TextPipe Pro 7.8 Final Beta available - release candidate

[DataMystic Support]

The latest TextPipe Pro Beta can be downloaded from:
http://www.datamystic.com/textpipeprobeta.exe

and from the pre-release link in the secure area.
http://www.datamystic.com/secure

What's New in TextPipe v7.8
===========================

*
New option to show special characters such as spaces, tabs and line feeds.
*
New option - triple-clicking a line selects the whole line.
*
New filter to convert Filters\Convert\UTF16-LE to Escaped Unicode (ASCII +
UCN \U)
*
The file list can now contain http://, ftp:// and gopher:// urls to download
directly.
*
The user is now given the option to manually start Word or Excel if they are
not running when TextPipe needs them (with the Convert Word/Excel Document to
Text filters).
*
Pattern matches can now name captured variables for use in the replacement
text e.g. searching for (?P<dollars>\d+) would then allow the macro @dollars
to be used in the replacement, and also as a global anywhere else in the
filter list.
*
Far more complex patterns can be processed, with support for up to 99
captured variables, from $1-$99.
*
New 'File\Send to...' function for filter lists.
*
New Fixed Width to Delimited Output wizard - makes converting fixed width
files to CSV, Tab simple.
*
Mainframe Copybook Wizard now allows underscores (_) and single quotes (') in
field names in 'XXX_XXXX 4 AN 1-4'-type copybooks.
Wizard now removes the content of fields called FILLER* .
Wizard now copes with repeated single-fields, as well as GROUPED OCCURS.
EBCDIC packed decimal fields can now have either a leading sign, trailing
sign or no sign.
New option to trim leading spaces and zeroes from fields.
New option to only expand numeric fields - Alpha fields are left unchanged.
This is useful when the target application can accept EBCDIC data but can't
handle packed decimals.
New option to allow COMP-3 to be unsigned.
Wizard can now cope with removing line numbers and commented copybook lines
(those with a '* ' in column 7).
Wizard can now cope with field definitions spread over multiple lines.
Wizard now remembers settings betting uses in a single session.
Expand Binary Number to EBCDIC filter now outputs a fixed width result.
Expand Unsigned EBCDIC filters now output a space for invalid digits A-F.
*
EasyPatterns now produce optimized perl regex output. Added EasyPattern
EBCDIC('') for matching EBCDIC values. Fixed parsing error with ASCII()
function. Now handles back references to groups 1-99. Fixed DayNumber
insisting on leading zero. EBCDIC() and ASCII() functions can now be used in
conjunction with not ... as character classes.
*
Hex displays now all use '.' to replace space.
Query replace window now shows hex offsets for Hex Dump display.
*
Convert Word Document to Text/Convert Excel Spreadsheets to Text now support
Unicode output as an option. Follow this with Convert UTF-16 to UTF-8 to
process these files using normal filters.
*
New File Shred filter to completely overwrite file contents. Run repeatedly
for more shredding.
*
Unicode property tables for pattern matching updated to Unicode 4.1.0, and
matching engine updated to PCRE 6.4.
*
Split filter now can split at a pattern.
New Split Filter option to avoid overloading directories. When more than
10000 files are created, TextPipe starts a new folder.
Fixed bug where \\ could appear in split output filename.
*
Filter list tree now auto-scrolls.
Filter list tree now allows mass enable/disable of filters with select and
right-click.
*
New Filters to Add File Headers/Footers. These filters do not output anything
when a restriction changes, so they only output once per file.
*
Now can drag and drop filters between filter lists in different windows.
*
Filters Add Header/Footer/Left Margin/Right Margin now always check for
@macros embedded in the text, regardless of whether the text comes from the
memo or an external file.
*
Fixed bug with search/replaces where an internal conversion from ansi to
Unicode and back could result in high-ansi characters not being mapped
correctly back to their original characters. In particular, the ansi (tm)
symbol mapped back to TWO characters.