Unix file pattern check and replacement

Shell Programming and Scripting
1

HI Guys ,
Using UNIX ,I intend to check with correct file pattern Access_file_Record.YYYYMM in path /tmp

If the file exist in correct format come out from code .

If not found check with different file patterns for same month period YYYYMM ( Like Access_file_Record_YYYYMM.txt or Access_file_Record_.YYYYMM)

Any different pattern file should be renamed to expected file Access_file_Record.YYYYMM
Every month YYYYMM pattern shoud be identified as 201201 for jan , 201202 for feb etc when script is ran for that month

Please can anyone help me on this.
Thanks

2

If the script encounters for month 201202 (lets say) 2 files like :
Access_file_Record_201202.txt and Access_file_Record_.201202

What should the script do ?

According to what you say it should rename them with Access_file_Record_201202 but in this case after renaming Access_file_Record_201202.txt in Access_file_Record_201202 when the script encounters Access_file_Record_.201202 it will overwrite the file Access_file_Record_201202 (which has the content of Access_file_Record_201202.txt) with the content of Access_file_Record_.201202 .

this thing will happen in ANY OS. you need to clarify this first, what exactly you want. You cannot have more than 1 file with the same name on the same path.

3

HI As i said , the correct file pattern should be Access_file_Record.201202 .
If pattern matched exit;

If Not,then check different possible formats Access_file_Record_201202.txt or Access_file_Record_201202 etc (Access_file_Record prefix is common) are present if any . Then rename that pattern to Access_file_Record.201202 and come out . Hope i am clear with requirement.

4

My question is "What if you have more than one file with wrong pattern? "

5

HI , There will be only one wrong file pattern avaliable . Either it may be
Access_file_Record_201202.txt or Access_file_Record_201202

6

The content of the script :

#!/bin/bash

target_date=`date +%Y%m`
cd /tmp
the_existing_file=`ls *$target_date*`
the_desired_file="Access_file_Record.${target_date}"

if [ $the_existing_file = $the_desired_file ]; then
echo "The file was correctly named"
else
echo The file had the name "$the_existing_file" and it was renamed "$the_desired_file"
mv $the_existing_file $the_desired_file
fi

Create with vi a script and copy paste the content from above, afterwards add execution rights to the scripts. The script works for Linux environments (because of the command "date +%Y%m") which can have different outputs in different UNIXes . If you have problems with the script tell exactly what kind of Linux you use (use the uname command), usually this should be provided in the first request.

1 Like
7

Thanks dude ..I tried and it perfectly worked .

8

For existing entries (and apologies not the slickest production solution):

ls -1 /tmp/Access_file_Record*.* | while read line
do
  IFS=.
  set $line
  IFS=" "
  if [ $2 = "txt" ]
  then
    echo "mv $line $1"
    mv $line $1
  else
    echo "mv $line $1$2"
    mv $line $1$2
  fi
done
9

you may also want to look at this thread problem with sed * wildcard to have more clarity....