inputs/*/Source/map.csv for non-herbaria: Mapped sampleType

inputs/.herbaria/herbaria/map.csv: Set sampleType to "specimen"

mappings/VegCore-VegBIEN.csv: Mapped sampleType

mappings/VegCore.csv: Added sampleType

schemas/vegbien.sql: source: Added sampletype field to indicate a plot or specimen datasource

schemas/vegbien.sql: Added sampletype enum

root Makefile: $(postgresReload-*): Confirm the operation before continuing, since it involves changing PostgreSQL config files in nontrivial ways. Added instructions for setting kernel.shmmax to at least 4GB minus 1 byte on Linux, to work with the shared_buffers setting in postgresql.conf.

schemas/postgresql.conf: shared_buffers: Documented that it must be less than ~95% of SHMMAX

schemas/vegbien.sql: analytical_stem_view: identifiedBy: Fixed bug where need to use party.fullname instead of name components because the name is now mapped to fullname

schemas/vegbien.sql: analytical_stem_view, darwin_core_view: dateCollected: Use the parent plot event's obsstartdate when the subplot event does not have its own obsstartdate

schemas/vegbien.sql: analytical_stem_view: Don't filter out rows without a date or non-current taxondeterminations

schemas/vegbien.sql: analytical_stem_view: Don't filter out rows without a date

schemas/vegbien.sql: Added darwin_core_view

schemas/vegbien.sql: analytical_stem_view: identifiedBy: Fixed bug where need to use party.fullname instead of name components because the name is now mapped to fullname

schemas/vegbien.sql: sync_analytical_*_to_view(): Added CREATE INDEX statements

README.TXT: Data import: Added steps to publish analytical DB on nimoy.bien_web

schemas/vegbien.sql: analytical_stem_view: Changed JOINs to LEFT JOINs to include occurrences without taxondeterminations

export_analytical_db: Use 'NULL' as the NULL value instead of \N, because MySQL has problems with \N

publish_analytical_db: Load to bien3_adb instead of bien_web

README.TXT: Data import: Added step to export analytical DB

root Makefile: $(postgres-Linux): Fixed bug where need $(asAdmin) before commands to rename existing *.conf

root Makefile: $(postgres-Linux): Also install postgresql-contrib, which contains the hstore extension

Added inputs/NVS/

inputs/CVS/Organism/map.csv: Mapped accordingTo to "Weakley 2006"

inputs/NY/Specimen/map.csv: Omit UniqueNYInternalRecordNumber to avoid confusion since this is an internal-only ID. This makes InstitutionCode+CollectionCode+CatalogNumber the globally unique identifier instead.

README.TXT: Added Datasource refreshing section with instructions for refreshing VegBank

schemas/vegbien.sql: Renamed taxonconcept.concept_source_id back to concept_reference_id

schemas/vegbien.sql: Renamed soilobs to soilsample per working group discussion

input.Makefile: SVN: add: verify: Fixed bug where need to use $ prefix before string to parse newline

input.Makefile: SVN: add: verify: Fixed bug where need to use $ prefix before string to parse newline

inputs/NY/verify/: svn:ignore .csv files

input.Makefile: SVN: add: Also svn:ignore .csv files

export_analytical_db: Export NULL as \N to work with MySQL

schemas/vegbien.sql: analytical_*: Added index on NOT NULL columns, starting with institutionCode

schemas/vegbien.sql: analytical_*: Removed primary keys and NOT NULL constraints on columns that sometimes have NULL values

publish_analytical_db: Added CSV dialect information

root Makefile: PostgreSQL: $(postgresReload-*): Rename existing *.conf to *.conf.old

publish_analytical_db: Use LOAD DATA LOCAL INFILE instead of LOAD DATA INFILE to avoid needing FILE permissions on bien_web

Added publish_analytical_db

export_analytical_db: Append the public schema version to the CSV filename

backups/Makefile: $(rsyncBackups): Added *.csv

Added export_analytical_db

backups/: Ignore _* and *.csv

make_analytical_db: mk_analytical_table(): Use explicit schema references everywhere. This fixes a bug where the TRUNCATE/INSERT steps on the public schema's table would reference the analytical_db view instead because they were not schema-scoped.

