From dae2880c061d30e4f05759e2e1b2b2d4bab44c19 Mon Sep 17 00:00:00 2001
From: Aaron Fiore <founder@evergreencrypto.co>
Date: Fri, 13 Feb 2026 11:32:37 -0800
Subject: [PATCH] container: hledger-flow: paypal: add uniform quotes

- Allows undesirable characters in variable-text columns
- Related refactoring
---
 .../accounts/paypal/paypal-shared.bash        | 67 ++++++++++++-------
 1 file changed, 42 insertions(+), 25 deletions(-)

diff --git a/container/src/hledger-flow/accounts/paypal/paypal-shared.bash b/container/src/hledger-flow/accounts/paypal/paypal-shared.bash
index 03a67d5..8d2dd14 100755
--- a/container/src/hledger-flow/accounts/paypal/paypal-shared.bash
+++ b/container/src/hledger-flow/accounts/paypal/paypal-shared.bash
@@ -2,7 +2,7 @@
 
 # docker-finance | modern accounting for the power-user
 #
-# Copyright (C) 2021-2025 Aaron Fiore (Founder, Evergreen Crypto LLC)
+# Copyright (C) 2021-2026 Aaron Fiore (Founder, Evergreen Crypto LLC)
 #
 # This program is free software: you can redistribute it and/or modify
 # it under the terms of the GNU General Public License as published by
@@ -47,10 +47,10 @@ function parse_report()
 
   # Paypal is known to allow commas within description ("Name") and amounts
   # NOTE: using custom timezone offset instead of provided timezone
-  # TODO: refactor xan/sed -> gawk (the sed line should remove non-csv commas and quotations)
   xan select "Date,Time,Name,Type,Status,Currency,Amount,Fees,Total,Exchange Rate,Receipt ID,Balance,Transaction ID,Item Title" "$global_in_path" \
-    | sed -e 's:, : :g' -e 's:,",:,:g' -e 's:,"\([0-9]*\),:,\1:g' -e 's:,"-\([0-9]*\),:,-\1:g' -e 's:"::g' \
-    | gawk -v global_year="$global_year" -v global_subaccount="$global_subaccount" \
+    | gawk --csv \
+      -v global_year="$global_year" \
+      -v global_subaccount="$global_subaccount" \
       '{
          if (NR<2 || $1 !~ global_year)
            next
@@ -71,32 +71,44 @@ function parse_report()
          # Print
          printf timestamp OFS  # timestamp
 
+         # description (Name)
          sub(/%/, "%%", $3)
-         printf $3 OFS         # description (Name)
+         gsub(/"/, "", $3)
+         printf "\"" $3 "\"" OFS
 
-         printf $4 OFS         # type
-         printf $5 OFS         # status
-         printf $6 OFS         # currency
+         printf "\"" $4 "\"" OFS  # type
+         printf "\"" $5 "\"" OFS  # status
+         printf "\"" $6 "\"" OFS  # currency
 
+         # amount
          direction=($7 ~ /^-/ ? "OUT" : "IN")
          sub(/^-/, "", $7)
-         printf $7 OFS         # amount
+         gsub(/,/, "", $7)
+         printf "\"" $7 "\"" OFS
 
+         # fees
          sub(/^-/, "", $8)
-         printf $8 OFS         # fees
+         gsub(/,/, "", $8)
+         printf "\"" $8 "\"" OFS
 
+         # total
          sub(/^-/, "", $9)
-         printf $9 OFS         # total
+         gsub(/,/, "", $9)
+         printf "\"" $9 "\"" OFS
 
-         printf $10 OFS        # exchange rate
+         # exchange rate
+         gsub(/,/, "", $10)
+         printf "\"" $10 "\"" OFS
 
-         printf $11 OFS        # code (Receipt ID)
-         printf $12 OFS        # balance
+         printf "\"" $11 "\"" OFS  # code (Receipt ID)
+         printf "\"" $12 "\"" OFS  # balance
 
-         printf $13 OFS        # txid (Transaction ID)
+         printf $13 OFS            # txid (Transaction ID)
 
+         # title (Item Title)
          sub(/%/, "%%", $14)
-         printf $14 OFS        # title (Item Title)
+         gsub(/"/, "", $14)
+         printf "\"" $14 "\"" OFS
 
          printf direction OFS
          printf global_subaccount OFS
@@ -110,7 +122,7 @@ function parse_report()
                      # fiat_value
 
          printf "\n"
-      }' FS=, OFS=, >"$global_out_path"
+      }' OFS=, >"$global_out_path"
 }
 
 # NOTE: crypto transactions CSV, not crypto statement CSV
@@ -122,7 +134,9 @@ function parse_crypto()
 
   # TODO: refactor into gawk after assert_header is fixed
   xan select "DateTime,Transaction Type,Asset In (Quantity),Asset In (Currency),Asset Out (Quantity),Asset Out (Currency),Transaction Fee (Quantity),Transaction Fee (Currency),Market Value (USD)" <(tail +4 "$global_in_path") \
-    | gawk -v global_year="$global_year" -v global_subaccount="$global_subaccount" \
+    | gawk --csv \
+      -v global_year="$global_year" \
+      -v global_subaccount="$global_subaccount" \
       '{
          if (NR<2 || $1 !~ global_year)
            next
@@ -150,16 +164,19 @@ function parse_crypto()
          printf OFS            # direction
          printf global_subaccount OFS
 
-         printf $3 OFS  # in_value
-         printf $4 OFS  # in_ticker
-         printf $5 OFS  # out_value
-         printf $6 OFS  # out_ticker
-         printf $7 OFS  # tx_fee_value
-         printf $8 OFS  # tx_fee_ticker
+         printf $3 OFS            # in_value
+         printf "\"" $4 "\"" OFS  # in_ticker
+
+         printf $5 OFS            # out_value
+         printf "\"" $6 "\"" OFS  # out_ticker
+
+         printf $7 OFS            # tx_fee_value
+         printf "\"" $8 "\"" OFS  # tx_fee_ticker
+
          printf $9      # fiat_value
 
          printf "\n"
-      }' FS=, OFS=, >"$global_out_path"
+      }' OFS=, >"$global_out_path"
 }
 
 function main()