- Change:
dataAddress
as uniform way to specify where to read data from / write data to. Remove now obsoletesheetName
,startColumn
,endColumn
,skipFirstRows
. - Feature: Append to existing files.
Only the range implicitly or explicitly specified via
dataAddress
will be overwritten. - Change: Remove
preHeaderLines
. This is superseded by writing into a file that contains all the required pre-headers and footers.
- Bugfix: Shade commons-compress. Fixes nightscape#93
- Change: Update to Apache POI 4.0.0
- Feature: Add
workbookPassword
option for reading encrypted excel spreadsheets
- Bugfix: Handle multi-line column headers properly
- Bugfix: Improve handling of columns where the first row doesn't contain any data
- Feature: Add
preHeader
option for writing rows/cells before the column headers - Feature: Add
skipFirstRows
option for skipping rows before the column headers
- Feature: Add handling of floats
- Feature: Improve performance by upgrading dependencies and instantiating date formatting lazily
- Bugfix: Infer schema for FORMULA cells
- Bugfix: Make sure files are closed after reading
- Feature: Added support for FORMULA cells when typing to String or Numeric types
- Bugfix: Shade Jackson dependency to avoid version conflicts
- Feature:
excerptSize
option determines how many rows to read when inferring schema
- Feature: Read file using streaming by specifying
maxRowsInMemory
- Feature: Serialize BigDecimals
- Feature: Detect date formatted columns when inferring schema #28
- Internal: Use scalafmt for code formatting
- Feature: Reading and writing timestamps
- Bugfix:
null
values in DataFrames are serialized as empty Strings. Thanks to @slehan
- Feature: Writing Excel files
- Change: Providing the path to the file is now either done via
.option("path", thePath)
or.load(thePath)
- Change: Some previously required parameters are now optional and have a default
- Feature: Respecting user-provided schema
- Bugfix: Several fixes to parsing