wildcardinal's Replies

Forum Replies Created

Viewing 4 replies - 1 through 4 (of 4 total)

Forum: Plugins
In reply to: [EntryWizard] Renames File – Use Username?

(@wildcardinal)

Hey @allythacker – we had the same need. Renaming natively is quite difficult (as @mbsharp says, the official way is to rename once downloaded). I created this which runs on the server and allows the Zip file created to have files within it renamed. There is no way that I would ever expect this to be rolled into the core of EntryWizard, but if you have some willingness to invest a bit of time (and understand a bit of shell scripting and you run on a Unix server) this may be able to do what you are wanting.

The first bit is a small update to wp-content/plugins/entrywizard/classes/ewz-webform.php to call a new script rather than just zipping the files together. About 1/3 of the way down this file, replace the line

$fp = popen("zip -0 -j -q - $fnames", 'r');

with…

$fp = popen("/usr/local/bin/createZipForCompetition.sh $fnames", 'r');

So you get this…

        // options for linux "zip" command:
        // -0 = no compression
        // -j = strip path, use only filename
        // -q = no extra output info
        // $fp = popen("zip -0 -j -q - $fnames", 'r');        // see testing notes
        $fp = popen("/usr/local/bin/createZipForCompetition.sh $fnames", 'r');  
        if( !$fp ){                                        // see testing notes
            error_log("EWZ: Zip command failed");
            throw new EWZ_Exception("Zip command failed");
        }

This means that rather than just creating a zip we are intercepting and calling our own script to so the renaming for us.

Then create the shell script to do the real work. This takes the image filenames and csv (passed in as arguments – as they would have been passed into the zip command) and then does all the magic. All the fields in the CSV are available (which includes username, category, image name etc). You just need to extract them from the CSV and use them in the rename – follow the pattern in the script!

You may need to make a few other updates to get the structures/names exactly as you want.

I have created this as /usr/local/bin/createZipForCompetition.sh

#!/bin/bash

# Given a list of files, create a zip suitable for download
# One of the files will be a CSV

TMPDIR=/tmp/CompetitionZIP_$$
mkdir -p $TMPDIR
mkdir -p $TMPDIR/orig
mkdir -p $TMPDIR/out
#mkdir -p $TMPDIR/out/NoType/fullname
#mkdir -p $TMPDIR/out/NoType/titleonly
#mkdir -p $TMPDIR/out/PDI/fullname
#mkdir -p $TMPDIR/out/PDI/titleonly
#mkdir -p $TMPDIR/out/Print/fullname
#mkdir -p $TMPDIR/out/Print/titleonly

CSVFILETITLE=EntryList-TitlesOnly.csv
CSVFILE=EntryList.csv

# store arguments in a special array
args=("$@")
# get number of elements
ELEMENTS=${#args[@]}

# echo each element in array
# for loop
for (( i=0;i<$ELEMENTS;i++)); do
	# echo "Processing ${args[${i}]}"
	# echo cp "${args[${i}]}" $TMPDIR/orig/
	cp "${args[${i}]}" $TMPDIR/orig/
done

# We've now copied all the files into the target location
# Next we need to process the CSV file

LINECOUNT=cat $TMPDIR/orig/*.csv | wc -l
# echo "Lines to handle = $LINECOUNT"

#Create the data files
cat $TMPDIR/orig/*.csv | csvcut -c "Display Name" > $TMPDIR/orig/_DISPLAYNAME.TXT
cat $TMPDIR/orig/*.csv | csvcut -c "Filename" > $TMPDIR/orig/_FILENAME.TXT
cat $TMPDIR/orig/*.csv | csvcut -c "Title" > $TMPDIR/orig/_TITLE.TXT
cat $TMPDIR/orig/*.csv | csvcut -c "Tp" > $TMPDIR/orig/_TYPE.TXT

