On 29 Jan 2014, at 18:48, Christian Schoenebeck <schoeneb...@crudebyte.com> wrote:
> Hi everybody! Hi! > I am currently working on a shell and wrote custom C++ code that takes > Bison's > symbol stack at a certain point (i.e. when an error occured) and walks ahead > the Bison generated yyfoo tables / tree of the shell's grammar. I added that > custom code for two purposes: 1) providing very user friendly error messages > on syntax errors and 2) providing the user visual auto completion & auto > correction while typing in the shell. > > It is already working, however to make those syntax error messages and auto > completion suggestions even more appropriate & human friendly, I would need > to > stick some additional custom meta informations to the individual grammar > rules. It seems that your should use LR(1); see the Bison manual, sec. 5.8.1, "LR Table Construction". Bison by default uses LALR(1), which merges the states in such a way that when an error occurs, though no further tokens will be read, some additional reductions may be applied before an error is issued. Using LR(1) guarantees that that will not happen; so you can check the lookahead tokens in the state, and that is also the set of valid completions. I do not know if LR(1) works with C++, though. And you may need to write some code in order to find the lookahead tokens; you might want having this code integrated into the Bison distribution, so that the feature does not break with updates. Hans _______________________________________________ help-bison@gnu.org https://lists.gnu.org/mailman/listinfo/help-bison
