The Parsely configuration editor will be displayed after selecting a file to upload, and choosing to create a new configuration. The configuration allows Riffyn to extract the subsections of the data file that contain the information you want to upload. These parts are called "Data Blocks." Data that has been arranged into multiple Data Blocks across one or more worksheets can be extracted. A Parsely configuration saves all the instructions provided on the locations of the Data Blocks and how to parse them.
Usually a single configuration will be sufficient to upload the data, but multiple configurations can be created if needed and executed together as a Collection.
Parsely can recognize 3 kinds of data organizations: a table, a matrix, and key-value.
1) Table Data Block
A Table Data Block (see screen shot above) contains both a Data Region (light blue in the data viewer) and a Header Region (dark blue). If the data is organized in columns, the table is said to have a Standard orientation. If the data is organized in rows, the table is said to have a Transposed orientation. Although a Parsely configuration assumes the header region of a table has the same dimensions from file to file, you may allow the table length to vary by adding/removing rows of data (if Standard) or allow the table width to vary by adding/removing columns of data (if Transposed).
By default Parsely assumes that imported data tables contain at least one row or column with headers. It is also possible to parse data tables which do not contain explicit headers in the file by selecting the "My table has no headers" option.
2) Matrix Data Block
A Matrix Data Block (see screen shot below) illustrates an example of data that was written to a file as a Matrix Data Block. Matrix data blocks can also be used to upload plate-formatted data as described in another article here.
A Matrix data block has up to 5 data regions. When you first create a Matrix data block Parsely will highlight a Values Region (gray), a Top Region (light green) and a Left Region (light blue).
The dimensions of the Matrix Values Region can also be modified from the default selected region. This can allow for headers to be defined using multiple rows (top region), or multiple columns (left region) to capture more complex data sets such as time series data. Each extra row or column header will be assigned as a distinct data field. The Matrix Values Region can be further modified to reveal more possible regions on the bottom and right hand side.
If there is data at the corner of the Matrix Data Block where the Top and Left Regions meet, those values can be used as File Data Field Names for either the Top or Left Regions when you check the box marked Includes Field Headers. Since there is no Header Region for a Matrix data type, some or all of its field names are implicitly created. You can rename them by editing the default field names created by Parsely.
3) Key:Value Data Block
A Key:Value data block has just a single region:
Every cell within this region is assumed to have two pieces of information separated by the same Delimiter (e.g., ":"). To the left of the delimiter is a Key that is interpreted by Parsely as a field data name. To the right of the delimiter is its corresponding Value. Key:Value data is therefore intrinsically single-valued. Parsely provides options to specify the delimiter used within each Key:Value block and to treat consecutive instances of the same delimiter as one (e.g., treat "::" as ":").
Steps For Creating Data Blocks In Parsely
1) Navigate to the worksheet containing the data of interest. Parsely will display multiple worksheets if present.
2) Click the "Create Data Blocks" button at the top of the screen. By default Parsely will assume the new data block is a Table format and will attempt to determine its orientation and dimensions automatically.
3) Name the data block if desired
4) Change the Data Block Type from Table to Matrix or Key:Value, if required.
5) Update the data region that has been selected in the worksheet, if needed. There are a few ways to do this:
- Click and drag the cursor over the region to be selected.
- Click on a cell in one corner of the data region, then hold shift while pressing arrow keys until the required region is selected.
- Click on a cell in one corner of the data region, then hold shift while clicking on the opposite corner of the region you wish to select
Note: the entire region does not need to be selected; it is possible to take advantage of the check boxes for variable size data.
6) Save the updated data region before proceeding
7) For Matrix data blocks, update the values region and save the changes, if needed
8) For Key:Value data blocks, edit the delimiter, if required
9) Rename the data field names, if required. Renaming data field names is recommended to disambiguate any duplicated data field names across the data blocks in your configuration. The file data field names that you specify here will be used throughout Parsely's Run Rules, Resource Rules and Property Mapping sections.
10) Update the Data Types of your data fields if needed to upload your data successfully. There are two common situations where modifying data types may be necessary:
- Changing type from "Number" to "Character", if attempting to write numeric data to character data fields
- If attempting to write date/time data, check that Riffyn has identified the correct date/time format to upload your data successfully.