# cat  $TMPDIR/orig/*.csv

# echo "======"
# cat  $TMPDIR/orig/_TYPE.TXT
# echo "======"

HASNOTYPE="0"
HASPRINT="0"
HASPDI="0"


for (( i=2;i<=$LINECOUNT;i++)); do
	IMAGEAUTHOR=cat $TMPDIR/orig/_DISPLAYNAME.TXT | sed -n -e 1p -e ${i}p | tail -n 1 |  tr -d "<>:\"/\\|?*_"
	IMAGEFILE=cat $TMPDIR/orig/_FILENAME.TXT | sed  -n -e 1p -e ${i}p | tail -n 1
	IMAGETITLE=cat $TMPDIR/orig/_TITLE.TXT | sed  -n -e 1p -e ${i}p | tail -n 1 | tr -d "<>:\"/\\|?*_"
	IMAGETYPE=cat $TMPDIR/orig/_TYPE.TXT | sed  -n -e 1p -e ${i}p | tail -n 1 |  tr -d "<>:\"/\\|?*_"
	RINDEX=echo ${IMAGEAUTHOR}${IMAGETITLE} |  md5sum | tr -d "&#091;a-z0]" | cut -c 3-9

	if [ -z "$IMAGETYPE" ]
	then
		IMAGETYPE="NoType"
		HASNOTYPE="1"
	fi

	if [ "$IMAGETYPE" = "P" ]
	then
		IMAGETYPE="Print"
		HASPRINT="1"
	fi

	if [ "$IMAGETYPE" = "D" ]
	then
		IMAGETYPE="PDI"
		HASPDI="1"
	fi



	  # echo "===== ENTRY $i"
	  # echo "Author     : >$IMAGEAUTHOR<"
	  # echo "Filename   : >$IMAGEFILE<"
	  # echo "Title      : >$IMAGETITLE<"
	  # echo "Type       : >$IMAGETYPE<"

	if [ -z "$IMAGEAUTHOR" ]
	then
		#  echo "***** NO AUTHOR - giving up"
		continue
	fi

	if [ -z "$IMAGEFILE" ]
	then
		#  echo "***** NO FILE - giving up"
		continue
	fi

	if [ -z "$IMAGETITLE" ]
	then
		#  echo "***** NO TITLE - giving up"
		continue
	fi

	mkdir -p ${TMPDIR}/out/${IMAGETYPE}/fullname
	mkdir -p ${TMPDIR}/out/${IMAGETYPE}/titleonly

       	TARGETFULL="${IMAGETITLE}_${IMAGEAUTHOR}.jpg"
       	TARGETSHORT="${RINDEX}_${IMAGETITLE}.jpg"

	SOURCEFILE=$(printf '%q' "$IMAGEFILE")

	TARGETFILEFULL=$(printf '%q' "$TARGETFULL")
	MYCMDFULL="cp $TMPDIR/orig/${SOURCEFILE} $TMPDIR/out/${IMAGETYPE}/fullname/${TARGETFILEFULL}"
	eval $MYCMDFULL

	TARGETFILESHORT=$(printf '%q' "$TARGETSHORT")
	MYCMDSHORT="cp $TMPDIR/orig/${SOURCEFILE} $TMPDIR/out/${IMAGETYPE}/titleonly/${TARGETFILESHORT}"
	eval $MYCMDSHORT

	echo ${RINDEX},\"${IMAGEAUTHOR}\",\"${IMAGETITLE}\" >> $TMPDIR/out/${IMAGETYPE}/${CSVFILE}
	echo ${RINDEX},\"${IMAGETITLE}\" >> $TMPDIR/out/${IMAGETYPE}/${CSVFILETITLE}

done

# cp $TMPDIR/orig/*.csv ${TMPDIR}/out


if [ "$HASNOTYPE" = "1" ]
then
	cd ${TMPDIR}/out/NoType

	cat $TMPDIR/out/NoType/${CSVFILE}      | sort --numeric-sort --field-separator=, --key 1 > $TMPDIR/out/NoType/${CSVFILE}.tmp
	cat $TMPDIR/out/NoType/${CSVFILETITLE} | sort --numeric-sort --field-separator=, --key 1 > $TMPDIR/out/NoType/${CSVFILETITLE}.tmp

	echo "Identifier,Author,Title" | cat - $TMPDIR/out/NoType/${CSVFILE}.tmp      > $TMPDIR/out/NoType/${CSVFILE}
	echo "Identifier,Title"        | cat - $TMPDIR/out/NoType/${CSVFILETITLE}.tmp > $TMPDIR/out/NoType/${CSVFILETITLE}

	rm $TMPDIR/out/NoType/${CSVFILE}.tmp
	rm $TMPDIR/out/NoType/${CSVFILETITLE}.tmp
fi

