How to use textscan to read data specific out of an ASCII-File

Hey Guys,

I need your help. I want to export the following file content from an ASCII file. However, here I am encountering difficulties so far due to the formatting of the text. I have tried different approaches. But so far in vain. Thus I turn to you.

My file has the following structure.

0.001465    SV: 4 0 1 ::AudioData::_audiobufferDecoded = D50 1.43516 1.76174 0.823262 ... 80 Samples. \n
0.001465    SV: 4 0 1 ::AudioData::_audiobufferDecoded = D50 1.43516 1.76174 0.823262 ...  80 Samples. \n ... and so on

I would like to read out exclusively the data of the audiobuffer with textscan (starts after 'D50 ' for every line).So far without success. Could you help me ?

Leo

[Stephen23]

@Leonard Haensel: please upload a sample data file by clicking the paperclip button.

[Leonard Haensel]

Sorry @Stephen23 that probably makes it easier to understand.

Answer 1

[Walter Roberson]

If each line has the same format and you are positioned after any header lines, then start the format with 8 occurrences of '%*s' to skip the first 8 columns (including the = )

If those lines are mixed in with other lines and you need to ignore the other lines then I would recommend against textscan for that kind of situation, and would suggest using regexp instead.

[Leonard Haensel]

Thank you for your quick reply mr Roberson. If I modify the command according to your specification. Is skipped within the data stream of the audio data also the following 8 values. However, I always want to skip only the first 8 entries at the beginning of each line. In the bufferstream then every single value should be taken. Thank you for your help.

The values written in bold are recorded accordingly, 8 subsequent values are skipped.

::AudioData::_audiobufferDecoded = D50 1.43516 1.76174 0.823262 0 0.536776 1.12253 1.43516 0 0.823262 1.43516

my command looks like this

C = textscan(fileID,'%*s %*s %*s %*s %*s %*s %*s %*s %f','HeaderLines',5);

[Walter Roberson]

you should be using as many %f as there are numeric fields on the line. repmat('%f', 1,80)

[Leonard Haensel]

It was so simple. Thank you for your patience and help