schemas/util.sql: added pg_get_viewdef() wrapper, which unexpands expanded * expressions

bugfix: schemas/util.sql: mk_drop_from_create(): need to match first rather than last CREATE

schemas/util.sql: recreate_view(): support omitting the view_query if the view has already been modified (eg. for public.*_view, which allow changing the view as a separate step)

bugfix: schemas/util.sql: recreate(): need to handle case where util.mk_drop_from_create() is NULL

bugfix: schemas/util.sql: mk_drop_from_create(): only match CREATE if no custom DROP came before it

fix: schemas/util.sql: explain2notice_msg_if_can(): also need to catch invalid_cursor_definition ("cannot open multi-query plan as cursor")

schemas/util.sql: added copy()

schemas/util.sql: added copy_data()

schemas/util.sql: added seq__reset()

schemas/util.sql: added seq__create()

fix: schemas/util.sql: try_cast(), is_castable(): also catch invalid_schema_name, thrown by `'pg_temp.__'::regclass`

fix: schemas/util.sql: mk_drop_from_create(): also support CREATE queries that include the SELECT statement on the same line as the CREATE

schemas/util.sql: create_if_not_exists(): print message if already exists, so the function doesn't inexplicably appear not to have run at all

schemas/util.sql: added recreate_view(), a special case of util.recreate()

fix: schemas/util.sql: recreate(): usage: use `schema` instead of `schemas`

schemas/util.sql: recreate(): perform the correct DROP VIEW in the function itself so that the caller does not have to worry about forming it properly

bugfix: schemas/util.sql: mk_drop_from_create(): added `DROP`

schemas/util.sql: added mk_drop_from_create()

schemas/util.sql: added regexp_match()

fix: schemas/util.sql: force_recreate(): renamed to just recreate(), because "force" normally implies that things will be deleted, which this function does not do

schemas/util.sql: show_create_view(): use new mk_set_relation_metadata()

schemas/util.sql: added mk_set_relation_metadata()

schemas/util.sql: force_recreate(): documented usage, which is somewhat complex and has several important parts

bugfix: schemas/util.sql: restore_views(): need to specify view name for a manual existence check, in case the view definition becomes invalid, which would produce a nonstandard (uncatchable) exception

bugfix: schemas/util.sql: explain(): don't display any log messages printed by IMMUTABLE functions in the query, which EXPLAIN apparently runs

schemas/util.sql: create_if_not_exists(): added relation param, which can be used to ensure that a standard exception is always generated if the relation exists, even if the table definition would be invalid (which generates a variety of exceptions)

bugfix: schemas/util.sql: try_cast(), is_castable(): also need to handle syntax_error_or_access_rule_violation, which is thrown by ::regclass

schemas/util.sql: added relation_exists()

schemas/util.sql: save_drop_views()/restore_views(): store the view path in addition to the definition so it can be used in restoring

bugfix: schemas/util.sql: in_reverse(): can't use `SELECT *` for composite types because this expands them

bugfix: schemas/util.sql: array_reverse(): can't use `SELECT *` for composite types because this expands them

schemas/util.sql: save_drop_views()/restore_views(): use opaque return type so it can be changed without changing the local var type in functions that use this

schemas/util.sql: force_recreate(): recreate_users_cmds: renamed to restore_views_info since this is now whatever is needed by util.restore_views()

schemas/util.sql: added restore_views() and use it in force_recreate()

bugfix: schemas/util.sql: save_drop_views(): views must be dropped in reverse dependency order, but returned in forward dependency order

bugfix: schemas/util.sql: force_recreate(): dependent views need to be re-created in forward dependency order. this is done by reversing the dependencies only for save_drop_views().

bugfix: schemas/util.sql: mk_set_comment(): need to use VIEW instead of TABLE for views

bugfix: schemas/util.sql: show_create_view(): also need to include view comment

fix: schemas/util.sql: mk_set_comment(): added ;

fix: schemas/util.sql: show_set_comment(): don't display a COMMENT ON statement if no comment, because this will be appended to table defs, etc. and would create clutter

bugfix: schemas/util.sql: mk_set_comment(): need to handle NULL comment properly

schemas/util.sql: added show_set_comment(regclass)

schemas/util.sql: set_comment(): use util.mk_set_comment()

schemas/util.sql: added mk_set_comment()

bugfix: schemas/util.sql: force_recreate(): recreate_users_cmds: use create_if_not_exists() rather than eval(), because cmd might manually re-create a deleted dependent view, causing it to already exist

bugfix: schemas/util.sql: save_drop_view(): util.show_create_view(): don't include `OR REPLACE` because that causes nonuniform errors (eg. invalid_table_definition), instead of the standard duplicate_table exception caught by util.create_if_not_exists()

fix: schemas/util.sql: show_create_view(): removed extra trailing ; because pg_get_viewdef() now includes one

schemas/util.sql: show_create_view(): made inclusion of `OR REPLACE` configurable

schemas/util.sql: save_drop_view(): use util.drop_view()

schemas/util.sql: save_drop_views(): return text[] instead of concatenated text so that the queries will be run individually and be easier to debug