if [ "$HASPRINT" = "1" ]
then
	cd ${TMPDIR}/out/Print

	cat $TMPDIR/out/Print/${CSVFILE}      | sort --numeric-sort --field-separator=, --key 1 > $TMPDIR/out/Print/${CSVFILE}.tmp
	cat $TMPDIR/out/Print/${CSVFILETITLE} | sort --numeric-sort --field-separator=, --key 1 > $TMPDIR/out/Print/${CSVFILETITLE}.tmp

	echo "Identifier,Author,Title" | cat - $TMPDIR/out/Print/${CSVFILE}.tmp      > $TMPDIR/out/Print/${CSVFILE}
	echo "Identifier,Title"        | cat - $TMPDIR/out/Print/${CSVFILETITLE}.tmp > $TMPDIR/out/Print/${CSVFILETITLE}

	rm $TMPDIR/out/Print/${CSVFILE}.tmp
	rm $TMPDIR/out/Print/${CSVFILETITLE}.tmp
fi

if [ "$HASPDI" = "1" ]
then
	cd ${TMPDIR}/out/PDI

	cat $TMPDIR/out/PDI/${CSVFILE}      | sort --numeric-sort --field-separator=, --key 1 > $TMPDIR/out/PDI/${CSVFILE}.tmp
	cat $TMPDIR/out/PDI/${CSVFILETITLE} | sort --numeric-sort --field-separator=, --key 1 > $TMPDIR/out/PDI/${CSVFILETITLE}.tmp

	echo "Identifier,Author,Title" | cat - $TMPDIR/out/PDI/${CSVFILE}.tmp      > $TMPDIR/out/PDI/${CSVFILE}
	echo "Identifier,Title"        | cat - $TMPDIR/out/PDI/${CSVFILETITLE}.tmp > $TMPDIR/out/PDI/${CSVFILETITLE}

	rm $TMPDIR/out/PDI/${CSVFILE}.tmp
	rm $TMPDIR/out/PDI/${CSVFILETITLE}.tmp
fi


cd ${TMPDIR}/out
zip -r -q -0 ${TMPDIR}/generatedzip.zip *
cat ${TMPDIR}/generatedzip.zip

cd /

rm -rf ${TMPDIR}

Forum: Plugins
In reply to: [EntryWizard] Rename File on upload

wildcardinal

(@wildcardinal)

5 years, 8 months ago

To Mcollen – yes, it’s the same as you are doing but getting it done on the server first. As long as there is a field in the CSV you can use it as part of the filename creation.

To Ronc – the down side is that when ever Josie puts out an update I have to go back in and make the update. Not great, but this approach does make it easier for us. I did raise thread asking if the command used to create the zip could be configurable, but Josie wasn’t keen (understandably). Thread here.

Forum: Plugins
In reply to: [EntryWizard] Rename File on upload

wildcardinal

(@wildcardinal)

5 years, 8 months ago

I have a work-around for this which does well for us. I’m not sure if you’ll want to dive in to this level though.

In wp-content/plugins/entrywizard/classes/ewz-webform.php
Change the line
$fp = popen("zip -0 -j -q - $fnames", 'r'); // see testing notes
To
$fp = popen("/usr/local/bin/createZipForCompetition.sh $fnames", 'r'); // see testing notes

Then create the file /usr/local/bin/createZipForCompetition.sh as follows

#!/bin/bash

# Given a list of files, create a zip suitable for download
# One of the files will be a CSV

TMPDIR=/tmp/CompetitionZIP_$$
mkdir $TMPDIR
mkdir $TMPDIR/orig
mkdir $TMPDIR/out
mkdir $TMPDIR/out/fullname
mkdir $TMPDIR/out/titleonly

# store arguments in a special array
args=("$@")
# get number of elements
ELEMENTS=${#args[@]}

# echo each element in array
# for loop
for (( i=0;i<$ELEMENTS;i++)); do
        # echo "Processing ${args[${i}]}"
        # echo cp "${args[${i}]}" $TMPDIR/orig/
        cp "${args[${i}]}" $TMPDIR/orig/
done

# We've now copied all the files into the target location
# Next we need to process the CSV file

LINECOUNT=<code>cat $TMPDIR/orig/*.csv | wc -l</code>
# echo "Lines to handle = $LINECOUNT"

