read a formated text

How can I read the Information from the following text file?

<content d="2000-11-02" o="21.410" h="21.800" c="21.600" l="21.310" v="21014" bl="" />
<content d="2000-11-03" o="21.610" h="21.680" c="21.420" l="21.380" v="7412" bl="" />
<content d="2000-11-06" o="21.420" h="21.450" c="21.370" l="21.330" v="8795" bl="" />
<content d="2000-11-07" o="21.380" h="21.500" c="21.360" l="21.280" v="9994" bl="" />

What interests me are

"2000-11-06", 21.420, 21.450, 21.370, 21.330, 8795

I tried

textscan(fid, '\t<content d="%4d-%2d-%2d" o="%f" h="%f" c="%f" l="%f" v="%f" bl="" />')

or

textscan(fid, '\t<content d="%q" o="%f" h="%f" c="%f" l="%f" v="%f" bl="" />')

But it doesn't work. Anyone can help?

BTW, converting dates using

dates = cellfun(@(x)datenum(x, 'yyyy-mm-dd'), dates);

is kind of slow. Anyone knows a faster way?

Answer 1

[dpb]

Forget the '\t' in the first format string; textscan uses it as one of the default delimiters, anyway.

I pasted one line in to command window --

 textscan(s, '<content d="%4d-%2d-%2d" o="%f" h="%f" c="%f" l="%f" v="%f" bl="" />')
ans = 
    [2000]    [11]    [2]    [21.4100]    [21.8000]    [21.6000]    [21.3100]    [21014]
>>

[Song]

S = ['<?xml version="1.0" encoding="UTF-8"?> />' sprintf('\n') ...
'<control> />' sprintf('\n') ...
'  <content d="2000-01-04" o="24.980" h="25.780" c="25.570" l="24.750" v="44961" bl="" />' sprintf('\n') ...
'  <content d="2000-01-05" o="25.570" h="25.980" c="25.280" l="25.150" v="52528" bl="" />' sprintf('\n') ...
'  <content d="2000-01-06" o="25.180" h="26.300" c="25.990" l="25.050" v="62297" bl="" />' sprintf('\n') ...
'  <content d="2000-01-07" o="26.300" h="27.500" c="26.900" l="26.120" v="213553" bl="" />' sprintf('\n') ...
'  <content d="2000-01-10" o="27.000" h="27.850" c="27.250" l="26.710" v="165397" bl="" />' sprintf('\n') ...
'  <content d="2000-01-11" o="27.250" h="27.300" c="26.200" l="26.120" v="93908" bl="" />' sprintf('\n') ...
'  <content d="2000-01-12" o="26.000" h="26.000" c="25.120" l="24.800" v="352749" bl="" />' sprintf('\n') ...
'  <content d="2000-01-13" o="25.000" h="25.250" c="24.900" l="24.800" v="79756" bl="" />' sprintf('\n') ...
'  <content d="2000-01-14" o="24.880" h="25.000" c="24.200" l="23.910" v="178619" bl="" />' sprintf('\n') ...
'  <content d="2000-01-17" o="24.080" h="24.440" c="24.400" l="23.750" v="81015" bl="" />' sprintf('\n') ...
'  <content d="2000-01-18" o="24.500" h="24.570" c="24.140" l="23.880" v="76933" bl="" />' sprintf('\n') ...
'  <content d="2000-01-19" o="24.140" h="24.290" c="24.130" l="23.980" v="46584" bl="" />' sprintf('\n') ...
'  <content d="2000-01-20" o="24.120" h="24.650" c="24.440" l="24.100" v="51144" bl="" />' sprintf('\n') ...
'</control>'];

textscan(S, '<content d="%4d-%2d-%2d" o="%f" h="%f" c="%f" l="%f" v="%f" bl="" />')

I tried your code, but still not working ...

Answer 2

[Song]

As an example to test:

S = ['<?xml version="1.0" encoding="UTF-8"?> />' sprintf('\n') ...
'<control> />' sprintf('\n') ...
'  <content d="2000-01-04" o="24.980" h="25.780" c="25.570" l="24.750" v="44961" bl="" />' sprintf('\n') ...
'  <content d="2000-01-05" o="25.570" h="25.980" c="25.280" l="25.150" v="52528" bl="" />' sprintf('\n') ...
'  <content d="2000-01-06" o="25.180" h="26.300" c="25.990" l="25.050" v="62297" bl="" />' sprintf('\n') ...
'  <content d="2000-01-07" o="26.300" h="27.500" c="26.900" l="26.120" v="213553" bl="" />' sprintf('\n') ...
'  <content d="2000-01-10" o="27.000" h="27.850" c="27.250" l="26.710" v="165397" bl="" />' sprintf('\n') ...
'  <content d="2000-01-11" o="27.250" h="27.300" c="26.200" l="26.120" v="93908" bl="" />' sprintf('\n') ...
'  <content d="2000-01-12" o="26.000" h="26.000" c="25.120" l="24.800" v="352749" bl="" />' sprintf('\n') ...
'  <content d="2000-01-13" o="25.000" h="25.250" c="24.900" l="24.800" v="79756" bl="" />' sprintf('\n') ...
'  <content d="2000-01-14" o="24.880" h="25.000" c="24.200" l="23.910" v="178619" bl="" />' sprintf('\n') ...
'  <content d="2000-01-17" o="24.080" h="24.440" c="24.400" l="23.750" v="81015" bl="" />' sprintf('\n') ...
'  <content d="2000-01-18" o="24.500" h="24.570" c="24.140" l="23.880" v="76933" bl="" />' sprintf('\n') ...
'  <content d="2000-01-19" o="24.140" h="24.290" c="24.130" l="23.980" v="46584" bl="" />' sprintf('\n') ...
'  <content d="2000-01-20" o="24.120" h="24.650" c="24.440" l="24.100" v="51144" bl="" />' sprintf('\n') ...
'</control>'];

textscan(S, '<content d="%4d-%2d-%2d" o="%f" h="%f" c="%f" l="%f" v="%f" bl="" />')

Answer 3

[Thorsten]

 i1 = findstr(S, '<content');
 i2 = findstr(S, '/>');
 i2 = i2(find(i2>i1(1), 1, 'first'):end) + 1;

 for i=1:numel(i1)
  R(i,:) = textscan(S(i1(i):i2(i)), '<content d="%4d-%2d-%2d" o="%f" h="%f" c="%f" l="%f" v="%f"  bl="" />')
 end

[Song]

Thanks, I figured out the same way.