projectAnonymizer bug


when using projectAnonymizer with some strings there is a bug that where does not replace the strings as it should but mixes them around instead and partly even duplicates part of the string and leaves it unchanged for 1 copy of it.

There is no difference whether to use a complex rule to catch all the strings to be converted or multiple smaller rules to catch them one by one. So I either leave these rules out or rearrange strings in Notepad++ manually...


Rules applied:

for seg 3 and 5:

{\d, date, long}|((\()?{[0-9], plural, (=1|one|other) {(#|[0-9])?)|(} other(\s)?{(#(\s)?)?)|(}}(\))?)

for seg 66:

({\d, select, (WITH_NAME|ADMIN){({\d})?)|(} other{)|(({{|{)(?!(font|page|credit)(s)?)((#)?\d+|[a-zA-Z]+((\s|_)[a-zA-Z]+)*))?(}}|})|(}}|})


Starting point:




I guess you can see what went wrong. Maybe the devs can find out why it ran wrong.

I'm setting up my regex expressions with RegexBuddy before entering them into projectAnonymizer and I also found out that projectAnonymizer sometimes seems to handle the rules differently.




Parents Reply

  • In fact I'm wrong... just realised you said 2017. The SDL Data Protection Suite is only available for 2019 and we have not implemented any of the fixes that went into this version in 2017. This applies to many of the enhancements we introduced prior to your posts as well. We also won't be doing this for 2017.

    The sourcecode is available if any developer wishes to do this, but we will not. The work is integrated into the SDL Data Protection Suite and we are only updating this version.