearchServlet extends HttpServlet {
private static final long serialVersionUID = 7835985518515673885L;
// GSA date formatter (short form of ISO8601 date format)
private static final String PATTERN_GSAFS = "yyyy-MM-dd";
public static final SimpleDateFormat FORMAT_GSAFS = new SimpleDateFormat(PATTERN_GSAFS, Locale.US);
/** GSA date formatter (short form of ISO8601 date format) */
private static final String PATTERN_GSAFS = "uuuu-MM-dd";
public static final DateTimeFormatter FORMAT_GSAFS = DateTimeFormatter.ofPattern(PATTERN_GSAFS)
.withLocale(Locale.US).withZone(ZoneId.systemDefault());
private final static GSAResponseWriter responseWriter = new GSAResponseWriter();
@ -273,10 +276,11 @@ public class GSAsearchServlet extends HttpServlet {
* @see ISO8601Formatter
*/
public final Date parseGSAFS(final String datestring) {
try {
return FORMAT_GSAFS.parse(datestring);
} catch (final ParseException e) {
return null;
}
try {
return Date
.from(LocalDate.parse(datestring, FORMAT_GSAFS).atStartOfDay(ZoneId.systemDefault()).toInstant());
} catch (final RuntimeException e) {
return null;
}
}
}

19
source/net/yacy/kelondro/blob/MapDataMining.java
Unescape View File

@ -30,15 +30,17 @@ package net.yacy.kelondro.blob;
import java.io.File;
import java.io.IOException;
import java.lang.reflect.Array;
import java.text.ParseException;
import java.text.SimpleDateFormat;
import java.time.DateTimeException;
import java.time.LocalDateTime;
import java.time.ZoneOffset;
import java.time.ZonedDateTime;
import java.util.Collection;
import java.util.HashMap;
import java.util.Iterator;
import java.util.Locale;
import java.util.Map;
import java.util.concurrent.ConcurrentHashMap;
import net.yacy.cora.date.GenericFormatter;
import net.yacy.cora.document.encoding.UTF8;
import net.yacy.cora.order.Base64Order;
import net.yacy.cora.order.ByteOrder;
@ -411,15 +413,13 @@ public class MapDataMining extends MapHeap {
super.close();
}
private static final String shortDateFormatString = "yyyyMMddHHmmss";
private static final SimpleDateFormat shortFormatter = new SimpleDateFormat(shortDateFormatString, Locale.US);
private static final long minutemillis = 60000;
private static long date2000 = 0;
static {
try {
date2000 = shortFormatter.parse("20000101000000").getTime();
} catch (final ParseException e) {}
date2000 = ZonedDateTime.of(2000, 1, 1, 0, 0, 0, 0, ZoneOffset.UTC).toInstant().toEpochMilli();
} catch (final DateTimeException e) {}
}
private static final byte[] plainByteArray = new byte[256];
@ -453,9 +453,10 @@ public class MapDataMining extends MapHeap {
if (s == null || s.isEmpty() || s.charAt(0) == '-') return 0;
try {
long l = 0;
if (s.length() == shortDateFormatString.length()) {
if (s.length() == GenericFormatter.PATTERN_SHORT_SECOND.length()) {
// try a date
l = ((shortFormatter.parse(s).getTime() - date2000) / minutemillis);
l = ((LocalDateTime.parse(s, GenericFormatter.FORMAT_SHORT_SECOND).toInstant(ZoneOffset.UTC)
.toEpochMilli() - date2000) / minutemillis);
if (l < 0) l = 0;
} else {
// try a number

Write Preview
Loading…
Cancel
Save