Upgrade deps, note use of upstream Clojure

Previously import/read-csv opened a file on the filesystem, making it hard to
test. Now it takes a string or java.io.Reader.

The warning about non-zero records with category "PFML ER PU" has now been
generalised to flag any non-zero records with an unknown category.

README.md

@@ -18,13 +18,13 @@ Run a demo with two example employees, Jack and Jill Citizen:
 
 Provide your own payroll data with:
 
-    java -jar payroll-importer-x.x.x-standalone.jar --csv resources/example-paychex-pay-item-details.csv --total-fees 206.50
+    java -jar payroll-importer-x.x.x-standalone.jar --total-fees 206.50 resources/example-paychex-pay-item-details.csv
 
 In the above, various values such as the date, time period covered and
 receipt/invoice values show "TODO" placeholders that you are expected to fill in
 later. If you prefer, you can provide any/all of these explicitly:
 
-    java -jar payroll-importer-x.x.x-standalone.jar --csv resources/example-paychex-pay-item-details.csv --date 2023-12-29 --period 'December 2023' --total-fees 206.50 --pay-receipt-no rt:19462/674660 --pay-invoice-no rt:19403/675431 --fees-receipt-no rt:19459/675387 --fees-invoice-no rt:19459/674887 --retirement-receipt-no rt:19403/676724 --retirement-invoice-no rt:19403/675431
+    java -jar payroll-importer-x.x.x-standalone.jar --date 2023-12-29 --period 'December 2023' --total-fees 206.50 --pay-receipt-no rt:19462/674660 --pay-invoice-no rt:19403/675431 --fees-receipt-no rt:19459/675387 --fees-invoice-no rt:19459/674887 --retirement-receipt-no rt:19403/676724 --retirement-invoice-no rt:19403/675431 resources/example-paychex-pay-item-details.csv
 
 You can test the output in Beancount by adding the following header entries to define the accounts:
 
@@ -64,7 +64,7 @@ Run tests with:
 
 You can run without building using:
 
-    bin/dev --csv resources/example-paychex-pay-item-details.csv --total-fees 206.50
+    bin/dev --total-fees 206.50 resources/example-paychex-pay-item-details.csv
 
 The project is set up for development in Emacs and CIDER-mode. Open a source
 file and run `cider-jack-in`.

bin/dev

@@ -1,4 +1,4 @@
 #!/usr/bin/env sh
 
 # Run the program without building it
-clojure -M -m core "$@"
+clojure -M:dev "$@"

build.clj

@@ -14,8 +14,6 @@
 
 (defn uber [_]
   (clean nil)
-  (b/copy-dir {:src-dirs ["src" "resources"]
-               :target-dir class-dir})
   (b/copy-dir {:src-dirs ["src" "resources"]
                :target-dir class-dir})
   (b/compile-clj {:basis @basis

deps.edn

@@ -1,12 +1,19 @@
 {:paths ["src" "resources" "private"]  ;; Private is not included in the build
  :deps {
-        org.clojure/clojure {:mvn/version "1.11.1"}
-        org.clojure/data.csv {:mvn/version "1.0.1"}
-        org.clojure/tools.cli {:mvn/version "1.1.230"}}
+        ;; Note that the clojure version below will be installed from upstream
+        ;; rather than using the system version.
+        org.clojure/clojure {:mvn/version "1.12.4"}
+        org.clojure/data.csv {:mvn/version "1.1.1"}
+        org.clojure/tools.cli {:mvn/version "1.3.250"}}
  :aliases
- {:dev {:extra-deps {lambdaisland/deep-diff2 {:mvn/version "2.10.211"}}}
+ {:dev {:extra-deps {lambdaisland/deep-diff2 {:mvn/version "2.12.219"}}
+        :main-opts ["-m" "core"]
+        ;; Saves ~ 1 second of startup time - 1.5 sec on my laptop. After
+        ;; building an uberjar and running with these options, it drops to about
+        ;; 750ms.
+        :jvm-opts ["-XX:TieredStopAtLevel=1" "-XX:+TieredCompilation"]}
   :test {:extra-deps {lambdaisland/kaocha {:mvn/version "1.87.1366"}}
          :main-opts ["-m" "kaocha.runner"]}
-    ;; Run with clj -T:build function-in-build
-  :build {:deps {io.github.clojure/tools.build {:mvn/version "0.9.6"}}
+  ;; Run with clj -T:build function-in-build
+  :build {:deps {io.github.clojure/tools.build {:mvn/version "0.10.12"}}
           :ns-default build}}}

