How to read large data file with different data types

1 次查看（过去 30 天）

显示更早的评论

Vida 2012-10-30

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/52289-how-to-read-large-data-file-with-different-data-types

I am having difficulty reading a large file 1GB. which looks like this

"-----------------------------------------------

per 8/23/2012 10:07:48 AM 93 122.00000

per 8/23/2012 10:07:48 AM 96 123.00000

Start 8/23/2012 10:07:49 AM 99 1

Order 8/23/2012 10:08:09 AM 99 Start session OK (Please bla and time, the difference between device time, please adjust time): 8/23/2012 10:08:09 AM

per 8/23/2012 10:08:09 AM 99 122.00000

per 8/23/2012 10:08:09 AM 102 122.00000

----------------------------------------------------"

The whole file is generally consistent with this foramt "per 8/23/2012 10:08:09 AM 102 122.00000". But sometimes there are lines that have sentences. I want to capture the whole data and ignore or store those sentences in the 6th column or somewhere else.

I am trying this

format = '%s %s %s %s %s %s' [segarray idx]= textscan(file_id, format, block_size, 'delimiter', '\n');

This stores the whole row

'Start 8/23/2012 10:07:48 AM 0 1'

'per 8/23/2012 10:07:48 AM 15 122.00000'

'per 8/23/2012 10:07:48 AM 33 121.00000'

where I want to have them separate Cell array columns (1 to 6 colums).In this way I can get rid of the long sentence.

when I try this without delimiter [ segarray idx]= textscan(file_id, format, block_size); I get the long sentence in different indeces of cell array. I want to conserve the indexing of data. 1-'per'

2-'per'

3-'Order'

*4-'session'

6-'and'

7-'IPG''s'

8-'over'

9-'time):'*

10-'per'

11-'per' Where the second column doesnt match

1-'8/23/2012'

2-'8/23/2012'

3-'8/23/2012'