directory-layout.txt

├── documents
│   ├── Assets
│   │   ├── Banks
│   │   │   ├── Bank-of-USA <--- *.pdf, *.jpg, etc. (documents)
│   ├── Expenses
├── ingest
|   ├── ingest_output
│   ├── filingcabinet <--- bean-file's root directory
│   │   ├── Assets
│   │   │   ├── Banks
│   │   │   │   ├── Bank-of-USA <--- *.ofx, *.csv, *.xlsx, etc.
│   │   │   ├── ...
│   │   ├── Income
│   │   │   └── Employment
│   │   └── Liabilities
│   │       └── Credit-Cards
│   │           └── ...
├── prices <--- prices.bc
├── main.bc <--- include "source/Assets.Banks.xyz.bc", etc.
├── source <--- Assets.Banks.[...].bc, ...

process_all_files.zsh

#!/bin/zsh
setopt extendedglob
# exit when any command fails
# set -e
# keep track of the last executed command
trap 'last_command=$current_command; current_command=$BASH_COMMAND' DEBUG
# echo an error message before exiting
trap 'echo "\"${last_command}\" command filed with exit code $?."' EXIT

# https://stackoverflow.com/questions/192249/how-do-i-parse-command-line-arguments-in-bash
while [[ "$#" -gt 0 ]]; do
    case $1 in
        -x|--unimplemented) target="$2"; shift ;;  # just for documentation
        -n|--nofile) nofile=1 ;;
        *) echo "Unknown parameter passed: $1"; exit 1 ;;
    esac
    shift
done

INGEST_ROOT=$BEAN_ROOT/ingest
INGEST_OUTPUT=$INGEST_ROOT/ingest_output
ROOT_DIR=~/Downloads/
SCRIPT_DIR=$(dirname -- "$0")

alias bi="bean-identify ${INGEST_ROOT}/my.import "
alias be="bean-extract -o --dir ${INGEST_OUTPUT} ${INGEST_ROOT}/my.import "
alias bf="bean-file -o ${INGEST_ROOT}/filingcabinet ${INGEST_ROOT}/my.import "

files=("${(@f)$(find $ROOT_DIR -maxdepth 1 -iname "*.qfx" -o -iname "*.ofx" -o -iname "*.xml" -o -iname "*.csv" -o -iname "*.xlsx")}")
echo "Ingesting ${#files[@]} files."

rm -frv ${INGEST_OUTPUT}
mkdir -p ${INGEST_OUTPUT}

# So zerosum doesn't run: both for performance and correctness (smart_importer)
if be -f <(echo 'plugin "beancount.plugins.auto_accounts"'; cat ${INGEST_ROOT}/../source/* ) $files ; then
  echo "Return value of bean-extract: $?"
  if [[ "$nofile"x != "1x" ]]; then
    bf $files
  fi
fi


echo "-----------------------------------"
echo "${#files[@]} files processed. Creating prices.db."

# Optional: Collect all price entries in a separate prices.bc file
touch ${INGEST_OUTPUT}/noaccount.bc
grep price ${INGEST_OUTPUT}/noaccount.bc >> ${BEAN_ROOT}/prices/prices.bc || echo ''
echo 'price' >> ${INGEST_OUTPUT}/noaccount.bc
echo '' >> ${INGEST_OUTPUT}/noaccount.bc
sed -i '/price/d' ${INGEST_OUTPUT}/noaccount.bc
sed -i '/^$/d' ${INGEST_OUTPUT}/noaccount.bc
[[ ! -s ${INGEST_OUTPUT}/noaccount.bc ]] && rm -fv ${INGEST_OUTPUT}/noaccount.bc

echo "Import done. Editing all imported files."
vim -p ${INGEST_OUTPUT}/**/*.bc
echo "Append to master? (Ctrl+c to exit)"
read
${INGEST_ROOT}/append_to_master.sh
rmdir --ignore-fail-on-non-empty ~/Downloads/skipped

I've been a bit confused about ingest_output. Is this a directory that would be in directory-layout next to filingcabinet? Is this where the main.beancount would be?

The -o --dir options of bean-extract are erroring out for me so I've been trying something like be -f $files > $(INGEST_OUTPUT)/test.beancount

ingest_output is a sibling of filingcabinet. I've updated the gist to illusrate this, and main.bc

-o and --dir need a patch to be applied. I have the patch in another gist. Search for patch in this article.

Hope that helps!