notid/integreat
1
0
Fork 0
Code Issues Pull Requests 2 Actions Packages Projects Releases Wiki Activity

first version of sample parser.

Browse Source
This commit is contained in:
Bryce Covert
2017-12-06 17:55:09 -08:00
commit 0c442b29e0
10 changed files with 176 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

10
.gitignore vendored Normal file
View File

@@ -0,0 +1,10 @@
/target
/lib
/classes
/checkouts
pom.xml
pom.xml.asc
*.jar
*.class
/.lein-*
/.nrepl-port

19
README.md Normal file
View File

@@ -0,0 +1,19 @@
# auto-ap
FIXME
## Prerequisites
You will need [Leiningen][] 2.0.0 or above installed.
[leiningen]: https://github.com/technomancy/leiningen
## Running
To start a web server for the application, run:
lein ring server
## License
Copyright © 2017 FIXME

13
project.clj Normal file
View File

@@ -0,0 +1,13 @@
(defproject auto-ap "0.1.0-SNAPSHOT"
:description "FIXME: write description"
:url "http://example.com/FIXME"
:min-lein-version "2.0.0"
:dependencies [[org.clojure/clojure "1.8.0"]
[compojure "1.6.0"]
[ring/ring-defaults "0.2.1"]
[ring/ring-json "0.4.0"]]
:plugins [[lein-ring "0.9.7"]]
:ring {:handler auto-ap.handler/app}
:profiles
{:dev {:dependencies [[javax.servlet/servlet-api "2.5"]
[ring/ring-mock "0.3.0"]]}})

BIN
resources/public/chefswarehouse.PDF Normal file
View File

Binary file not shown.

BIN
resources/public/cintas.pdf Normal file
View File

Binary file not shown.

BIN
resources/public/golden-gate-meats.pdf Normal file
View File

Binary file not shown.

37
resources/public/index.html Normal file
View File

@@ -0,0 +1,37 @@
<!doctype html>
<html lang="en">
<head>
<script src="https://cdn.jsdelivr.net/npm/dropzone@5.2.0/dist/dropzone.min.js"></script>
<script src="http://code.jquery.com/jquery-3.2.1.min.js" integrity="sha256-hwg4gsxgFZhOsEEamdOYGBf13FyQuiTwlAQgxVSNgt4=" crossorigin="anonymous"></script>
<link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/bulma/0.6.1/css/bulma.css" />
<style>.dz-error-mark { display:none} .dz-details {display:none}
form { border: 3px solid lightgray; padding: 25px; width: 100%;}
.dz-success-mark {display:none} </style>
</head>
<body>
<div id="app" class="container">
<h1>Invoice Parsing Demo</h1>
<form action="/pdf-upload" id="my-dropzone">
<h3>Drop invoice pdfs here</h3>
<input type="file" name="file" style="display:none"/>
</form>
<h2 style="display:none">Found invoices:</h2>
<ul>
</ul>
</div>
<script>
var myDropzone = new Dropzone("#my-dropzone");
myDropzone.on("success", function(file, a) {
$("h2").show();
JSON.parse(a).map(function(x) {
$("ul").append($("<li>").text(x));
});
});
</script>
</body>
</html>

29
src/auto_ap/handler.clj Normal file
View File

@@ -0,0 +1,29 @@
(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 "/" [] (response/resource-response "index.html" {:root "public"}))
(POST "/pdf-upload"
{{ files "file"} :params :as params}
(let [{:keys [filename tempfile]} (second files)]
(io/copy tempfile (io/file "resources" "public" filename))
(for [{:keys [total date invoice-number customer-identifier]} (parse/parse-file (str "resources/public/" filename))]
(do
(println (str "An invoice #" invoice-number " on " date " for " total))
(str "An invoice for customer " customer-identifier " #" invoice-number " on " date " for " 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/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))

14
test/auto_ap/handler_test.clj Normal file
View File

@@ -0,0 +1,14 @@
(ns auto-ap.handler-test
(:require [clojure.test :refer :all]
[ring.mock.request :as mock]
[auto-ap.handler :refer :all]))
(deftest test-app
(testing "main route"
(let [response (app (mock/request :get "/"))]
(is (= (:status response) 200))
(is (= (:body response) "Hello World"))))
(testing "not-found route"
(let [response (app (mock/request :get "/invalid"))]
(is (= (:status response) 404)))))