Github user rikles commented on a diff in the pull request:
https://github.com/apache/commons-lang/pull/75#discussion_r30460736
--- Diff: src/main/java/org/apache/commons/lang3/StringUtils.java ---
@@ -3277,6 +3277,164 @@ public static String substringBetween(final String
str, final String open, final
return list.toArray(new String[list.size()]);
}
+ /**
+ * <p>Split a String into an array, using an array of fixed string
lengths.</p>
+ *
+ * <p>If not null String input, the returned array size is same as the
input lengths array.</p>
+ *
+ * <p>A null input String returns {@code null}.
+ * A {@code null} or empty input lengths array returns an empty array.
+ * A {@code 0} in the input lengths array results in en empty
string.</p>
+ *
+ * <p>Extra characters are ignored (ie String length greater than sum
of split lengths).
+ * All empty substrings other than zero length requested, are returned
{@code null}.</p>
+ *
+ * <pre>
+ * StringUtils.splitByLength(null, *) = null
+ * StringUtils.splitByLength("abc") = []
+ * StringUtils.splitByLength("abc", null) = []
+ * StringUtils.splitByLength("abc", []) = []
+ * StringUtils.splitByLength("", 2, 4, 1) = [null, null, null]
+ *
+ * StringUtils.splitByLength("abcdefg", 2, 4, 1) = ["ab", "cdef",
"g"]
--- End diff --
Like said in the next line : `StringUtils.splitByLength("abcdefg", 2, 2)`
will return `["ab", "cd" ]`.
`StringUtils.splitByLength("abcdefghij", 2, 4, 1) = ["ab", "cdef", "g"]`
I asked myself the question during development. Do we discard the extra
characters ?
I think it would be nice to let users decide. Moreover, depending on use
case, it could be useful to keep/discard the "first extra characters" (like
parsing a single line commented out string).
I propose to :
* add a private `splitByLengthWorker(String string, boolean splitFromEnd,
boolean discardExtraChar, int ... lengths)`
* keep this `splitByLength(String, int ...)` method logic as default :
`return splitByLengthWorker(string, false, true, lengths)`. So, by default, the
returned array is same size as the `int ... lengths` array param and this
behavior is interesting on parsing "fixed column lengths" strings.
* add a `splitByLengthKeepExtraChar(String, int ...)` : `return
splitByLengthWorker(string, false, false, lengths)`
* add a `splitByLengthFromEnd(String, int ...)` : `return
splitByLengthWorker(string, true, false, lengths)`
* add a `splitByLengthFromEndKeepExtraChar(String, int ...)` : `return
splitByLengthWorker(string, true, true, lengths)`
A question : For _split from end_ methods, which call do you think is more
logic : _right aligned/end to start_ lengths, _reversed/not reversed_ result ?
* `StringUtils.splitByLengthFromEndKeepExtraChar("__abcdef", 1, 2, 3) =
["__", "a", "bc", "def"]` - (RA, NR)
* `StringUtils.splitByLengthFromEndKeepExtraChar("__abcdef", 1, 2, 3) =
["def", "bc", "a", "__"]` - (RA, R)
* `StringUtils.splitByLengthFromEndKeepExtraChar("__abcdef", 1, 2, 3) =
["f", "de", "abc", "__"]` - (E2S, R)
* `StringUtils.splitByLengthFromEndKeepExtraChar("__abcdef", 1, 2, 3) =
["__", "abc", "de", "f"]` - (E2S, NR)
I think the first one is more readable, we can visually understand the
splitting, but may be less intuitive :
```
StringUtils.splitByLengthFromEnd("ABCDEFGHIJKLM", 3, 4, 5) = ["BCD",
"EFGH", "IJKLM"]
[3][4_][_5_]
ABCDEFGHIJKLM
```
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at [email protected] or file a JIRA ticket
with INFRA.
---
