# Instructions for MatrixOne Load Data Support

The complete syntax of [MySQL 8.0](https://dev.mysql.com/doc/refman/8.0/en/load-data.html) is as follows:

```sql
LOAD DATA
    [LOW_PRIORITY | CONCURRENT] [LOCAL]
    INFILE 'file_name'
    [REPLACE | IGNORE]
    INTO TABLE tbl_name
    [PARTITION (partition_name [, partition_name] ...)]
    [CHARACTER SET charset_name]
    [{FIELDS | COLUMNS}
        [TERMINATED BY 'string']
        [[OPTIONALLY] ENCLOSED BY 'char']
        [ESCAPED BY 'char']
    ]
    [LINES
        [STARTING BY 'string']
        [TERMINATED BY 'string']
    ]
    [IGNORE number {LINES | ROWS}]
    [(col_name_or_user_var_var
        [, col_name_or_user_var] ...)]
    [SET col_name={expr | DEFAULT}
        [, col_name={expr | DEFAULT}] ...]
```

## Fields not supported by MatrixOne

In the above complete syntax structure, MatrixOne does not yet support the following fields:

- REPLACE
- PARTITION
- CHARACTER SET: Currently only UTF8 is supported
- FIELDS TERMINATED BY 'string': simdcsv does not accept \r,\n,Unicode replacement character (0xFFFD).
- FIELDS [OPTIONALLY] ENCLOSED BY 'char': simdCSV fixed is `"`.
- FIELDS ESCAPED BY 'char': simdcsv is not supported yet.
- LINES STARTING BY 'string': simdcsv is not supported. Lines with the first character are filtered out.
- LINES TERMINATED BY 'string': simdcsv does not support modification. Fixed '\n', `\r\n`.
- user_var: Not supported. The data columns corresponding to the variable are filtered out.
- SET col_name={expr | DEFAULT}: Not supported.
- LOW_PRIORITY: Not supported
- CONCURRENT: Not supported
- LOCAL: Not supported. local requires data to be transferred from the client side to the server side.

## MatrixOne supports syntax and formats

### grammar

|Grammar | Description |
|---|--|
|use T;||
|load data||
|infile 'pathto/file'||
|ignore|Optional. <br>Add `ignore`, means that when a field is listed in a table, the entire `load` will not be terminated. The number of `warnings` will be recorded, but the content of `warnings` will not be recorded; <br>When `ignore`, the field is listed in a table, the entire `load`|
|INTO TABLE tableA||
|FIELDS TERMINATED BY ','| Must represent the field splitter. Usually ','. And ssb's is '|'|
|IGNORE number LINES|Optional, indicating that the number line at the beginning of the file is ignored|
|(col0,col1,...,coln);|Optional, represents the list of column names in the table. For example, `coli` corresponds to the `i` column in the table. <br> For example, the data row `field0 field1 field2 ... fieldN`, corresponding to the table column `(col0, col1, col2 ,...,colN)`<br>If the column coli starts with '@', then field is discarded; <br>(col0,col1,...,coln) may not be the same as the column names in the table. fieldi binds to the corresponding column name. <br>If the data row has more fields than (col0,col1,...,coln) or the number of columns in the table, the excess fields are discarded. <br>If the data row has fewer fields than (col0,col1,...,coln) or the number of columns in the table, the few columns are supplemented with NULL. <br>If there is no (col0,col1,...,coln) in the load data syntax, it is equivalent to all columns loaded into the table. |

### csv format requirements

|character | Description |
|---|--|
|Field splitter | is usually ',', or other characters (not `"`,'\r','\n'). For example: the SSB dataset is '|'|
|Field contains symbol |`"`|
|Line End Character |`\r\n` or '\n'|
|Field NULL Value | For value \N , it means that the field is NULL, and any other case is considered a string input|
|Comment symbols|'#' lines as the first character will be filtered out|