Issue No. 06 - November/December (2001 vol. 18)
DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/52.965809
Constructing analysis and modification tools for software assets is laborious because you first need to implement the underlying parser for the software's specific programming language. These implementations are generally not in the public domain. So, parser development for any of the 500+ languages in use today implies a major up-front investment. The authors propose a solution that will work for virtually all languages: the rapid development of renovation parsers by stealing the grammars. They also share lessons learned.
C. Verhoef and R. Lämmel, "Cracking the 500-Language Problem," in IEEE Software, vol. 18, no. , pp. 78-88, 2001.