i'm using ant replaceregexp task replace text in on 10,000 files. each file has multiple search/replace patterns need perform. of right now, have multiple blocks, each fileset of "**/*.java".
as list of things search/replace grows, takes forever run since replaceregexp #1 on 10k files, replaceregexp2 on 10k files, , on. have 15 replaceregexp blocks, task taking 45 minutes.
is there way each file processed/accessed once, , performs 15+ of regex search/replaces on file rather file having processed 15+ individual times?
in essence, looking logically this:
<replaceregexp match="firstmatch" replace="firstraplce" match="secondmatch" replace="secondreplace" etc.. byline="true"> <fileset dir="."> <include name="**/*.java"/> </fileset> </replaceregexp>
that's invalid xml, shows conceptually i'm looking for. use 2 groups - 1 "firstmatch/firstreplace" , 1 "secondmatch/secondreplace" on same set of files.
thanks.
the following works:
<filterchain> <tokenfilter> <replacestring from="oldstr1" to="newstr1"/> <replacestring from="oldstr2" to="newstr2"/> </tokenfilter> </filterchain>
Comments
Post a Comment