Implement `parse-form-urlencoded'

3 files changed, 95 insertions, 3 deletions

diff --git a/tests/request.scm b/tests/request.scm
new file mode 100644
index 0000000..0b68ea1
--- /dev/null
+++ b/tests/request.scm
@@ -0,0 +1,60 @@
+;;; Weeb --- Web microframework for GNU Guile
+;;; Copyright © 2026 Jakob Kreuze <zerodaysfordays@sdf.org>
+;;;
+;;; This file is part of Weeb.
+;;;
+;;; Weeb is free software: you can redistribute it and/or modify it
+;;; under the terms of the GNU Affero General Public License as
+;;; published by the Free Software Foundation, either version 3 of the
+;;; License, or (at your option) any later version.
+;;;
+;;; Weeb is distributed in the hope that it will be useful, but WITHOUT
+;;; ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
+;;; or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Affero General
+;;; Public License for more details.
+;;;
+;;; You should have received a copy of the GNU Affero General Public
+;;; License along with Weeb. If not, see <https://www.gnu.org/licenses/>.
+
+(use-modules (srfi srfi-64)
+             (weeb request))
+
+(test-runner-current (test-runner-simple))
+
+(test-begin "weeb-request")
+
+(test-group "urldecode"
+  (test-equal "hello" (urldecode "hello"))
+  (test-equal "hello world" (urldecode "hello+world"))
+  (test-equal "hello world" (urldecode "hello%20world"))
+  (test-equal "!@#$" (urldecode "%21%40%23%24"))
+  (test-equal "a b c" (urldecode "a+b+c"))
+  (test-equal "a b c" (urldecode "a%20b%20c"))
+  (test-equal "" (urldecode ""))
+  (test-equal "invalid%hex" (urldecode "invalid%hex"))
+  (test-equal "%" (urldecode "%"))
+  (test-equal "100%" (urldecode "100%25")))
+
+(test-group "parse-form-urlencoded"
+  (test-equal '(("key" . "value"))
+              (parse-form-urlencoded "key=value"))
+  (test-equal '(("a" . "1") ("b" . "2"))
+              (parse-form-urlencoded "a=1&b=2"))
+  (test-equal '(("name" . "John Doe"))
+              (parse-form-urlencoded "name=John+Doe"))
+  (test-equal '(("query" . "foo bar"))
+              (parse-form-urlencoded "query=foo%20bar"))
+  (test-equal '(("user_id" . "42") ("role" . "admin"))
+              (parse-form-urlencoded "user_id=42&role=admin"))
+  (test-equal '(("empty" . ""))
+              (parse-form-urlencoded "empty="))
+  (test-equal '(("foo-bar" . "baz"))
+              (parse-form-urlencoded "foo-bar=baz"))
+  (test-equal '(("a" . "1") ("b" . "2") ("c" . "3"))
+              (parse-form-urlencoded "a=1&b=2&c=3"))
+  (test-equal '(("foo" . "bar"))
+              (parse-form-urlencoded "foo=bar")) ; Verifies (string->symbol key)
+  (test-equal '()
+              (parse-form-urlencoded "")))
+
+(test-end)
diff --git a/weeb/main.scm b/weeb/main.scm
index 7487f19..b6e47ec 100644
--- a/weeb/main.scm
+++ b/weeb/main.scm
@@ -110,6 +110,11 @@
     (match-lambda
       ("GET"  'GET)
       ("POST" 'POST)))
+  (define (strip-query-string uri)
+    (let ((match-object (string-match "^(/.*)\\?" uri)))
+      (if match-object
+          (match:substring match-object 1)
+          uri)))
   ;; According to RFC 9110, header names must be treated as case-insensitive.
   (define headers
     (filter-map
@@ -125,8 +130,9 @@
              #f)))
      (environ)))
   (let* ((method (string->method (getenv "REQUEST_METHOD")))
-         (uri    (getenv "REQUEST_URI")))
-    (make-struct <request> method uri '() headers (fcgi-read))))
+         (query  (parse-form-urlencoded (getenv "QUERY_STRING")))
+         (uri    (strip-query-string (getenv "REQUEST_URI"))))
+    (make-struct <request> method uri query headers (fcgi-read))))
 
 (define (handle-errors thunk)
   (catch #t
diff --git a/weeb/request.scm b/weeb/request.scm
index 42cd933..3d18b0a 100644
--- a/weeb/request.scm
+++ b/weeb/request.scm
@@ -17,8 +17,34 @@
 ;;; License along with Weeb. If not, see <https://www.gnu.org/licenses/>.
 
 (define-module (weeb request)
+  #:use-module (ice-9 match)
   #:use-module (weeb struct)
-  #:export (<request>))
+  #:export (<request>
+            urldecode
+            parse-form-urlencoded))
 
 (define-struct <request>
   method uri parameters headers body)
+
+(define (urldecode str)
+  (let loop ((i 0) (acc ""))
+    (cond
+     ((>= i (string-length str)) acc)
+     ((char=? (string-ref str i) #\+)
+      (loop (+ i 1) (string-append acc " ")))
+     ((and (char=? (string-ref str i) #\%)
+           (<= (+ i 2) (string-length str)))
+      (let* ((hex-part (substring str (+ i 1) (+ i 3)))
+             (code (string->number hex-part 16)))
+        (if code
+            (loop (+ i 3) (string-append acc (string (integer->char code))))
+            (loop (+ i 1) (string-append acc (string (string-ref str i)))))))
+     (else
+      (loop (+ i 1) (string-append acc (string (string-ref str i))))))))
+
+(define (parse-form-urlencoded encoded)
+  (map (lambda (encoded-pair)
+         (match (map urldecode (string-split encoded-pair #\=))
+           ((key value) `(,key . ,value))))
+       (filter (lambda (x) (not (zero? (string-length x))))
+               (string-split encoded #\&))))