container: migrate from xsv to xan

`xsv` is no longer maintained and the author recommends `xan` instead. Fortunately, `xan` is very suitable for our use-case; is very efficient, and produces no differences in our expected output.
2025-09-30 17:43:25 -07:00
parent 5e02aa4825
commit 3383b5c73f
9 changed files with 31 additions and 31 deletions
--- a/container/src/finance/lib/internal/lib_meta.bash
+++ b/container/src/finance/lib/internal/lib_meta.bash
@@ -65,7 +65,7 @@ function lib_meta::__parse_args()

 \e[32mAvailable columns:\e[0m

-$(echo $global_meta_header | xsv headers -)
+$(echo $global_meta_header | xan headers -)

 \e[32mExamples:\e[0m

@@ -123,7 +123,7 @@ function lib_meta::__meta()
    # Nth iteration (select column and pattern)
    local _tmp_file
    _tmp_file="$(mktemp -p $_tmp_dir ${_arg}_XXX)"
-    xsv search -i -s \
+    xan search -i -s \
      "${_arg%${global_arg_delim_2}*}" "${_arg#*${global_arg_delim_2}}" \
      "$_base_file" >"$_tmp_file"

--- a/container/src/finance/lib/internal/lib_taxes.bash
+++ b/container/src/finance/lib/internal/lib_taxes.bash
@@ -430,7 +430,7 @@ function lib_taxes::__taxes_print()

      if [[ "$_tag" != "MATCH" && "$_tag" != "PARTIAL_TRADE" ]]; then
        "${_hledger[@]}" \
-          | xsv select '"posting-comment"' \
+          | xan select '"posting-comment"' \
          | tail -n +2 \
          | sed -e 's:"::g' -e '/^$/d' \
          | gawk -v tag="$_tag" -v is_trades="$_is_trades" \
@@ -607,7 +607,7 @@ function lib_taxes::__taxes_print()

      if [[ "$_tag" == "MATCH" ]]; then
        "${_hledger[@]}" \
-          | xsv select '"posting-comment"' \
+          | xan select '"posting-comment"' \
          | tail -n +2 \
          | sed -e 's:"::g' -e '/^$/d' \
          | sort -u \
@@ -856,7 +856,7 @@ function lib_taxes::__taxes_print()

      if [[ "$_tag" == "PARTIAL_TRADE" ]]; then
        "${_hledger[@]}" \
-          | xsv select '"posting-comment"' \
+          | xan select '"posting-comment"' \
          | tail -n +2 \
          | sed -e 's:"::g' -e '/^$/d' \
          | sort -u \
@@ -1002,7 +1002,7 @@ function lib_taxes::__taxes_write()
  if [ -z "$global_arg_year" ]; then
    lib_utils::print_normal "Capturing the year of the oldest tagged record ..."

-    local -r _oldest_year="$("${_base_hledger[@]}" | head -n2 | xsv select \"date\" | tail -n1 | cut -d'-' -f1)"
+    local -r _oldest_year="$("${_base_hledger[@]}" | head -n2 | xan select \"date\" | tail -n1 | cut -d'-' -f1)"
    [ -z "$_oldest_year" ] && lib_utils::die_fatal "No records available to write"

  elif [[ "$global_arg_year" -gt "$_current_year" ]]; then
@@ -1247,7 +1247,7 @@ function lib_taxes::__reports_patch_verify()
  [ ! -f "$_file" ] && lib_utils::die_fatal "File not found: '${_file}'"

  if [ -s "$_file" ]; then
-    xsv select "Action" "$_file" \
+    xan select "Action" "$_file" \
      | sort -u \
      | tail -n +2 \
      | while read _line; do
@@ -1295,7 +1295,7 @@ function lib_taxes::__reports_obfs_gen()

    # Generate obfs/raw keymap
    local _raw_column
-    _raw_column="$(xsv select $_column $_in_file | tail -n +2 | sort -u)"
+    _raw_column="$(xan select $_column $_in_file | tail -n +2 | sort -u)"

    local _count
    _count=$(echo "$_raw_column" | wc -l)
@@ -1352,7 +1352,7 @@ function lib_taxes::__reports_obfs_gen()

    # Generate a temp file of the given raw column
    local _tmp_file="${_tmp_dir}/${_styled_column}"
-    xsv select "$_column" "$_in_file" -o "$_tmp_file"
+    xan select "$_column" "$_in_file" -o "$_tmp_file"

    # Obfuscate given raw column with keymap data
    tail -n +2 "$_keymap_file" \
@@ -1364,7 +1364,7 @@ function lib_taxes::__reports_obfs_gen()
  # Join the obfuscated temp files into out file and then delete the temp files.
  # NOTE: selecting will remove the raw (unobfuscated) columns
  csvjoin -I --snifflimit 0 "${_tmp_dir}"/* "$_in_file" \
-    | xsv select "$_all_columns" -o "$_out_file"
+    | xan select "$_all_columns" -o "$_out_file"
 }

 # vim: sw=2 sts=2 si ai et
--- a/container/src/hledger-flow/accounts/bittrex/bittrex-shared.bash
+++ b/container/src/hledger-flow/accounts/bittrex/bittrex-shared.bash
@@ -2,7 +2,7 @@

