Ferrite^TM offers businesses and users the quickest and easiest way to parse text files for tabular data, process the data and export to HTML or XML format.

• Parse text for tabular data and export as CSV.
• Filter unwanted lines/rows using powerful pattern matching
• Add columns by computing the value using the other columns
• Remove numbered columns or columns matching a pattern
• Search/replace within columns using regular expressions.
• One-click filters to convert columns to uppercase, lowercase or capitalization (uppercase first letter).
• Filter rows using criteria depending upon column data.
• Extract a subset of columns: numbered columns or use a regular expression pattern to select columns to extract.
• Convert to XML/HTML.
• Save editing rules and re-apply with a single click

If you have tabular data in a format other than CSV, you can parse the file and export to CSV. Similar to Microsoft Excel, Ferrite supports parsing files with a single-character field delimiter such as star(*), semi-colon(;), etc. However, if the delimiter is a string or a regular expression pattern, you are out of luck with Excel. In this case, you can use Ferrite to parse the text and export the data as CSV.

In addition to using a delimiter pattern, you can also extract tabular data from files with fixed-width columns by specifying the field positions.

Files containing useful data in tabular form usually include explanatory headers and footers, page breaks, etc which interfere with parsing, and require cumbersome manual editing. For this reason, applications such as Microsoft Excel offer options to ignore a certain number of header lines, and perhaps footers.

Ferrite offers users a lot more power and flexibility for cleaning up such data. Thus, you can use Ferrite as a "one-size-fits-all" solution for cleaning up and parsing.

• Remove unwanted headers and footers. For example, 4 header lines, and 6 footer lines.

  
  

  

• Remove lines/records matching a pattern. For example, blank lines, lines containing page numbers or page breaks, etc.

  

• Remove records that do not "fit in". For example, remove lines that do not have a specified number of fields.

  

You can add a column containing a constant string or number. However, more powerful and useful is to specify a Javascript expression to compute the column value using data from the other columns or the complete record.

For example, the following screenshot illustrates adding a column by computing the MD5 Digest from a field containing a customer's email address, perhaps to generate a coupon code for the customer.

Need to remove columns 4, 5 and 6 because they contain sensitive customer data? Use the numbered column removal option.

Or perhaps, delete columns containing URLs? Use the pattern-based column removal option.

How about removing columns whose value is within a certain range? All these operations and more are trivial when processing tabular data using Ferrite.

Need to perform search and replace within specific columns? That can be done in an application such as Microsoft Excel too. Ferrite, on the other hand, supports the use of pattern search and replace using regular expressions. If that does not fit the need, search for a pattern and specify a Javascript function to be invoked to compute the replacement!

While Javascript can be used to perform almost any kind of field data manipulation, it may not be everyone's cup of tea. With this in mind, we have included many one-click extensions for common tasks most of which require no Javascript programming. For example, need to upcase or downcase data in one or more columns? How about capitalizing (first letter uppercase) it?

Rows can be included or excluded in the output depending on column values. This functionality is similar to using SQL to filter rows on a database. Ferrite offers a similar functionality for any kind of tabular data including data loaded from CSV files. In addition, the language used for filtering is Javascript, which in our opinion is easier to use than SQL.

Extraction of a subset of columns is a frequent requirement when working with tabular data. Ferrite supports options to extract columns based on

• Column index, for example columns numbered 2, 3, 5 and 6
• Regular expression pattern. For example include columns matching an email address.
• Javascript expression. For example, columns with the values in a certain range: greater than 0 and less than 19.99.

Ferrite includes an easy-to-use extension to convert tabular data to HTML or XML. This extension only requires the names of the elements to wrap the row and column data in. For example, generate a HTML table by specifying table, tr and td for the parameters. Of course, once the Javascript code is generated, it can be customized as required.

Advanced Usage

When the built-in XML serializer does not fit the need, the Javascript for converting to XML can be customized; for example, to output some fields as attributes and some as columns. The following screenshot shows the code generated by the built-in XML serializer which can be further customized based on need.

Need to perform the same set of editing operations on one or more files? Ferrite is built for this usage pattern. Every editing operation is a filter and saved (as a workflow) along with whatever parameters are required, and can be re-applied with a single click.