From e8f889eb05eabfa99a36dadefa1d76f5e44a85a8 Mon Sep 17 00:00:00 2001
From: Matthew Flatt <mflatt@racket-lang.org>
Date: Mon, 17 Dec 2012 06:56:38 -0700
Subject: [PATCH] net/url: add `current-url-encode-mode'

Changing `current-url-encode-mode' from 'recommended to 'unreserved
causes `url->string' to encode !, *, ', (, and ) using %, which
can avoid confusing some parsers.

See also https://github.com/plt/racket/pull/198

original commit: b8b8260379716cf38136ff7f5fec1b33a824e151
---
 collects/net/scribblings/url.scrbl | 31 +++++++++++++++++++++++++-----
 collects/tests/net/url.rkt         |  7 +++++--
 2 files changed, 31 insertions(+), 7 deletions(-)

diff --git a/collects/net/scribblings/url.scrbl b/collects/net/scribblings/url.scrbl
index 26c9322858..9e24e5662f 100644
--- a/collects/net/scribblings/url.scrbl
+++ b/collects/net/scribblings/url.scrbl
@@ -81,7 +81,7 @@ a UNC path is represented by a first element that is @racket[""] and
 then successive elements complete the drive components that are
 separated by @litchar{/} or @litchar{\}.}
 
-@defstruct[path/param ([path (or/c string? (one-of/c 'up 'same))]
+@defstruct[path/param ([path (or/c string? (or/c 'up 'same))]
                        [param (listof string?)])]{
 
 A pair that joins a path segment with its params in a URL.}
@@ -169,7 +169,10 @@ The @racket[url->string] procedure uses
 @racket[alist->form-urlencoded] when formatting the query, so it is
 sensitive to the @racket[current-alist-separator-mode] parameter for
 determining the association separator. The default is to separate
-associations with a @litchar{&}.}
+associations with a @litchar{&}.
+
+The encoding of path segments and fragment is sensitive to the
+@racket[current-url-encode-mode] parameter.}
 
 
 @defproc[(path->url [path (or/c path-string? path-for-some-system?)])
@@ -179,19 +182,36 @@ Converts a path to a @racket[url].}
 
 
 @defproc[(url->path [URL url?]
-                    [kind (one-of/c 'unix 'windows) (system-path-convention-type)])
+                    [kind (or/c 'unix 'windows) (system-path-convention-type)])
          path-for-some-system?]{
 
 Converts @racket[URL], which is assumed to be a @racket["file"] URL,
 to a path.}
 
 
-@defparam[file-url-path-convention-type kind (one-of/c 'unix 'windows)]{
+@defparam[file-url-path-convention-type kind (or/c 'unix 'windows)]{
 
 Determines the default conversion to and from strings for
 @racket["file"] URLs. See @racket[string->url] and @racket[url->string].}
 
 
+@defparam[current-url-encode-mode mode (or/c 'recommended 'unreserved)]{
+
+Determines how @racket[url->string] encodes @litchar{!}, @litchar{*}, @litchar{'}, @litchar{(},
+and @litchar{)} in path segments and fragments:
+@racket['recommended] leave them as-is, while @racket['unreserved]
+encodes them using @litchar{%}. The @racket['recommended] mode corresponds
+to the recommendations of RFC 2396 @cite["RFC2396"], but @racket['unreserved]
+avoids characters that are in some contexts mistaken for delimiters around
+URLs.
+
+Internally, @racket['recommended] mode uses
+@racket[uri-path-segment-encode] and @racket[uri-encode], while
+@racket['unreserved] mode uses
+@racket[uri-path-segment-unreserved-encode] and
+@racket[uri-unreserved-encode].}
+
+
 @deftogether[(
 @defproc[(get-pure-port [URL url?]
                         [header (listof string?) null]
@@ -428,7 +448,8 @@ Note that @racket[net/url] does not provide the
 @defsignature[url^ ()]{
 
 Includes everything exported by the @racketmodname[net/url] module
-except @racket[current-https-protocol].  Note that the exports of
+except @racket[current-https-protocol] and @racket[current-url-encode-mode].
+Note that the exports of
 @racketmodname[net/url] and the @racket[url^] signature do not include
 @racket[current-connect-scheme].}
 
diff --git a/collects/tests/net/url.rkt b/collects/tests/net/url.rkt
index 91371eb6ec..8c47a3ce1f 100644
--- a/collects/tests/net/url.rkt
+++ b/collects/tests/net/url.rkt
@@ -110,8 +110,11 @@
 
   ;; test the characters that need to be encoded in paths vs those that do not need to
   ;; be encoded in paths
-  (test-s->u #("http" #f "www.drscheme.org" #f #t (#("a:@!$&'()*+,=z") #("/?#[];") #("")) () #f)
-             "http://www.drscheme.org/a:@!$&'()*+,=z/%2F%3F%23%5B%5D%3B/")
+  (test-s->u #("http" #f "www.drscheme.org" #f #t (#("a:@!$&'()*+,=z") #("/?#[];") #("")) () "@!$&'()*+,=z")
+             "http://www.drscheme.org/a:@!$&'()*+,=z/%2F%3F%23%5B%5D%3B/#%40!%24%26'()*%2B%2C%3Dz")
+  (parameterize ([current-url-encode-mode 'unreserved])
+    (test-s->u #("http" #f "www.drscheme.org" #f #t (#("a:@!$&'()*+,=z") #("/?#[];") #("")) () "@!$&'()*+,=z")
+               "http://www.drscheme.org/a:@%21$&%27%28%29%2A+,=z/%2F%3F%23%5B%5D%3B/#%40%21%24%26%27%28%29%2A%2B%2C%3Dz"))
 
   (test-s->u #("http" #f "www.drscheme.org" #f #t (#(".") #("..") #(same) #(up) #("...") #("abc.def")) () #f)
              "http://www.drscheme.org/%2e/%2e%2e/./../.../abc.def")