4.0.1

• Variables in cookie definitions didn't get properly replaced. (#23)

4.0.0

• Upgrade code base to XLT 7.x / HtmlUnit 3.x
• Update pom.xml to publish library artifact at Maven Central
• Update dependencies where possible

3.2.0

• Cookie and Header response extractors support optional regex/group parameters now. (#16)

3.1.0

• xlt-nocoding checks HTTP responses for status code 200 by default. If this is not wanted, you can now switch that off by setting com.xceptance.xlt.nocoding.defaultHttpCodeValidation = false. (#15)
• Parser instances now allow to read scripts not only from files, but also from arbitrary sources. (#14)

3.0.0

• Requires Java 11 to compile and run.
• Based on XLT 6.x.x.

2.3.0

• #9 Added Store: Delete to be able to clean the entire data pool. Thanks to @m-31 for providing the base PR.
• #11 Updated dependencies of XLT, CSV, and GSON
• #12 Beanshell not longer silently fails but breaks to report the reason of the problem for easier diagnostics, this might break some scripts because commands now have to succeed and non-existing variables might cause complains here as well incorrect Java code (Beanshell)

Please note, that Beanshell currently has issues to support dynamic argument lists for String.format and MessageFormat.format. It will always complains about a missing signature, because it looks for the specific class instead for Object.. as second parameter.

2.2.2

• The error message when the actual count differs from the expected one now shows both, the actual count and the expected one.

2.2.1

• Validation errors now show which action they belonged to.
• CsvParser now throws an error if a record doesn't have enough elements instead of skipping it.
• Errors during parsing CSV files now show the line number.
• Actions don't create a new WebClient anymore.

2.2.0

• Introduced a new mode with better performance called request
  • No pages are rendered, only HTTP request and responses are sent and received. This makes this mode even faster than the light mode, which renders a lightweight page.
  • Some XLT settings are ignored. For example, since no page is created static content is not downloaded even with the appropriate setting enabled.
  • XPath is still available in this mode. However, if it is used, a page is created once for the associated request, which is very costly.

2.1.0

In response to issue #4, version 2.0.2 used Snakeyaml to resolve Yaml anchors and aliases. The resulting structure was then transformed back to Yaml and parsed with Jackson, the parser also used in 2.0.0 and 2.0.1, to avoid rewriting a parser from scratch. However, this introduced wrong line numbers in error messages since Yaml files were parsed twice.
Hence, version 2.1.0 changed the parser from Jackson to Snakeyaml.
Furthermore, error messages during parsing were also reworked and now show some context to an error and the actual error.