Summary

This operator allows to import data from a text file where values are separated by comma (comma separated values, CSV) or another symbol.

It can only be applied to an empty data node.

The text file has to be available in the TIS File on the TIS Server. Therefore, a file from a local storage has to be uploaded to the TIS Server. 

Configuration

Settings

Name

Value

Opt.

Description

Example

Select file

System.String

-

Enter a Path or Browse through the TIS-File system and choose a file


Text encoding

System.String

  • AUTODETECT
  • ASCII
  • WINDOWS1252
  • UTF8
  • UTF7
  • UTF32
  • UNICODE

opt.

Select text encoding type


First row is a header row

System.Boolean

-

Transform the first row in a header row


Delete trailing blanks

System.Boolean

-

Deletes trailing blanks of each entry


Column separator

System.String

  • ;
  • ,
  • -
  • .
  • |
  • #
  • *
  • '
  • TAB
  • SPACE

opt.

Select a column separator for the selected file


Text delimiter

System.String

  • BLANK
  • ""
  • '

opt.

Select a text delimeter for the selected file 


Decimal symbol

System.String

  • ,
  • .

-

Select the decimal symbol separator


Behavior if parsing fails

System.String

  • Set to NULL
  • Use default value
  • Error

-

Behavior for cells where data cannot be parsed or converted into the desired data type


System.String

  • SETNULCOLUMNMAPPINGL
  • Use default value
  • Error

-

Behavior for cells where no or blank data is indicated


Column mapping

System.String

opt.

After selecting a file, the column and rows will be mapped and displayed


Data node for error output

System.Int32

opt.

Data node in which possible error messages are saved.


Want to learn more?

Imports data from a text file.

Screenshot



Examples

Example: Import text file of the type "Date/Time/Time"

Situation

The text file Import_1.txt is to be imported into a data node.

Step 1

Go to Editor Homepage and choose the TIS File tab; upload file.

Step 2

Create an empty data node. To this node, add the operation "Import Text or CSV file 3.0".

Click on the folder symbol next to "Name" and select the import file.

Alternatively, files can be defined via parameters under "Path (manual):"


Then choose the following settings:

Then click "Execute" to let TIS suggest the columns based on your suggestions. You will need to correct the suggested column "Bis" to Semantics: "Time (end)".

  • You get into the editing mode by clicking the icon
  • You need to confirm changes by clicking the icon
  • Columns that are not supposed to be imported can be deleted with the icon
  • You can change the order of columns (how they will be displayed in the result table) with the icons und

As of version 5.8 a specific error node can optionally be defined. That gives the possibility that error nodes can be displayed in TIS Board.

Create an empty data node and choose the node as data node in the import operation.

Result


Further steps

The resulting table should be converted to date+time/date+time to allow operations such as Scaling 8.0. 

TIS Project

Confluence Op Import Text or CSV file.gzip


Example 2: Import file of type date+time/date+time


Situation

Read in the text file Import_2a.txt

First, save the file on your PC and then import it into TIS (see TIS:Datenmanagement (link to old wiki)).

Create an empty node in the TIS project and add operation "Import text or csv-file".

Operation setting

Click on the folder symbol next to "Name" and select the import file.

Alternatively, files can be defined via parameters under "Path (manual):"

Then, chose the following settings:

Click "Execute" to let TIS suggest the columns based on your suggestions. You will need to correct the suggested column "Bis" to Semantics: "Date+Time (end)".

Please note: Click on the button "Edit" to change settings. Click on the button "Apply" to confirm changes.

As of version 5.8 a specific error node can optionally be defined. That gives the possibility that error nodes can be displayed in [TIS]Board.

Define an empty data node and choose the node as data node in the import operation.

Result

TIS Project

Confluence Op Import Text or CSV file.gzip

Example 3: Import file of type date+time/date+time with overlap to the next day


Situation

Read in the text file

Datum;Von;Bis
01.01.2007;06:00;14:30
02.01.2007;06:00;13:00
03.01.2007;14:00;23:00
04.01.2007;14:00;22:00
05.01.2007;22:00;06:00
06.01.2007;22:00;06:00
09.01.2007;06:00;14:00
10.01.2007;06:00;15:00
11.01.2007;06:00;25:00
12.01.2007;06:00;05:00+
13.01.2007;06:00;06:00+
14.01.2007;22:00;06*00

Please note: Times overlapping to the next day are indicated by different formats here: 25:00, 5:00+, and 6*00. This example will demonstrate how to read in each of these formats.

Operation setting

One solution for this is to apply the Formula operator (row-by-row) (5.0) to create new columns which calculate the correct date and time:

Formulas are:

1. NextDay [Ja/Nein]: #Bis#.Substring(#Bis#.Length-1 > 0 ? #Bis#.Length - 1 : 0) == "+" || #Bis#.Contains("*")
2. Bis2 [Text]: #Bis#.Replace("+","").Replace("*",":")
3. h [ganze Zahl]: System.Convert.ToInt32(#Bis2#.Split(':')[0])
4. m [ganze Zahl]: System.Convert.ToInt32(#Bis2#.Split(':')[1])
5. Bis3 [DateTime]: #Datum#.AddDays(#Nextday# ? 1 : 0).AddHours(#h#).AddMinutes(#m#)
6. Von2 [DateTime]: #Datum#.Add(#Von#.TimeOfDay)

Result

TIS Project

DateTimeImport.gzip


Troubleshooting

Problem

Frequent Cause

Solutions

Everything is suggested as Identifier although I wanted to read in Text.


  • This can be changed manually, but it is usually not necessary.
  • All operations which work with Text, also work with Identifier. (Identifier = special case of text)
  • If columns remain after conversion, and this is not wanted, simply delete those columns.

How can I make changes to the "Mapping" table?


  • You get into the editing mode by clicking the button
  • You need to confirm changes by clicking the button
  • Columns that are not supposed to be imported can be deleted with the button
  • You can change the order of columns (how they will be displayed in the result table) with the buttons and

"Execute" is suggesting only 1 column with all column titles in "Mapping"

Commong error: Wrong column separator has been chosen (e.g., semicolon instead of tabulator)

Check column separator, change it and click "execute"

The operation does not read in the decimal points (or not in every column)

  • Decimal symbol has not been selected correctly.
  • After clicking "Execute" the suggested columns have been applied without further checking. However, TIS interprets the values in the import table by reading the values in the first row. If the first row contains numbers without decimal numbers, TIS suggests Integer as data type. 
  • Check if the decimal symbol has been chosen correctly.
  • Check the data type for each column. For decimal numbers you need to select "Double" as data type.


24:00 is not imported correctly

24:00 is no allowed format, because in most software systems it means that it is already the next day.

1. Trick: Replacement of illegal values in combination with formula operator:

  • State in the import operation that illegal values should be replaced by 23:59:59:
  • Then import, with replacement of illegal values.
  • Convert the table to a date+time/date+time format with the Operation Convert "From-Date/From-Time/To-Time"
  • Then, apply the Formula operator (row-by-row) (5.0) and change the corrected date (attention: don't forget to choose Date as the data type) with the following formula: 

If the column that needs to be changed is named "Datum", then you can copy this formula:

#Datum#.AddSeconds(#Datum#.Second == 59 && #Datum#.Minute ==59 && #Datum#.Hour==23 ? 1 : 0)

A time overlapping to the next day is included (e.g., 25:00)

25:00 is no allowed format, because in most software systems it means that it is already the next day.

Possible solutions please see Example 3 above

Another trick: Use data type "Text" for time

Resulting in:



Related topics