 # docker-finance | modern accounting for the power-user
 #
-# Copyright (C) 2021-2024 Aaron Fiore (Founder, Evergreen Crypto LLC)
+# Copyright (C) 2021-2025 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
@@ -43,7 +43,7 @@ source "${DOCKER_FINANCE_CONTAINER_REPO}/src/hledger-flow/lib/lib_preprocess.bas
 function parse_deposit()
 {
  # NOTE: headers can be variable, not possible to assert
-  xsv select "id,completedAt,currencySymbol,quantity,cryptoAddress,txId" "$global_in_path" \
+  xan select "id,completedAt,currencySymbol,quantity,cryptoAddress,txId" "$global_in_path" \
    | gawk -v global_year="$global_year" -v global_subaccount="$global_subaccount" \
      '{
         if (NR<2 || $2 !~ global_year)
@@ -79,7 +79,7 @@ function parse_trade()
  #    WARNING: don't use limit because market orders will be empty

  # NOTE: headers can be variable, not possible to assert
-  xsv select "id,closedAt,direction,marketSymbol,fillQuantity,proceeds,commission" "$global_in_path" \
+  xan select "id,closedAt,direction,marketSymbol,fillQuantity,proceeds,commission" "$global_in_path" \
    | gawk -v global_year="$global_year" -v global_subaccount="$global_subaccount" \
      '{
         if (NR<2 || $2 !~ global_year)
@@ -130,7 +130,7 @@ function parse_trade()
 function parse_withdrawal()
 {
  # NOTE: headers can be variable, not possible to assert
-  xsv select "id,completedAt,currencySymbol,quantity,txCost,cryptoAddress,txId" "$global_in_path" \
+  xan select "id,completedAt,currencySymbol,quantity,txCost,cryptoAddress,txId" "$global_in_path" \
    | gawk -v global_year="$global_year" -v global_subaccount="$global_subaccount" \
      '{
         if (NR<2 || $2 !~ global_year)
--- a/container/src/hledger-flow/accounts/celsius/celsius-shared.bash
+++ b/container/src/hledger-flow/accounts/celsius/celsius-shared.bash
@@ -2,7 +2,7 @@

 # docker-finance | modern accounting for the power-user
 #
-# Copyright (C) 2021-2024 Aaron Fiore (Founder, Evergreen Crypto LLC)
+# Copyright (C) 2021-2025 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
@@ -44,7 +44,7 @@ function parse_ui()
 {
  lib_preprocess::assert_header "Internal id, Date and time, Transaction type, Coin type, Coin amount, USD Value, Original Reward Coin, Reward Amount In Original Coin, Confirmed"

-  xsv search -s " Date and time" "$global_year" "$global_in_path" \
+  xan search -s " Date and time" "$global_year" "$global_in_path" \
    | sed -r 's/ ([0-9]+),/ \1/' \
    | gawk -v global_subaccount="$global_subaccount" -M -v PREC=100 \
      '{
@@ -116,8 +116,8 @@ function parse_api()
 {
  lib_preprocess::assert_header "id,amount,amount_precise,amount_usd,coin,state,nature,time,tx_id,original_interest_coin,interest_amount_in_original_coin"

-  xsv select "id,time,nature,coin,amount_precise,amount_usd,original_interest_coin,interest_amount_in_original_coin,state,tx_id" "$global_in_path" \
-    | xsv search -s "time" "$global_year" \
+  xan select "id,time,nature,coin,amount_precise,amount_usd,original_interest_coin,interest_amount_in_original_coin,state,tx_id" "$global_in_path" \
+    | xan search -s "time" "$global_year" \
    | gawk -v global_subaccount="$global_subaccount" -M -v PREC=100 \
      '{
         if (NR<2)
--- a/container/src/hledger-flow/accounts/coinbase-pro/coinbase-pro-shared.bash
+++ b/container/src/hledger-flow/accounts/coinbase-pro/coinbase-pro-shared.bash
@@ -2,7 +2,7 @@

 # docker-finance | modern accounting for the power-user
 #
-# Copyright (C) 2021-2024 Aaron Fiore (Founder, Evergreen Crypto LLC)
+# Copyright (C) 2021-2025 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
@@ -57,7 +57,7 @@ function parse_trade()
  cat "$global_in_path" \
    | if [ ! -z "$_details_transfer_id" ]; then csvjoin -I --snifflimit 0 - <(echo "$_details_transfer_id"); else cat; fi \
    | if [ ! -z "$_details_transfer_type" ]; then csvjoin -I --snifflimit 0 - <(echo "$_details_transfer_type"); else cat; fi \
-    | xsv select "id,amount,balance,created_at,type,details_transfer_id,details_transfer_type,details_order_id,details_product_id,details_trade_id" \
+    | xan select "id,amount,balance,created_at,type,details_transfer_id,details_transfer_type,details_order_id,details_product_id,details_trade_id" \
    | gawk -v account_id="$account_id" -v account_currency="$account_currency" \
      -v global_year="$global_year" -v global_subaccount="$global_subaccount" \
      '{
--- a/container/src/hledger-flow/accounts/coinbase/coinbase-shared.bash
+++ b/container/src/hledger-flow/accounts/coinbase/coinbase-shared.bash
@@ -60,7 +60,7 @@ function add_to_headers()
 function join_to_header()
 {
  # WARNING: csvjoin *MUST* use -I or else satoshis like 0.00000021 will turn into 21000000 BTC!
-  # TODO: `xsv join` is significantly faster but does not produce same results
+  # TODO: will `xan join` be faster and produce the same results?
  csvjoin -I --snifflimit 0 "$1" <(echo "$2")
 }

