User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:22.214.171.124) Gecko/2008071615 Fedora/3.0.1-1.fc9 Firefox/3.0.1 Build Identifier: elsa changeset 2432 XmlLexer::read_xml_string() is calling XmlLexer::yyunderflow() to fetch next char from buffer. If yy_c_buf_p is pass EOB it return 0. That is, read_xml_string() function is incrementing yy_c_buf_p variable before calling yyunderflow(). The responsibility of yyunderflow function is to update the buffer when it is reached. Which is when *yy_c_buf_p == 0. Reproducible: Always Steps to Reproduce: 1. in oink directory 2. make check #should fail 3. Actual Results: it fail on my system for previous reasons
now, i am reaching following failure # qual-check-resrz/../elsa/in/t0148.cc ./qual -fq-no-explain-errors -fq-no-names ../elsa/in/t0148.cc -o-srz /dev/shm/oink.buildtmp/test/t0148.cc.1.qdir -tr nohashlines -q-config ../libqual/config/taint.lattice -fq-no-compactify-graph qual: Assertion failed: !obj->target->type->asRval()->isCompoundType() && "53f8abcb-a6e1-40a9-9b95-53adb8aa3b86", file dataflow_visitor.cc line 1018 make: *** [qual-check-resrz/../elsa/in/t0148.cc] Error 64
Yeah. I'd like to find this regression. Unfortunately the fact that oink consists of so many repositories and that the initial svn import screwed up things a little makes doing a bisect impossible. Even a manual equivalent of bisect does not appear practical at this point. I'll debug this isCompoundType thing by hand and then investigate your other fix. I don't get that failure here(but I also run an older flex).
At this time, Mozilla is no longer pursuing the use of Elsa and Pork for static analysis and rewriting. Therefore, there are no longer any intentions to fix or support any outstanding bugs in these tools.
Status: UNCONFIRMED → RESOLVED
Last Resolved: 8 years ago
Resolution: --- → WONTFIX
You need to log in before you can comment on or make changes to this bug.