InfoVeg

A number of problems are encountered in Caroline Bernard's hand error checking:

1) The database crashes frequently in this mode in Access 2007 on Caroline's computer. Copying to local drive didn't seem to help. It seems to be an Access problem working with this amount of data. Perhaps some indexing could be added to help it work with the large amount of data, or perhaps we should load only one plot at a time. As a temporary workaround, we have upgraded project 89 to Access 2007 file format, and so far this seems to have solved the intermittent problem.

2) On the tree stems, some error is causing the database to crash from function: select0IfThere() which is a trivial function that should simply ignore errors. MTL to add error ignoring to that function.

3) there is also evidence in the logs of errors with too much data trying to be inserted when adding a new species in FormAddSpecies sci_name_afterUpdate. MTL to try to reproduce this bug and fix it.

4) log dates are included in the diff between versions of error embedding and these can't be caught and should not be included in the diff. Last update plot, row and person last updating plot and row are all included in the diff.

5) DBA mode takes precedence over error embedding, which is backwards. Error embedding should take precedence so that no one ever embeds errors in the real data.

In general, some hammering on the error embedding should help.