manifest.scm

@@ -4,10 +4,10 @@
 ;; develop/build the program. Use it with `guix shell --manifest=manifest.scm`.
 (specifications->manifest
  (list
-  ;; No issues running this OpenJDK 21 program on Debian Stable (OpenJDK 17).
-  "openjdk@21"
-  ;; Works fine with clojure-tools from Guix.
+  ;; No issues running this OpenJDK 21 program on Debian Bookworm (OpenJDK 17).
+  "openjdk@21:jdk"
   "clojure-tools"
+  "rlwrap"
   "clj-kondo"
   "beancount"
   ))

src/core.clj

@@ -6,17 +6,15 @@
   (:require [clojure.java.io :as io]
             [clojure.set :as set]
             [clojure.string :as str]
-            [clojure.tools.cli :refer [parse-opts]]
+            [clojure.tools.cli :as cli]
             [import :as import])
   (:gen-class))
 
 (def cli-options
-  [[nil "--csv FILE" "Pay Item Details CSV report"
-    :validate [#(-> % io/file .exists) "File does not exist"]]
-   [nil "--date DATE" "Date used for the transactions (YYYY-MM-DD)"
+  [[nil "--date DATE" "Date used for the transactions (YYYY-MM-DD)"
     :validate [#(re-matches #"\d{4}-\d{2}-\d{2}" %) "Must be of format YYYY-MM-DD"]
     :default "TODO-DATE"]
-   [nil "--period PERIOD" "Month/year covered by the pay run eg. \"December 2023\""
+   [nil "--period PERIOD" "Month/year of the pay run eg. \"December 2023\""
     :default "TODO-PERIOD"]
    [nil "--total-fees NUM" "Total fee charged by Paychex, eg. \"206.50\""
     :parse-fn bigdec
@@ -33,13 +31,13 @@
     :default "TODO-RETIREMENT-RECEIPT"]
    [nil "--retirement-invoice-no REFERENCE" "Retirement receipt number, eg. \"rt:111/222\""
     :default "TODO-RETIREMENT-INVOICE"]
-   [nil "--project EMPLOYEE:PROJECT" "Allocate an employee to a specific project, eg. \"Doe-Jane:Outreachy\". Use once for each employee."
+   [nil "--project EMPLOYEE:PROJECT" "Allocate employee to project, eg. \"Doe-Jane:Outreachy\""
     :multi true
     :validate [#(= 2 (count %)) "Must be of the form \"name:project\""]
     :parse-fn #(str/split % #":")
     :default {}
     :assoc-fn (fn [m k [name proj]] (assoc-in m [k name] proj))]
-   [nil "--demo" "Produce demo output based made-up payroll data. Useful for documentation."]
+   [nil "--demo" "Produce demo output based made-up payroll data"]
    ["-h" "--help"]])
 
 (defn unmatched-employees
@@ -57,9 +55,10 @@
   "Run the import with a map of options."
   [options]
   (let [options (if (:demo options) (merge options demo-options) options)
-        {:keys [date period pay-receipt-no pay-invoice-no total-fees project]} options
+        {:keys [csv date period pay-receipt-no pay-invoice-no total-fees project]} options
         {:keys [fees-receipt-no fees-invoice-no retirement-receipt-no retirement-invoice-no]} options
-        records (import/read-csv (:csv options))
+        records (with-open [reader (io/reader csv)]
+                  (import/read-csv reader))
         imported (concat (import/net-pay date period pay-invoice-no project records)
                          (import/individual-taxes date period pay-invoice-no retirement-invoice-no project records)
                          (import/employer-taxes date period pay-invoice-no project records)
@@ -67,32 +66,49 @@
                          (import/taxes-ach-debit date period pay-receipt-no pay-invoice-no project records)
                          (import/fees date period fees-receipt-no fees-invoice-no total-fees project records)
                          (import/retirement date period retirement-receipt-no retirement-invoice-no records))
-        unmatched (unmatched-employees records project)]
-    [imported unmatched]))
+        unmatched (unmatched-employees records project)
+        warnings (import/warnings records)]
+    [imported unmatched warnings]))
+
+(defn usage [summary]
+  (str
+   "Usage: java -jar payroll-importer.jar [OPTIONS] [CSV]\n\n"
+    "Options include:\n\n"
+    summary
+    "\n\n"
+    "Use --project once for each employee."))
 
