Create a timeline of jpg images according to modification date using linux

Question

I'm new to Bash scripting and am having a difficult time with creating a script that takes all the jpg files from a directory (assuming all of them are jpg) and creating a timeline of images according to their modification date.

I've tried so far:

ls -Rt *.jpg | convert -append jpgfile

and

convert -append $(ls -Rtr $(find MontrealTest -name *.jpg)) MontrealTest.jpg

The bash program should create a timeline of images in q3_image_sorter.bash from oldest to newest. convert -append can be used to create the timeline image made of N image name arguments. N-1 are treated as inputs, to be read and stacked vertically and the final (Nth) argument is the output

filename. time-line image should match the path given as the first argument, but _ to replace any slashes. ex. $ bash q3_image_sorter.bash Q3/SimpleTest outputs file Q3_SimpleTest.jpg. * recommended to useeog to view the files

These are the list of jpg files, and I need to convert them to the images. For example, when I run bash q3_image_sorter.bash Q3/MontrealTest I will get the 8 images from oldest to newest.

Answer 1

Doing this robustly (that is, supporting unusual or hostile file or directory names safely -- except those with newlines, which we discard to avoid a dependency on GNU sort) might look like the following:

#!/usr/bin/env bash

# Start with a version check; this would need changes to support older bash
case $BASH_VERSION in ''|[123].*) echo "ERROR: Bash 4.0 required" >&2; exit 1;; esac

dir=$1
[[ $dir ]] || { echo "Usage: ${0##*/} dirname" >&2; exit 1; }

# telling find to ignore files with newlines so we don't need to use GNU sort
readarray -t files < <(
  find "$dir" -name $'*\n*' -prune -o -type f -name '*.jpg' -printf '%T@ %p\n' \
    | sort -n | sed -e 's/^[^ ]* //'
)

(( ${#files[@]} )) || { echo "ERROR: No .jpg files found in $dir" >&2; exit 1; }

outname=${dir//'/'/_}.jpg  ## replace slashes with underscores and add .jpg to get outname

convert "${files[@]}" -append "$outname"

References for the syntax used in each location:

• readarray reads a list of items into an array. By default, it expects newline-delimited content; in bash 4.4, the -d '' argument can be used to specify NUL delimiters instead. readarray is also named mapfile, and is documented at http://wiki.bash-hackers.org/commands/builtin/mapfile
• (( )) enters an arithmetic context, where a zero result evaluates as false and a nonzero result identifies as true. See http://wiki.bash-hackers.org/syntax/arith_expr for more on arithmetic expressions in bash.
• ${#arrayname[@]} expands to the number of items in the array named arrayname. See BashFAQ #5 for a general introduction to arrays in bash, or http://wiki.bash-hackers.org/syntax/arrays
• ${varname//search/replace} expands to the contents of the variable named varname, with all instances of search replaced with replace. This is an example of parameter expansion, described in detail at http://wiki.bash-hackers.org/syntax/pe.
• "${arrayname[@]}" expands to the individual items from the array named arrayname.
• <( ... ) is a process substitution -- syntax which creates a file-like object which, when read, will contain the output of the command ....