aeaready

#!/bin/bash
os=$(uname)
apperror=0
version=1.1
#
# A few alternative looks
# CSSURL=https://gist.githubusercontent.com/ryangray/1882525/raw/2a6e53f645b960f0bed16d686ba3df36505f839f/buttondown.css
#set -e
#CSSURL=https://gist.githubusercontent.com/dashed/6714393/raw/ae966d9d0806eb1e24462d88082a0264438adc50/github-pandoc.css
CSSURL=https://raw.githubusercontent.com/AEADataEditor/editor-scripts/refs/heads/main/.aeaready.css
CSSLOCAL=.aeaready.css
LOGOLOCAL=.logo-aea-88x88.jpg
DOCKERIMG=aeadataeditor/html-to-pdf:latest
dockerbin=$(which docker1 2>/dev/null || which docker)
#pdfengine="--pdf-engine wkhtmltopdf"
#pdfengine=""
pdfengine="docker"
case $pdfengine in
   "")
      WKHTMLOPTS="-s Letter --disable-smart-shrinking --enable-local-file-access"
      ;;
   "docker")
      DOCKEROPTS=""
      ;;
   *)
      WKHTMLOPTS="-V margin-top=1in -V margin-left=1in -V margin-right=1in -V margin-bottom=1in"
      ;;
esac
# other stuff
ICPSRMSG="Details in the full report, which you will receive via ScholarOne shortly. Please provide your response to the items listed above via the openICPSR Project Communication log, specifying AEAREP-xxx. Other items in the report may need to be addressed via ScholarOne."
PANDOCOPTS="--fail-if-warnings -s"
template_app="template/REPLICATION_appendix.md"
appendix="generated/REPLICATION_appendix.md"

echo "$0 version $version"

# validating inputs
if [[ -z $1 ]]
then
	echo "Please add issue number"
	exit 2
fi
num=$1
shift
case $1 in
  a*|approve)
     preapprove="Approved. Ready to submit."
     decision=approve
     shift
     ;;
  p*|pre-approve)
     preapprove="Preapproved. Ready for approval."
     decision=pre-approve
     shift
     ;;
  *)
    echo "Please define approval status: [a]pproved or [p]reapproved"
    exit 2
    ;;
esac
nopdf=0
if [[ "$1" == "nopdf" ]]
then
   nopdf=1
   shift
fi

extramsg=$*

exit_on_error() {
    exit_code=$1
    exit_message=${@:2}
    if [ $exit_code -ne 0 ]; then
        >&2 echo "${exit_message}, exit code ${exit_code}."
        exit $exit_code
    fi
}

set -o history -o histexpand


case $os in
	    Linux)
          pythonbin=$(which python3.12)
          [[ -f $pythonbin ]] || pythonbin=$(which python3)
          [[ -f $pythonbin ]] || pythonbin=$(which python)
          [[ -f $pythonbin ]] || apperror=1
		    pandoc=$(which pandoc)
          [[ -f $pandoc ]] || apperror=1
          if [[ "$pdfengine" == "" ]]
          then
            wkhtmltopdf=$(which wkhtmltopdf)
            [[ -f $wkhtmltopdf ]] || apperror=1
          elif [[ "$pdfengine" == "docker" ]]
          then
            [[ -f $dockerbin ]] || apperror=1
          fi
	       export QT_STYLE_OVERRIDE=fusion 
	       # get the CSS
          if [[ -f $(dirname $0)/$CSSLOCAL ]]
          then
            cp $(dirname $0)/$CSSLOCAL $CSSLOCAL
            else
            curl $CSSURL > $CSSLOCAL 
          fi
          # get the logo
          if [[ -f $(dirname $0)/$LOGOLOCAL ]]
          then
            cp $(dirname $0)/$LOGOLOCAL $LOGOLOCAL
          fi
	       
