Activity
From 05/23/2012 to 06/21/2012
06/21/2012
- 10:24 PM Revision 3013: import.stats.xls: Changed \/row (ms)" to "ms/row"
- 09:52 PM Revision 3012: sql.py: mk_insert_select(): INSERT IGNORE: Used typed vars to store each column of the select query because the select query's column names are not known, so it would be unknown which record var column to use for each output column. (Add types so there are no unknown-type values that PostgreSQL will say can't be cast.)
- 09:21 PM Revision 3011: sql.py: insert_select(): In ignore mode, always recover from errors
- 09:14 PM Revision 3010: sql.py: mk_insert_select(): INSERT IGNORE: Implemented simple cursor loop that just inserts each row, without ignoring duplicate keys
- 08:26 PM Revision 3009: sql.py: mk_insert_select(): Added ignore param to do an INSERT IGNORE equivalent operation. Added basic structure for INSERT IGNORE. Support embeddable queries without RETURNING columns (which is the case for most INSERT IGNORE queries).
- 06:50 PM Revision 3008: sql.py: run_query_into(): Always set into.schema to None, so that table is created in the temp schema in debug_temp mode
- 06:25 PM Revision 3007: sql.py: DbConn.DbCursor.execute(): Don't cache ADD COLUMN queries because the rest of the query must be unique in order to use caching
- 06:24 PM Revision 3006: sql.py: add_col(): Don't ignore already-existing columns because sometimes name truncation causes collisions, requiring the caller to version the name. add_index_col(): Version the column name to avoid collisions.
- 06:08 PM Revision 3005: sql.py: add_index(): Version index names to avoid collisions
- 05:58 PM Revision 3004: sql.py: DbConn.DbCursor.execute(): Always cache structural changes, not just if they throw an exception
- 05:40 PM Revision 3003: sql.py: add_index(): Fixed bug where expr could not be deep-copied until ensure_not_null() had been run on it, because ensure_not_null() modifies the index_cols of the table and this needs to be modified on the original table
- 05:39 PM Revision 3002: sql.py: add_index(): Fixed bug where expr could not be deep-copied until ensure_not_null() had been run on it, because ensure_not_null() modifies the index_cols of the table and this needs to be modified on the original table
- 05:12 PM Revision 3001: sql_gen.py: EnsureNotNull: Added to_str() which uses the column's index col if available. This works around a PostgreSQL bug where indexes on expressions are sometimes not used, but indexes on plain columns are used.
- 05:09 PM Revision 3000: sql_gen.py: Store index cols in the table instead of in the column, because columns are often recreated from the table and name but tables are generally not copied
- 04:21 PM Revision 2999: sql_gen.py: Added is_indexed_col()
- 04:15 PM Revision 2998: sql.py: add_index(): Use new sql.ensure_not_null(), which creates a separate column to store the index expr where possible
- 04:12 PM Revision 2997: sql.py: Added add_not_null(), add_index_col(), ensure_not_null()
- 04:10 PM Revision 2996: sql_gen.py: Added is_temp_col()
- 03:57 PM Revision 2995: sql_gen.py: concat(): Also preserve cast suffix
- 03:44 PM Revision 2994: sql_gen.py: Col: Added index_col attr
- 03:19 PM Revision 2993: sql_gen.py: Added suffixed_col()
- 02:59 PM Revision 2992: sql.py: run_query_into(): Set the into table to be a temp table
- 02:59 PM Revision 2991: sql_gen.py: Table: Store whether table is temp table
- 01:50 PM Revision 2990: sql_gen.py: concat(): Also preserve appended column names
- 01:48 PM Revision 2989: sql_gen.py: Col.to_str(): Take for_str param which does both concat() and str()/clean_name() instead of use_concat param which doesn't remove quotes from the strings before concatenating, causing strings to be incorrectly truncated
06/20/2012
- 09:46 PM Revision 2988: sql_gen.py: EnsureNotNull: Cast the null sentinel to the type it's the sentinel for, so that COALESCE() returns the same type as the provided type, instead of potentially incorrectly assuming the constant's type. This fixes a bug where COALESCE() returned an incorrect type.
- 09:38 PM Revision 2987: sql.py: cast(): Use new sql_gen.Cast
- 09:38 PM Revision 2986: sql_gen.py: Added Cast
- 09:37 PM Revision 2985: sql_gen.py: concat(): Consider ")"s part of the existing suffix, so that parentheses match up
- 09:05 PM Revision 2984: Removed extra trailing whitespace
- 08:36 PM Revision 2983: main Makefile: Added explicit schemas/temp/reinstall target because vegbiendev (Ubuntu 12.04) seems to ignore %/reinstall
- 08:24 PM Revision 2982: vegbien_dest: Added temp to schemas list so the intermediate tables will be created there, where they are easy to remove with `make schemas/temp/reinstall`
- 08:23 PM Revision 2981: main Makefile: Added %/reinstall, which calls the corresponding uninstall and install targets
- 08:11 PM Revision 2980: Added temp schema to store intermediate tables permanently when desired for debugging. It is not part of schemas/all because it should not be remade to include current intermediate tables when remaking other schema exports.
- 08:04 PM Revision 2979: schemas: Moved *.sql.make into Makefile to take advantage of % pattern matching
- 07:47 PM Revision 2978: Removed no longer used schemas/vegbien_empty.sql
- 07:46 PM Revision 2977: main Makefile: Removed empty_db, because `make schemas/reinstall` has the same effect and is simpler
- 07:40 PM Revision 2976: README.TXT: Changed documentation to use make schemas/reinstall to empty the DB, since that command is simpler. Added how to archive the last import.
- 07:11 PM Revision 2975: db_xml.py: put_table(): Removed `if not db.debug_temp` check because that is done by sql.empty_temp()
- 07:10 PM Revision 2974: sql.py: put_table(): Use new empty_temp()
- 07:06 PM Revision 2973: import.stats.xls: Added comments for estimated numbers. Added "," separators to large numbers.
- 06:21 PM Revision 2972: sql.py: empty_temp(): In debug_temp mode, leave temp tables there for debugging
- 06:06 PM Revision 2971: sql.py: empty_temp(): Don't output at log_level 2 because it's an internal query, not part of the core algorithm
- 06:06 PM Revision 2970: sql.py: truncate(): Added kw_args to pass to run_query()
- 05:52 PM Revision 2969: Added inputs/import.stats.xls, which compares row-based and column-based import. This shows that column-based import is slowed down by table locking when run simultaneously, so we will need a new INSERT IGNORE replacement that doesn't lock tables.
- 03:14 PM Revision 2968: inputs: Ignore OpenOffice.org lock files
- 02:19 PM Revision 2967: sql.py: empty_temp(): Don't print log message if not emptying any tables
- 02:16 PM Revision 2966: db_xml.py: put_table(): Empty unneeded temp tables to free up memory
- 02:14 PM Revision 2965: sql.py: Added empty_temp()
- 02:14 PM Revision 2964: sql.py: Use new lists.mk_seq()
- 02:13 PM Revision 2963: lists.py: Added mk_seq()
- 02:11 PM Revision 2962: lists.py: is_seq(): Also return true for sets
06/19/2012
- 03:02 PM Revision 2961: schemas/vegbien.sql: specimenreplicate: Added indexes using COALESCE() to match what sql_gen does
- 02:08 PM Revision 2960: sql.py: put_table(): Getting output table pkeys of existing/inserted rows: Do a DISTINCT ON the input pkey (row_num) in case the plain JOIN matched multiple output table rows for one input table row
- 01:44 PM Revision 2959: sql.py: put_table(): Empty unneeded temp tables to free up memory and avoid running out of memory (the temp tables seem to be in-memory only)
- 01:30 PM Revision 2958: sql_gen.py: null_sentinels: Added value for type timestamp with time zone. Put each type on its own line for clarity.
- 01:03 PM Revision 2957: schemas/vegbien.sql: locationdetermination: Changed indexes to use COALESCE() to match what sql_gen now does
- 12:23 PM Revision 2956: schemas/vegbien.sql: location: Added indexes using COALESCE() to match what sql_gen does
- 12:06 PM Revision 2955: sql.py: cast_temp_col(): Add an index on the created column
- 11:55 AM Revision 2954: sql_gen.py: null_sentinels: Added value for type double precision
- 11:52 AM Revision 2953: sql_gen.py: ensure_not_null(): Warn of no null sentinel for type, even if caller catches error
- 10:11 AM Revision 2952: schemas/py_functions.sql: Added plain function _namePart() and use it in trigger function _namePart()
- 09:42 AM Revision 2951: schemas/py_functions.sql: Added plain functions _dateRangeStart() and _dateRangeEnd() and use them in trigger functions _dateRangeStart() and _dateRangeEnd()
- 09:28 AM Revision 2950: schemas/functions.sql: _label(): Ensure that label is NOT NULL so it doesn't NULL out the entire string
- 09:23 AM Revision 2949: schemas/functions.sql: Added plain function _nullIf() and use it in trigger function _nullIf()
- 08:56 AM Revision 2948: sql.py: DbConn.DbCursor._cache_result(): Corrected comment to reflect why different types of queries are cached differently
- 08:46 AM Revision 2947: sql.py: add_col(): Catch DuplicateExceptions so that columns that already existed are ignored
- 08:43 AM Revision 2946: sql.py: run_query(): DuplicateException: Also match "column already exists" errors
- 08:20 AM Revision 2945: sql.py: Merged DuplicateTableException and DuplicateFunctionException into one exception DuplicateException, with a type variable for the type of duplicate item. Added ExceptionWithNameType.
- 08:05 AM Revision 2944: schemas/functions.sql: Fixed bug where external function calls needed to be schema-qualified in case functions schema is not in the search_path
- 07:59 AM Revision 2943: schemas/functions.sql: Added plain function _label() and use it in trigger function _label()
- 07:50 AM Revision 2942: sql.py: put_table(): Support plain SQL functions in addition to relational functions
06/18/2012
- 11:08 PM Revision 2941: sql_gen.py: Added NamedArg. FunctionCall: Support named arguments (http://www.postgresql.org/docs/9.0/static/sql-syntax-calling-funcs.html).
- 10:54 PM Revision 2940: schemas/functions.sql: Added plain function _merge() and use it in trigger function _merge()
- 10:49 PM Revision 2939: schemas/functions.sql: Added plain function _alt() and use it in trigger function _alt()
- 10:37 PM Revision 2938: schemas/functions.sql: Removed no longer used ensure_not_null()
- 10:22 PM Revision 2937: sql.py: put_table(): MissingCastException: Use cast_temp_col() so that cast will occur before any main insert, which locks the output table and should take as little time as possible
- 10:18 PM Revision 2936: sql.py: Added cast_temp_col()
- 10:17 PM Revision 2935: sql.py: add_col(): Support additional run_query() kw_args. add_row_num(): Use new add_col().
- 10:09 PM Revision 2934: sql.py: Added add_col()
- 08:17 PM Revision 2933: sql_gen.py: Col.__str__(): Truncate any table name using concat() to ensure that the full column name is included in the string
- 07:59 PM Revision 2932: strings.py, sql_gen.py: Renamed add_suffix() to concat() to reflect that this is a fixed-length replacement for +
- 07:49 PM Revision 2931: sql.py: put_table(): Moved MissingCastException to the top of the exceptions list because it's more of a core exception than the others, and will be raised before any rows are even inserted
- 06:20 PM Revision 2930: sql.py: DbConn.with_savepoint(): Always release savepoint, because after ROLLBACK TO SAVEPOINT, "The savepoint remains valid and can be rolled back to again" (http://www.postgresql.org/docs/8.3/static/sql-rollback-to.html). Moved `self._savepoint -= 1` to the main try block's new finally block.
- 05:59 PM Revision 2929: sql.py: put_table(): Lock output table right before, and in the same nested transaction as, the insert statement that needs lock, so that it is not released in a prior autocommit and is held for as little time as possible
- 05:38 PM Revision 2928: db_xml.py: put_table(): Removed no longer needed commit param
- 05:16 PM Revision 2927: bin/map: Removed rollback() call before closing the connection because PostgreSQL does this automatically
- 05:14 PM Revision 2926: csv2db: Log inserts with log_level=5 so they are not shown for verbosity 4, which is used to see the savepoints and autocommits
- 05:13 PM Revision 2925: Removed unnecessary db.db.commit() calls because commits are now done automatically by DbConn's autocommit mode
- 04:54 PM Revision 2924: sql.py: DbConn.do_autocommit(): Output the "Autocommitting" debug message with level=4 so that it doesn't clutter up the logging output for normal verbosities
- 04:50 PM Revision 2923: DbConn: autocommit mode defaults to True so that all scripts get the benefit of automatic commits
- 04:49 PM Revision 2922: input.Makefile: Staging tables: import/install-%: Include the table name in the log file name so that successive tables for the same datasource don't overwrite the same log file
- 04:39 PM Revision 2921: sql.py: DbConn: Don't always autocommit in debug_temp mode, because this could cause autocommit mode to be turned on when the user does not expect it
- 04:36 PM Revision 2920: bin/map: connect_db(): Autocommit in commit mode to avoid the need for manual commits. This should also reduce the time that table locks are held, to avoid unnecessary contention when multiple processes are trying to insert into the same output table. (The program always uses nested transactions to support rollbacks, so there is no problem autocommitting whenever a top-level nested transaction or top-level query completes.)
- 04:29 PM Revision 2919: sql_gen.py: Removed TempFunction because that functionality is now provided by DbConn.TempFunction()
- 04:28 PM Revision 2918: sql.py: Use new DbConn.TempFunction()
- 04:28 PM Revision 2917: sql.py: DbConn: Added TempFunction()
- 04:25 PM Revision 2916: sql.py: Use new DbConn.debug_temp config option to control whether temporary objects should instead be permanent
- 04:20 PM Revision 2915: sql.py: DbConn: Added config option debug_temp
- 04:12 PM Revision 2914: sql.py: function_exists(): Fixed bug where trigger functions needed to be excluded, since they cannot be called directly
- 03:49 PM Revision 2913: sql.py: Added function_exists()
- 03:49 PM Revision 2912: sql_gen.py: Made Function an alias of Table so that isinstance(..., Function) will always work correctly
- 03:45 PM Revision 2911: sql_gen.py: Added as_Function()
06/15/2012
- 06:16 AM Revision 2910: sql.py: put_table(): Lock the output table in EXCLUSIVE mode *before* getting its pkey so that an ACCESS SHARE lock is not acquired before EXCLUSIVE (causing a lock upgrade and deadlock). This race condition may not have been previously noticeable because pkey() is cached, so calling it doesn't necessarily execute a query or acquire an ACCESS SHARE lock.
- 05:52 AM Revision 2909: sql.py: put_table(): Document that must be run at the *beginning* of a transaction
- 05:49 AM Revision 2908: sql.py: put_table(), mk_select(): Switched back to having put_table() acquire the EXCLUSIVE locks, but right at the beginning of the transaction, in order to avoid lock upgrades which cause deadlocks
- 05:35 AM Revision 2907: sql.py: with_autocommit(): Only allow turning autocommit on, because the opposite is not meaningful and may conflict with the session-global isolation level
- 05:33 AM Revision 2906: sql.py: DbConn: Set the transaction isolation level to READ COMMITTED using set_isolation_level() so that the isolation level affects all transactions in the session, not just the current one
- 05:21 AM Revision 2905: sql.py: DbConn: Always set the transaction isolation level to READ COMMITTED so that when a table is locked for update, its contents are frozen at that point rather than earlier. This ensures that no concurrent duplicate keys were inserted between the time the table was snapshotted (at the beginning of the transaction for SERIALIZABLE) and the time it was locked for update.
- 05:02 AM Revision 2904: sql.py: put_table(): Removed locking output tables to prevent concurrent duplicate keys because that is now done automatically by mk_select()
- 05:01 AM Revision 2903: sql.py: mk_select(): Filtering on no match: Lock the joined table in EXCLUSIVE mode to prevent concurrent duplicate keys when used with INSERT SELECT
- 04:59 AM Revision 2902: sql_gen.py: Added underlying_table() and use it in underlying_col()
- 04:39 AM Revision 2901: main Makefile: schemas/rotate: Fixed bug where needed to run schemas/public/install, not full schemas/install, after renaming public schema
- 04:32 AM Revision 2900: sql.py: put_table(): Lock output tables to prevent concurrent duplicate keys
- 04:31 AM Revision 2899: sql.py: Added lock_table()
- 03:53 AM Revision 2898: bin/map: connect_db(): Only use autocommit mode if verbosity > 3, to avoid accidentally activating it if you want debug output in normal import mode
- 03:45 AM Revision 2897: bin/map: connect_db(): Only use autocommit mode if verbosity > 2, because it causes the intermediate tables to be created as permanent tables, which you don't want unless you're actually debugging (verbosity = 2 is normal for column-based import)
- 03:25 AM Revision 2896: sql.py: put_table(): remove_all_rows(): Changed log message to "Ignoring all rows" because NULL is not necessarily the pkey value that will be returned for the rows
- 03:17 AM Revision 2895: sql.py: put_table(): Don't add index on columns that will have values filtered out, because indexes have already been added on all columns in the iteration's input table by flatten()
- 03:12 AM Revision 2894: sql.py: DbConn._db(): Setting serializable isolation level: Always set this (if self.serializable is set), even in autocommit mode, because autocommit mode is implemented by manual commits in the DbConn wrapper object rather than using the underlying connection's autocommit mode (which does not allow setting the isolation level)
- 03:08 AM Revision 2893: sql.py: DbConn._db(): Setting search_path: Use `SET search_path` and `SHOW search_path` instead of combining the old and new search_paths in SQL itself using `SELECT set_config('search_path', ...)`
- 02:47 AM Revision 2892: csv2db: ProgressInputStream: Use default progress message 'Read %d line(s)' because there is not necessarily one CSV row per line, due to embedded newlines
- 01:47 AM Revision 2891: input.Makefile: Staging tables: import/install-%: Only output to the log file if log option is non-empty (which it is by default)
- 01:46 AM Revision 2890: csv2db: Support reinstalling just the errors table using new errors_table_only option
- 01:45 AM Revision 2889: sql.py: Added drop_table()
- 01:20 AM Revision 2888: schemas/vegbien.sql: method: Changed indexes to use `COALESCE(..., E'\\N')` to match what sql_gen now does
- 01:16 AM Revision 2887: schemas/vegbien.sql: specimenreplicate: Added indexes using COALESCE() to match what sql_gen does
- 01:12 AM Revision 2886: schemas/vegbien.sql: locationevent: Added indexes using COALESCE() to match what sql_gen does
- 12:57 AM Revision 2885: schemas/vegbien.ERD.mwb: Synced with schema
- 12:54 AM Revision 2884: schemas/vegbien.sql: party: Changed indexes to use `COALESCE(..., E'\\N')` to match what sql_gen now does
- 12:38 AM Revision 2883: Wrap sys.stderr.write() calls in strings.to_raw_str() to avoid UnicodeEncodeErrors when stderr is to a file and the default encoding is ASCII
- 12:37 AM Revision 2882: strings.py: Added to_raw_str()
- 12:12 AM Revision 2881: bin/map: When logging the row # being processed, add 1 because row # is interally 0-based, but 1-based to the user
- 12:05 AM Revision 2880: bin/map: Log the row # being processed with level=1.1 so that the user can see a status report if desired
06/14/2012
- 11:35 PM Revision 2879: exc.py: str_(): Fixed bug where UnicodeEncodeError would be raised when msg contains non-ASCII chars, by wrapping e.args[0] in strings.ustr()
- 11:23 PM Revision 2878: exc.py: print_ex(): Wrap msg in strings.to_unicode() to try to avoid UnicodeEncodeError when msg contains non-ASCII chars
- 11:20 PM Revision 2877: sql.py: create_table(): Don't set pkey.nullable to False because the caller should make sure the pkey has the appropriate type
- 11:20 PM Revision 2876: csv2db: Use sql_gen.TypedCol.nullable instead of manually adding 'NOT NULL' to the type. Ensure that pkeys are properly NOT NULL.
- 11:15 PM Revision 2875: csv2db: Adding indexes: Create plain indexes using ensure_not_null=False because the indexes will primarily be used by the user to search for specific values, rather than by the mapping script which uses the ensure_not_null
- 11:13 PM Revision 2874: sql.py: DbConn.col_info(): Run query with log_level=4 because it gathers information about database structure, and should have the same log_level as other queries that do that
- 11:09 PM Revision 2873: csv2db: Adding indexes: Fixed bug where col.to_Col() could not be used because sql.add_index() does not support name-only columns (plain strings are OK, though)
- 11:00 PM Revision 2872: sql.py: create_table(): has_pkey: Use new TypedCol.constraints to store 'PRIMARY KEY'
- 10:59 PM Revision 2871: sql_gen.py: TypedCol: Added constraints instance var
- 10:38 PM Revision 2870: sql_gen.py: EnsureNotNull: Made coalesce() all uppercase to match how pg_dump spells it
- 10:36 PM Revision 2869: schemas/vegbien.sql: namedplace: Fixed bug where parent_id needed to be included in UNIQUE CONSTRAINT (now UNIQUE INDEX), since there can be more than one e.g. city of the same name if they are in different states
- 10:32 PM Revision 2868: schemas/vegbien.sql: plantname: Changed indexes to use `COALESCE(..., 2147483647)` to match what sql_gen now does
- 10:22 PM Revision 2867: schemas/py_functions.sql: _dateRangeStart, _dateRangeEnd: Changed indexes to use `COALESCE(..., E'\N')` to match what sql_gen now does
- 10:19 PM Revision 2866: schemas/py_functions.sql: _namePart: Changed indexes to use `COALESCE(..., E'\N')` to match what sql_gen now does
- 10:16 PM Revision 2865: schemas/functions.sql: _nullIf: Changed indexes to use `COALESCE(..., E'\N')` to match what sql_gen now does
- 10:15 PM Revision 2864: schemas/functions.sql: _nullIf: Require a non-NULL null-equivalent value
- 10:12 PM Revision 2863: schemas/functions.sql: _label: Changed indexes to use `COALESCE(..., E'\N')` to match what sql_gen now does
- 10:09 PM Revision 2862: schemas/functions.sql: _label: Require a non-NULL label
- 10:04 PM Revision 2861: sql_gen.py: null_sentinels: Removed types where a sentinel doesn't make sense (unknown types, boolean) because types with no sentinel are now handled gracefully by users of ensure_not_null()
- 10:00 PM Revision 2860: sql.py: add_index(): ensure_not_null: Handle unknown types gracefully
- 09:52 PM Revision 2859: sql_gen.py: MockDb: Added col_info()
- 09:48 PM Revision 2858: sql_gen.py: Use as_*() functions where the auto-wrapping was previously done manually
- 09:44 PM Revision 2857: sql_gen.py: CompareCond.to_str(): Use ensure_not_null()'s new type_ param to apply same function to both sides but not if the right side is already NOT NULL
- 09:26 PM Revision 2856: sql_gen.py: null_sentinels: Added value for character varying type
- 08:45 PM Revision 2855: sql_gen.py: ensure_not_null(): Support non-column inputs if type_ is set
- 08:32 PM Revision 2854: sql_gen.py: null_sentinels: Added value for USER-DEFINED type
- 08:31 PM Revision 2853: sql.py: mk_select(): Joins: Filtering on no match: Use '~=' sql_gen.CompareCond operator so that IS NULL is always used, regardless of the not-null column's nullability
- 08:24 PM Revision 2852: sql_gen.py: null_sentinels: Added value for boolean type
- 08:21 PM Revision 2851: sql_gen.py: ensure_not_null(): Added type_ param to override the underlying column's type
- 08:09 PM Revision 2850: sql_gen.py: EnsureNotNull: Take a type param instead of a null param so that the EnsureNotNull object stores the underlying column's type
- 08:00 PM Revision 2849: sql_gen.py: underlying_col(): Support non-Col inputs
- 07:42 PM Revision 2848: sql_gen.py: EnsureNotNull: Removed default value for null param to remind user that default value depends on value's type and will not always be a string
- 07:39 PM Revision 2847: sql.py: add_index(): Added ensure_not_null param to disable the ensure_not_null functionality to force a plain index
- 07:14 PM Revision 2846: sql.py: flatten(): Add indexes on the created table so its columns can be used in an O(n) merge join
- 07:12 PM Revision 2845: sql_gen.py: null_sentinels: Added value for integer type
- 06:52 PM Revision 2844: sql_gen.py: CompareCond.to_str(): Always wrap the left-side column if it's nullable. Wrap the right-side value if the left side was wrapped, rather than if both the left and right side are nullable. This causes coalesce() indexes to be used to look up NULL values using the value NULL gets coalesced to, rather than doing a sequential scan.
- 06:43 PM Revision 2843: sql_gen.py: Run truncate() on all identifiers so that literal-string-based lookups for an identifier (such as in db.col_info()) don't use the untruncated value
- 06:40 PM Revision 2842: sql_gen.py: Added truncate()
- 06:40 PM Revision 2841: sql.py: put_table(): Resolving default value column: Fixed bug where default value of None was used as a key for mapping, even though this is an invalid Col name
- 05:30 PM Revision 2840: sql_gen.py: ensure_not_null(): If input column cannot be ensured to be NULL, pass any raised exception through rather than suppressing it and leaving the column in a nullable state
06/13/2012
- 11:11 PM Revision 2839: schemas/functions.sql: _merge: Changed indexes to use `COALESCE(..., E'\\N')` to match what sql_gen now does
- 11:08 PM Revision 2838: schemas/functions.sql: _alt: Changed indexes to use `COALESCE(..., E'\\N')` to match what sql_gen now does
- 10:59 PM Revision 2837: sql_gen.py: CompareCond.to_str(): Handle nullable columns using ensure_not_null()
- 10:46 PM Revision 2836: sql_gen.py: ensure_not_null(): Raise NoUnderlyingTableException if can't ensure not null for that reason
- 10:20 PM Revision 2835: sql_gen.py: is_underlying_table(): Support non-Table inputs
- 10:15 PM Revision 2834: sql_gen.py: NamedValues: Call set_cols_table() with the created table, not just the name, so that is_underlying_table() works properly
- 09:49 PM Revision 2833: sql_gen.py: underlying_col(): If no underlying table, raise NoUnderlyingTableException
- 09:45 PM Revision 2832: sql_gen.py: Added is_underlying_table()
- 09:34 PM Revision 2831: sql_gen.py: ensure_not_null(): Call underlying_col() on the column to remove all renamings
- 09:33 PM Revision 2830: sql_gen.py: Added underlying_col()
- 09:24 PM Revision 2829: sql_gen.py: Join.to_str(): join(): Removed no longer needed `*_table = *_table.to_Table()`
- 09:22 PM Revision 2828: sql_gen.py: Col: Support Table objects that are not just names, by calling `.to_Table()` on the table before stringifying it
- 09:22 PM Revision 2827: sql_gen.py: Col: Support Table objects that are not just names, by calling `.to_Table()` on the table before stringifying it
- 08:59 PM Revision 2826: sql_gen.py: ensure_not_null(): Added ignore_unknown_type param
- 08:55 PM Revision 2825: sql_gen.py: CompareCond.to_str(): Put handling nullable columns as a separate step so it can be expanded
- 08:36 PM Revision 2824: csv2db: Errors table: Removed no longer needed sql_gen.EnsureNotNull() because this is now added automatically
- 08:35 PM Revision 2823: sql.py: add_index(): Handle nullable columns by using sql_gen.ensure_not_null() to automatically add a coalesce() wrapper where needed
- 08:33 PM Revision 2822: sql_gen.py: Added ensure_not_null()
- 08:29 PM Revision 2821: sql.py: DbConn.col_info(): Fixed bug where is_nullable needed to be cast to a boolean
- 08:06 PM Revision 2820: sql.py: cast(): Support string column name inputs
- 07:12 PM Revision 2819: sql.py: DbConn: Renamed col_default() to col_info() and have it return a sql_gen.TypedCol object containing all the TypedCol info about the column, not just the default value
- 06:55 PM Revision 2818: sql_gen.py: TypedCol: Added default and nullable params
- 06:53 PM Revision 2817: dicts.py: Import util after items that util depends on have been defined, to avoid unsatisfied circular dependency
- 06:30 PM Revision 2816: sql.py: DbConn.col_default(): Pass the connection to sql_gen.as_Code() so it fixes the syntax on values returned by PostgreSQL
- 06:29 PM Revision 2815: sql_gen.py: as_Code(): Added optional db param, which causes the function to run db.std_code() on the value to fix the syntax
- 06:27 PM Revision 2814: sql.py: DbConn: Added std_code()
- 03:02 PM Revision 2813: db_xml.py: Removed into_table_name() because this functionality is now handled by sql.into_table_name()
- 02:59 PM Revision 2812: sql.py: into_table_name(): Also parse hierarchical tables (mappings with a rank column) using a special syntax
- 02:44 PM Revision 2811: sql.py: put_table(): Fixed bug where distinct_on included columns that were not in the input table, and were thus incorrectly taken from the LEFT JOINed output table
06/12/2012
- 10:32 PM Revision 2810: sql.py: track_data_error(): Do nothing if cols are empty, because mk_track_data_error() requires at least one col. mk_track_data_error(): Assert that cols are not empty because VALUES clause requires at least one row.
- 10:26 PM Revision 2809: bin/map: by_col: Pass on_error to db_xml.put_table() that calls ex_tracker.track()
- 10:14 PM Revision 2808: sql.py: put_table(): No handler for exception: Pass exception to on_error() instead of raising a warning, so that error message can be formatted
- 10:08 PM Revision 2807: db_xml.py: put_table(): Pass on_error to sql.put_table()
- 10:07 PM Revision 2806: db_xml.py: put_table(): Take on_error param like row-based put()
- 10:06 PM Revision 2805: sql.py: put_table(): Take on_error param
- 09:48 PM Revision 2804: sql.py: get_cur_query(): Removed no longer used input_params parameter
- 09:46 PM Revision 2803: sql.py: Removed unused mogrify()
- 09:42 PM Revision 2802: sql.py: DbConn.DbCursor.execute(): Removed no longer used params parameter
- 09:32 PM Revision 2801: sql.py: with_autocommit(): Use isolation_level attr and set_isolation_level() method of connection instead of autocommit attr to support older versions of psycopg2
- 09:18 PM Revision 2800: sql.py: DbConn.DbCursor.execute(): Only fetch all rows for empty SELECT query, to support older versions of Python that would give a "no results to fetch" error for other types of queries
- 09:13 PM Revision 2799: csv2db: When reraising exception, use `raise` instead of `raise e` to preserve whole stack trace
- 09:12 PM Revision 2798: sql.py: Removed no longer used _query_lookup()
- 09:05 PM Revision 2797: sql.py: DbConn: Cache queries without params, as params are no longer used
- 09:03 PM Revision 2796: sql.py: DbConn.is_cached(): Removed no longer used params parameter
- 09:01 PM Revision 2795: sql.py: Removed no longer used run_raw_query()
- 09:00 PM Revision 2794: sql.py: run_query(): Call db.run_query() directly instead of via run_raw_query()
- 08:56 PM Revision 2793: sql.py: DbConn.run_query(): Removed no longer used params parameter
- 08:50 PM Revision 2792: sql.py: DbConn._db(): Setting search_path: Use esc_value() instead of params
- 08:43 PM Revision 2791: sql.py: run_query(): Removed no longer used params parameter
- 08:37 PM Revision 2790: sql.py: run_query_into(): Moved main case (into != None) outside of if statement because the special-case if statement contains `return`
- 08:35 PM Revision 2789: sql.py: run_query_into(): Removed no longer used params parameter
- 08:32 PM Revision 2788: sql.py: mk_insert_select(): Removed no longer used params parameter
- 08:25 PM Revision 2787: sql.py: mk_insert_select(): Return just the query instead of the query plus empty params
- 08:22 PM Revision 2786: sql.py: mk_select(): Return just the query instead of the query plus empty params
- 08:12 PM Revision 2785: sql.py: tables(): Use select() instead of a custom run_query() to avoid using params, which will be deprecated to make it easier to support old versions of Python
- 08:11 PM Revision 2784: sql.py: DbConn.DbCursor.execute(): Require that params are empty, to ensure that code uses db.esc_value() instead. This keeps literal values in the same place as the rest of the query, so that they do not need to be maintained and passed around separately in a params list.
- 08:06 PM Revision 2783: sql.py: constraint_cols(): Use db.esc_value() instead of params
- 08:05 PM Revision 2782: sql.py: index_cols(): Use db.esc_value() instead of params
- 08:00 PM Revision 2781: sql.py: add_pkey(): Use simpler `ADD PRIMARY KEY` syntax to avoid having to create a name for the primary key
- 07:57 PM Revision 2780: db_xml.py: put_table(): Subsetting in_table: Add pkey to created temp table to facilitate joining it with intermediate tables
- 07:47 PM Revision 2779: sql.py: tables(): Use select() instead of a custom run_query() to avoid using params, which will be deprecated to make it easier to support old versions of Python
- 07:14 PM Revision 2778: schemas/postgresql.nimoy.conf: shared_buffers: Fixed syntax error where decimals were not supported
- 07:03 PM Revision 2777: sql.py: truncate(): Re-added support for string tables using sql_gen.as_Table(). This fixes empty_db(), which relied on this functionality.
- 07:01 PM Revision 2776: sql_gen.py: as_Table(): Added schema param to use as default schema
- 06:54 PM Revision 2775: inputs/SALVIAS: Switched to using CSV exports of the DB, so that staging tables could be created for column-based import
- 06:48 PM Revision 2774: sql.py: run_query_into(): Added add_indexes_ param which causes the function to add indexes on the created table
- 06:18 PM Revision 2773: sql.py: create_table(): Use new add_indexes()
- 06:15 PM Revision 2772: sql.py: Added add_indexes()
- 05:38 PM Revision 2771: sql.py: get_cur_query(): Fixed bug where strings.ustr() needed to be used instead of str() when ensuring that get_cur_query() returns a string
- 05:35 PM Revision 2770: sql.py: cast(): Removed conditional checks for save_errors, since it's now always true if the function got passed the `not save_errors` special case
- 05:32 PM Revision 2769: sql.py: cast(): Only convert errors to warnings if errors will be saved in errors_table, so that import will always be aborted if user supplied invalid values in the mappings, even if these values are passed through a relational function
- 04:58 PM Revision 2768: sql.py: put_table(): Support inserting tables with all default values, by providing the pkey's default value for all rows so that the SELECT query has at least one column
- 04:57 PM Revision 2767: sql_gen.py: is_table_col(): Check that input is a Col object
- 04:31 PM Revision 2766: sql.py: put_table(): Assert that mapping is non-empty
- 04:23 PM Revision 2765: sql.py: mk_select(): Assert that fields list is non-empty
- 04:18 PM Revision 2764: sql.py: DbConn.DbCursor.execute(): Set _is_insert only if query *starts with* INSERT, so that function definitions containing INSERT are not cached as INSERT statements (exceptions only) themselves
- 04:10 PM Revision 2763: sql.py: DbConn.DbCursor.execute(): Fixed bug where params == None would apparently turn off the mogrifier completely, causing "%"s to be excessively escaped, by just setting params to None if it was [] or () and not using strings.esc_for_mogrify() at all
- 03:51 PM Revision 2762: sql.py: DbConn.DbCursor.execute(): If not using params, escape the query using strings.esc_for_mogrify() in case any literals contained "%"s
- 03:49 PM Revision 2761: strings.py: Added esc_for_mogrify()
- 03:05 PM Revision 2760: sql.py: create_table(): Add indexes on all non-pkey columns, unless turned off or deferred using new param col_indexes
- 02:46 PM Revision 2759: csv2db: Add column indexes on errors table. Use typed_cols and `.to_Col()` to iterate over columns to add indexes on, for the main and errors tables.
06/11/2012
- 11:13 PM Revision 2758: sql.py: Added track_data_error(). put_table(): ignore(): Take extra e param for the exception. Use track_data_error() to store the invalid value in the errors table.
- 11:11 PM Revision 2757: sql_gen.py: Join.to_str(): Add newline before and after right table if it's been renamed (and therefore takes up multiple lines)
- 11:10 PM Revision 2756: exc.py: ExceptionWithCause: Store the cause in an instance variable for later use
- 10:47 PM Revision 2755: sql.py: mk_track_data_error(): Rename the errors_table to make the generated SQL less verbose
- 10:42 PM Revision 2754: sql.py: mk_insert_select(): Run sql_gen.remove_table_rename() on table to get just the actual name in the DB
- 10:42 PM Revision 2753: sql_gen.py: Added remove_table_rename()
- 10:40 PM Revision 2752: sql_gen.py: Col: Run `.to_Table()` on table to get just the reference to the table, not any SQL code that defines it
- 10:10 PM Revision 2751: sql.py: Added mk_track_data_error() and use it in cast(). This also ensures that if only one source column's row in the CROSS JOIN violates a unique constraint, other source columns' rows are still inserted.
- 09:59 PM Revision 2750: sql_gen.py: with_default_table(): Added overwrite param to overwrite the table (if it isn't a NamedCol)
- 09:57 PM Revision 2749: sql_gen.py: Join.to_str(): join(): Get just the table name of left_table and right_table using `.to_Table()`. Moved order switching of tables inside join() because the order reversing only applies to an individual condition.
- 09:34 PM Revision 2748: sql_gen.py: Renamed set_default_table() to with_default_table() and copy col before modifying it so don't modify input
- 09:14 PM Revision 2747: sql_gen.py: Added set_default_table(). as_ValueCond(): Use set_default_table() instead of as_Col() so that any name-only column also gets its table set. Join.to_str(): Parse left side using set_default_table() instead of as_Col() so that any name-only column also gets its table set.
- 07:31 PM Revision 2746: sql_gen.py: Join: mapping param defaults to {} for e.g. CROSS JOINs. to_str(): Omit join_cond if mapping is empty, rather than if join is a specific type.
- 07:24 PM Revision 2745: sql_gen.py: NamedValues: Change cols to Col objects with the table set to `name`
- 07:23 PM Revision 2744: sql_gen.py: Added set_cols_table()
- 07:00 PM Revision 2743: sql.py: mk_insert_select(): returning: Use sql_gen.to_name_only_col()
- 06:52 PM Revision 2742: sql_gen.py: NamedTable: cols: Use sql_gen.Col objects or name strings instead of pre-rendered SQL code
- 06:35 PM Revision 2741: sql_gen.py: NamedTable: Wrap nested code in Expr if needed
- 06:27 PM Revision 2740: sql_gen.py: Added NamedValues
- 06:04 PM Revision 2739: sql_gen.py: Values: Support multiple rows
- 05:54 PM Revision 2738: sql.py: insert(): Use new sql_gen.Values
- 05:53 PM Revision 2737: sql_gen.py: Added Values and default
- 05:26 PM Revision 2736: sql_gen.py: Join.to_str(): Don't add join condition for CROSS JOINs
- 05:03 PM Revision 2735: sql.py: put_table(): Factored out errors_table name setting so it can be used by ignore()
- 04:38 PM Revision 2734: bin/map: If doing full import, clear errors table
- 04:37 PM Revision 2733: sql.py: truncate(): Support sql_gen.Table objects
- 04:21 PM Revision 2732: sql.py: Moved truncate() to Database structure queries section
- 04:11 PM Revision 2731: sql.py: tables(): Run query with log_level=4 because it's a low-level structure-determining query
- 04:07 PM Revision 2730: sql.py: table_exists(): Use new tables() exact param so that LIKE special chars in table name are not interpreted specially
- 04:07 PM Revision 2729: sql.py: tables(): Added exact param to check for exact matches only
- 04:03 PM Revision 2728: sql.py: put_table(): MissingCastException: Use new errors_table()
- 04:02 PM Revision 2727: csv2db: Use new sql.errors_table()
- 04:02 PM Revision 2726: sql.py: Added table_exists() and errors_table()
06/08/2012
- 09:59 PM Revision 2725: sql.py: DbConn.print_notices(): Fixed bug where it should not do anything for a MySQL connection, because that doesn't store notices the way Postgres does
- 09:54 PM Revision 2724: sql.py: put_table(): MissingCastException: Debug message: Added Redmine formatting
- 09:48 PM Revision 2723: schemas/functions.sql, vegbien.sql: Removed no longer needed cast functions, which are now created on the fly by column-based import
- 09:45 PM Revision 2722: schemas/functions.sql: _nullIf(): Ignore uncastable value, because a value that's invalid for the given type is still well-defined as not matching the nullif() criterion
- 09:25 PM Revision 2721: sql.py: put_table(): MissingCastException: Debug message: Removed "'s" so it wouldn't mess up syntax highlighting when pasting debug output into a SQL file
- 09:08 PM Revision 2720: sql.py: cast(): Made errors table also store SQLSTATE in error_code column
- 08:40 PM Revision 2719: sql.py: cast(): Documented that the value and error are inserted for *each* source column (hence the CROSS JOIN)
- 08:39 PM Revision 2718: sql.py: cast(): Version the function name if using an errors table, to avoid collisions with other cast functions when the function name is truncated (or, more rarely, collisions with casts to the same type and on the same input columns but of a different table)
- 08:25 PM Revision 2717: sql.py: cast(): function_name: Fixed bug where sql_gen.FunctionCall()'s parameters needed to be passed with *args syntax
- 08:23 PM Revision 2716: sql.py: mk_flatten_mapping(): Propagate the original columns' sources to the flattened columns so they won't be lost in the flattening
- 08:20 PM Revision 2715: sql.py: put_table(): MissingCastException: Use in_tables0's source table to locate the errors table in case in_tables0 has been subset into a temp table (which removes the schema name)
- 08:19 PM Revision 2714: db_xml.py: put_table(): Track in_table's source so its original schema can be obtained and auxiliary tables located
- 08:17 PM Revision 2713: sql_gen.py: Derived.set_srcs(): Added optional overwrite param so that a default srcs value can be set only if one isn't already set
- 08:04 PM Revision 2712: sql_gen.py: Made Table a Derived element so that a row-subset temp table could retain the schema of the table it came from, and any auxiliary tables in that schema could be located properly
- 08:01 PM Revision 2711: sql_gen.py: Moved srcs-related functionality from Col to new superclass Derived
- 07:53 PM Revision 2710: sql.py: cast(): save_errors: Fixed bug where srcs needed to have their names extracted before being wrapped in sql_gen.Literals. Fixed bug where errors table INSERT needed to prefix the CROSS JOIN-ed VALUES statements with SELECT * FROM because the CROSS JOIN makes it a whole SELECT query, not just a VALUES statement.
- 07:40 PM Revision 2709: schemas/vegbien.ERD.mwb: Adjusted lines to make less intersections happen on the first page
- 07:27 PM Revision 2708: sql.py: put_table(): MissingCastException: Fixed bug where errors_table needed to have the same schema as in_tables0 as well as part of the same name
- 07:23 PM Revision 2707: sql_gen.py: Added suffixed_table()
- 07:14 PM Revision 2706: sql.py: MissingCastException: Print log message that it's casting the column, to introduce the SQL function definition that follows
- 07:06 PM Revision 2705: sql.py: put_table(): MissingCastException: Use new cast() instead of relying on existing cast functions in the database
- 07:04 PM Revision 2704: sql.py: Added cast()
- 06:00 PM Revision 2703: sql_gen.py: Added wrap() and use it in wrap_in_func()
- 05:36 PM Revision 2702: sql.py: put_table(): Track the input column(s) a column is derived from, so that error messages can be attributes to the proper input column(s)
- 05:35 PM Revision 2701: sql_gen.py: Col: Support tracking the column(s) a column is derived from, so that error messages can be attributes to the proper input column(s)
- 05:33 PM Revision 2700: objects.py: BasicObject: Allow subclasses to customize which attrs are compared on, by adding _compare_on() method
- 04:45 PM Revision 2699: lists.py: uniqify(): Document that it will work on any iterable, not just lists
- 04:06 PM Revision 2698: sql.py: mk_insert_select(): embeddable: Use sql_gen.TempFunction and sql_gen.FunctionCall
- 04:04 PM Revision 2697: sql_gen.py: Added TempFunction
- 03:26 PM Revision 2696: schemas/functions.sql, vegbien.sql: Cast functions: Removed `RETURN new;` at end (artifact of when they were relational functions). Made the EXCEPTION block the main block of the function, to avoid unnecessary nesting.
- 02:26 PM Revision 2695: csv2db: Errors table: index_cols: Remove no longer needed sql_gen.Col() (now done by EnsureNotNull)
- 02:25 PM Revision 2694: sql_gen.py: EnsureNotNull: Run value through as_Col() so FunctionCall won't default it to a Literal
- 02:19 PM Revision 2693: csv2db: Use sql_gen.EnsureNotNull instead of the ensure_not_null() function in the functions schema to avoid a dependency on the functions schema, which would cause the UNIQUE index to be dropped whenever the functions schema is reinstalled
- 02:18 PM Revision 2692: sql_gen.py: Added EnsureNotNull
- 02:17 PM Revision 2691: sql_gen.py: Added InternalFunction
- 02:08 PM Revision 2690: sql_gen.py: FunctionCall: Ensure all args are Code objects using as_Value()
- 01:51 PM Revision 2689: csv2db: Errors table: Add UNIQUE index on all columns
- 01:50 PM Revision 2688: sql.py: add_index(): Support multiple column(s) or expression(s). Support separate table (not extracted from cols). Support UNIQUE indexes.
06/07/2012
- 09:41 PM Revision 2687: sql.py: add_index(): Fixed bug where expr needed to be deep copied so that any column nested in it (e.g. inside a FunctionCall) wouldn't be modified when col.table is set to None
- 09:35 PM Revision 2686: sql.py: add_pkey(): Support multiple, custom columns
- 09:24 PM Revision 2685: csv2db: Vacuum the created table
- 09:23 PM Revision 2684: sql.py: Added vacuum()
- 09:23 PM Revision 2683: sql.py: DbConn: Added with_autocommit()
- 08:58 PM Revision 2682: csv2db: Create errors table for use by column-based import
- 08:57 PM Revision 2681: sql.py: create_table(): Added has_pkey param to disable making the first column the primary key
- 08:21 PM Revision 2680: csv2db: Use verbosity-based logging like bin/map. Use sql.create_table(). Add indexes on the columns to speed up column-based import and to speed up searching the table for particular values.
- 08:00 PM Revision 2679: sql.py: create_table(): Don't add indexes on columns, because that shouldn't happen until after the table's rows have been inserted
- 07:55 PM Revision 2678: sql.py: DbConn._db(): Output 'SET TRANSACTION ISOLATION LEVEL SERIALIZABLE' with log_level=4 because that should not be shown when the search_path is shown, which has log_level=3
- 07:43 PM Revision 2677: sql.py: cleanup_table(): Use update(), which also fixes some formatting bugs
- 07:24 PM Revision 2676: sql.py: DbConn._db(): Output connection configuration statements with log_level=3
- 07:15 PM Revision 2675: sql.py: Added create_table()
- 07:14 PM Revision 2674: sql_gen.py: Added TypedCol
- 06:59 PM Revision 2673: sql.py: insert_select(): Pass log_level to run_query_into()
- 05:15 PM Revision 2672: streams.py: LineCountInputStream: Fixed bug where EOF was incorrectly considered a line, causing the final line count (used by ProgressInputStream) to be off by one
- 04:25 PM Revision 2671: sql.py: DbConn: Added print_notices() and call it after running a query
- 04:12 PM Revision 2670: dicts.py: Added IdCompared
- 03:50 PM Revision 2669: lists.py: Added clear()
- 03:24 PM Revision 2668: sql.py: DbConn.do_autocommit(): Fixed typo in 'Autocommiting' debug message
- 03:20 PM Revision 2667: sql_gen.py: ColDict: Extend dicts.DictProxy instead of UserDict.DictMixin because that already provides some of the functionality needed by ColDict
06/06/2012
- 09:46 PM Revision 2666: sql.py: run_query(): Only pass non-None debug_msg_ref to DbConn.run_query() if filtering with log_ignore_excs or can't mogrify() (and therefore can't print queries before they're run)
- 09:36 PM Revision 2665: sql.py: DbConn: Added can_mogrify() and use it in mogrify()
- 09:35 PM Revision 2664: sql.py: DbConn.run_query(): Log query before running if no debug_msg_ref specified. Documented debug_msg_ref param.
- 08:43 PM Revision 2663: sql.py: DbConn: Added mogrify() and use it in esc_value()
- 07:58 PM Revision 2662: schemas/functions.sql: _nullIf(): Fixed bug where wrong var name was used to retrieve type param. Reenabled _nullIf now that it's working.
- 07:25 PM Revision 2661: sql_gen.py: ColDict.__setitem__(): Translate NULL values to the appropriate default value of the key column (which is often NULL, but not always) so that select query filtering/joins work correctly
- 07:21 PM Revision 2660: sql.py: DbConn.col_default(): Fixed bug where returned string needed to be wrapped in sql_gen.as_Code() because it's sometimes a SQL expression and sometimes NULL
- 07:20 PM Revision 2659: sql_gen.py: Added as_Code(). Split SQL code objects into separate sections so unparameterized classes would be separate from general classes.
- 07:08 PM Revision 2658: sql_gen.py: Code.to_str(): Fixed bug where needed to raise NotImplementedError, not NotImplemented
- 06:58 PM Revision 2657: dicts.py: is_dict(): Fixed bug where lists also have __getitem__() methods, so keys() was checked for instead
- 06:44 PM Revision 2656: sql.py: put_table(): Fixed bug where mapping was still a plain dict because that's what dicts.join() returns, by moving the wrapping of it in a sql_gen.ColDict after dicts.join()
- 06:28 PM Revision 2655: sql_gen.py: ColDict: Fixed bug where copy() needed to be implemented
- 06:14 PM Revision 2654: sql.py: into_table_name(): Fixed bug where key needed to be passed through str() now that ColDict uses Col objects for everything
- 06:11 PM Revision 2653: sql_gen.py: ColDict: Extend UserDict.DictMixin instead of dict because its non-core function implementations route all inner dict accesses to the core functions __getitem__() and __setitem__(). sql.py: put(): DuplicateKeyException: Wrap util.dict_subset_right_join() in a sql_gen.ColDict because the dict returned by util.dict_subset_right_join() is just a plain dict. (This change must happen at the same time because the previous functionality relied on a bug in ColDict.)
- 06:04 PM Revision 2652: util.py: DefaultDict: Use dicts.DictProxy instead of collections.defaultdict so that it provides a view of the given dict instead of copying it
- 06:03 PM Revision 2651: dicts.py: Added DictProxy
- 05:37 PM Revision 2650: sql.py: mk_select(): conds: Support dict-like objects other than dict itself (such as sql_gen.ColDict)
- 05:16 PM Revision 2649: dicts.py: Added is_dict()
- 05:06 PM Revision 2648: sql.py: DbConn.DbCursor.execute(): When setting self.query, provide default values for query, params because some debug printing statements use self.query directly instead of _add_cursor_info()
- 05:01 PM Revision 2647: sql.py: insert(): Fixed bug where column remaing needed to be removed and sql_gen.Literal value unwrapped
- 03:40 PM Revision 2646: sql.py: mk_insert_select(): cols: Fixed bug where table name needed to be removed from columns with a table
- 03:40 PM Revision 2645: sql_gen.py: ColDict: Only allow dict inputs to constructor so update() can later be used. Ensure keys_table is a Table object.
- 02:55 PM Revision 2644: sql_gen.py: ColDict: Translate NULL values to the key column's actual default value. This avoids incorrectly comparing against NULL in join/filter conditions when the column's actual default value must be used.
- 02:52 PM Revision 2643: sql.py: DbConn: Added col_default()
- 02:52 PM Revision 2642: sql_gen.py: ColDict: Take a db param for later use in getting default column values
- 02:16 PM Revision 2641: sql_gen.py: ColDict: Take a keys_table param for operations that require the output table. Use this to convert all string keys to sql_gen.Col objects.
- 01:55 PM Revision 2640: sql.py: put(): Use a sql_gen.ColDict for row, just like in put_table(), so that put() will benefit from planned enhancements to sql_gen.ColDict, such as translating NULL to the default value of a NOT NULL column
- 01:38 PM Revision 2639: schemas/functions.sql: _nullIf: Fixed bug where NOT NULL parameters were not supported, because an input NULL value would not match an existing DEFAULT value in a select query, by temporarily disabling _nullIf until this can be supported. Removed previous workarounds.
06/05/2012
- 03:21 PM Revision 2638: bin/map: out_is_db, row-based mode: Disabled using DB relational functions instead of XML functions because they were causing problems
- 03:18 PM Revision 2637: sql.py: put(): DuplicateKeyException: Fixed bug where select() call needs to be surrounded by value_or_none(), not value(), because sometimes there is no entry
- 03:17 PM Revision 2636: schemas/functions.sql: Added indexes on all relational function parameters since the unique index will not be used for lookups
- 02:29 PM Revision 2635: schemas/py_functions.sql: Added _namePart relational function
- 01:06 PM Revision 2634: schemas/py_functions.sql: Added _dateRangeEnd relational function
- 12:59 PM Revision 2633: main Makefile: schemas/%/uninstall: Removed delete confirmation message because it's repetitive to keep being asked to confirm deletion when running schemas/reinstall. schemas/%/uninstall is not intended to be run by itself, so the public schema's delete confirmation message will protect the entire schemas/uninstall sequence.
- 12:55 PM Revision 2632: schemas/py_functions.sql: _dateRangeStart(): Fixed bug where needed to get the first element of the array returned by parse_date_range()
- 12:32 PM Revision 2631: main Makefile: VegBIEN DB: Schemas: Changed to only allow reinstalling all schemas at once (using the usual `make schemas/reinstall` command), because reinstalling just one schema would silently remove anything in other schemas that depends on that schema
- 11:52 AM Revision 2630: schemas/py_functions.sql: Added _dateRangeStart relational function
- 10:53 AM Revision 2629: vegbien_dest: $schemas: Added py_functions
- 10:49 AM Revision 2628: pg_dump_vegbien: Fixed bug where needed space before $noOwners options
- 10:37 AM Revision 2627: main Makefile: VegBIEN DB: DB and bien user: mk_db: Also `createlang plpythonu`
- 10:26 AM Revision 2626: pg_dump_vegbien: Log the pg_dump command being run
- 10:23 AM Revision 2625: main Makefile: VegBIEN DB: DB and bien user: Added schemas/py_functions/reset. db: Create py_functions schema.
- 10:22 AM Revision 2624: main Makefile: VegBIEN DB: DB and bien user: Added schemas/py_functions/reset. db: Create py_functions schema.
- 10:16 AM Revision 2623: schemas/py_functions.sql.make: Fixed bug where owners needed to be included because schema is imported as superuser so that untrusted PL/Python functions can be created
- 10:15 AM Revision 2622: pg_dump_vegbien: Support optionally including owners
- 09:59 AM Revision 2621: main Makefile: VegBIEN DB: DB and bien user: Factored $(confirmRm<schema>) functions message text out into $(confirmRmSchema) function
- 09:52 AM Revision 2620: schemas/Makefile, py_functions.sql.make: Generate py_functions.sql from vegbien's py_functions schema
- 09:32 AM Revision 2619: main Makefile: postgres-Linux: Install postgresql-plpython
- 09:27 AM Revision 2618: main Makefile: python-Linux, postgres-Linux: Fixed bug where apt-get installs needed to each be run in a separate command, so that if any package was not found, the other packages would still install. (apt-get aborts on the first invalid package name.)
- 09:18 AM Revision 2617: db_dump_localize: Use new pg_version
- 09:18 AM Revision 2616: Added pg_version
- 08:05 AM Revision 2615: sql.py: into_table_name(): If relational function has a value argument, don't include other arguments, to save space
- 08:03 AM Revision 2614: sql.py: add_pkey(): Version the index name just in case add_suffix() doesn't correctly preserve a needed version #
- 08:01 AM Revision 2613: sql_gen.py: add_suffix(): Fixed bug where only strings already at the max length had the version preserved, even though appending the suffix could bring it past the max length and still cause the version to be overwritten. Fixed bug where last # in str, not first, should be considered to precede the version.
- 07:46 AM Revision 2612: sql.py: put_table(): mapping param: Fixed documentation of supported key/value types
- 07:09 AM Revision 2611: db_xml.py: put_table(): Removed no longer accurate comment about handling _simplifyPath
- 07:01 AM Revision 2610: schemas/functions.sql: Added _nullIf relational function
- 06:39 AM Revision 2609: sql_gen.py: add_suffix(): Preserve version so that it won't be truncated off the string, leading to collisions
06/04/2012
- 03:35 PM Revision 2608: sql_gen.py: identifier_max_len: Fixed bug where PostgreSQL's max length was actually 63, not 64
- 03:18 PM Revision 2607: schemas/functions.sql: _label(): Fixed bug where some Python syntax had not been translated to PostgreSQL
- 03:07 PM Revision 2606: schemas/functions.sql: Added _label relational function
- 03:06 PM Revision 2605: db_xml.py: put_table(): Subsetting in_table: Fixed bug where in_table was not being ordered by the row_num, because order_by was set to None when it should have been omitted so it would default to the pkey
- 02:51 PM Revision 2604: csv2db: Increased frequency of "Processed .. row(s)" messages to match slower, more common INSERT case instead of faster, less used COPY FROM case
- 02:40 PM Revision 2603: schemas/functions.sql: _merge(): Fixed bug where values were ordered by value instead of by sort order (column name)
- 02:17 PM Revision 2602: xml_func.py: process(): Refactored to emphasize special handling for row-based and column-based modes. In row-based mode, always use a DB relational function over a local XML function when possible, to faciliate testing of DB relational functions in row-based mode. (The shadowed local XML version will still be tested in non-DB modes, such as outputting to intermediate XML files.)
- 01:01 PM Revision 2601: bin/map: Move retrieval of out_db's relational functions outside of process_input() so they can also be used by the non-by_col case
- 12:52 PM Revision 2600: bin/map: out_is_db: Don't evaluate relational functions in xml_func.process() because these will be evaluated by db_xml.put()
- 12:41 PM Revision 2599: xml_func.py: Removed no longer used strip()
- 12:40 PM Revision 2598: bin/map: Use xml_func.process(..., strip=True) instead of xml_func.strip()
- 12:39 PM Revision 2597: xml_func.py: process(): Added strip()'s functionality via strip option
- 12:10 PM Revision 2596: schemas/functions.sql: Added _merge relational function
- 11:48 AM Revision 2595: schemas/functions.sql: Added join_strs() aggregate
- 10:21 AM Revision 2594: sql.py: Renamed index_pkey() to add_pkey() to be consistent with add_index()
- 10:07 AM Revision 2593: sql.py: into_table_name(): In function args, omit column name for function result columns
- 09:57 AM Revision 2592: sql.py: into_table_name(): In function args, keep the input table name for input columns to identify where they came from, except for the *main* input table name because it makes the string too long
- 09:22 AM Revision 2591: sql_gen.py: esc_name(): Don't return plain name if is_safe_name(), because this makes the SQL inconsistent when some names have "_"s and some don't
- 09:17 AM Revision 2590: sql.py: index_pkey(): Use sql_gen.add_suffix() to ensure index name isn't too long
- 09:15 AM Revision 2589: sql.py: put_table(): insert_out_pkeys, insert_in_pkeys: Use sql_gen.add_suffix() to ensure name isn't too long
- 09:07 AM Revision 2588: sql.py: next_version(): Use new sql_gen.add_suffix(). Removed identifier_max_len because it is now in sql_gen.
- 09:07 AM Revision 2587: sql_gen.py: Added identifier_max_len and add_suffix()
- 09:04 AM Revision 2586: next_version(): Append the version # so it looks more natural. Take into account the max identifier length.
- 09:03 AM Revision 2585: strings.py: Added add_suffix()
- 08:51 AM Revision 2584: sql.py: put_table(): Name the in_table just "in" plus the version #, and the insert_in_pkeys/insert_out_pkeys based on in_table, so that they don't take up so much space in the SQL
- 08:50 AM Revision 2583: sql_gen.py: is_safe_name(): Fixed bug where keywords were incorrectly considered safe
- 08:40 AM Revision 2582: strings.py: repr_no_u(): Fixed bug where "u" prefix was removed even in reprs of non-strings
- 08:32 AM Revision 2581: db_xml.py: into_table_name(): Removed no longer necessary handling of simple functions, which is now done by sql.into_table_name(). Ensure that rank params in functions (not tables) are not treated specially as hierarchical.
- 08:21 AM Revision 2580: sql.py: put_table(): If into == None: For function calls, include the arguments in the into table name
- 08:17 AM Revision 2579: sql_gen.py: to_name_only_col(): Support non-Col Code inputs
- 07:42 AM Revision 2578: sql_gen.py CompareCond.to_str(), callers of combine_conds(): Removed unnecessary grouping () to make SQL clearer
- 07:31 AM Revision 2577: sql_gen.py: Added combine_conds() and use it in Join.to_str() and sql.py mk_select()
- 07:18 AM Revision 2576: sql_gen.py Join.to_str(), sql.py mk_select(): Combining conditions: Don't add newlines where not needed, so that output is less vertically spread out
- 07:10 AM Revision 2575: sql_gen.py: is_safe_name(): Fixed bug where names starting with a digit were incorrectly considered safe
- 07:06 AM Revision 2574: sql.py: put_table(): Separate temp table names from into table name with "_" instead of "-" so that quoting the table name will usually be unnecessary
- 07:03 AM Revision 2573: sql.py: esc_name_by_module(): Remove unused param ignore_case
- 06:59 AM Revision 2572: sql_gen.py: esc_name(): If is_safe_name(), just return name, to avoid escessive escaping in debug output for Redmine
- 06:55 AM Revision 2571: sql_gen.py: is_safe_name(): Don't consider uppercase letters safe because they would cause inconsistent behavior in PostgreSQL if quoted vs. not quoted (only unquoted identifiers are case-insensitive)
- 06:51 AM Revision 2570: sql.py: Removed no longer needed check_name()
- 06:50 AM Revision 2569: sql.py: esc_name_by_module(): psycopg2: If ignore_case is set but name is unsafe, just escape it instead of raising an exception
- 06:49 AM Revision 2568: sql_gen.py: Added is_safe_name()
- 06:39 AM Revision 2567: sql.py: put_table(): col_ustr(): Removed no longer needed sql_gen.as_Col() because mapping and join_cols now ensure that their contents are sql_gen.Col objects
06/01/2012
- 08:29 PM Revision 2566: schemas/functions.sql: Added _alt relational function
- 08:28 PM Revision 2565: sql.py: put_table(): Make mapping and join_cols a sql_gen.ColDict so that literal values will always be turned into sql_gen.Col objects. DuplicateKeyException: Use dict_subset_right_join() instead of dict_subset() so that all columns in a constraint are included in joins on out_table (such as for a relational function with omitted arguments).
- 08:25 PM Revision 2564: sql_gen.py: Added ColDict
- 08:19 PM Revision 2563: sql_gen.py: as_Col(): Added optional name param to specify that non-Col input will be renamed using NamedCol with the given name
- 07:06 PM Revision 2562: sql.py: put_table(): FunctionValueException: Fixed bug where only function calls, not plain columns, were handled, by using sql_gen.unwrap_func_call() to remove any function call *only if* there was one
- 07:04 PM Revision 2561: sql_gen.py: Added unwrap_func_call()
- 06:47 PM Revision 2560: bin/map: by_col: Stripping XML functions not in the DB: Fixed bug where preserve_funcs.add() was used when `preserve_funcs |=` should have been used to add the entire iterable that sql.tables() returns
- 06:45 PM Revision 2559: sql.py: not_null_col: Changed value to 'not_null_col' so that column doesn't seem like a status indicator of whether some value is not null (in fact it's just a column that is always not null)
- 06:05 PM Revision 2558: xml_func.py: Replaced xpath.get_1() with xpath.get_value() where possible, for simplicity
- 05:59 PM Revision 2557: xml_func.py: strip(): Evaluate structural functions like _ignore and _ref by process() instead of removing them. Store structural functions' names in structural_funcs module var. This ensures that _ref targets are still expanded in column-based import.
- 05:56 PM Revision 2556: xpath.py: get(): Create attrs: Put keys last so that any lookahead assertion's path will be created last as it would have without the assertion. This ensures that any value argument of an XML function will always go last even if a lookahead assertion would otherwise have caused it to be created with the element's keys, which previously were created before the attributes.
- 04:55 PM Revision 2555: sql.py: put_table(): If is_func, default into table name ends in () instead of '-pkeys'
- 04:54 PM Revision 2554: schemas/vegbien.sql, functions.sql: Made cast functions STRICT to enable the RETURNS NULL ON NULL INPUT optimization
- 04:33 PM Revision 2553: db_xml.py: put_table(): Pass is_func to sql.put_table()
- 04:32 PM Revision 2552: sql.py: put_table(): Added is_func param for whether out_table is the name of a SQL function, not a table
- 04:09 PM Revision 2551: db_xml.py: put_table(): Treat every node name that starts with "_" as a function, not just members of put_table_special_funcs. This ensures that DB function args are always treated as values, not children with fkeys to parent.
- 03:40 PM Revision 2550: bin/map: by_col: Strip only XML functions that are not in the DB
- 03:39 PM Revision 2549: db_xml.py: put_table(): Make special_funcs externally available as module constant put_table_special_funcs
- 03:38 PM Revision 2548: sql.py: tables(): Changed schema param to schema_like and filter the schema using LIKE so that all schemas can be selected
- 01:56 PM Revision 2547: to_do/timeline.doc: Updated to reflect the month we spent on optimization and column-based import
- 12:54 PM Revision 2546: sql.py: put_table(): in_table name: Remove '-pkeys' suffix from the into table name before adding '-input' so that the name is shorter and clearer
- 12:43 PM Revision 2545: sql.py: put_table(): Wrap repr() calls for debug messages in strings.as_tt() to add Redmine formatting
- 12:39 PM Revision 2544: sql.py: put_table(): Output "Adding index" debug message with level=2.5 so it's not part of the Redmine steps
05/31/2012
- 03:39 PM Revision 2543: schemas/vegbien.sql, functions.sql: Cast functions: Fixed bug where invalid value exceptions were not being caught, because implicit conversions to the return type apparently only happen outside the block containing the RETURN statement (i.e. at the end of the function). Fixed by adding explicit type conversion to return type, so that type conversion would happen inside try block.
- 03:31 PM Revision 2542: sql.py: put_table(): Re-enabled FunctionValueException handling, by just filtering out the value on all input columns that use the named function (since the error message does not specify which column it was that had the invalid value). This is in some ways better, anyway, because that way the invalid value is filtered out right away in all columns that could contain it, instead of potentially once for each column (if the value appears in more than one input column).
- 03:18 PM Revision 2541: sql.py: add_index(): Fixed bug where expressions could not be converted to a string until their table name had been removed
- 03:17 PM Revision 2540: sql_gen.py: Added Expr
- 03:13 PM Revision 2539: sql.py: add_index(): Fixed bug where expressions needed to be enclosed in () to distinguish them from plain columns
- 03:06 PM Revision 2538: sql.py: add_index(): Support simple expressions as well as columns
- 02:37 PM Revision 2537: sql.py: Renamed index_col() to add_index() so its name isn't similar to index_cols()
- 02:33 PM Revision 2536: sql_gen.py: FunctionCall: Removed __repr__() because it's a Code object and its to_str() does not take extra arguments
- 02:12 PM Revision 2535: sql.py: run_query(): FunctionValueException: Expanded parsing to include regular function calls, not just relational functions' trigger functions. put_table(): Disabled FunctionValueException handling because this expands FunctionValueException beyond what put_table() could handle.
- 01:38 PM Revision 2534: sql.py: put_table(): MissingCastException: Fixed bug where renaming of cast literal value was not properly propagated to the returned value of the function call, causing the query to assume that a DISTINCT ON column referred to column in one of the joined tables instead of a named column in the SELECT columns list. This logic error would have been very difficult to catch without inspecting the code!
- 01:33 PM Revision 2533: sql_gen.py: Added wrap_in_func()
- 01:25 PM Revision 2532: sql_gen.py: FunctionCall: Filter args through remove_col_rename() to remove any renamings from the function args
- 01:20 PM Revision 2531: sql.py: put_table(): No handler for exception: Print full exception instead of just first line to assist in debugging
- 01:06 PM Revision 2530: schemas/vegbien.sql, functions.sql: Removed _to* relational functions because type casting for those types is now automatic
- 01:02 PM Revision 2529: mappings/DwC2-VegBIEN.specimens.csv: Removed _to* relational functions because type casting for those types is now automatic
- 12:59 PM Revision 2528: schemas/functions.sql: Added cast functions for _to* relational functions
- 12:58 PM Revision 2527: schemas/vegbien.sql: Changed cast functions' input types to text because type must match exactly, not just be implicitly castable
- 12:47 PM Revision 2526: sql.py: run_query(): MissingCastException parsing: Support multiple-word types
- 12:38 PM Revision 2525: sql.py: put_table(): Handle MissingCastExceptions by attempting to call a function with the name of the type on the column
- 12:33 PM Revision 2524: sql_gen.py: Added Functions section with Function and FunctionCall
- 11:56 AM Revision 2523: sql.py: Added MissingCastException and parse it in run_query()
- 11:36 AM Revision 2522: schemas/vegbien.sql: Added cast functions for enum types which map invalid values to NULL
- 10:57 AM Revision 2521: sql.py: put_table(): Fixed bug where some exceptions with no handler would not even allow insertion of no rows into the out_table (due to type mismatch issues), by creating an empty pkeys table as a special case
- 10:49 AM Revision 2520: sql.py: put_table(): Preparing to insert new rows: Fixed bug where main_select needed to be generated *after* distinct_on was set in the if statement
- 10:48 AM Revision 2519: sql.py: put_table(): log_exc(): Fixed bug where the exception strings rather than the exceptions themselves needed to be put in the set, because exceptions are not comparable with ==
- 10:25 AM Revision 2518: sql.py: put_table(): Moved mk_main_select() call out of try block since it is not related to the exceptions that may be thrown
- 10:17 AM Revision 2517: sql.py: put_table(): log_exc(): Check if exception already caught before to avoid infinite loops
- 09:35 AM Revision 2516: Added debug2redmine and helper file debug2redmine.csv
- 09:20 AM Revision 2515: sql.py, db_xml.py: Removed unnecessary calls to sql_gen.clean_name() now that str() handles this automatically
- 09:14 AM Revision 2514: sql_gen.py: sql_gen classes inherit from new base class BasicObject, whose __str__() calls clean_name() on the object's repr(). Changed the main debug-repr producing method to be repr() instead of str().
- 08:45 AM Revision 2513: Moved clean_name() from sql.py to sql_gen.py because it's DB-general and so that it can be used by sql_gen.py without circular dependencies
- 08:41 AM Revision 2512: db_xml.py: into_table_name(): Handle hierarchical tables specially by including their rank in the into table. Interpret any table with a value column as a function, regardless of out_table name.
05/30/2012
- 11:07 PM Revision 2511: sql.py: put_table(): Log "Default value column does not exist in mapping" error with level 2.1 so that it doesn't appear in Redmine output
- 11:05 PM Revision 2510: db_xml.py: put_table(): Pass next as sql.put_table()'s default param now that it is supported
- 11:04 PM Revision 2509: sql.py: put_table(): Changed default param to be an *output* column because that is what would be passed in by db_xml.put_table(), and because there is already a mapping that resolves that to a flattened input column
- 10:37 PM Revision 2508: sql.py: put_table(): Added default param for the value or input column to use as the pkey for missing rows
- 10:20 PM Revision 2507: sql.py: put_table(): Use single quotes rather than double quotes around strings where possible
- 10:18 PM Revision 2506: db_xml.py: Added internal next param used by _simplifyPath. put_table_(): Refactored to use outer parent_ids_loc var and modify that as needed rather than having to pass parent_ids_loc as a param to put_table_().
- 09:55 PM Revision 2505: sql.py: put_table(): When calling strings.as_*table(), pass custom ustr that removes col renames and adds double quotes on plain strings
- 09:53 PM Revision 2504: strings.py: as_*table(): Added ustr param to override the method (by default ustr()) used to convert each value to a string
- 09:15 PM Revision 2503: sql_gen.py: MockDb.esc_value(): Use new strings.repr_no_u()
- 09:14 PM Revision 2502: strings.py: Added repr_no_u()
- 09:09 PM Revision 2501: sql.py: clean_name(): Also remove '`' (which is used by MySQL)
- 09:06 PM Revision 2500: sql.py: esc_name_by_module(): Use new sql_gen.esc_name()
- 09:03 PM Revision 2499: sql_gen.py: Added esc_name() and use it in MockDb.esc_name()
- 09:00 PM Revision 2498: sql.py: next_version(): Use special chars in version part of name string for clarity
- 08:53 PM Revision 2497: sql.py: mk_insert_select(): embeddable: function_name is first line of query for clarity, and to reduce length from including the column names. This also fixes the problem of double quotes around column names in the previous function_name.
- 08:47 PM Revision 2496: sql.py: esc_name_by_module(): Double embedded quotes to escape them instead of removing them
- 08:35 PM Revision 2495: sql.py: put_table(): Use "-" to separate temp table suffixes from into table name
- 08:26 PM Revision 2494: db_xml.py: into_table_name(): Format relational functions' into table names as a function call on the value column, using special chars for readability
- 08:19 PM Revision 2493: sql.py: run_query(): Exception parsing: Use "(.+?)" wherever possible to match names containing special chars
- 07:52 PM Revision 2492: sql.py: clean_name(): For clarity, just remove '"'s, so that "."s are preserved and show the path structure of the input name
- 07:38 PM Revision 2491: db_xml.py: put_table(): sql.put_table(): Name the into table ...literal instead of ...value if the value column is a literal value
- 07:08 PM Revision 2490: bin/map: Logging: log(): Remove extra debug info from DB query messages and format level 1.5 (summary) messages as Redmine list items
- 06:50 PM Revision 2489: sql.py: put_table(): Renamed temp_prefix param to into and allow it to be a sql_gen.Table object. Use into directly as the pkeys table, and make its default value be `out_table.name+'_pkeys'`.
- 06:31 PM Revision 2488: db_xml.py: put_table(): Pass custom temp_prefix to sql.put_table() for relational funcs, so that their value param's input column name is included in the temp table name
- 06:19 PM Revision 2487: sql.py: put_table(): Added optional param temp_prefix for the prefix of generated temp tables
- 06:13 PM Revision 2486: sql.py: put_table(): Made debug messages more self-documenting
- 05:44 PM Revision 2485: sql.py: put_table(): Changed "Setting missing rows' pkeys to NULL" to "Setting pkeys of missing rows to NULL" to avoid having single quote in debug output, which messes up text editor SQL syntax highlighting
- 05:40 PM Revision 2484: sql.py: Parsed exceptions: Use strings.as_tt() to format Python values
- 05:37 PM Revision 2483: strings.py: Split as_table() into as_table() and as_inline_table() depending on whether the table needs to be inlined in an ordered list item or not
- 05:36 PM Revision 2482: strings.py: Split as_table() into as_table() and as_inline_table() depending on whether the table needs to be inlined in an ordered list item or not
- 05:03 PM Revision 2481: strings.py: as_table(): Changed to use <pre> formatting because Redmine tables can't be embedded in ordered lists without restarting the numbering
- 03:58 PM Revision 2480: strings.py: as_table(): Fixed bug where table was not ended properly, by adding a space after the last \n and having rstrip() string only newlines
05/29/2012
- 09:19 PM Revision 2479: sql.py: mk_select(): Columns: Separate columns with newlines
- 09:10 PM Revision 2478: sql.py: put_table(): Use new strings.as_table() to format mappings as tables
- 09:09 PM Revision 2477: strings.py: Added as_tt() and as_table()
- 09:09 PM Revision 2476: bin/map: Logging: log(): Strip trailing newlines from msg
- 08:40 PM Revision 2475: strings.py: as_code(): Added multiline param to disable multiline formatted output
- 08:33 PM Revision 2474: sql.py: put_table(): "Ignoring existing rows, comparing on" debug message: Wrap the mapping in strings.as_code() so it will have Redmine syntax-highlighting
- 08:26 PM Revision 2473: sql.py: put_table(): "Putting columns" debug message: Wrap the mapping in strings.as_code() so it will have Redmine syntax-highlighting
- 08:22 PM Revision 2472: sql.py: DbConn.run_query(): Query debug message: Wrap the query in strings.as_code() so it will have Redmine syntax-highlighting
- 08:20 PM Revision 2471: strings.py: Added as_code()
- 08:04 PM Revision 2470: sql.py: DbConn.run_query(): Prepend "DB query" before the query debug message so it can be identified as a DB query
- 07:43 PM Revision 2469: db_xml.py: put_table(): Subset in_table: Document that in_table will be shadowed (hidden) by the created temp table, rather than versioned, now that the table is (almost) always created as a temp table
- 07:40 PM Revision 2468: sql.py: Create temp items as permanent in autocommit mode rather than in debug mode so that temp items are only permanent if actually committing result. This ensures that the generated SQL in test mode matches what would actually get run in regular commit mode, and the SQL is only altered to make the temp items visible if actually debugging (autocommit mode).
- 07:30 PM Revision 2467: sql.py, sql_gen.py: Reformatted generated SQL for presentability by adding newlines
- 07:14 PM Revision 2466: sql.py: DbConn.run_query(): Put a newline before the query in the debug message so that multiline queries have all rows at the left edge rather than the first row prefixed by other text
- 07:09 PM Revision 2465: sql.py: DbConn.run_query(): Don't put generated query debug message all on one line, so that embedded newlines are preserved
- 06:59 PM Revision 2464: sql.py: Fixed bug where queries with versioned identifiers which threw an exception (not related to name collisions) were being output with a too-high log_level, because *all* exceptions were output with the higher exc_log_level, by making the following changes: DbConn.run_query(): Changed exc_log_level param to log_ignore_excs param so that only certain exceptions would cause the query to be output with a higher log_level. Moved the code that actual emits the query debug message from DbConn.run_query() to module-level run_query() so it would apply the log_ignore_excs filter after the exception had already been parsed into specific types.
- 03:16 PM Revision 2463: Moved "Putting columns" debug message from db_xml.py put_table() to sql.py put_table() to put it in the same place as the other debug messages
- 03:12 PM Revision 2462: sql_gen.py: Added remove_col_rename() and use it where `if isinstance(value, NamedCol): value = value.code` was used
- 03:10 PM Revision 2461: sql_gen.py: CompareCond.to_str(): If left_value has been renamed as a NamedCol, unwrap it
- 02:53 PM Revision 2460: sql_gen.py: Join.to_str(): Fixed bug where USING should be used if all columns are join_same_not_null, rather than join_same, because USING uses plain = for comparison. sql.py: put_table(): input_joins now can use sql_gen.join_same_not_null in order to use USING syntax.
05/25/2012
- 07:14 PM Revision 2459: db_xml.py: put_table(): Output debug messages with a level of 1.5 to match sql.put_table()'s level for summary messages
- 07:01 PM Revision 2458: bin/map: Fixed bug where verbosity needed to be 1 outside of test mode so that profiling and errors stats would be printed at end of import. Verbosity defaults to 0.5 rather than 1 in test mode so profiling and errors stats do not clutter up the test output when running automated tests.
- 06:55 PM Revision 2457: bin/map: Only display verbose_errors in test mode, but with any nonzero verbosity. They should not be displayed outside of test mode because verbose errors make the log files huge.
- 06:52 PM Revision 2456: bin/map: Renamed verbose param to verbosity because it's now a number, not a boolean
- 06:51 PM Revision 2455: bin/map: Removed no longer used debug param (verbose=2 is used instead)
- 06:48 PM Revision 2454: bin/map: Fixed bug where verbose_errors' default value depended on debug var, which was not yet set. Removed verbose_errors param and instead turn verbose_errors on whenever verbosity >= 1. Verbosity defaults to 1 in test mode.
- 06:33 PM Revision 2453: bin/map: Logging: Don't set sql.run_raw_query.debug, because it is not used anymore (sql.connect(log_debug=...) is used instead)
- 06:29 PM Revision 2452: bin/map: Logging: Print debug messages (level > 1) prefixed with their level, to distinguish higher- and lower-level debug messages
- 06:22 PM Revision 2451: sql.py: put_table(): Only display warning for exceptions with no handler (which are unexpected), not missing mappings for NOT NULL columns (which are normal in datasources without those columns)
- 06:15 PM Revision 2450: sql.py: put_table(): Log summarizing debug messages with a level of 1.5 so they will be displayed even when the major SQL queries (which have a level of 2) are not shown
- 06:08 PM Revision 2449: bin/map: Provide a log_debug() function to sql.connect() if verbosity > 1 rather than >= 2, to support fractional verbosities
- 06:04 PM Revision 2448: sql.py: log_debug_none: Fixed bug where needed to take kw arg level to work with verbosity-based logging
- 05:57 PM Revision 2447: bin/map: Allow fractional verbosity values
- 05:56 PM Revision 2446: sql.py: Functions that version created tables, functions, etc. if they already exist: Use (default) exc_log_level=4 to hide the unsuccessful attempts to create items that already exist and show only the successful attempt
- 05:43 PM Revision 2445: sql.py: DbConn.run_query(): Added exc_log_level param to specify a different log_level if the query throws an exception. This will useful for functions that version created tables, functions, etc. if they already exist.
- 05:34 PM Revision 2444: sql.py: DbConn.run_query(): Removed no longer accurate doc comment, because that functionality is now in module-level run_query()
- 05:31 PM Revision 2443: sql.py: Specify log_levels for minor queries so they can be excluded from the debug output
- 05:16 PM Revision 2442: sql.py: select(): Pass log_level to run_query()
- 05:13 PM Revision 2441: sql.py: DbConn.run_query(): Added log_level param and pass it to self.log_debug(). run_query(): Pass extra kw_args to DbConn.run_query() (via run_raw_query()) so that caller can specify log_level.
- 04:54 PM Revision 2440: sql.py: run_query_into(): Fixed bug where "temporary tables cannot specify a schema name"
- 04:42 PM Revision 2439: bin/map: Switched to verbosity-level-based system of logging. verbose is now an integer, and debug sets the minimum verbosity to 2.
- 04:37 PM Revision 2438: input.Makefile: Configuration: Removed debug var since it's not used in the Makefile
- 04:09 PM Revision 2437: db_xml.py: put_table(): put_table_(): Fixed bug where row_ins_ct_ref needed to be passed recursively to put_table() as keyword arg, because the in_row_ct_ref is not passed recursively
- 04:07 PM Revision 2436: db_xml.py: put_table(): _simplifyPath: Parse "next" XPath param to extract col name of next level's pkey
- 03:26 PM Revision 2435: bin/map: by_col: xml_func.strip(): Don't remove _simplifyPath because it is now handled by db_xml.put_table()
- 03:25 PM Revision 2434: db_xml.py: put_table(): Added basic special handling for structural XML functions, which for now just skips the function
- 03:21 PM Revision 2433: xml_func.py: strip(): Added preserve param for XML functions not to remove
- 02:49 PM Revision 2432: db_xml.py: put_table(): Handle forward pointers in translation-to-sql_gen step instead of in XML-tree-parsing step, so that special handling for structural XML functions can use the parsed tree before any sql.put_table() processing takes place
- 02:44 PM Revision 2431: xml_dom.py: Added is_node()
- 02:22 PM Revision 2430: sql.py: table_row_count(): Pass start=0 to mk_select() to avoid "SELECT statement missing a WHERE, LIMIT, or OFFSET clause" warnings
- 02:12 PM Revision 2429: sql.py: put_table(): Handle unknown exceptions by returning NULL for all rows. Refactored Missing mapping for NOT NULL column handling to use new helper function remove_all_rows().
- 01:54 PM Revision 2428: sql.py: put_table(): Assert that insert_out_pkeys and insert_in_pkeys have same row count. Assert that pkeys and in_table have same row count.
- 12:57 PM Revision 2427: db_xml.py: put_table(): Use new sql.table_row_count()
- 12:56 PM Revision 2426: sql.py: Added table_row_count()
- 12:52 PM Revision 2425: db_xml.py: put_table(): Use new sql_gen.row_count
- 12:47 PM Revision 2424: sql_gen.py: Added row_count
- 12:41 PM Revision 2423: db_xml.py: put_table(): Count # rows and update in_row_ct_ref once all columns have been processed. Don't pass in_row_ct_ref to recursive calls because it should only be increased once.
- 12:28 PM Revision 2422: db_xml.py: put_table(): Added in_row_ct_ref param to store the # of input rows processed. Renamed row_ct_ref param to row_ins_ct_ref to distinguish it from new in_row_ct_ref param.
05/24/2012
- 09:26 PM Revision 2421: sql_gen.py: MockDb.esc_name(): Don't use sql.esc_name_by_module() to avoid circular dependency on sql module
- 09:20 PM Revision 2420: sql.py: put_table(): Factored out mk_select() calls in calls to run_query_into_pkeys() into new helper function insert_into_pkeys()
- 09:09 PM Revision 2419: sql.py: put_table(): run_query_into_pkeys() calls use order_by=None in their select statements because there is a pkey, so order (row #) does not matter
- 09:05 PM Revision 2418: db_xml.py: put_table(): Subset in_table if limit != None or start != 0. start param defaults to 0 again to avoid subsetting the table when starting from row 0 (with no limit).
- 08:46 PM Revision 2417: db_xml.py: put_table(): Don't pass limit, start recursively, because the table subsetting will happen only once in the first invocation of the function. Moved limit, start params to end since they are not passed recursively. start param no longer defaults to 0 because this is not needed since sql.put_table() now sets start to 0 where needed.
- 08:38 PM Revision 2416: sql.py: put_table(): Removed limit and start params because they were never fully implemented, and because it's simpler to just have the caller subset their input table
- 08:27 PM Revision 2415: lists.py: Added uniqify()
- 08:08 PM Revision 2414: sql.py: Moved mk_flatten_mapping(), flatten() to Basic queries section since they don't involve database structure info
- 08:06 PM Revision 2413: sql.py: put_table(): Use single quotes rather than double quotes around strings where possible
- 07:59 PM Revision 2412: schemas/functions.sql, vegbien.sql: Changed CAST-related relational functions to return NULL on data exceptions and convert the exceptions to warnings. This helps column-based import by mapping invalid values to NULL instead of aborting the whole query on the first invalid value.
- 07:33 PM Revision 2411: sql.py: index_col(): Cache the query so it doesn't try to add an index on the same column multiple times
- 07:18 PM Revision 2410: sql.py mk_select(), sql_gen.py Join.to_str(): Fixed bug where conditions needed to be wrapped in () before being AND-ed together to ensure the proper operator precedence
- 06:49 PM Revision 2409: sql.py: put_table(): Add index on columns with invalid values to enable fast filtering
- 06:47 PM Revision 2408: sql.py: Added index_col()
- 06:18 PM Revision 2407: sql.py: put_table(): Add pkey on returned pkeys table to enable fast joins
- 06:17 PM Revision 2406: sql.py: Added index_pkey()
- 05:41 PM Revision 2405: sql.py: mk_update(): When running sql_gen.to_name_only_col(), check that the col's table is table
- 05:38 PM Revision 2404: sql.py: put_table(): Renamed *_pkeys to insert_*_pkeys to distinguish them from the full set of pkeys on the input table
- 05:27 PM Revision 2403: sql.py: put_table(): FunctionValueException: Change invalid values to NULL using UPDATE instead of filtering them out using WHERE, to avoid adding lots of conditions to the SELECT statement
- 05:11 PM Revision 2402: sql.py: Added mk_update() and update()
- 05:10 PM Revision 2401: sql_gen.py: Added to_name_only_col()
- 04:56 PM Revision 2400: sql_gen.py: Added as_Value()
- 04:29 PM Revision 2399: sql.py: mk_select(): conds: Use new sql_gen.ColValueCond instead of sql_gen.as_ValueCond(). Documented that Code and ValueCond are sql_gen objects.
- 04:28 PM Revision 2398: sql_gen.py: Added ColValueCond
- 03:59 PM Revision 2397: sql.py: mk_flatten_mapping(): Filter str(col) through clean_name() to remove quotes, etc.
- 03:58 PM Revision 2396: sql.py: Added clean_name()
- 03:43 PM Revision 2395: sql.py: put_table(): Join together input tables into new table for speed and so don't modify input if values edited
- 03:37 PM Revision 2394: sql.py: mk_flatten_mapping(): Take as_items param to return a list of dict items instead of a dict. Sort preserve cols before other cols. flatten(): Turn on as_items so that cols list is sorted in input order, with preserve cols first. This ensures that if a pkey is provided in preserve, it will be the first col in the generated table.
- 03:24 PM Revision 2393: sql.py: mk_flatten_mapping(), flatten(): Take list of cols to select instead of using all cols in all tables to join
- 02:58 PM Revision 2392: sql.py: mk_flatten_mapping(), flatten(): Renamed flat_table param to into to be consistent with run_query_into() and put it first because it is the output param
- 02:55 PM Revision 2391: sql.py: Added flatten()
- 02:38 PM Revision 2390: sql.py: mk_flatten_mapping(): preserve Col objects will have tables changed to flat_table to work with flattened table
- 02:29 PM Revision 2389: sql.py: mk_flatten_mapping(): Added preserve param for list of columns not to rename
- 02:18 PM Revision 2388: sql.py: esc_name_by_module(): Support module value None, and use default module psycopg2 for it
05/23/2012
- 09:58 PM Revision 2387: sql.py: put_table(): Renamed *pkeys_ref to *pkeys to reflect that they are now objects rather than an array-based references
- 09:54 PM Revision 2386: sql.py: run_query_into(): Renamed into_ref param to into to reflect that it's now an object rather than an array-based reference
- 09:51 PM Revision 2385: sql.py: run_query_into(): Made into_ref a sql_gen.Table instead of an array containing a table name to improve flexibility and clarity
- 09:34 PM Revision 2384: dicts.py: Added join()
- 09:20 PM Revision 2383: sql.py: Added mk_flatten_mapping()
- 08:28 PM Revision 2382: sql.py: put_table(): Renamed the copy of in_tables that gets modified to in_tables_, so that the original list can eventually be reused in joining together the input tables into a temp table
- 07:10 PM Revision 2381: sql.py: run_query(): FunctionValueException: Also match "date/time field value out of range" errors
- 07:04 PM Revision 2380: sql.py: put_table(): conds: Use a set instead of a list for faster checking of the "cond not in conds" assertion
- 06:55 PM Revision 2379: sql.py: mk_select(): conds: Support containers of any iterable type
- 06:52 PM Revision 2378: sql.py: put_table(): Made conds a list so that there can be multiple conditions on the same column
- 06:36 PM Revision 2377: sql.py: mk_select(): conds is list of (key, value) tuples instead of dict (dict still supported for compatibility), so that there can be multiple conditions on the same column
- 06:35 PM Revision 2376: sql.py: mk_select(): conds is list of (key, value) tuples instead of dict (dict still supported for compatibility), so that there can be multiple conditions on the same column
- 06:28 PM Revision 2375: util.py: NamedTuple inherits from objects.BasicObject so that it's comparable and hashable. This fixes a bug in dicts.make_hashable() where the NamedTuple created for a dict would appear to be hashable but would always compare as unequal.
- 06:15 PM Revision 2374: sql.py: DbConn.esc_value(): Run strings.to_unicode() on the generated string so that if it contains unescaped non-ASCII characters, these will not cause problems when concatenated with plain strings
- 05:58 PM Revision 2373: sql.py: run_query(): FunctionValueException: Unpack match.groups() into vars to make code clearer
- 05:56 PM Revision 2372: exc.py: str_(): Avoid traceback exception-formatting functions when possible because they escape non-ASCII characters
- 05:11 PM Revision 2371: sql.py: get_cur_query(): If no raw query: Use strings.ustr() instead of repr() to ensure that if the exception is parsed, embedded quotes will not be double-escaped. Prefix the query by [input] to show that it's not the raw query.
- 04:59 PM Revision 2370: sql_gen.py: Non-Code objects: __str__() passes informative placeholder string to self.to_str() instead of empty string
- 04:41 PM Revision 2369: sql.py: ExceptionWithNameValue: Use repr() instead of strings.ustr() on the value
- 04:38 PM Revision 2368: sql.py: run_query(): Exception parsing: Use non-greedy qualifier "?" in regexps wherever possible to avoid matching closing quotes later in the error message
- 04:32 PM Revision 2367: sql_gen.py: MockDb.esc_value(): Use repr() instead of strings.ustr() so the quotes around the value are included
- 04:30 PM Revision 2366: sql_gen.py: ValueCond and Join class hierarchies inherit from objects.BasicObject like Code does
- 04:24 PM Revision 2365: sql.py: put_table(): ignore(): Fixed bug where value needed to be filtered through repr(). NullValueException: Fixed bug where value passed to ignore() was the string 'NULL' instead of the value None.
- 04:14 PM Revision 2364: mappings/DwC2-VegBIEN.specimens.csv: plantname.rank: Filter through _toTaxonrank
- 04:03 PM Revision 2363: sql.py: put_table(): ignore(): Avoid infinite loops by asserting that in_col is not in conds
- 03:58 PM Revision 2362: objects.py: BasicObject: Fixed bug where util needed to be imported. Added __eq__() and __hash__().
- 03:47 PM Revision 2361: strings.py: Removed no longer used DebugPrintable (that functionality is now in objects.BasicObject)
- 03:46 PM Revision 2360: sql_gen.py: Code: Inherit from new objects.BasicObject
- 03:46 PM Revision 2359: Added objects.py
- 03:37 PM Revision 2358: sql.py: put_table(): Renamed log_ignore() to ignore() and factored common conds-modifying code into it
- 03:29 PM Revision 2357: sql.py: put_table(): Moved post-insert code outside while loop because it will now always be run (there are no longer special cases where the postprocessing doesn't happen)
- 03:25 PM Revision 2356: sql.py: put_table(): Missing mapping for NOT NULL column: Just create an empty pkeys table, since the missing rows' pkeys will be set to NULL later
- 03:17 PM Revision 2355: sql.py: put_table(): Joining together output and input pkeys: Use new sql_gen.join_same_not_null
- 03:14 PM Revision 2354: sql.py: put_table(): Setting missing rows' pkeys to NULL: Use new sql_gen.join_same_not_null
- 03:14 PM Revision 2353: sql_gen.py: Join: Added join_same_not_null. to_str(): Refactored to switch order of left and right tables and cols because left_table is on the right in the comparison, and using the sides of the comparison instead of the sides of the join makes the code clearer.
- 02:51 PM Revision 2352: sql_gen.py: Renamed join_using to join_same to reflect that it can also be used without USING
- 02:48 PM Revision 2351: sql.py: put_table(): Set missing rows' pkeys to NULL
- 02:10 PM Revision 2350: sql.py: put_table(): NullValueException: no mapping for missing col: Fixed bug where run_query_into_pkeys() was still using insert_joins instead of input_joins
- 02:06 PM Revision 2349: sql_gen.py: Added MockDb. All __str__() methods: Use self.to_str() with mockDb.
- 01:59 PM Revision 2348: sql_gen.py: Use db.esc_name() instead of sql.esc_name(db, ...) so passed-in db can be a mock object
- 01:58 PM Revision 2347: sql.py: DbConn: Added esc_name()
- 01:51 PM Revision 2346: db_xml.py: put_table(): Debug-print which columns are being put
- 01:50 PM Revision 2345: sql.py: ConstraintException, NullValueException: Improved error messages
- 01:31 PM Revision 2344: sql.py: put_table(): FunctionValueException: Fixed bug where out_table was still assumed to be an escaped string, but is now a Table object
- 01:29 PM Revision 2343: sql.py: mk_select(): joins: Use new table_not_null_col() instead of pkey() to get a non-NULL column to filter out on
Also available in: Atom