#!/bin/bash
infile=$1
outfile=$1".mod.wts"
# add version and root begin tags to beginning of input file
echo '<?xml version="1.0" encoding="ASCII"?>' > temp1
echo "<net>" >> temp1
cat temp1 $1 > temp2
# format information tags (end with />)
sed 's/<Fmt *\([a-z,A-Z]*\) *>/<Fmt Fmt="\1"\/>/g' temp2 | \
sed 's/<Name\ Network_0>/<Name\ Name="Network_0"\/>/g' | \
sed 's/<Epoch *\([0-9]*\) *>/<Epoch Epoch="\1"\/>/g' | \
# format tags with arguments of different name than tag
sed 's/<UgUn *\([^ ]\{1,9\}\) *>/<UgUn id="\1">/g' | \
sed 's/<Cn *\([0-9]*\) *>/<Cn num="\1">/g' | \
# format remaining tags where tag and arg have same value
perl -pe 's/<(\w+) (\w+)>/<\1 \1="\2">/' | \
# format tags with multiple args
awk '{ if ( $0 ~ /<Cg 0 Fm:Input>/ ) {
printf( "%s\n", "<Cg Cg=\"0\">" );
printf( "%s\n", "<Fm>Input</Fm>" );
} else {
print $0;
}
}' | \
awk '{ if ( $0 ~ /<Cg 0 Fm:Hidden>/ ) {
printf( "%s\n", "<Cg Cg=\"0\">" );
printf( "%s\n", "<Fm>Hidden</Fm>" );
} else {
print $0;
}
}' | \
# format multi row, multi are data
perl -p0e 's/(<Un>\n)(.*)/\1<Bias>\2<\/Bias>/g' | \
perl -pe 's/(\d+) ([\d.-]+)/<C\1>\2<\/C\1>/ if /<Cn/../<\/Cn/' > $outfile
# add root close tag to end of file
echo "</net>" >> $outfile
# cleanup
rm temp1 temp2
You can see that I had to sort of hard code the single line tags at the beginning (format <stuff stuff="stuff"/>). These will always be present, so that isn't a big deal. The network does not always have to be Network_0, but when I tried the last perl code you posted, it changed some other things that I didn't need changed. I guess these three sort of need to look for,
<Fmt
<Name
<Epoch
The script works, and quickly, but is sort of a mishmash of regex stuff. I guess that's not too unusual, but if you see anything that is a real issue, or could be much cleaner, I would appreciate a heads up.
I have attached the file this works on, and the converted file.
The network does not always have to be Network_0, but when I tried the last perl code you posted, it changed some other things that I didn't need changed.
Can you post the list of tags that shouldn't be changed?
where the second value in the original tag goes in the quotes, followed by the added "/". The second value could be anything and so needs to be read from the original tag.
There are other tags that have a single argument, but do not get the trailing "/",
Code:
<Lay Input>
<UgUn 0 >
<Cn 36>
I have converted some of these so that the argument name is not the same as the tag name,
Code:
<UgUn id="0">
<Cn num="36">
but perhaps that is not such a good idea. I guess they can just be handeled as two cases,
Code:
text tag, space, text argument (<Lay Input>)
text tag, space, int argument (<Cn 36>)
to be converted to,
Code:
<Lay Lay="Input">
<UgUn UgUn="0">
<Cn Cn="36">
note that, for some reason, there is a trailing space between the argument and the > in in <UgUn 0 >.
There are multi argument tags,
Code:
<Cg 0 Fm:Input>
to convert to,
Code:
<Cg Cg="0">
<Fm>Input</Fm>
I expect that these could be,
Code:
<Cg Cg="0" Fm="Input">
And lastly there is the data that was corrected by the perl in your first post.
OK, but is everything working fine now? Because I really don't feel like analyzing 1000+ lines XML file If something is not reformatted properly, then which particular tag is it?
in the script posted above, I see lot of sed and awk that are chained of list of commands to be executed within a bash wrapper. With increase in file size, this approach is going to terribly slow down the processing as its going to keep spawning multiple processes.
Have you considered writing it in perl with read line interface and processing, which will be way faster than the current approach.
Sorry for the long/weird title but I'm stuck on a problem I have. I have this XML file:
</member>
<member>
<name>TransactionID</name>
<value><string>123456789123456</string></value>
</member>
<member>
<name>Number</name>
... (9 Replies)
Hi Everyone,
I'm new here and I was checking this old post:
/shell-programming-and-scripting/180669-splitting-file-into-several-smaller-files-using-perl.html
(cannot paste link because of lack of points)
I need to do something like this but understand very little of perl.
I also check... (4 Replies)
Hello,
I have a requirement to extract the value from multiple xml node and print out the values to new file to compare.
Would be done using either awk/perl or some unix script.
For example sample input file:
.....
.....
<factories xmi:type="resources.jdbc:DataSource"... (2 Replies)
Hi All,
I need help from any of you.Would be so thankful for your help.
I/P
DDDD,1045,161,1557,429,1694,800,1911,1113,2460,1457,2917>
1609,3113,1869,3317,2732,3701,3727,4132,5857,5107>
9004,6496
DDDD,1125,157,1558,429,1694,800,1911,1117,2432,1444,2906>... (2 Replies)
I have a Unix file with 200,000 records, and need to remove all records from the file that have the character ‘I' in position 68 (68 bytes from the left). I have searched for similar problems and it appears that it would be possible with sed, awk or perl but I do not know enough about any of these... (7 Replies)
I hopefully have a simple request - I need to process multiple files reformatting the output based on tags at the beginning of each line. So the data for the new 3 lines of the output file are in the HDR line and then the details are in the DTL tagged lines.
for ifile in $indir
do
echo... (1 Reply)
Hello everyone,
Firstly i do not require alot of help.. i am right at the end of finishing my scipt but cannot find a solution to the last part.
What i need to do is, prompt the user for a file to work with, which i have done.
promt the user for an output file - which is done.
#!/bin/bash... (14 Replies)
hello,
new to this forum. but i have a requirement to extract the value from multiple xml node and print out the values to new file with comma seperated. would like to know how this would be done using either sed/perl or some unix script. an example would be tremendous...
sample input file:... (2 Replies)
Hi, I want to get data from Xml file by using sed or awk command. I want to get the following result :
mon titre 1;Createur1;Dossier1
mon titre 1;Createur1;Dossier1
and save it in cvs file (fichier.cvs).
FROM this Xml file (test.xml):
<playlist version="1">
<trackList>
<track>... (1 Reply)
I need to know the way. I have got parsing down some nodes. But I was unable to get the child node perfectly. If you have code please send it. It will be very useful for me. (0 Replies)
04-17-2011
Because I really don't feel like analyzing 1000+ lines XML file 
If something is not reformatted properly, then which particular tag is it?