		    ;;
       Darwin)
            pythonbin=$(which python)
            [[ -f $pythonbin ]] || apperror=1
            # for MacOS, might want to install Rstudio (comes with pandoc) and replace the line above with the outcome of
            #   find /Applications -name pandoc -type f
            pandoc=/Applications/RStudio.app/Contents/MacOS/pandoc/pandoc
            # later versions have it bundled in the quarto directory
            [[ ! -f $pandoc ]] && pandoc=/Applications/RStudio.app/Contents/MacOS/quarto/bin/pandoc
            [[ ! -f $pandoc ]] && pandoc=/Applications/RStudio.app/Contents/MacOS/quarto/bin/tools/pandoc
            [[ ! -f $pandoc ]] && pandoc=/Applications/RStudio.app/Contents/Resources/app/quarto/bin/tools/pandoc
            # let's see if its there
            if [[ ! -f $pandoc ]]
            then
              echo "We suggest using pandoc from the RStudio install. Searching for a different one..."
              pandoc=$(find /Applications -name pandoc -type f)
            fi
            echo "Found pandoc at $pandoc"
	         # get the CSS
          if [[ -f $(dirname $0)/$CSSLOCAL ]]
          then
            cp $(dirname $0)/$CSSLOCAL $CSSLOCAL
            else
            curl $CSSURL > $CSSLOCAL 
          fi
          # get the logo
          if [[ -f $(dirname $0)/$LOGOLOCAL ]]
          then
            cp $(dirname $0)/$LOGOLOCAL $LOGOLOCAL
          fi
            ;;
	    *)
		    echo "Please review the script to define your OS"
            apperror=1
		    ;;
esac

# update the CSS

if [[ -f $CSSLOCAL ]]
then
   
   case $pdfengine in
      "")
         WKHTMLOPTS="$WKHTMLOPTS --user-style-sheet $CSSLOCAL"
         ;;
      "docker")
         # 
         PANDOCOPTS="$PANDOCOPTS --css $CSSLOCAL" 
         ;;
      *)
         WKHTMLOPTS="$WKHTMLOPTS --css $CSSLOCAL"
         ;;
   esac
fi

# other stuff 

if [[ "$apperror" == "1" ]]
then
    case $pdfengine in
        "")
            echo "We need pandoc and wkhtmltopdf"
            echo "One or the other not found"
            ;;
        "docker")
            echo "We need pandoc and docker"
            echo "One or the other not found"
            ;;
        *)
            echo "We need pandoc and wkhtmltopdf"
            echo "One or the other not found"
            ;;
    esac
    case $os in
        Darwin)
            echo "Install Rstudio to get pandoc"
            if [[ "$pdfengine" == "" ]]
            then
                echo "Try: brew cask install wkhtmltopdf"
            elif [[ "$pdfengine" == "docker" ]]
            then
                echo "Try: brew install docker"
            fi
            ;;
        *)
            echo "Please install"
            ;;
    esac
    [[ "$nopdf" == "1" ]] || exit_on_error 2 "Required dependencies missing"
fi 

commitmsg="AEAREP-$num #comment $preapprove $extramsg"

## Test if we are where we think we should be

if [[ ! -d ".git" ]]
then 
   exit_on_error 2 Not in right directory
fi

if [[ ! -f "REPLICATION.md" ]]
then
   exit_on_error 2 Missing REPLICATION.md -required-
fi

# Only add checkboxes to lines with '-' that also contain 'REQUIRED' or 'SUGGESTED' and are not already checkboxes
if [[ -f "REPLICATION.md" ]]; then
  sed -i.bak -E '/^- *\[ \]/!{/^-.*(REQUIRED|SUGGESTED)/s/^(- *)/- [ ] /}' REPLICATION.md
fi

# If there is a line with "Automatically Generated Appendices", we remove it and everything after it.

tmpmain=$(mktemp)
tmpapp=$(mktemp)

cat REPLICATION.md > $tmpmain

if grep -q "Automatically Generated Appendices" $tmpmain
then
   sed  '/Automatically Generated Appendices/,$d' $tmpmain > $tmpapp
else
   cp $tmpmain $tmpapp
fi

# Fill in the appendix

python3 tools/replace_placeholders.py --infile ${template_app} --indir "generated" --outfile $appendix
git add $appendix

# Append the generated appendix to the base file
echo "" >> $tmpapp
cat $tmpapp $appendix > REPLICATION.md

# Query JIRA API for DOI information and inject it into REPLICATION.md
if [[ -f "tools/jira_get_info.py" ]]
then
    echo "Querying JIRA for DOI information..."
    doi_info=$($pythonbin tools/jira_get_info.py aearep-$num doi 2>/dev/null)

    if [[ -n "$doi_info" && "$doi_info" != "" ]]
    then
        echo "Found DOI: $doi_info"

