bug#24694: Document url--allow-chars for external use?

bug#24694: Document url--allow-chars for external use?

[Hong Xu]

[-- Attachment #1.1: Type: text/plain, Size: 515 bytes --]

Currently url-hexify-string has an optional argument allowed-chars,
which is a long vector. However, it is more natural to specify a list of
characters. Internally, the default value of allowed-chars is specified
by converting a list of chars to such a vector by using an internal
function url--allow-chars. I would like to suggest to document
url--allow-chars for external use, thus we can improve the interface of
the url-hexify-string function.

I can draft the document change, if you agree with me.


[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 819 bytes --]

bug#24694: Document url--allow-chars for external use?

[Eli Zaretskii]

> From: Hong Xu <hong@topbug.net>
> Date: Fri, 14 Oct 2016 11:24:52 -0700
> 
> Currently url-hexify-string has an optional argument allowed-chars,
> which is a long vector. However, it is more natural to specify a list of
> characters. Internally, the default value of allowed-chars is specified
> by converting a list of chars to such a vector by using an internal
> function url--allow-chars. I would like to suggest to document
> url--allow-chars for external use, thus we can improve the interface of
> the url-hexify-string function.
> 
> I can draft the document change, if you agree with me.

Isn't it better to teach url-hexify-string to accept lists as well?

bug#24694: Document url--allow-chars for external use?

[Hong Xu]

[-- Attachment #1.1: Type: text/plain, Size: 781 bytes --]

On 10/14/2016 12:02 PM, Eli Zaretskii wrote:
>> From: Hong Xu <hong@topbug.net>
>> Date: Fri, 14 Oct 2016 11:24:52 -0700
>>
>> Currently url-hexify-string has an optional argument allowed-chars,
>> which is a long vector. However, it is more natural to specify a list of
>> characters. Internally, the default value of allowed-chars is specified
>> by converting a list of chars to such a vector by using an internal
>> function url--allow-chars. I would like to suggest to document
>> url--allow-chars for external use, thus we can improve the interface of
>> the url-hexify-string function.
>>
>> I can draft the document change, if you agree with me.
> 
> Isn't it better to teach url-hexify-string to accept lists as well?
> 
Yes, that would be even better.


[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 819 bytes --]

bug#24694: Document url--allow-chars for external use?

[Hong Xu]

[-- Attachment #1.1: Type: text/plain, Size: 1011 bytes --]


On 2016-10-14 Fri 12:02 GMT-0800, Eli Zaretskii <eliz@gnu.org> wrote:

>> From: Hong Xu <hong@topbug.net>
>> Date: Fri, 14 Oct 2016 11:24:52 -0700
>> 
>> Currently url-hexify-string has an optional argument allowed-chars,
>> which is a long vector. However, it is more natural to specify a list of
>> characters. Internally, the default value of allowed-chars is specified
>> by converting a list of chars to such a vector by using an internal
>> function url--allow-chars. I would like to suggest to document
>> url--allow-chars for external use, thus we can improve the interface of
>> the url-hexify-string function.
>> 
>> I can draft the document change, if you agree with me.
>
> Isn't it better to teach url-hexify-string to accept lists as well?

The patch is attached.

Make url-hexify-string accept a list of allowed chars.

	* url-util.el (url-hexify-string): Accept a list of allowed chars.
	* url.texi (URI Encoding): Update url-hexify-string doc and index improvements.


[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1.2: url.patch --]
[-- Type: text/x-diff, Size: 2791 bytes --]

diff --git a/doc/misc/url.texi b/doc/misc/url.texi
index 91cb6b54a847..b5e6e2ce465a 100644
--- a/doc/misc/url.texi
+++ b/doc/misc/url.texi
@@ -221,6 +221,7 @@ URI Encoding
 @section URI Encoding
 
 @cindex percent encoding
+@findex url-generic-parse-url
   The @code{url-generic-parse-url} parser does not obey RFC 3986 in
 one respect: it allows non-@acronym{ASCII} characters in URI strings.
 