make_analytical_db: mk_analytical_table(): Factored table references in different schemas out into vars

schemas/vegbien.sql: analytical_stem_view: recordNumber: Combine identifying fields in taxonoccurrence, plantobservation, and stemobservation to ensure that this field is unique within the plot and not NULL

Regenerated vegbien.ERD exports

make_analytical_db: Moved set -x () around just psql_verbose_vegbien so embedded $() expressions wouldn't also be in set -x (verbose) mode

make_analytical_db: Fixed bug where need to use bash instead of sh because vegbien_dest requires it

make_analytical_db: Factored analytical_* table creation code out into mk_analytical_table() function

make_analytical_db: Create analytical_db views pointing to the analytical_* versions in the public schema

vegbien_dest: $schemas: Removed analytical_db because views that will be added to it were shadowing public schema tables with the same names during population of those tables in make_analytical_db

vegbien_dest: Export $public, to make sure it's available to any invoked scripts as an env var

vegbien_dest: $schemas: Added analytical_db

inputs/import.stats.xls: Added separate tab with stats for 2012-6~9. The Excel format apparently only supports 255 columns, so previous imports had been silently truncated off. Note that once the 2012-10 imports reach column 255, a new tab will need to be created with the 2012-10+ imports.

bin/map: in_is_db: by_col: Clearing errors table: Skip this if the table has been set to None because it didn't exist (and thus was a metadata-only map spreadsheet)

schemas/vegbien.sql: analytical_stem_view: scientificNameWithMorphospecies: Fixed bug where need to use the specific_epithet from the accepted_taxonverbatim rather than the parsed_taxonverbatim

schemas/vegbien.sql: analytical_stem_view: scientificNameWithMorphospecies: Include the family any time the genus is not specified, instead of just when accepted_taxonlabel.rank = 'family'. These should have the same effect since TNRS includes the rank, but using COALESCE is clearer.

schemas/vegbien.sql: analytical_stem_view: scientificNameWithMorphospecies: Changed to also include morphospecies when just the family is specified

schemas/vegbien.sql: analytical_stem_view: Fixed bug where location.authorlocationcode needed to be used as the plotName when location.sourceaccessioncode was not provided, to ensure that plotName would be NOT NULL

inputs/FIA/import_order.txt: Fixed bug where FIA_COND_unique needed to be explicitly included in import_order.txt now that we're using import_order.txt to import the Source metadata table before the data tables

inputs/import.stats.xls: Updated import times

root Makefile: PostgreSQL: $(postgresReload-Linux): Try chmoding both as your user and as the bien user

input.Makefile: Testing: $(runTest): Ignore failed diffs when the test is compared to another test's output (e.g. in by_col mode)

bin/map: in_is_db: If table does not exist, set table to None so that db_xml.put_table() doesn't try to access it. This fixes a bug in metadata-only map spreadsheets under column-based import.

db_xml.py: put_table(): Support None in_table by calling put() directly

Removed no longer used geoscrub.*.sql. Use geoscrub_output instead.

Removed no longer used geoscrub_cleaned_unique. Use geoscrub_output instead.

Removed no longer used geoscrub_cultivated. Use analytical_stem_view.cultivated instead.

Removed no longer used geoscrub_cultivated. Use analytical_stem_view.cultivated instead.

schemas/vegbien.sql: analytical_stem_view: cultivated: Removed BIEN2's geoscrub_cultivated, which has now been replaced by the primary corresponding scripts (and never had particularly many matches to the locations in any case)

schemas/vegbien.sql: analytical_stem_view: cultivated: Use OR instead of _or() to combine cultivated_family_locations.country IS NOT NULL with the other values, because this field's false value should not be used in place of NULL if all the other values are NULL, as it would be with _or(). (cultivated_family_locations.country IS NOT NULL can indicate presence, but not absence, of cultivated status.)