        # Check if the DOI line already exists and remove it
        sed -i '/^\s*- \[FYI\] \[EDITORIAL STAFF\] The predicted DOI of the replication package is /d' REPLICATION.md

        # Find the line with "Action Items (manuscript)" and insert blank line + DOI line after it
        sed -i "/### Action Items (manuscript)/a\\
\\
 - [FYI] [EDITORIAL STAFF] The predicted DOI of the replication package is $doi_info" REPLICATION.md

        echo "DOI information added to REPLICATION.md"
    else
        echo "No DOI information found in JIRA"
    fi
else
    echo "jira_get_info.py not found in tools/ subdirectory, skipping DOI injection"
    sleep 1
fi

# Query JIRA API for openICPSR project URL and inject it into REPLICATION.md
if [[ -f "tools/jira_get_info.py" ]]
then
    echo "Querying JIRA for openICPSR URL..."
    openicpsr_url=$($pythonbin tools/jira_get_info.py aearep-$num openicpsrurl 2>/dev/null)

    if [[ -n "$openicpsr_url" && "$openicpsr_url" != "" ]]
    then
        echo "Found openICPSR URL: $openicpsr_url"

        # Try to replace the placeholder URL with the actual URL
        # The line pattern to match (approximately):
        # > [REQUIRED] When returning proofs, confirm that all requested changes to the replication package, as outlined in the "Actions Items (openICPSR)"  section of the Data Editor report, have been completed. The URL to access the deposit is <https://www.openicpsr.org/openicpsr/workspace?goToPath=/openicpsr/xxxxx>.

        # Search for the template text in the 50 lines after "### Action Items (manuscript)"
        if sed -n '/### Action Items (manuscript)/,+50p' REPLICATION.md | grep -q "confirm that all requested changes to the replication package.*The URL to access the deposit is"
        then
            # Replace existing placeholder URL - use more specific pattern to avoid spurious matches
            sed -i "s|\(confirm that all requested changes to the replication package.*The URL to access the deposit is <\)https://www.openicpsr.org/openicpsr/workspace?goToPath=/openicpsr/[^>]*|\1$openicpsr_url|g" REPLICATION.md
            echo "openICPSR URL updated in REPLICATION.md"
        else
            # Template line not found, inject full text before "### Action Items (openICPSR)"
            sed -i "/### Action Items (openICPSR)/i\\
> [REQUIRED] When returning proofs, confirm that all requested changes to the replication package, as outlined in the \"Actions Items (openICPSR)\"  section of the Data Editor report, have been completed. The URL to access the deposit is <$openicpsr_url>.\\
\\
" REPLICATION.md
            echo "openICPSR URL line injected into REPLICATION.md"
        fi
    else
        echo "No openICPSR URL found in JIRA"
    fi
else
    echo "jira_get_info.py not found in tools/ subdirectory, skipping openICPSR URL injection"
    sleep 1
fi

# Query JIRA API for DCAF_Access_Restrictions_V2 and inject private data notice if applicable
if [[ -f "tools/jira_get_info.py" ]]
then
    echo "Querying JIRA for DCAF private data restriction..."
    dcaf_private=$($pythonbin tools/jira_get_info.py aearep-$num dcaf_private 2>/dev/null)

    if [[ -n "$dcaf_private" && "$dcaf_private" == "yes" ]]
    then
        echo "Found DCAF private data restriction"

        # Check if the notice already exists to avoid duplication
        if ! grep -q "The article uses restricted-access data, which the authors provided privately not-for-publication" REPLICATION.md
        then
            # Find the line starting with "In assessing compliance with our" and insert notice above it
            if grep -q "^In assessing compliance with our" REPLICATION.md
            then
                sed -i "/^In assessing compliance with our/i\\
> The article uses restricted-access data, which the authors provided privately not-for-publication. We will be in touch separately to confirm deletion of the data.\\
\\
" REPLICATION.md
                echo "Private data restriction notice added to REPLICATION.md"
            else
                echo "WARNING: Could not find 'In assessing compliance with our' line in REPLICATION.md"
            fi
        else
            echo "Private data restriction notice already exists in REPLICATION.md"
        fi
    else
        echo "No DCAF private data restriction found in JIRA"
    fi
else
    echo "jira_get_info.py not found in tools/ subdirectory, skipping DCAF private data check"
    sleep 1
fi