#Create the data files
cat $TMPDIR/orig/*.csv | csvcut -c "Display Name" > $TMPDIR/orig/_DISPLAYNAME.TXT
cat $TMPDIR/orig/*.csv | csvcut -c "Filename" > $TMPDIR/orig/_FILENAME.TXT
cat $TMPDIR/orig/*.csv | csvcut -c "Title" > $TMPDIR/orig/_TITLE.TXT
cat $TMPDIR/orig/*.csv | csvcut -c "Tp" > $TMPDIR/orig/_TYPE.TXT

for (( i=2;i<=$LINECOUNT;i++)); do
        IMAGEAUTHOR=<code>cat $TMPDIR/orig/_DISPLAYNAME.TXT | sed -n -e 1p -e ${i}p | tail -n 1 |  tr &quot;<>:\&quot;/\\|?*_&quot; @</code>
        IMAGEFILE=<code>cat $TMPDIR/orig/_FILENAME.TXT | sed  -n -e 1p -e ${i}p | tail -n 1</code>
        IMAGETITLE=<code>cat $TMPDIR/orig/_TITLE.TXT | sed  -n -e 1p -e ${i}p | tail -n 1 | tr &quot;<>:\&quot;/\\|?*_&quot; @</code>
        IMAGETYPE=<code>cat $TMPDIR/orig/_TYPE | sed  -n -e 1p -e ${i}p | tail -n 1 |  tr &quot;<>:\&quot;/\\|?*_&quot; @</code>
        RINDEX=<code>od -A n -t d -N 4 /dev/urandom | tail -c 6</code>

        #  echo "===== ENTRY $i"
        #  echo "Author     : >$IMAGEAUTHOR<"
        #  echo "Filename   : >$IMAGEFILE<"
        #  echo "Title      : >$IMAGETITLE<"
        #  echo "Type       : >$IMAGETYPE<"

        if [ -z "$IMAGEAUTHOR" ]
        then
                #  echo "***** NO AUTHOR - giving up"
                continue
        fi

        if [ -z "$IMAGEFILE" ]
        then
                #  echo "***** NO FILE - giving up"
                continue
        fi

        if [ -z "$IMAGETITLE" ]
        then
                #  echo "***** NO TITLE - giving up"
                continue
        fi

        if [ -z "$IMAGETYPE" ]
        then
                # echo "***** NO TYPE - creating a standard name"
                TARGETFULL="${IMAGETITLE}_${IMAGEAUTHOR}.jpg"
                TARGETSHORT="${RINDEX}_${IMAGETITLE}.jpg"
        else
                # echo "***** HAS TYPE - creating an extended name"
                TARGETFULL="${IMAGETITLE}_${IMAGETYPE}_${IMAGEAUTHOR}.jpg"
                TARGETSHORT="${RINDEX}_{IMAGETYPE}_${IMAGETITLE}.jpg"
        fi

        SOURCEFILE=$(printf '%q' "$IMAGEFILE")

        TARGETFILEFULL=$(printf '%q' "$TARGETFULL")
        MYCMDFULL="cp $TMPDIR/orig/${SOURCEFILE} $TMPDIR/out/fullname/${TARGETFILEFULL}"
        eval $MYCMDFULL

        TARGETFILESHORT=$(printf '%q' "$TARGETSHORT")
        MYCMDSHORT="cp $TMPDIR/orig/${SOURCEFILE} $TMPDIR/out/titleonly/${TARGETFILESHORT}"
        eval $MYCMDSHORT
done

cd ${TMPDIR}/out

zip -r -q -0 ${TMPDIR}/generatedzip.zip *
cat ${TMPDIR}/generatedzip.zip

cd /

rm -rf ${TMPDIR}

You may need to update to grab the specific columns you want from the CSV file generated for inclusion in the filename.

We run this on a Linux hosted environment and not hit any problems yet. We strip out the problematic characters from the titles before renaming the file (tell people that they can’t have things like * < > | etc).

Works for us, but we need to remember to re-modify the plugin code every time we update.

Forum: Plugins
In reply to: [EntryWizard] Custom export

Thread Starter wildcardinal

(@wildcardinal)

6 years, 3 months ago

Just to mention – I realise that the script is not currently production strength. Entry titles could be set to contain characters that can break the script, do nasty things etc.
I need to work on protection etc – but not yet got there!

Viewing 4 replies - 1 through 4 (of 4 total)