@@ -233,6 +234,7 @@ URI Encoding
 @acronym{ASCII} characters must also be percent encoded when they
 appear in URI components.)
 
+@findex url-encode-url
   The function @code{url-encode-url} can be used to convert a URI
 string containing arbitrary characters to one that is properly
 percent-encoded in accordance with RFC 3986.
@@ -244,11 +246,13 @@ URI Encoding
 previously uppercase.
 @end defun
 
+@findex url-hexify-string
+@findex url-unhex-string
   To convert between a string containing arbitrary characters and a
 percent-encoded all-@acronym{ASCII} string, use the functions
 @code{url-hexify-string} and @code{url-unhex-string}:
 
-@defun url-hexify-string string &optional allowed-chars
+@defun url-hexify-string string &optional allowed-chars list-chars
 This function performs percent-encoding on @var{string}, and returns
 the result.
 
diff --git a/lisp/url/url-util.el b/lisp/url/url-util.el
index 1ae2213eee65..96fc741a1700 100644
--- a/lisp/url/url-util.el
+++ b/lisp/url/url-util.el
@@ -398,18 +398,23 @@ url--allowed-chars
     vec))
 
 ;;;###autoload
-(defun url-hexify-string (string &optional allowed-chars)
+(defun url-hexify-string (string &optional allowed-chars list-chars)
   "URI-encode STRING and return the result.
 If STRING is multibyte, it is first converted to a utf-8 byte
 string.  Each byte corresponding to an allowed character is left
 as-is, while all other bytes are converted to a three-character
 string: \"%\" followed by two upper-case hex digits.
 
-The allowed characters are specified by ALLOWED-CHARS.  If this
+The allowed characters are specified by ALLOWED-CHARS. If this
 argument is nil, the list `url-unreserved-chars' determines the
-allowed characters.  Otherwise, ALLOWED-CHARS should be a vector
-whose Nth element is non-nil if character N is allowed."
-  (unless allowed-chars
+allowed characters.  Otherwise, if LIST-CHARS is nil or omitted,
+ALLOWED-CHARS should be a vector whose Nth element is non-nil if
+character N is allowed; if LIST-CHARS is non-nil, ALLOWED-CHARS
+should be a list of allowed chars.
+"
+  (if allowed-chars
+      (when list-chars
+        (setq allowed-chars (url--allowed-chars allowed-chars)))
     (setq allowed-chars (url--allowed-chars url-unreserved-chars)))
   (mapconcat (lambda (byte)
 	       (if (aref allowed-chars byte)

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 818 bytes --]

bug#24694: Document url--allow-chars for external use?

[Hong Xu]

On 12/1/16 1:28 PM, Hong Xu wrote:
> 
> On 2016-10-14 Fri 12:02 GMT-0800, Eli Zaretskii <eliz@gnu.org> wrote:
> 
>>> From: Hong Xu <hong@topbug.net>
>>> Date: Fri, 14 Oct 2016 11:24:52 -0700
>>>
>>> Currently url-hexify-string has an optional argument allowed-chars,
>>> which is a long vector. However, it is more natural to specify a list of
>>> characters. Internally, the default value of allowed-chars is specified
>>> by converting a list of chars to such a vector by using an internal
>>> function url--allow-chars. I would like to suggest to document
>>> url--allow-chars for external use, thus we can improve the interface of
>>> the url-hexify-string function.
>>>
>>> I can draft the document change, if you agree with me.
>>
>> Isn't it better to teach url-hexify-string to accept lists as well?
> 
> The patch is attached.
> 
> Make url-hexify-string accept a list of allowed chars.
> 
> 	* url-util.el (url-hexify-string): Accept a list of allowed chars.
> 	* url.texi (URI Encoding): Update url-hexify-string doc and index improvements.
> 

It's been a few years... I'm wondering whether this patch is still interesting?

bug#24694: Document url--allow-chars for external use?

