Using HAZEL to run a Automator Workflow

I am trying to use Hazel to run an Automator Workflow App. but when I try and point the rule to the appropriate Automator script, all the scripts are greyed out.

Do I need to convert the Workflow? I have just tried running the Run Shell Script directly with Hazel without success.

Ideas?

Thanks Paul

What's the extension on the file?

Try saving the workflow as a .workflow file instead.

changing the app to workflow allowed me to select the workflow in Hazel... however after I run it, hazel ran and done process the folder but did not make the changes I expected. When ran independently the workflow works as expected.

What did it do differently? Can you post the workflow?

does not run through and rename images...

#!/bin/zsh

# associate array for SHA key and duplicate filenames
typeset -gA dx=()
typeset -ga exifdata=()
typeset -giZ2 dupecnt=0
typeset -giZ3 nseq=1
typeset -g RPT="$HOME/Desktop/dupe_rpt.txt"

function rpt_writer () {
# print by SHA key, original file, and duplicates (in any child dir)
for k in ${(@k)dx};
do
print ${(k)dx[$k]} >> ${RPT}
print ${${(s:|:)dx[$k]}[1]} >> ${RPT}
printf '\t%s\n' ${${(s:|:)dx[$k]}[2,-1]} >> ${RPT}
printf '\n' >> ${RPT}
done
return
}

: <<'COMMENT'
Recursively processes child folders in specified parent directory. Extracts
image data using ExifTool, or where absent fabricate that data to provide a
standard TEMPLATE for file renaming. Encountered duplicates will add 2-digit
suffix to the TEMPLATE which forms the renaming OUTFMT string. Report writer
to audit duplicate images.
COMMENT

# parent folder passed into script on the command-line
STARTDIR="${1:a}"

# case-insensitive directive
setopt NOCASE_GLOB
for f in ${STARTDIR}/**/*.(raf|cr2|dng|jpg|jpeg|tif|tiff)(.N);
do
# ideally, place these three results into array if image supports them
exifdata=( $(/usr/local/bin/exiftool -s3 -d "%Y%m%d%H%M%S" -DateTimeOriginal \
-EXIF:FocalPlaneXResolution \
-EXIF:FocalPlaneYResolution "${f}") )

# check number of items in array. There will never be a 2 category
case ${#exifdata[@]} in
0)
# could not get DateTimeOriginal so wing it with creation (Birth) date
CD="$(stat -t '%Y%m%d%H%M%S' -f '%SB' "${f}")"
exifdata=(${CD} 0000 0000)
;;
1)
# got DateTimeOriginal but no FocalPlaneResolution data
exifdata+=(0000 0000)
;;
3)
# If FocalPlaneResolution is floating point,round to nearest integer
[[ ${exifdata[2]} =~ "." ]] && exifdata[2]=$(printf "%.0f" ${exifdata[2]})
[[ ${exifdata[3]} =~ "." ]] && exifdata[3]=$(printf "%.0f" ${exifdata[3]})
;;
*)
# should not get here as won't get just two items
continue
;;
esac
# yyyymmddHHMMSS fx fy seq
# 20200530221158|6532|6532|_001
# the (j) concatenates the three array elements as one string
# deleted from below to stop sequencing
TPLATE=${f:a:h}/${(j::)exifdata}_${nseq}

: <<'COMMENT'
# reset assoc array and duplicates counter when child folder changes
if [[ $curdir != $prevdir ]]; then
dupecnt=0
fi
COMMENT

# get 40 character SHA checksum from current file
dkey=$(shasum -a1 "${f}" | cut -d ' ' -f1) 2> /dev/null

# check if current file is a duplicate by key lookup
# use not (!) because non-duplicate branch will improve speed
if [[ ! ${dx[(k)$dkey]} ]]; then
# non-duplicate processing
OUTFMT="${TPLATE}.${f:e}"
# push new key/value pair with SHA key and original filename with delimiter
dx+=( $dkey "|${OUTFMT} => ${f}" )

else
# duplicate file processing
# return the count of duplicate filenames, excluding original file
dupecnt=$(( ${(ws:|:)#dx[(k)$dkey]} - 1))
(( dupecnt++ ))
OUTFMT="${TPLATE}-${dupecnt}.${f:e}"
# add filename to key's list of duplicate names
dx[$dkey]+="|${OUTFMT} => ${f}"
fi

# rename the image
mv "${f:a}" ${OUTFMT}
done
rpt_writer
# release associative array storage
dx=()
exit

That looks more like a shellscript. Is there a reason you are using Automator here?

there is no reason except that I use the Automator to ask for the folder to indicate the search...

Paul

Hazel passes in the file that matches the rule. Is there a reason why you need Automator to additionally prompt?

I used Automator to get this working as a proof of concept now I know the script works... I would like to run it with Hazel so that I can add other task in...

How should I run it in Hazel as part of a workflow?

Paul

Since it's a shellscript, use the "Run shellscript" action. You'll need to take into account that the file matching the rule is passed in as the first argument.

Thanks very much for your post...

Can you expand on passing in the first argument please...

Paul

Check the help. How you access the argument is different depending on the scripting language you use but the script editor should tell you which one.

I have been working on trying to get this to run, it seems I have complicated what I want to achieve.

so what I thought I will use the exiftool to correct the names of the extensions and also rename the names of the files.

I can run the following command lines through Terminal , however I I would like to create a Hazel work flow?

to rename a file:
exiftool '-FileName<DateTimeOriginal' -d "%Y%m%d_%H.%M.%S%%-c.%%e"

Rebuild and fix corrupted EXIF metadata
exiftool -all= -tagsfromfile @ -all:all -unsafe -icc_profile -F

Correct Extension
exiftool -ext '*' '-filename<%f.$fileTypeExtension' '

I have create a hazel work flow with tree rules one for each of the above, however I get errors when I try this out.

2020-07-04 14:28:36.892 hazelworker[8521] [Error] Shell script failed: Error processing shell script on file /Users/Paul/Pictures/Temp/test1.dng.
2020-07-04 14:28:36.893 hazelworker[8521] Shellscript exited with non-successful status code: 1
2020-07-04 14:28:41.118 hazelworker[8521] Caught signal: 15

what am I doing incorrectly...

Many Regards Paul