Buy Download Tour

This is a collection of frequently asked questions about TextPipe. 

Online Demonstrations

Please click here.

Other help resources

Questions

Answers

Why won't TextPipe run from the command line?

(Try using the Command line wizard in the Tools menu) There are a number of possible answers. 

Firstly, the evaluation edition of TextPipe always displays a dialog when it runs that requires operator intervention. This dialog is not in the registered versions.

Secondly, you may not have specified /G /Q at the end of the command line. This tells TextPipe to run the current filter and exit when it has finished e.g.

"C:\Program Files\DataMystic\TextPipe\textpipe.exe" /f=autorun.fll /g /q

Thirdly, you may have forgotten to correctly quote long filenames (e.g. those with spaces). You must use double quotes around the ENTIRE parameter. e.g.

"C:\Program Files\DataMystic\TextPipe\textpipe.exe" "/f=long filename filter.fll" /g /q

See the notes in TextPipe's help file under the command line topic.

Top

How can I search and replace within <form>...</form> tags?

If you want to search and replace inside <form> elements, use Filters\Restrict\HTML or XML element or attribute, and select Restrict to between tags <FORM> and </FORM>. Check 'Include start and end tags' if you also want to search/replace inside the attributes of the tag as well.

Next, add a new search and replace (Filters\Replace\Find pattern (perl style)) and drag it INSIDE the original. This new search and replace acts only on the text found by the original search and replace, i.e. all the text between the FORM tags

Top

How can I search and replace everything after a particular character?

If you want to search and replace everything after a particular character in a file, use Filters\Replace\Find pattern (perl style), for

(?<=CHAR).*?

Where CHAR is the particular character or string that needs to be found. Set the replacement text to

$0

This essentially performs no replacement at all - because '$0' is shorthand for 'the found text'. We are just using the search and replace to identify a section of text for us.

Next, add a new search and replace, and drag it INSIDE the original. This new search and replace acts only on the text found by the original search and replace, i.e. all the text after the particular character.

Top

How can I search and replace everything before a particular character?

If you want to search and replace everything before a particular character in a file, use Filters\Replace\Find pattern (perl style), for

.*(?=CHAR)

Where CHAR is the particular character or string that needs to be found. Set the replacement text to

$0

This essentially performs no replacement at all - because '$0' is shorthand for 'the found text'. We are just using the search and replace to identify a section of text for us.

Next, add a new search and replace, and drag it INSIDE the original. This new search and replace acts only on the text found by the original search and replace, i.e. all the text before the particular character.

Top

How do I data mine web page content?

In order to effectively data mine content from web pages, you first have to remove all the extraneous information such as color and formatting, extra spaces, graphics, forms, comments, styles, advertising and embedded frames. To perform this step, we link to a predefined TextPipe filter in web site mining\data mine.fll. To use it, in the File Menu, choose Link to Filter, and then select the filter file. This includes the filter without modifying it.

Next we need to simplify the html tags to change "<table border="3" padding="3" etc>" into just "<table>". This will make it much easier to search and replace later on. To do this, we use a filter called web site mining\simplify tags.fll. Again, to use it, in the File Menu, choose Link to Filter, and then select the filter file.

Finally, to convert data from html table format to a CSV (comma-separated value) format that we can easily import into Excel, we use the filter web site mining\data mine html tables.fll. Again, to use it, in the File Menu, choose Link to Filter, and then select the filter file.

Once this is done, just drag and drop the file onto TextPipe's window, set the Output Filter to save the result file somewhere like the Desktop, and then click 'Go'.

It's worth noting that you may need to remove other html tables from headers and footers near your data. This must be done manually, because there is no way the software can determine what is junk data and what is not. To remove a table, in the Special Menu, choose Find and Replace (Find Pattern). A new search and replace filter is added, ensure it has a find type of Pattern (perl). The add text like '<table>.*</table>'. This will find a start and end table tag with anything in-between.

You can use Offline Explorer Pro to download all or part of web sites on a scheduled basis and then feed them into TextPipe Pro automatically.

Top

Why doesn't my query/database driver work?

TextPipe Pro's database filter requires MDAC version 2.1 or later. MDAC is automatically installed with software such as Windows 98 / 2000 and Internet Explorer version 5 or later. If you have Windows 95 you may get a warning indicating that you need DCOM95 to install MDAC, since there is no support for OLE DB in Windows 95. Once you install DCOM95, you can proceed and install MDAC.

How can I improve the performance (like speed up query data retrieval)?
Avoid returning too many fields. Performance suffers as a larger number of fields are returned. For example using "SELECT * FROM TableName" when TableName has 40 fields, and you really need to modify only 2 or 3 fields.

How do I use multiword table / field names (spaces in Table or Field name)?
Enclose multiword names in [ ] brackets:
SELECT [Last Name], [First Name] FROM [Address Book]

How do I use constant fields in a SQL query?
SELECT ''2002'', [First Name], Salary FROM Employees

How should my SELECT statement look for an Excel spreadsheet?
SELECT * FROM [Sheet1$];

Top

I'm trying to automate Outlook but keep getting nag screens from it

Try using the Outlook Redemption Object Model, which provides access to blocked properties and methods:

https://www.dimastr.com/redemption/

Top

Dragging and dropping a file onto a short cut doesn't work

Unfortunately with a short cut, the dragged filename is always appended to the existing command line (hence it comes after the /g /q when it must be before). The best way around this is to use a batch file like this:

In T.bat:

@echo off
"C:\Program Files\TextPipe\textpipe.exe" "/f=C:\Documents and Settings\Simon Carter\My Documents\TextPipe\My Filters/ipMon.fll" %1 %2 %3 %4 %5 %6 /g /q

Then add a short cut on the desktop for t.bat, and drag and drop files onto it. You might need to add double quotes around the %1 if you are using long filenames.

Top

Remove blank lines at the end of the file

Use a Perl pattern to search for:

(\r\n)*\z

And replace with nothing.

Top

How to extract just lines 1,830,500 to 1,830,600 from a 1.7 GB file?

Use the Remove Line Range filter to first remove the lines from 1 to 1,830,499, and then follow it with another to remove lines 100 and onwards.

Top

How to extract SQL Server/Oracle bigint, decimal & numeric data

We cannot SELECT data from bigint, decimal & numeric fields. However, you can work around this with a cast e.g. select cast(bigint as varchar) from tablename. Also note that data from Binary fields often contain ASCII 0's - use a Map to replace these nulls with space or similar.

Top

Buy Download Tour