[Eli Zaretskii]

> From: Hong Xu <hong@topbug.net>
> Cc: 24694@debbugs.gnu.org
> Date: Fri, 4 Oct 2019 12:31:27 -0700
> 
> > Make url-hexify-string accept a list of allowed chars.
> > 
> > 	* url-util.el (url-hexify-string): Accept a list of allowed chars.
> > 	* url.texi (URI Encoding): Update url-hexify-string doc and index improvements.
> > 
> 
> It's been a few years... I'm wondering whether this patch is still interesting?

I think it is, yes.  Can you send a patch relative to the current
master branch?

Thanks.

bug#24694: Document url--allow-chars for external use?

[Robert Pluim]

>>>>> On Fri, 11 Oct 2019 15:38:03 +0300, Eli Zaretskii <eliz@gnu.org> said:

    >> From: Hong Xu <hong@topbug.net>
    >> Cc: 24694@debbugs.gnu.org
    >> Date: Fri, 4 Oct 2019 12:31:27 -0700
    >> 
    >> > Make url-hexify-string accept a list of allowed chars.
    >> > 
    >> > 	* url-util.el (url-hexify-string): Accept a list of allowed chars.
    >> > 	* url.texi (URI Encoding): Update url-hexify-string doc and index improvements.
    >> > 
    >> 
    >> It's been a few years... I'm wondering whether this patch is still interesting?

    Eli> I think it is, yes.  Can you send a patch relative to the current
    Eli> master branch?

Api nit: if the first argument is going to be a list or a vector, you
could just keep one argument and check 'vectorp'.

Robert

bug#24694: Document url--allow-chars for external use?

[Hong Xu]

[-- Attachment #1: Type: text/plain, Size: 904 bytes --]

On 10/11/19 7:32 AM, Robert Pluim wrote:
>>>>>> On Fri, 11 Oct 2019 15:38:03 +0300, Eli Zaretskii <eliz@gnu.org> said:
> 
>      >> From: Hong Xu <hong@topbug.net>
>      >> Cc: 24694@debbugs.gnu.org
>      >> Date: Fri, 4 Oct 2019 12:31:27 -0700
>      >>
>      >> > Make url-hexify-string accept a list of allowed chars.
>      >> >
>      >> > 	* url-util.el (url-hexify-string): Accept a list of allowed chars.
>      >> > 	* url.texi (URI Encoding): Update url-hexify-string doc and index improvements.
>      >> >
>      >>
>      >> It's been a few years... I'm wondering whether this patch is still interesting?
> 
>      Eli> I think it is, yes.  Can you send a patch relative to the current
>      Eli> master branch?
> 
> Api nit: if the first argument is going to be a list or a vector, you
> could just keep one argument and check 'vectorp'.
> 

Thanks, I've updated the patch as attached.

[-- Attachment #2: 0001-Make-url-hexify-string-accept-a-list-of-allowed-char.patch --]
[-- Type: text/x-patch, Size: 2431 bytes --]

From 4f56c4081f44ab04e3095b5f3b5332d5c3eb1cdf Mon Sep 17 00:00:00 2001
From: Hong Xu <hong@topbug.net>
Date: Fri, 11 Oct 2019 19:17:23 -0700
Subject: [PATCH] Make url-hexify-string accept a list of allowed chars.

* url-util.el (url-hexify-string): Accept a list of allowed chars.
* url.texi (URI Encoding): Update url-hexify-string doc and index
improvements.
---
 doc/misc/url.texi    |  4 ++++
 lisp/url/url-util.el | 10 +++++++---
 2 files changed, 11 insertions(+), 3 deletions(-)

diff --git a/doc/misc/url.texi b/doc/misc/url.texi
index e72d9bfe3d2e..ff6c547bec1e 100644
--- a/doc/misc/url.texi
+++ b/doc/misc/url.texi
@@ -221,6 +221,7 @@ URI Encoding
 @section URI Encoding
 
 @cindex percent encoding
+@findex url-generic-parse-url
   The @code{url-generic-parse-url} parser does not obey RFC 3986 in
 one respect: it allows non-@acronym{ASCII} characters in URI strings.
 
