Screen shots showing how you might use ParseRat
(The images may be slow to load. This is necessary to preserve quality in the full-size images.)
|
Overview |
 Click images to enlarge. |
ParseRat works by:
 Selecting or finding fields, records or data blocks in the input file or on the system clipboard.
Parsing, processing, combining or otherwise obtaining information from those fields to create output fields.
Formatting the output fields to your specifications.
Exporting those fields to an output file in a format which you select or to the system clipboard for pasting into another program (e.g. into an Excel spreadsheet). |
| Finding input fields | |
|
If the input file is record-oriented, the user may choose to treat it as fixed format or as delimited. When ParseRat recognises common delimiters it will select them by default, but the user may over-ride with his/her choice of record format and may select any delimiter character. For example, if the input data is being pasted from the system clipboard after capture from another matrix-oriented program like Excel the data will be tab-delimited. ParseRat's initial assumption will recognise this and separate the input fields accordingly. |
|
If the input file is not record-oriented (e.g. it may be a redirected print file from another process or some other type of multi-line data blocks), ParseRat permits data to be extracted from the section or page headers, footers and single or multiple data blocks on the page. An example would be sheets of labels or of catalogue entries. The data may be in fixed locations on the page or it may be identified by its position in relation to specifed "tags" (useful for extracting data from emails). |
|
Emails, web pages and XML files are examples of tag-oriented data. In the case of XML files, ParseRat will make an automatic initial selection of tags which may be edited by the user if the ones chosen are not appropriate. In the case of other "tagged" input formats such as emails or HTML tables in pages saved from websites or captured by users on the system clipboard, the user identifies the tags of interest. |
|
Many files consist of binary data, with no delimiters. Frequently such files consist of an initial header block, usually giving details of the structure of the remainder of the file. Following the header (if any) are records of a fixed length but with no delimiters. These records may consist of text-only material or binary numbers or a combination of both. Even if you do not know the formal structure of the file format, in many cases you can determine the structure of a given file by inspecting it in ParseRat (you can usually see how "columns" line up, even if you do not have the file definition). You may then export its data in a manner similar to that for fixed format files. ParseRat is able to extract data from most structured binary files, correctly handling all binary data representations including zoned and packed decimal in both Intel and Motorola conventions (i.e. both "endianisms") ASCII and EBCDIC. |
| Parsing and processing the data from the fields | |
|
Optionally, input data blocks may be included in the output only if a user-selected field equals or contains a user-specified value. |
|
The data in the input fields may be combined in various sequences, further split into components or processed in other ways. If there is not a predefined process which suits the user's needs, a "plugin" program may be written in virtually any computer language to further process the data. In the example at the left, the predefined combination process assembles fields from the input data in three different ways. Additionally a plugin program called "reverse.exe" was written to take the value of "Combination 1", reverse it and return it as "Combination 4". This gives great flexibility when data needs to be generated from the contents of the input fields in ways determined only by the user. For example, it could be used to create passwords or registration codes by "hashing" values from selected input fields. |
|
Fields (or groups of fields combined) containing name data may be parsed in sophisticated manners. Titles (if any), first name, middle names, last names and suffices (if any) are parsed out into individual elements. Different (user specified) input formats may be re-assembled into other formats for output. e.g. a "Phone Book" pattern name may be converted to a "Mailing List" pattern format. Gender is automatically determined for 10,000 pre-programmed first names and users may add up to another 10,000 names. Gender-correct titles and salutations are generated suitable for use in letters, with automatic recognition of "couples" if male and female names are both present in the name. Words and word sequences to be omitted in the generation of name data may be specified (e.g. "Fax Line", "Teen Phone", etc). Multi word surnames are recognised with most prefix words pre-programmed (e.g. Van, De, Di, etc) and provision for the user to add more. |
|
Street address data is intelligently parsed into its essential components and may be re-assembled into a user specified standard address sequence to suit a particular postal authority. Standard abbreviations are generated for address elements accordng to rules which may be edited by the user. The generation of Odd/Even codes and the ability for ParseRat to divide a numeric field by 100 makes easy the generation of canvasser lists by hundred-block and side of street. Similar parsing is performed for City, State/Province and postal code. Postal code patterns for many countries are automatically recognised if present in a field. |
|
Date and time information is handled for all formats. Many formats are automatically recognised, including dBase dates, Julian Date (days since 4713 BC) and Calendar Date (Seconds since January 1st, 1970 - used in many computer applications, including Internet bookmark data). Dates may be exported in every format including numeric, full alphabetic, abbreviated alpha, US, UK and metric sequences. |
|
Measurement Unit Conversion is provided for fields selected or assembled using the Combo panel. Over 2,300 built-in conversions are provided or the user may specify any linear conversion (i.e. of the Y=A + BX pattern). In the example shown at left, the Combo1 field has been selected from an input field containing the number 63. The conversion "Hectares to Acres" has been selected and the result 155.676 is placed in the Convert1 parser generated field for placement in the output record. |
| Other parsing and processing options are available and are best evaluated by downloading the free thirty day evaluation version. This may be accomplished by clicking the button at the left. |
| Assembling and exporting data records in a variety of formats. | |
|
The user may choose from many choices of output format, for the file as a whole as well as for individual fields. The user may optionally insert header and trailer blocks ahead of and behind the database records (combining this with fixed text in some fields allows the automatic generation of web pages with tables of catalog items, links, etc).
To assist in mailing lists for "multiple up" forms, there is provision to "interleave" records in multiple sequences within the file. |
Rated
at TUCOWS
