> The one thing I wish they would do, though, is improve the error messages when a query fails. Yea, we definitely need to do better there. But it's harder tha

> The one thing I wish they would do, though, is improve the error messages when a query fails. Yea, we definitely need to do better there. But it's harder than one might think initially :(. My suspicion is that we (the PG developers) will have to write our own parser generator at some point... We don't really want to go to a manually written recursive descent parser, to a large part because the parser generator ensuring that the grammar doesn't have ambiguities is very helpful. > At the very least, if I know which position in the text is the beginning of an error, I can fix it. For many syntax errors and the like we actually do output positions. E.g. postgres[1429782][1]=# SELECT * FROM FROM blarg; ERROR: 42601: syntax error at or near "FROM" LINE 1: SELECT * FROM FROM blarg; ^ (with a mono font the ^ should line up with the second FROM) If you have a more concrete example of unhelpful errors that particularly bug you, it'd be helpful. In some cases it could be an easy change.