Now a simple re-frame app

This commit is contained in:
Bryce Covert
2017-12-07 11:23:57 -08:00
parent 275119c362
commit 5b578c11e8
13 changed files with 618 additions and 40 deletions

View File

@@ -0,0 +1,32 @@
(ns auto-ap.handler
(:require [compojure.core :refer :all]
[compojure.route :as route]
[clojure.java.io :as io]
[auto-ap.parse :as parse]
[ring.middleware.multipart-params :as mp]
[ring.util.response :as response]
[ring.middleware.defaults :refer [wrap-defaults site-defaults]]
[ring.middleware.json :refer [wrap-json-response]]))
(defroutes app-routes
(GET "/hi" [] "hello")
(GET "/" [] (response/resource-response "index.html" {:root "public"}))
(POST "/pdf-upload"
{{ files "file"} :params :as params}
(let [{:keys [filename tempfile]} (second files)]
(println tempfile)
#_(io/copy tempfile (io/file "resources" "public" filename))
(for [{:keys [total date invoice-number customer-identifier]} (parse/parse-file (.getPath tempfile))]
{"customer-identifier" customer-identifier
"invoice-number" invoice-number
"date" date
"total" total})))
(route/resources "/")
(route/not-found "Not Found"))
#_(defroutes routes
(GET "/" [] (resource-response "index.html" {:root "public"}))
(resources "/"))
(def app
(wrap-json-response (mp/wrap-multipart-params app-routes)))

54
src/clj/auto_ap/parse.clj Normal file
View File

@@ -0,0 +1,54 @@
(ns auto-ap.parse
(:require [clojure.java.io :as io]
[clojure.string :as str]
[clojure.java.shell :as sh]))
(def templates
[{:keywords [#"CHEF'S WAREHOUSE"]
:extract {:total #"2 WKS C\.C\.\s+([\d.,]+)"
:customer-identifier #"\n([A-Z][A-Z ]+)\s{2,}"
:date #"\s+([0-9]+/[0-9]+/[0-9]+)"
:invoice-number #"\s+[0-9]+/[0-9]+/[0-9]+\s+([0-9]+)"}}
{:keywords [#"Golden Gate Meat"]
:extract {:total #"Invoice Total\:\s+\$([\d.,]+)"
:customer-identifier #"Bill To\s*:\s*([\w ]+)\s{2,}"
:date #"Printed:\s+([0-9]+/[0-9]+/[0-9]+)"
:invoice-number #"Invoice\s+[^\n]+\n[^\n]+\n\s+([0-9]+)"}}
{:keywords [#"CINTAS CORPORATION"]
:extract {:invoice-number #"INVOICE\s#\s+([\d.,]+)"
:customer-identifier #"BILL TO\s*:\s{2,}([\w ]+)\s{2,}"
:date #"INVOICE DATE\s*\n.*\s+([0-9]+/[0-9]+/[0-9]+)"
:total #"INVOICE TOTAL\s+([0-9.]+)"}
:multi #"\f\f"}])
(defn template-applies? [text {:keys [keywords]}]
(every? #(re-find % text) keywords))
(defn extract-template [text template]
(if (:multi template)
(mapcat
#(extract-template % (dissoc template :multi))
(str/split text (:multi template)))
[(->> template
:extract
(reduce-kv
(fn [result k v]
(assoc result k (some-> (first (map second (re-seq v text)))
str/trim )))
{}))]))
(defn parse [text]
(->> templates
(filter (partial template-applies? text))
first
(extract-template text)))
(defn parse-file
[file]
(-> (sh/sh "pdftotext" "-layout" file "-")
:out
(doto println)
parse))

View File

@@ -0,0 +1,9 @@
(ns auto-ap.server
(:require [auto-ap.handler :refer [app]]
[config.core :refer [env]]
[ring.adapter.jetty :refer [run-jetty]])
(:gen-class))
(defn -main [& args]
(let [port (Integer/parseInt (or (env :port) "3000"))]
(run-jetty app {:port port :join? false})))