How many spaces will Java String.trim() remove?

All of them.

Returns: A copy of this string with leading and trailing white space removed, or this string if it has no leading or trailing white space.

~ Quoted from Java 1.5.0 docs

(But why didn't you just try it and see for yourself?)

From the source code (decompiled) :

  public String trim()
  {
    int i = this.count;
    int j = 0;
    int k = this.offset;
    char[] arrayOfChar = this.value;
    while ((j < i) && (arrayOfChar[(k + j)] <= ' '))
      ++j;
    while ((j < i) && (arrayOfChar[(k + i - 1)] <= ' '))
      --i;
    return (((j > 0) || (i < this.count)) ? substring(j, i) : this);
  }

The two while that you can see mean all the characters whose unicode is below the space character's, at beginning and end, are removed.

When in doubt, write a unit test:

@Test
public void trimRemoveAllBlanks(){
    assertThat("    content   ".trim(), is("content"));
}

NB: of course the test (for JUnit + Hamcrest) doesn't fail

One thing to point out, though, is that String.trim has a peculiar definition of "whitespace". It does not remove Unicode whitespace, but also removes ASCII control characters that you may not consider whitespace.

This method may be used to trim whitespace from the beginning and end of a string; in fact, it trims all ASCII control characters as well.

If possible, you may want to use Commons Lang's StringUtils.strip(), which also handles Unicode whitespace (and is null-safe, too).

See API for String class:

Returns a copy of the string, with leading and trailing whitespace omitted.

Whitespace on both sides is removed:

Note that trim() does not change the String instance, it will return a new object:

 String original = "  content  ";
 String withoutWhitespace = original.trim();

 // original still refers to "  content  "
 // and withoutWhitespace refers to "content"