 (defn -main
   "Run the CLI interface."
   [& args]
-  (let [{:keys [options errors summary]} (parse-opts args cli-options)
-        no-csv-or-demo? (not (or (contains? options :csv) (contains? options :demo)))
+  (let [{:keys [options arguments errors summary]} (cli/parse-opts args cli-options)
+        csv (first arguments)
+        options (assoc options :csv csv)
+        demo? (contains? options :demo)
+        neither-csv-or-demo? (and (not csv) (not demo?))
+        csv-doesnt-exist? (and (not demo?) csv (not (-> csv io/file .exists)))
         errors (cond-> errors
-                 no-csv-or-demo? (conj "Please provide a CSV file with \"--csv FILE\" or try \"--demo\""))]
+                 neither-csv-or-demo? (conj "Please provide a CSV file argument or try \"--demo\"")
+                 csv-doesnt-exist? (conj (str "CSV file \"" csv "\" does not exist")))]
     (when (:help options)
-      (println summary)
+      (println (usage summary))
       (System/exit 0))
     (when errors
       (println
        (str "The following errors occurred:\n\n"
             (str/join \newline errors)))
       (System/exit 1))
-    (let [[imported unmatched] (run options)]
+    (let [[imported unmatched warnings] (run options)]
       (when (seq unmatched)
         (println
          (str "Could not find these employees in the payroll:\n\n"
               (str/join ", " unmatched)))
         (System/exit 1))
       (doseq [i imported]
-        (println (import/render-transaction i))))))
+        (println (import/render-transaction i)))
+      (when (seq warnings)
+        (.println System/err "WARNINGS:\n")
+        (.println System/err (str/join "\n" warnings))))))
 
 (comment
   ;; Examples to exercise the importer during development.
@@ -103,7 +119,9 @@
   ;; These examples are not included with the code for privacy reasons.
   (require '[examples])
 
-  (def records (import/read-csv "/home/ben/Downloads/2024-01-29_Pay-Item-Details_2024-01.csv"))
+  (def records (with-open [reader (io/reader "/home/ben/downloads/2026-01-28_Pay-Item-Details_2026-01.csv")]
+                 (import/read-csv reader)))
+
   (def imported
     (concat (import/net-pay "2024-01-31" "January 2024" "rt:19462/685751" {} records)
             (import/individual-taxes "2024-01-31" "January 2024" "rt:19462/685751" "rt:19403/685602" {} records)

src/import.clj

@@ -19,42 +19,67 @@
 
 (defn read-csv
   "Read in CSV and return a vector of maps with only the fields we want.
-  Merges the various number fields into a single \"amount\" field."
-  [filename]
-  (with-open [reader (io/reader filename)]
-    (doall
-     (for [[_ name _ category type & totals] (csv/read-csv reader)]
-       {:name (employee-name->entity-tag name)
-        :category category
-        :type type
-        :amount (apply max (map bigdec (remove str/blank? totals)))}))))
+  Merges the various number fields into a single \"amount\" field. Per
+  clojure.data.csv, `reader` can be a string or java.io.Reader."
+  [reader]
+  (doall
+   (for [[_ name _ category type & totals] (csv/read-csv reader)]
+     {:name (employee-name->entity-tag name)
+      :category category
+      :type type
+      :amount (reduce max 0 (map bigdec (remove str/blank? totals)))})))
+
+(def cat->payroll
+  {"Fed Income Tax" "US:Tax:Income"
+   "Medicare" "US:Tax:Medicare"
+   "IL Income Tax" "US:IL:Tax:Income"
+   "NY Income Tax" "US:NY:Tax:Income"
+   "NYC Income Tax" "US:NY:Tax:NYC"
+   "OR Income Tax" "US:OR:Tax:Income"
+   "OH Income Tax" "US:OH:Tax:Income"
+   "PNTSD Income Tax" "US:OH:Tax:PNTSD"
+   "COLMB Income Tax" "US:OH:Tax:COLUMB"
+   "Social Security" "US:Tax:SocialSecurity"
+   "NY Disability" "US:NY:Disability"
+   "OR Disability PFL" "US:OR:Disability:PFL"
+   "NY Disability PFL" "US:NY:Disability:PFL"
+   "OR TRANS STT" "US:OR:Tax:STT"
+   "Fed Unemploy" "US:Unemployment"
+   "IL Unemploy" "IL:Unemployment"
+   "NY Unemploy" "NY:Unemployment"
+   "OR Unemploy" "OR:Unemployment"
+   "NY Re-empl Svc" "US:NY:Reempt"})
 
