Bases: object
The basis-class for all type of sources provides some general functions. And evaluate the ‘data_type’ data_type can be “int”, “float”, hex” or “fooBar” in case you don’t know the type of your data.
In diaGrabber every generated basis-dimension will added to the basis. Use this method to define an alternative list of basis-dimensions.
In diaGrabber every generated merge-dimension will added to the basis. Use this method to define an alternative list of merge-dimensions.
Readout values from a plain-text saved in file. The format has to be similar to the following example:
value00 value01 value02
value10 value11 value12
...
Where the values of all dimensions are written in one line, seperated by one or more signs like spaces ‘ ‘ or tab ‘ ‘.
For a list of all possible Keyword-arguments have a look at setArgs()
includes diaGrabber.source._dimension.basisDimension to the source
includes diaGrabber.source._dimension.mergeDimension to the source
Required kwargs (“keyword arguments”) are:
Keyword | Type | Example | Description |
---|---|---|---|
file | string | ‘myFile.txt’ | the name of your plainText-file |
dimSeperator | string | ‘ ‘ (space) | the sign, that seperates the values in one line e.g. ‘ ‘ for a tab |
Optional kwargs (“keyword arguments”) are:
Keyword | Type | Default | Description |
---|---|---|---|
dataType | string | “unknown” | for all possible values are defined see diaGrabber.source._source.source._evalDataType() |
nFileLines | int | 0 | if you know the number of lines in your file you can avoid the (time-consuming) counting of it through diaGrabber |
readoutEveryNLine | int | 1 | if you don’t need every line of your file you can spare time by readout only every n line of it |
folder | string | “” | the name of the folder of the file |
buffer | int | 10e6 | max. size(byte) loaded into RAM insert ‘0’ to read the whole file |
silentReadout | bool | False | set True if you dont want to see a statusBar or similar while readOut |
Readout values from a process which produces unlimited or limited output to RAM. The process output after the following scheme:
time0 --> "value00 value01 value02"
time1 --> "value10 value11 value12"
...
Where the values of all dimensions are written in one line, seperated by one or more signs like spaces ‘ ‘ or tab ‘ ‘.
includes diaGrabber.source._dimension.basisDimension to the source
includes diaGrabber.source._dimension.mergeDimension to the source
Required kwargs (“keyword arguments”) are:
Keyword | Type | Example | Description |
---|---|---|---|
command | string | ‘python’ OR ‘./’ | The command to start the stream |
file | string | ‘test.py’ | The name of the stream |
dimSeparator | string | space: ‘ ‘ OR TAB: ‘ ‘ | One or more signs which seperates the values in every line |
stopVia | string | ‘done’ | If the stream is limited this string let’s diaGrabber when the stream finishes |
Optional kwargs (“keyword arguments”) are:
Keyword | Type | Default | Description |
---|---|---|---|
folder | string | “” | Name of the parent folder of the stream |
readoutEveryNLine | string/int | 1 | if you don’t need every line of your file you can spare time by readout only every n line of it. Most likely the stream generates faster new output than diaGrabber can process, In this case this value has to be greater than 1, Type ‘calc’ to let diaGrabber decide this value corresponding to the development of the amount of unprocessed output. |
showCalcNLine | bool | False | if readoutEveryNLine == ‘calc’ Setting this value to True print out every new calculated value of readoutEveryNLine |
infoEveryNLine | int | 1000 | Prints every n reaout-lines a status. |
dataType | string | “unknown” | for all possible values are defined see diaGrabber.source._source.source._evalDataType() |
keyToEndProcess | string | “” | This entry will finish the stream-output |
runInShell | bool | False | Decide whether a process has to be called out of a shell. Set this value to True if the stream doesn’t start. |
silentReadout | bool | False | set True if you dont want to see a statusBar or similar while readOut |
Allow to read values from a libre/open-office-calc document For a list of all possible Keyword-arguments have a look at setArgs()
Add a new basis-dimension to the source. See diaGrabber.source._dimension.basisDimension for further explanation and options. Replaces keyword ‘index’ with ‘cellRange’:
Parameters: | cellRange (string) – [“START:STOP”] range of the cells in the ods-file e.g. “C2:C1000” or “A1:E10” |
---|
Add a new merge-dimension to the source. See diaGrabber.source._dimension.mergeDimension for further explanation and options. Replaces keyword ‘index’ with ‘cellRange’:
Parameters: | cellRange (string) – “START:STOP” range of the cells in the ods-file e.g. “C2:C1000” or “A1:E10” |
---|
Required kwargs (“keyword arguments”) are:
Keyword | Type | Example | Description |
---|---|---|---|
file | string | “myFile.ods” | Name of the ODS-File |
sheet | string/int | “sheet1” OR 1 | Sheet-name or number in the ODS-file |
Optional kwargs (“keyword arguments”) are:
Keyword | Type | Default | Description |
---|---|---|---|
folder | string | “” | Name of the parent-folder of the file |
readoutEveryNLine | int | 1 | read every n line from the file |
dataType | string | “unknown” | for all possible values are defined see diaGrabber.source._source.source._evalDataType() |
ignoreEmptyCells | bool | True | Set to False to exit, when cells inthe cellRange of thedimensions are empty |
silentReadout | bool | False | set True if you dont want to see a statusBar or similar while readOut |