Hi all, I've looked through the code base of [text] and I already did some clean up. However there are a few things, that need more work IMHO:
1. IP clearance! There are several comments talking about code adapted from PHP libraries. This needs to be listed in NOTICE.txt. For an example see the NOTICE.txt in [lang]. 2. The names package needs work. Currently the HumanNameParser takes a Name object (which is just a wrapper around a String) as parameter. The parse result is then saved in fields (e.g. firstName). This makes the parser unusable after it has parsed a name. I think the API should be reworked such as: - The constructor of the parser takes configuration options which can be reused for several names to parse - the parse method takes a string as parameter, containing a name - the parse method returns an immutable Name objects which has getters for firstName, lastName etc. 3. There are a some useless JavaDocs, which can be dropped. For example: /** * Middle name. */ private String middle; Better rename the variable to middleName and drop the comment. 4. Several classes have no tests. The overall test coverage looks good, so this may not be a problem. 5. I don't think the shade configuration is correct. To me it looks like commons-text depends on [lang] and also shades some classes. Do we really need the explicit dependency? Furthermore, why do we need to shade so many classes? Why do we need anything beside StringUtils? Nevertheless, this library looks very good. Kudos to Bruno for pushing this forward! Regards, Benedikt -- http://people.apache.org/~britter/ http://www.systemoutprintln.de/ http://twitter.com/BenediktRitter http://github.com/britter
