Spring/spring-batch
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

Revert the behavior of DelimitedLineTokenizer to pre-4

Browse Source 
DelimitedLineTokenizer was refactored to improve its performance but its behavior was also slightly changed.
When the input is an empty csv string with double quotes, e.g. the java
String line = "\"\""; the tokenizer returns the same String, whereas
before it was returning an empty string.

This is due to the condition checking for quotes to remove not stripping
the double quotes from the empty csv string (because its length is 2 and
the code check for strictly greater than 2).
The previous code is there 973fe44d6c/spring-batch-infrastructure/src/main/java/org/springframework/batch/item/file/transform/DelimitedLineTokenizer.java (L202)  and checking for
double quotes at the start and end of the string without special
handling for the empty string.

At test is also added to ensure this won't break again.

Resolves BATCH-2657
This commit is contained in:
Olivier Bourgain
2017-11-03 15:32:54 +01:00
committed by Michael Minella
parent 31c7e8f83f
commit 62a580f52d
2 changed files with 7 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
spring-batch-infrastructure/src/main/java/org/springframework/batch/item/file/transform/DelimitedLineTokenizer.java
View File

@@ -222,7 +222,7 @@ public class DelimitedLineTokenizer extends AbstractLineTokenizer
String value;
if ((chars.length > 2) && (chars[start] == quoteCharacter) && (chars[start + len - 1] == quoteCharacter)) {
if ((chars.length >= 2) && (chars[start] == quoteCharacter) && (chars[start + len - 1] == quoteCharacter)) {
value = new String(chars, start + 1, len - 2);
if (value.contains(escapedQuoteString)) {
value = StringUtils.replace(value, escapedQuoteString, quoteString);