René's URL Explorer Experiment

{"@context":"https://schema.org","@type":"DiscussionForumPosting","headline":"Check datatype compatibility pre vs post #514","articleBody":"With PR #514, we now actively restrict what MySQL attribute data types are formally supported in DataJoint. This unfortunately means that if users were using \"unsupported\" data types prior to this PR, their code will cease to work (or worse yet, they will continue to be able to use the table, but will encounter an issue the next time they try to re-declare the table). To minimize issues caused by this, I've listed all datatypes present under MySQL so that we can explicitly rule them in vs out of our supported data types list. Following is the suggested criterion for the selection:\r\n\r\n1. you could insert into an attribute of that datatype\r\n2. you could fetch from an attribute of that datatype\r\n3. Fetched value from that datatype can be inserted back directly without any inconsistency or error\r\n4. the use of the datatype in DataJoint makes sense/fits the expected use of DataJoint\r\n\r\nwith that here are the data types collected from [MySQL 8.0 reference manual](https://dev.mysql.com/doc/refman/8.0/en/numeric-type-overview.html). Checked item means that the decisions are already made for support/non-support and have been verified:\r\n\r\n### Numeric types\r\n- [x] INTEGER (supported?)\r\n- [x] (TINY/SMALL/MEDIUM/BIG)INT (**supported**)\r\n- [x] DECIMAL (**supported**)\r\n- [ ] NUMERIC\r\n- [x] FLOAT (**supported**)\r\n- [ ] REAL\r\n- [x] DOUBLE (**supported**)\r\n- [x] BOOL\r\n- [x] BOOLEAN\r\n- [ ] BIT\r\n- [ ] BIT(M)\r\n\r\n### Date and time types\r\n- [x] DATE\r\n- [x] TIME\r\n- [x] DATETIME (**supported**)\r\n- [x] TIMESTAMP\r\n- [x] YEAR\r\n\r\n### String types\r\n- [x] CHAR\r\n- [x] VARCHAR\r\n- [ ] BINARY\r\n- [ ] VARBINARY\r\n- [x] (TINY/MEDIUM/LONG)BLOB\r\n- [x] (TINY/MEDIUM/LONG)TEXT\r\n- [x] ENUM\r\n- [ ] SET\r\n\r\n\r\n\r\n\r\n","author":{"url":"https://github.com/eywalker","@type":"Person","name":"eywalker"},"datePublished":"2018-12-08T21:46:17.000Z","interactionStatistic":{"@type":"InteractionCounter","interactionType":"https://schema.org/CommentAction","userInteractionCount":6},"url":"https://github.com/544/datajoint-python/issues/544"}