schemas/util.sql: added eval(queries text[])

bugfix: schemas/util.sql: force_recreate(): need to drop users in reverse to avoid dependencies that block the drop

schemas/util.sql: added array_reverse()

schemas/util.sql: added in_reverse()

bugfix: schemas/util.sql: force_recreate(): need to use :lower:, not [:lower:]

schemas/util.sql: force_recreate(): debug-print PG_EXCEPTION_DETAIL, users to assist in debugging

fix: schemas/util.sql: debug_print_value(): added trailing newline to visually distinguish the printed value

schemas/util.sql: added debug_print_var()

schemas/util.sql: debug_print_return_value(): use new util.debug_print_value()

schemas/util.sql: added debug_print_value()

fix: schemas/util.sql: util.force_recreate(): also support dropping things besides tables

bugfix: schemas/util.sql: explain2notice_msg_if_can(): don't use util.is_explainable() because the list provided by Postgres (http://www.postgresql.org/docs/9.3/static/sql-explain.html#AEN77691) excludes some query types that are in fact EXPLAIN-able

bugfix: schemas/util.sql: materialize_query(): contents generated from: need to include search_path so query is runnable

schemas/util.sql: added loop_ignore_errors()

schemas/util.sql: added _km_to_m() for use when public._km_to_m() is not defined yet

bugfix: schemas/util.sql: is_castable(): need to pass NULL through, for proper NULL propagation

schemas/util.sql: added is_castable()

schemas/util.sql: added try_cast()

schemas/util.sql: added util.cast(), which allows casting to an arbitrary type without eval()

bugfix: schemas/util.sql: remake_diff_table(): need to rm_freq() type_table, because left/right_table don't have freq yet

schemas/util.sql: auto_rm_freq(): use new rm_freq()

schemas/util.sql: added rm_freq(regclass[])

schemas/util.sql: remake_diff_table(): result table comment: documented how to display NULL values that are extra or missing

fix: schemas/util.sql: trim(): by default, cascadingly drop dependent columns so that they don't prevent trim() from succeeding. note that this requires the dependent columns to then be manually re-created.

bugfix: schemas/util.sql: is_constant(util.col_ref): updated to include standard newline at beginning of comment (applies to newly-imported staging tables)

fix: schemas/util.sql: explain2notice_msg(): don't include EXPLAIN output for simple, single-value queries, to avoid cluttering up the log output

schemas/util.sql: added fold_explain_msg()

bugfix: schemas/util.sql: set_col_types(): need to COALESCE the executed SQL to '' because util.eval() does not support NULL (and shouldn't, because this indicates a missing COALESCE in constructing the statement)

schemas/util.sql: set_col_types(): use simpler util.eval() instead of manual EXECUTE/util.debug_print_sql()

schemas/util.sql: set_col_types(): use string_agg() instead of array_to_string(ARRAY) for clarity

schemas/util.sql: added mk_not_null()

bugfix: drop_column(regclass[]): need to run `SELECT NULL::void;` at end of function to avoid folding away functions called in previous query

fix: schemas/util.sql: diff(regclass, regclass): moved try_create() of copies column in parent table to auto_rm_freq() so that it would only happen if both tables actually contain a copies column (otherwise, the try_create() will create an empty copies column if both tables are empty)

schemas/util.sql: try_create(): also handle "child table is missing column" errors

schemas/util.sql: added coalesce(anyarray), which can be used to force evaluation of all values of a COALESCE

schemas/util.sql: EXCEPTION blocks with multiple exception types: use OR to merge exception types into one WHEN block

schemas/util.sql: mk_keys_func(regtype, util.col_cast[]): indicate in the type comment that the keys() type is autogenerated, so it can be distinguished from custom keys() types when bulk-regenerating keys() types

bugfix: schemas/util.sql: show_relations_like(): also need to include composite types, as these are also relations (and are expected to be included by callers of show_relations_like())

schemas/util.sql: added debug_print_func_call(text) and use it where applicable

schemas/util.sql: drop_relations_like(): debug-print the regexps so that you can tell which tables it's trying to match

schemas/util.sql: removed no longer used to_freq(regclass, drop_if_always_1). use to_freq(regclass) and auto_rm_freq() instead.

bugfix: schemas/util.sql: diff(regclass, regclass): only drop freq column if all tables have all 1s

schemas/util.sql: auto_rm_freq(): accept multiple tables, so the freq column is only dropped if all tables have all 1s

schemas/util.sql: added freq_always_1(regclass[])

schemas/util.sql: added drop_column(regclass[])

schemas/util.sql: added parent(regclass)

schemas/util.sql: try_create(): also handle not_null_violation, which is thrown when trying to add a NOT NULL column to a parent table, which cascades to a child table whose values for the new column will be NULL

bugfix: schemas/util.sql: diff(text, text): also need to cast left_/right_ to base type for the IS DISTINCT FROM filter, because the WHERE clause apparently does not use columns from the SELECT list, even though GROUP BY and ORDER BY do

schemas/util.sql: added to_freq(regclass, drop_if_always_1)

schemas/util.sql: added auto_rm_freq(regclass)