@@ -318,9 +318,9 @@ function parse()

  # NOTE: prepends account_id to header (this will now be the first column)

-  xsv select "$selected_header" "$global_in_path" \
+  xan select "$selected_header" "$global_in_path" \
    | join_to_header - "$universal_header" \
-    | xsv select "$universal_header" \
+    | xan select "$universal_header" \
    | sed -e "s:^:${_account_id},:g" -e "1 s:^${_account_id},:account_id,:g" \
    | __parse >"$global_out_path"
 }
--- a/container/src/hledger-flow/accounts/gemini/gemini-shared.bash
+++ b/container/src/hledger-flow/accounts/gemini/gemini-shared.bash
@@ -45,7 +45,7 @@ source "${DOCKER_FINANCE_CONTAINER_REPO}/src/hledger-flow/lib/lib_preprocess.bas
 function parse_transfers()
 {
  # NOTE: header can be variable, ineffective to assert_header
-  # TODO: optimize: do successive test_headers and whichever wins, use that string so `xsv select` is avoided
+  # TODO: optimize: do successive test_headers and whichever wins, use that string so `xan select` is avoided
  local _header=""

  if lib_preprocess::test_header "info_feeCurrency"; then
@@ -59,7 +59,7 @@ function parse_transfers()
    # Off-chain (withdrawal)
    lib_preprocess::test_header "info_method" && _header+=",info_method"

-    xsv select "$_header" "$global_in_path" \
+    xan select "$_header" "$global_in_path" \
      | gawk -v global_year="$global_year" -v global_subaccount="$global_subaccount" -M -v PREC=100 \
        '{
           if (NR<2)
@@ -130,7 +130,7 @@ function parse_transfers()
    # Off-chain withdrawal
    lib_preprocess::test_header "info_method" && _header+=",info_method"

-    xsv select "$_header" "$global_in_path" \
+    xan select "$_header" "$global_in_path" \
      | gawk -v global_year="$global_year" -v global_subaccount="$global_subaccount" -M -v PREC=100 \
        '{
           if (NR<2)
--- a/container/src/hledger-flow/accounts/paypal-business/paypal-business-shared.bash
+++ b/container/src/hledger-flow/accounts/paypal-business/paypal-business-shared.bash
@@ -41,8 +41,8 @@ function parse()
  lib_preprocess::assert_header "$_header"

  # Paypal is known to allow commas within description ("Name") and amounts
-  # TODO: refactor xsv/sed -> gawk (the sed line should remove non-csv commas and quotations)
-  xsv select "$(echo "$_header" | sed 's:"::g')" $global_in_path \
+  # TODO: refactor xan/sed -> gawk (the sed line should remove non-csv commas and quotations)
+  xan select "$(echo "$_header" | sed 's:"::g')" $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" \
      '{
--- a/container/src/hledger-flow/accounts/paypal/paypal-shared.bash
+++ b/container/src/hledger-flow/accounts/paypal/paypal-shared.bash
@@ -47,8 +47,8 @@ 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 xsv/sed -> gawk (the sed line should remove non-csv commas and quotations)
-  xsv select "Date,Time,Name,Type,Status,Currency,Amount,Fees,Total,Exchange Rate,Receipt ID,Balance,Transaction ID,Item Title" "$global_in_path" \
+  # TODO: refactor xan/sed -> gawk (the sed line should remove non-csv commas and quotations)
+  xan select "Date,Time,Name,Type,Status,Currency,Amount,Fees,Total,Exchange Rate,Receipt ID,Balance,Transaction ID,Item Title" "$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" \
      '{
@@ -121,7 +121,7 @@ function parse_crypto()
  #lib_preprocess::assert_header "DateTime,Transaction Type,Asset In (Quantity),Asset In (Currency),Asset Out (Quantity),Asset Out (Currency),Transaction Fee (Quantity),Transaction Fee (Currency),Market Value (USD)" "$(sed -n '4p' $global_in_path)"

  # TODO: refactor into gawk after assert_header is fixed
-  xsv 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 "$global_in_path") \
+  xan select "DateTime,Transaction Type,Asset In (Quantity),Asset In (Currency),Asset Out (Quantity),Asset Out (Currency),Transaction Fee (Quantity),Transaction Fee (Currency),Market Value (USD)" <(tail +4 "$global_in_path") \
    | gawk -v global_year="$global_year" -v global_subaccount="$global_subaccount" \
      '{
         if (NR<2 || $1 !~ global_year)