 (defn- cat->payroll-type
   "Map the CSV withholding categories to Beancount payroll-type tags."
   [cat]
   ;; May need to become config, similar to names. Should prompt you if no
   ;; mapping exists to avoid mistakes.
-  (case cat
-    "Fed Income Tax" "US:Tax:Income"
-    "Medicare" "US:Tax:Medicare"
-    "IL Income Tax" "US:IL:Tax:Income"
-    "NY Income Tax" "US:NY:Tax:Income"
-    "NYC Income Tax" "US:NY:Tax:NYC"
-    "OR Income Tax" "US:OR:Tax:Income"
-    "OH Income Tax" "US:OH:Tax:Income"
-    "PNTSD Income Tax" "US:OH:Tax:PNTSD"
-    "COLMB Income Tax" "US:OH:Tax:COLUMB"
-    "Social Security" "US:Tax:SocialSecurity"
-    "NY Disability" "US:NY:Disability"
-    "OR Disability PFL" "US:OR:Disability:PFL"
-    "NY Disability PFL" "US:NY:Disability:PFL"
-    "OR TRANS STT" "US:OR:Tax:STT"
-    "Fed Unemploy" "US:Unemployment"
-    "IL Unemploy" "IL:Unemployment"
-    "NY Unemploy" "NY:Unemployment"
-    "OR Unemploy" "OR:Unemployment"
-    "NY Re-empl Svc" "US:NY:Reempt"
-    cat))
+  (get cat->payroll cat cat))
+
+(def known-category
+  "These are all CSV categories that we know about."
+  (into #{"403b ER match"
+          "403b EE Pretax"
+          "Exp Reimb Non Tax"
+          "Net Pay"
+          "Salary"}
+        (keys cat->payroll)))
+
+(defn non-zero-unknown-field?
+  "Detect non-zero unknown fields.
+  New field categories are probably safe enough if they have no amount or a zero
+  amount listed. If not, we probably need to investigate and figure out how to
+  incorporate these into the books."
+  [record]
+  (when (and (not (known-category (:category record)))
+             (not (zero? (:amount record))))
+    record))
+
+(defn warnings [records]
+  (->> records
+       (filter non-zero-unknown-field?)
+       (map #(str "Unexpected non-zero \"" (:category %) "\" category found for " (:name %)))))
 
 (defn- assoc-project
   "Conditionally adds a specific project tag to metadata.
@@ -167,7 +192,7 @@
   [date period pay-invoice-no retirement-invoice-no projects records]
   (for [[name employee-records] (group-by :name records)
         :let [template {:date date :desc (format "Monthly Payroll - %s - TAXES - %s" period name)
-                    :meta (assoc-project
+                        :meta (assoc-project
                            projects name
                            {:project "Conservancy"
                             :program "Conservancy:Payroll"

test/import_test.clj

@@ -3,19 +3,35 @@
 
 (ns import-test
   (:require [import]
-            [clojure.java.io]
+            [clojure.java.io :as io]
             [clojure.string :as str]
             [clojure.test :as t :refer [deftest is are]]))
 
+;; Run all our functions over two slightly different data examples. 2024
+;; uses "Liability" where 2025 uses "Liability Expense". 2026 includes blank
+;; PFML records.
+(def paychex-csv-2024
+  ;; clojure.java.io/resource locates a file in the classpath
+  (with-open [reader (io/reader (io/resource "example-paychex-pay-item-details-2024.csv"))]
+    (import/read-csv reader)))
+(def paychex-csv-2025
+  (with-open [reader (io/reader (io/resource "example-paychex-pay-item-details-2025.csv"))]
+    (import/read-csv reader)))
 
-;; Run all our functions over two slightly different data examples. 2024 uses
-;; "Liability" where 2025 uses "Liability Expense".
-(def paychex-csv-2024 (->> "example-paychex-pay-item-details-2024.csv"
-                           clojure.java.io/resource
-                           import/read-csv))
-(def paychex-csv-2025 (->> "example-paychex-pay-item-details-2025.csv"
-                           clojure.java.io/resource
-                           import/read-csv))
+(deftest handles-blank-amount
+  (let [amount (-> "Example Co Inc,\"Citizen, Jill B\",2,PFML ER PU,Fringe Benefits,,,,,,,,,,,\n"
+                   import/read-csv
+                   first
+                   :amount)]
+    (is (= amount 0))))
+
+(deftest warns-about-non-zero-pfml-records
+  (let [records [{:name "Doe-Jane"
+                  :category "PFML ER PU"
+                  :type "Fringe Benefits"
+                  :amount 1M}]
+        warnings (import/warnings records)]
+    (is (= (count warnings) 1) "Expected a non-zero PFML warning")))
 
 (deftest render-transaction
   (let [transaction '{:date "DATE"