@@ -233,6 +234,7 @@ URI Encoding
 @acronym{ASCII} characters must also be percent encoded when they
 appear in URI components.)
 
+@findex url-encode-url
   The function @code{url-encode-url} can be used to convert a URI
 string containing arbitrary characters to one that is properly
 percent-encoded in accordance with RFC 3986.
@@ -244,6 +246,8 @@ URI Encoding
 previously uppercase.
 @end defun
 
+@findex url-hexify-string
+@findex url-unhex-string
   To convert between a string containing arbitrary characters and a
 percent-encoded all-@acronym{ASCII} string, use the functions
 @code{url-hexify-string} and @code{url-unhex-string}:
diff --git a/lisp/url/url-util.el b/lisp/url/url-util.el
index a390723e73dc..7abd5c216eeb 100644
--- a/lisp/url/url-util.el
+++ b/lisp/url/url-util.el
@@ -395,9 +395,13 @@ url-hexify-string
 
 The allowed characters are specified by ALLOWED-CHARS.  If this
 argument is nil, the list `url-unreserved-chars' determines the
-allowed characters.  Otherwise, ALLOWED-CHARS should be a vector
-whose Nth element is non-nil if character N is allowed."
-  (unless allowed-chars
+allowed characters.  Otherwise, ALLOWED-CHARS should be either a
+vector whose Nth element is non-nil if character N is allowed or
+a list of allowed chars.
+"
+  (if allowed-chars
+      (unless (vectorp allowed-chars)
+        (setq allowed-chars (url--allowed-chars allowed-chars)))
     (setq allowed-chars (url--allowed-chars url-unreserved-chars)))
   (mapconcat (lambda (byte)
 	       (if (aref allowed-chars byte)
-- 
2.20.1

bug#24694: Document url--allow-chars for external use?

[Eli Zaretskii]

> Cc: 24694@debbugs.gnu.org
> From: Hong Xu <hong@topbug.net>
> Date: Fri, 11 Oct 2019 19:21:34 -0700
> 
> Thanks, I've updated the patch as attached.

I have a couple of minor comments:

> * url-util.el (url-hexify-string): Accept a list of allowed chars.
> * url.texi (URI Encoding): Update url-hexify-string doc and index
> improvements.

The log message should specify each file with leading directories
starting from the root of the Emacs tree.  E.g., url-util.el should be
shown as lisp/url/url-util.el.

Also, please always mention the bug number in the log message.

These and other stylistic issues are described in detail in
CONTRIBUTE.

> +allowed characters.  Otherwise, ALLOWED-CHARS should be either a
> +vector whose Nth element is non-nil if character N is allowed or
> +a list of allowed chars.                                     ^

Please insert a comma where indicated, it will make the sentence
easier to read.

Thanks.

bug#24694: Document url--allow-chars for external use?

[Hong Xu]

[-- Attachment #1: Type: text/plain, Size: 800 bytes --]

On 10/12/19 2:11 AM, Eli Zaretskii wrote:
> 
> I have a couple of minor comments:

I've attached an update.


> 
> The log message should specify each file with leading directories
> starting from the root of the Emacs tree.  E.g., url-util.el should be
> shown as lisp/url/url-util.el.

Done.

> 
> Also, please always mention the bug number in the log message.

Done.

> 
>> +allowed characters.  Otherwise, ALLOWED-CHARS should be either a
>> +vector whose Nth element is non-nil if character N is allowed or
>> +a list of allowed chars.                                     ^
> 
> Please insert a comma where indicated, it will make the sentence
> easier to read.

I swapped the two choices (the shorter one goes first) and hopefully this will be easier to read. Also updated the doc in url.texi.

