Parsing the output

ttsdinesh · 12-17-2009, 05:13 AM

I want to display particular line of the output. For eg, i executed "cat /proc/cpuinfo". The output is:

Quote:

[root@Ubuntu]# cat /proc/cpuinfo
processor : 0
vendor_id : GenuineIntel
cpu family : 6
model : 15
model name : Intel(R) Core(TM)2 Duo CPU T5270 @ 1.40GHz
stepping : 13
cpu MHz : 800.000
cache size : 2048 KB
physical id : 0
siblings : 2
core id : 0
cpu cores : 2
apicid : 0
initial apicid : 0
fdiv_bug : no
hlt_bug : no
f00f_bug : no
coma_bug : no
fpu : yes
fpu_exception : yes
cpuid level : 10
wp : yes
flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe nx lm constant_tsc arch_perfmon pebs bts pni dtes64 monitor ds_cpl est tm2 ssse3 cx16 xtpr pdcm lahf_lm ida
bogomips : 2791.74
clflush size : 64

I want to display CPU MHz alone. How to parse the Output? Here my CPU MHz is 800.00. I want to assign the value to a variable, say for eg. N, and write the value of N to a file named "cpu.txt". How to perform it? Thanks in advance.

HasC · 12-17-2009, 05:26 AM

use some shell scripting...

As an example:

$ grep "cpu MHz" /proc/cpuinfo | awk '{print $4}' | cut -d \. -f 1

will give you the "800".

And, perhaps you'd also like to use bc to round a floating-point MHz value.

druuna · 12-17-2009, 05:30 AM

Hi,

If you want the total line (ie: cpu MHz : 800.000): grep "cpu MHz" /proc/cpuinfo

If you only want the number (Ie: 800.000): awk '/cpu MHz/ { print $4 }' /proc/cpuinfo

Assigning it to a variable you can do as follows:

cpuMhz="`awk '/cpu MHz/ { print $4 }' /proc/cpuinfo`"
or
cpuMhz=$(awk '/cpu MHz/ { print $4 }' /proc/cpuinfo)

Writing it to a file (not using the variable): awk '/cpu MHz/ { print $4 }' /proc/cpuinfo > cpu.txt

Writing to a file using the variable: echo $cpuMhz > cpu.txt

Hope this helps.

evo2 · 12-17-2009, 05:31 AM

Code:

awk '{ if ($2=="MHz")print $4 }' /proc/cpuinfo

Or, no decimals

Code:

awk '{ if ($2=="MHz")printf "%.0f\n", $4 }' /proc/cpuinfo

Evo2.
PS. druunas matching method is probably better than the "if" I use here.

ttsdinesh · 12-17-2009, 05:50 AM

Thank u evo2, druuna, HasC. Let me try these snippets and get u later.

ghostdog74 · 12-17-2009, 05:53 AM

Quote:

Originally Posted by HasC

use some shell scripting...

As an example:

$ grep "cpu MHz" /proc/cpuinfo | awk '{print $4}' | cut -d \. -f 1

will give you the "800".

And, perhaps you'd also like to use bc to round a floating-point MHz value.

useless use of grep and cut

ghostdog74 · 12-17-2009, 05:56 AM

Code:

awk -F":" 'BEGIN{IGNORECASE=1}/^cpu MHz/{print $2} ' /proc/cpuinfo > file

druuna · 12-17-2009, 06:06 AM

@ghostdog74: Why the "complicated" awk command (that leaves the tab/space in front of the output)?

Isn't awk '/cpu MHz/ { print $4 }' /proc/cpuinfo the simpler solution.

HasC · 12-17-2009, 06:25 AM

the simplest solution is the one you like the most, I believe.
An Awk guru would use the ghostdog snippet, a Perl monger would program something else (in perl, obviusly :-). I prefer shell scripting

druuna · 12-17-2009, 06:32 AM

@HasC: Most people tend to go with what they know (hence the cat file | awk 'something' | awk 'somethingelse' constructs, which can probably done with one awk statement).

I noticed over time that ghostdog74 likes resource friendly, elegant and precise coding (which is a good thing!) and I'm a bit confused why he posted that specific command. Especially because the output isn't fully correct.

ghostdog74 · 12-17-2009, 06:42 AM

Quote:

Originally Posted by druuna

@ghostdog74: Why the "complicated" awk command (that leaves the tab/space in front of the output)?

Isn't awk '/cpu MHz/ { print $4 }' /proc/cpuinfo the simpler solution.

errmm, because you posted this solution earlier than me that's why.

As for the "complicated" version, i am being overly cautious, that's all. As for the output, changing -F":" to -F": " will correct that minor space problem.

druuna · 12-17-2009, 06:48 AM

@ghostdog74: LOL

BTW: The output of /proc/xyz is by convention (the describing part), no need to be overly cautious. But then again, conventions are broken all the time, so maybe it is a good idea to make it case insensitive.....