我有一个需要解析的错误消息。
DSNL027I AGENTNAME SERVER DISTRIBUTED AGENT WITH
LUWID=XX00000.X000.XX0000000X00=00000
THREAD-INFO=XXXX00:255.255.255.255:XXXX00:application_name:*:*:*:*
RECEIVED ABEND=00X
FOR REASON=00000000
我需要解析数以千计的这些消息。它们都遵循类似下面的模式。
DSNL027I agent-type DISTRIBUTED AGENT WITH
LUWID luw-id=token
THREAD-INFO thread-information
RECEIVED ABEND=abend-code
FOR REASON=reason-code
此外，这些模式都可以在IBM文档站点上找到。
我需要捕获此消息中的所有字段并创建XML对象，下面是给定错误消息的示例。
<agent-type>AGENTNAME</agent-type>
<luwid>XX00000.X000.XX0000000X00</luwid>
<token>00000</token>
<thread-information>XXXX00:255.255.255.255:XXXX00:application_name:*:*:*:*</thread-information>
<abend-code>00X</abend-code>
<reason-code>00000000</reason-code>
使用正则表达式匹配消息很容易，也很有意义。
是否有一个简单的编程方法可以采用模式（小写字符）并在实际消息中提取相应的值？
答案 0 :(得分：1)
试试这个：
\S* (\S*) DISTRIBUTED AGENT WITH\s*LUWID (\S*)=(\S*)\s*THREAD-INFO (\S*)\s*RECEIVED ABEND=(\S*)\s*FOR REASON=(\S*)$
（注意：在Java中，你将使用双反斜杠来转义字符：\\）
替换为：
<agent-type>$1</agent-type>\n<luwid>$2</luwid>\n<token>$3</token>\n<thread-information>$4</thread-information>\n<abend-code>$5</abend-code>\n<reason-code>$6</reason-code>
输出：
<agent-type>agent-type</agent-type>
<luwid>luw-id</luwid>
<token>token</token>
<thread-information>thread-information</thread-information>
<abend-code>abend-code</abend-code>
<reason-code>reason-code</reason-code>
答案 1 :(得分：1)
我写了一堂课来解决你的问题：
import java.util.ArrayList;
import java.util.regex.Matcher;
import java.util.regex.Pattern;
public class Regex {
public static String escape(final String pattern) {
return Pattern.quote(pattern);
}
public static String[] getLines(final String arg) {
if (arg == null) {
return new String[] {};
}
else {
final String[] temp = arg.split("[\r\n]{1,2}");
final int tempLength = temp.length;
final String[] output = new String[tempLength];
for (int i = 0; i < tempLength; i++) {
output[i] = new String(temp[i].trim());
}
return output;
}
}
public static boolean matches(final Object str, final Pattern pat) {
return new Regex(str, pat).matches();
}
public static boolean matches(final Object page, final String string) {
return new Regex(page, string).matches();
}
public static String replace(final String text, final String regex, final String replacement) {
return Pattern.compile(regex, Pattern.DOTALL | Pattern.MULTILINE).matcher(text).replaceAll(replacement);
}
private Matcher matcher;
private boolean memOpt = true;
public Regex(final Matcher matcher) {
if (matcher != null) {
this.matcher = matcher;
}
}
public Regex(final Object data, final Pattern pattern) {
this(data.toString(), pattern);
}
public Regex(final Object data, final String pattern) {
this(data.toString(), pattern);
}
public Regex(final Object data, final String pattern, final int flags) {
this(data.toString(), pattern, flags);
}
public Regex(final String data, final Pattern pattern) {
if (data != null && pattern != null) {
this.matcher = pattern.matcher(data);
}
}
public Regex(final String data, final String pattern) {
if (data != null && pattern != null) {
this.matcher = Pattern.compile(pattern, Pattern.CASE_INSENSITIVE | Pattern.DOTALL).matcher(data);
}
}
public Regex(final String data, final String pattern, final int flags) {
if (data != null && pattern != null) {
this.matcher = Pattern.compile(pattern, flags).matcher(data);
}
}
public int count() {
if (this.matcher == null) {
return 0;
}
else {
this.matcher.reset();
int c = 0;
final Matcher matchertmp = this.matcher;
while (matchertmp.find()) {