[-- Attachment #2: 0001-Make-url-hexify-string-accept-a-list-of-allowed-char.patch --]
[-- Type: text/x-patch, Size: 2965 bytes --]

From 01bb6f74a4e91534a253bffab0457a27f6cb6111 Mon Sep 17 00:00:00 2001
From: Hong Xu <hong@topbug.net>
Date: Fri, 11 Oct 2019 19:17:23 -0700
Subject: [PATCH] Make url-hexify-string accept a list of allowed chars
 (bug#26469)

* lisp/url/url-util.el (url-hexify-string): Accept a list of allowed
chars.
* doc/misc/url.texi (URI Encoding): Update url-hexify-string doc and
index improvements.
---
 doc/misc/url.texi    |  8 ++++++--
 lisp/url/url-util.el | 10 +++++++---
 2 files changed, 13 insertions(+), 5 deletions(-)

diff --git a/doc/misc/url.texi b/doc/misc/url.texi
index e72d9bfe3d2e..79dead185dd0 100644
--- a/doc/misc/url.texi
+++ b/doc/misc/url.texi
@@ -221,6 +221,7 @@ URI Encoding
 @section URI Encoding
 
 @cindex percent encoding
+@findex url-generic-parse-url
   The @code{url-generic-parse-url} parser does not obey RFC 3986 in
 one respect: it allows non-@acronym{ASCII} characters in URI strings.
 
@@ -233,6 +234,7 @@ URI Encoding
 @acronym{ASCII} characters must also be percent encoded when they
 appear in URI components.)
 
+@findex url-encode-url
   The function @code{url-encode-url} can be used to convert a URI
 string containing arbitrary characters to one that is properly
 percent-encoded in accordance with RFC 3986.
@@ -244,6 +246,8 @@ URI Encoding
 previously uppercase.
 @end defun
 
+@findex url-hexify-string
+@findex url-unhex-string
   To convert between a string containing arbitrary characters and a
 percent-encoded all-@acronym{ASCII} string, use the functions
 @code{url-hexify-string} and @code{url-unhex-string}:
@@ -263,8 +267,8 @@ URI Encoding
 argument is @code{nil}, the allowed characters are those specified as
 @dfn{unreserved characters} by RFC 3986 (see the variable
 @code{url-unreserved-chars}).  Otherwise, @var{allowed-chars} should
-be a vector whose @var{n}-th element is non-@code{nil} if character
-@var{n} is allowed.
+be either a list of allowed chars, or a vector whose Nth element is
+non-nil if character N is allowed.
 @end defun
 
 @defun url-unhex-string string &optional allow-newlines
diff --git a/lisp/url/url-util.el b/lisp/url/url-util.el
index a390723e73dc..e5b09d5a29db 100644
--- a/lisp/url/url-util.el
+++ b/lisp/url/url-util.el
@@ -395,9 +395,13 @@ url-hexify-string
 
 The allowed characters are specified by ALLOWED-CHARS.  If this
 argument is nil, the list `url-unreserved-chars' determines the
-allowed characters.  Otherwise, ALLOWED-CHARS should be a vector
-whose Nth element is non-nil if character N is allowed."
-  (unless allowed-chars
+allowed characters.  Otherwise, ALLOWED-CHARS should be either a
+list of allowed chars, or a vector whose Nth element is non-nil
+if character N is allowed.
+"
+  (if allowed-chars
+      (unless (vectorp allowed-chars)
+        (setq allowed-chars (url--allowed-chars allowed-chars)))
     (setq allowed-chars (url--allowed-chars url-unreserved-chars)))
   (mapconcat (lambda (byte)
 	       (if (aref allowed-chars byte)
-- 
2.20.1

bug#24694: Document url--allow-chars for external use?

[Lars Ingebrigtsen]

Hong Xu <hong@topbug.net> writes:

>> I have a couple of minor comments:
>
> I've attached an update.

Thanks; applied to the trunk now.

-- 
(domestic pets only, the antidote for overdose, milk.)
   bloggy blog: http://lars.ingebrigtsen.no