schemas/functions.sql, vegbien.sql: _and(), _or(): Added comment comparing the function and the corresponding logical operator

schemas/vegbien.sql: public: Added _or(), for use by analytical_stem_view

schemas/vegbien.sql: analytical_stem_view: cultivated: Also set if family/country combination found in cultivated_family_locations

schemas/vegbien.sql: cultivated_family_locations: Added data from nimoy:/home/boyle/bien2/geoscrub/cultivated/cult_by_taxon/flag_by_taxa.inc

schemas/vegbien.sql: Added cultivated_family_locations to store locations where various taxon families are considered cultivated

mappings/VegCore-VegBIEN.csv: Mapped locality description fields to location.iscultivated using _locationnarrative_is_cultivated()

xml_func.py: Simplifying functions: Added passthru entries for _and, _or

schemas/vegbien.sql: Added _locationnarrative_is_cultivated()

lib/PostgreSQL-MySQL.csv: Change text to varchar(255) because text columns can't be used in indexes in MySQL

lib/PostgreSQL-MySQL.csv: Resaved in Excel, which removed unnecessary quotes around fields

schemas/vegbien.sql: analytical_aggregate: Added identifiedBy, which is no longer a scoping field (which would prevent scientificNameWithMorphospecies from being unique) now that there is only one taxondetermination for each taxonoccurrence

schemas/vegbien.sql: analytical_stem_view: dateCollected: For plots data, use the locationevent obsstartdate instead of the collectiondate in order to group taxonoccurrences/stems from the same locationevent together

schemas/vegbien.sql: analytical_* pkeys: Added dateCollected because the records are actually unique within the location*event*, not the location

schemas/vegbien.sql: analytical_stem_view: Exclude records with no collectiondate or obsstartdate, which is required to uniquely identify a record

analytical_stem_view: dateCollected: Use locationevent.obsstartdate when aggregateoccurrence.collectiondate is not provided

schemas/vegbien.sql: analytical_stem_view: Include only the current taxondetermination for each taxonoccurrence, to avoid cross-joining taxondeterminations with stems and thus multiplying the number of rows for datasources that have multiple taxondeterminations per taxonoccurrence

schemas/vegbien.sql: taxondetermination: Added AFTER trigger to set the current taxondetermination for the taxonoccurrence

lib/PostgreSQL-MySQL.csv: Statements ending in ";": When matching any character, use .*? (with the (?s) flag) instead of [^;]* in order to allow embedded ; to be matched. This fixes a bug where a CREATE VIEW statement was not removed because it contained an embedded ; .

schemas/vegbien.sql: taxondetermination: Added unique index to ensure that there is only one current determination for each taxonoccurrence

lib/PostgreSQL-MySQL.csv: Remove indexes with WHERE clauses

schemas/vegbien.sql: analytical_aggregate: Added primary key on institutionCode, plotName, scientificNameWithMorphospecies, recordNumber. Note that this makes these fields NOT NULL, which should not be a problem because there are inner joins instead of LEFT JOINs on most of the tables which provide them, and LEFT JOINed tables have their identifying fields combined to create a NOT NULL value.

schemas/vegbien.sql: analytical_stem_view: recordNumber: Combine identifying fields in taxonoccurrence, plantobservation, and stemobservation to ensure that this field is unique within the plot and not NULL

lib/PostgreSQL-MySQL.csv: Only match a statement-terminating ; when it's at the end of a line

schemas/vegbien.sql: analytical_aggregate: Added primary key on institutionCode, plotName, scientificNameWithMorphospecies. Note that this makes these fields NOT NULL, which should not be a problem because there are inner joins instead of LEFT JOINs on the tables which provide them.

db_xml.py: put(): _setDefault(): Delay the evaluation of each col_default's value until the col_default is actually retrieved. This fixes a bug in the source table mappings where the explicit source entry was being created after the col_default source entry, causing the initial entry, which did not have the additional fields populated, to be used instead.

dicts.py: Added WrapDict, a dict that runs a function on each value retrieved

db_xml.py: put(): _setDefault(): Fixed bug where need to copy col_defaults before calling update() on it, to avoid modifying the input value (which may be reused by the caller, expecting it to be unmodified)

db_xml.py: put(): col_defaults param: Fixed bug where need to use None as default value, because col_defaults will be modified by put() and the {} default value is a global instance