how is sed generating .2.

[grazinggoat]

I have files that are log.20170321PDT 8 numeric spaces and log.2017032307PDT 10 spaces
the script does a conversion to add seconds to the time stamp to keep them unique
when zipping but for some reason sed adds the extension below

I'm at a loss on where the .2. is being genrated by sed
any thoughts?

echo log.2017032222PDT | sed -e s/\.[0-9][0-9][0-9][0-9][0-9][0-9][0-9][0-9]..T$/\.20170322225715PDT/

generates:

log.2.20170322225715PDT

i want the replacement to be log.20170322225715PDT for both 8 or 10 spaced files

if echo ${NAME} | egrep "log.[0-9][0-9][0-9][0-9][0-9][0-9][0-9][0-9]..T" > /dev/null 2>&1 ; then
NewName=$(echo ${NAME} | sed -e s/log\.[0-9][0-9][0-9][0-9][0-9][0-9][0-9][0-9]..T$/\.${EXTENSION}/)
else
NewName=$(echo ${NAME} | sed -e s/log\.[0-9][0-9][0-9][0-9]-[0-9][0-9]-[0-9][0-9]$/\.${EXTENSION}/)
fi


if echo ${NAME} | egrep "log.[0-9][0-9][0-9][0-9][0-9][0-9][0-9][0-9][0-9][0-9]..T" > /dev/null 2>&1 ; then
NewName=$(echo ${NAME} | sed -e s/log\.[0-9][0-9][0-9][0-9][0-9][0-9][0-9][0-9][0-9][0-9]..T$/\.${EXTENSION}/)
else
NewName=$(echo ${NAME} | sed -e s/log\.[0-9][0-9][0-9][0-9][0-9][0-9]-[0-9][0-9]-[0-9][0-9]$/\.${EXTENSION}/)

 

[SamBones]

That dot (\.) is matching ANY character, not a dot. And you have an incorrect number of numeric digits ([0-9]). So that dot is matching the first '2', not the dot. So it's not being matched the way you intended.

Try this...

# echo log.2017032222PDT | sed -e s/[.][0-9]*..T$/\.20170322225715PDT/
log.20170322225715PDT

The dot in brackets actually matches a dot. Then you match any number of numeric digits up to <anything><anything>T<end-of-line}.

 

[grazinggoat]

Ahh ok totally saw my errors you pointed out. This works like a charm! thanks much!