Hello again, So after discussions with Jan and further contemplation, I concluded that the only way to get a hold on the '#line ... ' issues in the generated parse.c is to write a dedicated tool to do it. The patch herein contains such a tool, named 'lineclean'. The tool detects sequences of the form
#if(n)def .... #line ... "parse.c" #else #line ... "parse.c" #endif and replaces them with a single line #line ... "parse.c" Also, lineclean fixes up all '#line ... "parse.c"' commands so that they actually refer to the next line in the file. (This was completely broken as a result of running the diff). Anyhoo, since I still haven't heard from any of the maintainers, I'll leave it at that. Should you decide to apply the patch (with attribution, of course :-)) and run into problems, please e-mail me directly (zlaski _at_ ziemas _dot_ net) since I'm not a member of sqlite-users. Thank you, --Zem _______________________________________________ sqlite-users mailing list sqlite-users@mailinglists.sqlite.org http://mailinglists.sqlite.org/cgi-bin/mailman/listinfo/sqlite-users