# Update timestamp directly in REPLICATION.md
timestamp=$(date '+%B %d, %Y  %H:%M %Z')
# Remove any existing timestamp lines
sed -i '/^\*Report last created on.*\*$/d' REPLICATION.md
# Add new timestamp before "Some useful links:" line
sed -i "/^> Some useful links:/i\\
*Report last created on $timestamp*\\
\\
" REPLICATION.md 

# convert to PDF, just in case
# requires: wkhtmltopdf and pandoc

if [[ "$nopdf" == "0" ]]
then
   case $pdfengine in
      "")
         echo "Using pandoc to generate HTML, then wkhtmltopdf to generate PDF"
         outfile=REPLICATION.html
         PANDOCXTRA=""
         ;;
      "docker")
         echo "Using pandoc to generate HTML, then Docker to generate PDF"
         outfile=REPLICATION.html
         PANDOCXTRA=""
         ;;
      *)
         echo "Using pandoc and wkhtmltopdf to generate PDF"
         outfile=REPLICATION.pdf
         PANDOCXTRA="$WKHTMLOPTS"
         ;;
   esac
   $pandoc $PANDOCOPTS REPLICATION.md -f gfm -t html5 -o $outfile --metadata pagetitle="Report for AEAREP-$num" $PANDOCXTRA 
   exit_on_error $? "Pandoc failed to generate $outfile"
   # now generate the PDF
   if [[ "$pdfengine" == "" ]]
   then
      $wkhtmltopdf $WKHTMLOPTS $outfile REPLICATION.pdf
      exit_on_error $? "wkhtmltopdf failed to generate REPLICATION.pdf"
   elif [[ "$pdfengine" == "docker" ]]
   then
      echo "Preparing for Docker execution..."
      echo "Pulling latest Docker image..."
      $dockerbin pull $DOCKERIMG || echo  "Failed to pull Docker image $DOCKERIMG"
      # Set directory permissions to allow Docker (others) to access
      chmod o+rwx .
      chmod o+r .[a-z]*css
      chmod -R o+rX *
      # Remove any pre-existing REPLICATION.pdf
      rm -f REPLICATION.pdf
      $dockerbin run -it --rm --user 10042:999 -v $(pwd):/home/pptruser/project -w /home/pptruser/project $DOCKERIMG
      exit_on_error $? "Docker failed to generate REPLICATION.pdf"
   fi
else
   echo "Skipping PDF generation as requested"
fi

rm $CSSLOCAL
[[ -f $LOGOLOCAL ]] && rm $LOGOLOCAL
[[ -f REPLICATION.html ]] && rm REPLICATION.html

# No need to restore from backup - timestamp is now permanent in REPLICATION.md

echo "-------- Current git status -----------"
git status

echo "---------------------------------------"
git add REPLICATION.md
git add REPLICATION.pdf
if [[ -f "for openICPSR.md" ]]
then 
   # check for standard language in ICPSR file
   grep ScholarOne "for openICPSR.md" 1>/dev/null
   icpsr_check=$?
   case $icpsr_check in
      0)
      # tag is already in
      echo "ICPSR file is ready for addition"
      ;;
      *)
      # tag needs to be added
      echo "" >> "for openICPSR.md"
      echo "" >> "for openICPSR.md"
      echo "$ICPSRMSG" | sed "s/xxx/$num/" >> "for openICPSR.md"
      echo "---------------------------------------"
      cat "for openICPSR.md"
      echo "---------------------------------------"
      echo "::: Does this look fine?"
      read
      ;;
   esac
   git add "for openICPSR.md"
   commitmsg="$commitmsg with notes for openICPSR"
fi
echo "-------- Updated git status -----------"
git status

echo "::: Ready to commit/ push? "
echo "::: Msg:"
echo "    $commitmsg"
read
git commit -m "$commitmsg"
git push

# Now modify the Jira issue accordingly. Ask first, and then call jira-approval-manager
echo "::: Ready to update Jira issue AEAREP-$num ? (Ctrl-C to abort)"
read

# check first if the script is in the path
if ! command -v jira-approval-manager &> /dev/null
then
   echo "jira-approval-manager could not be found in PATH. Not modifying Jira issue."
   echo "You may want to call   aeaopen   ."
   exit 0
fi
set -evx
jira-approval-manager aearep-$num $decision 0