Fine-Tuning Data Access
Resource location
The URI should be used to specify a CSV resource location. Set the URI property to specify one of the following sources:- An empty value automatically assigns the URI to a reference to the current directory, "./". The explicit path to the CSV folders depends on the environment of the running application.
- A path to a folder.
- A path to a .zip, .tar. or .gz archive file.
- Include the file, not just the containing directory. For example: C:\Users\Public\Documents\CSVdata.zip
- A path to a file or stream - in this case you can query the file by executing SELECT * FROM streamedtable.
Modeling tables
Set the following properties to control how the component models CSV as tables:
- IncludeColumnHeaders: Set this to get column names from the first line of the specified files (the default). Otherwise, the column names are the column numbers.
- FMT: Set this to the format to be used to parse the text files: CsvDelimited (the default) or TabDelimited.
- IncludeFiles: Set this to a comma-separated list of file extensions to include into the set of files modeled as tables. (By default, .txt, .tab, and .csv files are modeled.)
- Specify files by their file extensions in all-caps, without the '.'. For example: "CSV,TXT,TAB".
- Archive files are supported (ZIP, TAR, and GZ) and are modeled as if they were folders.
- RowScanDepth: Set this to automatically determine data types by scanning rows up to the specified depth.