c++;
}
return c;
}
}
public String[] getColumn(int x) {
if (this.matcher == null) {
return null;
}
else {
x++;
final Matcher matcher = this.matcher;
matcher.reset();
final java.util.List<String> ar = new ArrayList<String>();
while (matcher.find()) {
String tmp = matcher.group(x);
if (tmp != null && this.memOpt) {
tmp = new String(tmp);
}
ar.add(tmp);
}
return ar.toArray(new String[ar.size()]);
}
}
public String getMatch(final int group) {
if (this.matcher != null) {
final Matcher matcher = this.matcher;
matcher.reset();
if (matcher.find()) {
final String ret = matcher.group(group + 1);
if (ret != null && this.memOpt) {
return new String(ret);
}
return ret;
}
}
return null;
}
public String getMatch(int entry, final int group) {
if (this.matcher != null) {
final Matcher matcher = this.matcher;
matcher.reset();
// group++;
entry++;
int groupCount = 0;
while (matcher.find()) {
if (groupCount == group) {
final String ret = matcher.group(entry);
if (ret != null && this.memOpt) {
return new String(ret);
}
return ret;
}
groupCount++;
}
}
return null;
}
public Matcher getMatcher() {
if (this.matcher != null) {
this.matcher.reset();
}
return this.matcher;
}
public String[][] getMatches() {
if (this.matcher == null) {
return null;
}
else {
final Matcher matcher = this.matcher;
matcher.reset();
final java.util.List<String[]> ar = new ArrayList<String[]>();
while (matcher.find()) {
final int c = matcher.groupCount();
int d = 1;
String[] group;
if (c == 0) {
group = new String[c + 1];
d = 0;
}
else {
group = new String[c];
}
for (int i = d; i <= c; i++) {
String tmp = matcher.group(i);
if (tmp != null && this.memOpt) {
tmp = new String(tmp);
}
group[i - d] = tmp;
}
ar.add(group);
}
return ar.size() == 0 ? new String[][] {} : ar.toArray(new String[][] {});
}
}
public String[] getRow(final int y) {
if (this.matcher != null) {
final Matcher matcher = this.matcher;
matcher.reset();
int groupCount = 0;
while (matcher.find()) {
if (groupCount == y) {
final int c = matcher.groupCount();
final String[] group = new String[c];
for (int i = 1; i <= c; i++) {
String tmp = matcher.group(i);
if (tmp != null && this.memOpt) {
tmp = new String(tmp);
}
group[i - 1] = tmp;
}
return group;
}
groupCount++;
}
}
return null;
}
public boolean matches() {
final Matcher matcher = this.matcher;
if (matcher == null) {
return false;
}
else {
matcher.reset();
return matcher.find();
}
}
public void setMatcher(final Matcher matcher) {
this.matcher = matcher;
}
public Regex setMemoryOptimized(final boolean t) {
this.memOpt = t;
return this;
}
@Override
public String toString() {
final StringBuilder ret = new StringBuilder();
final String[][] matches = this.getMatches();
final int matchesLength = matches.length;
String[] match;
int matchLength;
for (int i = 0; i < matchesLength; i++) {
match = matches[i];
matchLength = match.length;
for (int j = 0; j < matchLength; j++) {
ret.append("match[");
ret.append(i);
ret.append("][");
ret.append(j);
ret.append("] = ");
ret.append(match[j]);
ret.append(System.getProperty("line.separator"));
}
}
this.matcher.reset();
return ret.toString();
}
}
在这里，您可以使用括号创建复杂的RegEx来创建节，并使用方法getMatch（int i）来提取您的信息：
private static final String PATTERN_ESTACION_ID = "(SALIDA DE[ ]*)(.{1,5})(:[ ]*)([a-zA-Z]*)";
Regex regex = new Regex(stringConsist, PATTERN_ESTACION_ID);
String tipoMovimiento = regex.getMatch(3).trim();
String estacion = regex.getMatch(1);
括号计数器从0开始。
我希望它可以帮到你。