It would be nice to know where a node ends. For the start of a token, we have node.coord. Similarly, it would be nice if we also had node.endcoord. The reason why I want this feature is that it would enable precisely underlined error messages like GCC has:
main.c: In function ‘main’:
main.c:2:13: warning: initialization of ‘int’ from ‘char *’ makes integer from pointer without a cast [-Wint-conversion]
2 | int x = "asdf";
| ^~~~~~
Here is where the coord is parameter is retrieved from yacc:
I am not sure if the end position can be obtained from yacc. In case this is not possible, here is a hacky idea: Find the end position by scanning backwards from the next token until a non-whitespace character is found.
I do not have the time to implement this right now, but I thought I should start a discussion here in case someone has a better idea.
It would be nice to know where a node ends. For the start of a token, we have
node.coord. Similarly, it would be nice if we also hadnode.endcoord. The reason why I want this feature is that it would enable precisely underlined error messages like GCC has:Here is where the
coordis parameter is retrieved from yacc:https://github.com/eliben/pycparser/blob/f7409953060f1f4d0f8988f1e131a49f84c95eba/pycparser/plyparser.py#L55
yacc defines e.g. lineno here:
https://github.com/eliben/pycparser/blob/f7409953060f1f4d0f8988f1e131a49f84c95eba/pycparser/ply/yacc.py#L258
I am not sure if the end position can be obtained from yacc. In case this is not possible, here is a hacky idea: Find the end position by scanning backwards from the next token until a non-whitespace character is found.
I do not have the time to implement this right now, but I thought I should start a discussion here in case someone has a better idea.