Last active
October 6, 2016 02:10
-
-
Save cattaka/3376e1275af688f16e67d42fa6ad25d1 to your computer and use it in GitHub Desktop.
Split and Join function for strings such as CSV, TSV
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
public class StringUtilsTest { | |
@Test | |
public void testSplit() { | |
assertThat(StringUtils.split(null, ',', '"'), is(Collections.EMPTY_LIST)); | |
assertThat(StringUtils.split("", ',', '"'), is(Collections.<String>singletonList(null))); | |
assertThat(StringUtils.split("\"\"", ',', '"'), is(Collections.singletonList(""))); | |
assertThat(StringUtils.split("\"\",,\"\"", ',', '"'), is(Arrays.asList("", null, ""))); | |
assertThat(StringUtils.split("a,b,c", ',', '"'), is(Arrays.asList("a", "b", "c"))); | |
assertThat(StringUtils.split("a\"b,c", ',', '"'), is(Arrays.asList("a\"b", "c"))); | |
assertThat(StringUtils.split("\"a\"\"b\",c", ',', '"'), is(Arrays.asList("a\"b", "c"))); | |
assertThat(StringUtils.split("\"a,b\",c", ',', '"'), is(Arrays.asList("a,b", "c"))); | |
} | |
@Test | |
public void testJoin() { | |
assertThat(StringUtils.join(null, ',', '"'), is(nullValue())); | |
assertThat(StringUtils.join(Collections.<String>emptyList(), ',', '"'), is(nullValue())); | |
assertThat(StringUtils.join(Collections.<String>singletonList(null), ',', '"'), is("")); | |
assertThat(StringUtils.join(Collections.singletonList(""), ',', '"'), is("\"\"")); | |
assertThat(StringUtils.join(Arrays.asList("", null, ""), ',', '"'), is("\"\",,\"\"")); | |
assertThat(StringUtils.join(Arrays.asList("a", "b", "c"), ',', '"'), is("a,b,c")); | |
assertThat(StringUtils.join(Arrays.asList("a\"b", "c"), ',', '"'), is("\"a\"\"b\",c")); | |
assertThat(StringUtils.join(Arrays.asList("a,b", "c"), ',', '"'), is("\"a,b\",c")); | |
} | |
} | |
public class StringUtils { | |
public static List<String> split(String src, char delim, char bracket) { | |
ArrayList<String> result = new ArrayList<>(); | |
if (src == null) { | |
return result; | |
} | |
StringBuilder sb = new StringBuilder(); | |
int mode = 0; | |
for (int i = 0; i <= src.length(); i++) { | |
char ch = (i < src.length()) ? src.charAt(i) : delim; | |
switch (mode) { | |
case 0: | |
if (ch == delim) { | |
if (sb.length() == 0) { | |
result.add(null); | |
} else { | |
result.add(sb.toString()); | |
} | |
sb.delete(0, sb.length()); | |
} else if (ch == bracket) { | |
mode = 1; | |
} else { | |
sb.append(ch); | |
mode = 3; | |
} | |
break; | |
case 1: | |
if (ch == delim) { | |
sb.append(ch); | |
} else if (ch == bracket) { | |
mode = 2; | |
} else { | |
sb.append(ch); | |
} | |
break; | |
case 2: | |
if (ch == delim) { | |
result.add(sb.toString()); | |
sb.delete(0, sb.length()); | |
mode = 0; | |
} else if (ch == bracket) { | |
sb.append(ch); | |
mode = 1; | |
} else { | |
// Error case | |
sb.append(delim); | |
sb.append(ch); | |
mode = 1; | |
} | |
break; | |
case 3: | |
if (ch == delim) { | |
result.add(sb.toString()); | |
sb.delete(0, sb.length()); | |
mode = 0; | |
} else if (ch == bracket) { | |
sb.append(ch); | |
} else { | |
sb.append(ch); | |
} | |
break; | |
default: | |
throw new RuntimeException("ERROR"); | |
} | |
} | |
return result; | |
} | |
public static String join(List<String> src, char delim, char bracket) { | |
if (src == null || src.size() == 0) { | |
return null; | |
} | |
String bracketStr = "" + bracket; | |
String bracketStr2 = "" + bracket + bracket; | |
StringBuilder sb = new StringBuilder(); | |
boolean first = true; | |
for (String s : src) { | |
if (first) { | |
first = false; | |
} else { | |
sb.append(delim); | |
} | |
if (s == null) { | |
// none | |
} else if (s.length() == 0) { | |
sb.append(bracket); | |
sb.append(bracket); | |
} else if (s.indexOf(delim) != -1 || s.indexOf(bracket) != -1) { | |
sb.append(bracket); | |
sb.append(s.replace(bracketStr, bracketStr2)); | |
sb.append(bracket); | |
} else { | |
sb.append(s); | |
} | |
} | |
return sb.toString(); | |
} | |
} | |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment