-
Notifications
You must be signed in to change notification settings - Fork 3
/
Changelog.txt
99 lines (89 loc) · 5.23 KB
/
Changelog.txt
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
-- 1.4 [2021-05-10]
- Changed the format of index and temp table names to have a column name in the beginning, to avoid issues with PostgreSQL cutting the last part of the name and causing duplicates.
- Fixed an issue with vertical alighment flags not working on all versions of wxPython
- Fixed granting select permission to a history table when it's not created
-- 1.3 [2020-07-29]
- Added a prompt for a password when connecting to the target database to import stage tables/columns instead of storin it in the metadata db.
-- 1.2 [2020-01-28]
- Added Comments for DWH tables and columns that will be used when creating tables and also useful for generating data model documentation
- Replaced double quotes with single quotes where possible
- Add generating Markdown documentation for the main tables
- Fix debug message formatting
-- 1.1 [2019-12-11]
- Updated Import Tables/Columns function to use system tables that don't require select permissions, to use an account without read access to data
- Added DB role names to grant select on tables
- Added a workaround for a problem with SQLite saving empty string values into INTEGER columns
- Updated deprecated function calls in WxPython
- Removed debugging messages from STDOUT
-- 1.0 [2019-08-27]
- Switched to Python 3
- Started to use UUID based on business key values instead of Surrogate Keys
- Renamed Batch Info tables to Metadata
- Removed PK Lookup tables, started to use Metadata tables to check if key exists
- Replaced MD5 with SHA1 for record hashes
- For Foreign Keys, adding an indexed UUID column
- Removed loading of inferred records
- Rename columns in History tables
- Generate FKs for columns named like column_id or column_key to just column_uuid
-- 0.6 [2018-10-16]
- Replace implicit type casting to explicit for business keys and hash column (failed on JSON field)
- Allow to manually set inferred entity names due to a UI bug on Mac OS not allowing to select from a dropdown
- Fix drop trigger statement for partitioned table DDL
- Comment out ANALYZE statements
- Add forgotten is_track_deleted column
- Update README.md
-- 0.5 [2016-06-23]
- Added an option to Rebuild Indexes after data load
- Added an option to partition table by date column for Postgres (using a trigger and a function)
- Removed schema name from index name, shortened long staging tables names
- Added auto-decode from timezone 'UTC' for Unix timestamps
-- 0.4 [2015-11-25]
- Changed Distkey/Sortkey from position to flags 1/0
- Use Distkey/Sortkey in Redshift
- Add indexes on "Distkey/Sortkey Pos != 0" columns to Postgres
- By default comment out pk_lookup DDL section for non-main entities
- Moved the FK BK expressions to a lower level in source table select, where BK and Hash are
- Fixed more than one date_updated columns
- Added "nulls last" to date_updated columns
- Added timestamp with/without timezone to Hash definitions to use formatting
- Added is_deleted flag in batch_info (is_track_deleted for snapshot load tables) for snapshot loads (move old row to history, keep pk_lookup, remove the row in main)
- Added session id to inferred stage table names, to avoid overwriting with parallel scripts
- Disabled dropping of temporary staging batch table, to avoid unnoticed collisions between running scripts
- Added begin/lock table/commit to avoid conlicts when loading inferred tables with no direct depencendies.
- Showing dropdown tables for FKs even from other stage DBs within one project
- Fixed that "Import Column" didn't populate target attribute name/type
- Disabled Analyze for inferred entities, because it takes too much time
- Showing stage table name in entities tree
- Allowing now to check Sortkey or Distkey for BK columns, because separate indexes might be required for composite (multi-column) indexes
- Removed intendation for "create index" statements
-- 0.3 [2015-06-29]
- Add "Track changes", "Keep history" flags
- When a column is ignored, still use it for last_updated
- Add optional tablespace to tables
- Add target schema to generated file name
- Add ".sql" to save file name in Linux
- Add "Is Truncate Stage" flag
- Add "Is Ignore Changes" flag for columns
- Use tablespace for stage tables and indexes
- Add indexes to stage tables (all entity_bks, row_number, batch_number_old ?)
- Add indexes on target tables DDL (entity_bk for pk_lookup, all entity_keys)
- Remove primary key from history tables
- Setting (use "Is Fk Mandatory") to ignore NULL for entity/inferred BK
-- 0.2 [2015-06-19]
- For inferred: subname of another entity (to populate a column in)
- is_delete_temp_tables flag
- Add DDL generation
- Fix bpchar and int4 -> timestamp for unix timestamps
- Add settings for temporary/permanent stage tables, drop after use or not.
- Fix numeric type without precision
--0.1 [2015-05-30]
- add entity bk
- build hash from all columns as char
- add order by updated date and hash
- build inferred section from each inferred row
- build middle lookup sections with entity name only
- build stage2 - entity key with required joins, then all other columns mixed with fk column keys plus left joins (alias by column name).
- in history, first add system columns, then all table columns.
- Add custom table expression field
- Add custom column expression field
- Add stage schema to stage_db table, use it in all scripts for staging tables