Edwin de Jonge

- generate a warning that weights can be set,(when no weights are specified) and mention `set.seed` 1) To make users aware that set.seed most of the times needs to be...

It seems useful to add (several) weight suggestion methods, that generate a weight matrix for a data set.

Allow [generalized Feligi Holt edits](https://unece.org/fileadmin/DAM/stats/documents/ece/ces/ge.44/2014/mtg1/Topic_2_NL_Scholtus.pdf) to be specified. - all mechanical components to handle those edits are available (inside) - first, the specification of soft rules should be exported (allowing...

Currently `errorlocate` issues a warning that some of the rules are skipped, because they are not record-based or non-linear. A novice user should be warned more clearly, that this can...

It would be nice to have rbind functionality to the results of locate_errrors. Simular to editrules.

Currently `errorlocate` ignores `lin_eq_eps` and `lin_ineq_eps`. Should be fixed, because errorlocate must obey validate semantics. If according to `validate` data is valid, errorlocate should not mark it as an error.

Since MIP functionality is desirable at multiple place, it may be handy to split of the translation to MIP and vice versa in a seperate MIP package. This should also...

`localize_errors` should allow for multiple error locaters, that each use a subset of the rules. unused rules are reported.

Current replace_errors changes errors into "NA". Preferably the user can provide different options, e.g. the suggestion generated by the mip model.