sql_io.py: put_table(): is_function: Fixed bug where need to add the pkeys table's test pkey constraint after the data is added rather than when the empty table is created, to avoid adding a pkey constraint that will later be violated by data which returns multiple output rows for an input row (such as calls to _split())

sql_io.py: put_table(): insert_into_pkeys(): Allow callers to override run_query_into()'s add_pkey_ param in case the initial version of the pkeys table should not yet have the test pkey constraint (e.g. because data is added after the table is created)

README.TXT: Data import: Checking for errors: Search for "Command exited with non-zero status" to find errors, which is faster than checking that each input's log ends in "Encountered 0 error(s)"

inputs/import.stats.xls: Updated import times

README.TXT: Data import: import_all: Corrected text of note about time until control is returned to the shell

README.TXT: Data import: Moved download of logs to right after the import is done, because this is a quick step that doesn't depend on the backup- and export-creation steps

mappings/VegCore-VegBIEN.csv: institutionCode: Removed mapping to sourcename.matched_source_id, which is now autopopulated. Split any list of institutionCodes apart using new _split().

schemas/vegbien.sql: sourcename: Added sourcename_set_matched_source_id() trigger

schemas/functions.sql: Added _split()

schemas/vegbien.sql: sourcelist_unique: Removed COALESCE around name because it's NOT NULL