* Prevent nnrss from generating text/plain parts for HTML text
@ 2006-01-02 17:41 Mark Plaksin
2006-01-02 17:51 ` Mark Plaksin
2006-01-04 3:36 ` Katsumi Yamaoka
0 siblings, 2 replies; 4+ messages in thread
From: Mark Plaksin @ 2006-01-02 17:41 UTC (permalink / raw)
[-- Attachment #1: Type: text/plain, Size: 789 bytes --]
nnrss would be better if it did not generate text/plain parts for items
that contain HTML. As-is, when you discourage HTML and an item contains
HTML you see the unrendered HTML first. K v will show you the HTML version
but it's extra keystrokes.
I use the this (from Katsumi :) to discourage HTML:
(setq gnus-buttonized-mime-types
'("multipart/alternative" "multipart/signed")
mm-discouraged-alternatives
'("text/html" "image/.*"))
Boing Boing's RSS items (from http://boingboing.net/index.xml) are always
HTML so I always have to hit K v to read them.
I don't know the best way to check for HTML in elisp. I've attached a
patch which uses a regexp to do the job. The patch works OK for me but I
don't think it's ideal. What's a better way to do this?
Thanks!
[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #2: nnrss textplain --]
[-- Type: text/x-patch, Size: 3597 bytes --]
--- nnrss.el.orig 2005-12-21 12:43:03.000000000 -0500
+++ nnrss.el 2006-01-02 12:30:51.000000000 -0500
@@ -208,40 +208,47 @@
rfc2047-encode-encoded-words body)
(when (or text link enclosure comments)
(insert "\n")
- (insert "<#multipart type=alternative>\n"
- "<#part type=\"text/plain\">\n")
- (setq body (point))
- (when text
- (insert text)
- (goto-char body)
- ;; See `nnrss-check-group', which inserts "<br /><br />".
- (if (search-forward "<br /><br />" nil t)
- (if (eobp)
- (replace-match "\n")
- (replace-match "\n\n")
- (let ((fill-column default-fill-column)
- (window (get-buffer-window nntp-server-buffer)))
- (when window
- (setq fill-column
- (max 1 (/ (* (window-width window) 7) 8))))
- (fill-region (point) (point-max))
- (goto-char (point-max))
- ;; XEmacs version of `fill-region' inserts newline.
- (unless (bolp)
- (insert "\n"))))
- (goto-char (point-max))
- (insert "\n"))
- (when (or link enclosure)
- (insert "\n")))
- (when link
- (insert link "\n"))
- (when enclosure
- (insert (car enclosure) " "
- (nth 2 enclosure) " "
- (nth 3 enclosure) "\n"))
- (when comments
- (insert comments "\n"))
- (setq body (buffer-substring body (point)))
+ (insert "<#multipart type=alternative>\n")
+ ;; If text contains HTML, don't generate a text/plain part
+ ;; FIXME: The regexp is not very smart. A regexp probably isn't
+ ;; the right thing to use anyhow. It means any article which has
+ ;; had "<br /><br />" inserted by nnrss-check-group will have no
+ ;; text/plain part.
+ (if (not (string-match "<[a-zA-Z]+[^>]*>" text))
+ (progn
+ (insert "<#part type=\"text/plain\">\n")
+ (setq body (point))
+ (when text
+ (insert text)
+ (goto-char body)
+ ;; See `nnrss-check-group', which inserts "<br /><br />".
+ (if (search-forward "<br /><br />" nil t)
+ (if (eobp)
+ (replace-match "\n")
+ (replace-match "\n\n")
+ (let ((fill-column default-fill-column)
+ (window (get-buffer-window nntp-server-buffer)))
+ (when window
+ (setq fill-column
+ (max 1 (/ (* (window-width window) 7) 8))))
+ (fill-region (point) (point-max))
+ (goto-char (point-max))
+ ;; XEmacs version of `fill-region' inserts newline.
+ (unless (bolp)
+ (insert "\n"))))
+ (goto-char (point-max))
+ (insert "\n"))
+ (when (or link enclosure)
+ (insert "\n")))
+ (when link
+ (insert link "\n"))
+ (when enclosure
+ (insert (car enclosure) " "
+ (nth 2 enclosure) " "
+ (nth 3 enclosure) "\n"))
+ (when comments
+ (insert comments "\n"))
+ (setq body (buffer-substring body (point)))))
(insert "<#/part>\n"
"<#part type=\"text/html\">\n"
"<html><head></head><body>\n")
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: Prevent nnrss from generating text/plain parts for HTML text
2006-01-02 17:41 Prevent nnrss from generating text/plain parts for HTML text Mark Plaksin
@ 2006-01-02 17:51 ` Mark Plaksin
2006-01-04 3:36 ` Katsumi Yamaoka
1 sibling, 0 replies; 4+ messages in thread
From: Mark Plaksin @ 2006-01-02 17:51 UTC (permalink / raw)
[-- Attachment #1: Type: text/plain, Size: 392 bytes --]
Mark Plaksin <happy@mcplaksin.org> writes:
> I don't know the best way to check for HTML in elisp. I've attached a
> patch which uses a regexp to do the job. The patch works OK for me but I
> don't think it's ideal. What's a better way to do this?
Oops--that patch failed when the text of an item was empty (e.g., some
items in del.icio.us feeds). Here's a new patch which fixes that.
[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #2: nnrss textplain --]
[-- Type: text/x-patch, Size: 3630 bytes --]
--- nnrss.el.orig 2005-12-21 12:43:03.000000000 -0500
+++ nnrss.el 2006-01-02 12:44:33.000000000 -0500
@@ -208,40 +208,48 @@
rfc2047-encode-encoded-words body)
(when (or text link enclosure comments)
(insert "\n")
- (insert "<#multipart type=alternative>\n"
- "<#part type=\"text/plain\">\n")
- (setq body (point))
- (when text
- (insert text)
- (goto-char body)
- ;; See `nnrss-check-group', which inserts "<br /><br />".
- (if (search-forward "<br /><br />" nil t)
- (if (eobp)
- (replace-match "\n")
- (replace-match "\n\n")
- (let ((fill-column default-fill-column)
- (window (get-buffer-window nntp-server-buffer)))
- (when window
- (setq fill-column
- (max 1 (/ (* (window-width window) 7) 8))))
- (fill-region (point) (point-max))
- (goto-char (point-max))
- ;; XEmacs version of `fill-region' inserts newline.
- (unless (bolp)
- (insert "\n"))))
- (goto-char (point-max))
- (insert "\n"))
- (when (or link enclosure)
- (insert "\n")))
- (when link
- (insert link "\n"))
- (when enclosure
- (insert (car enclosure) " "
- (nth 2 enclosure) " "
- (nth 3 enclosure) "\n"))
- (when comments
- (insert comments "\n"))
- (setq body (buffer-substring body (point)))
+ (insert "<#multipart type=alternative>\n")
+ ;; If text contains HTML, don't generate a text/plain part
+ ;; FIXME: The regexp is not very smart. A regexp probably isn't
+ ;; the right thing to use anyhow. It means any article which has
+ ;; had "<br /><br />" inserted by nnrss-check-group will have no
+ ;; text/plain part.
+ (if (and text
+ (not (string-match "<[a-zA-Z]+[^>]*>" text)))
+ (progn
+ (insert "<#part type=\"text/plain\">\n")
+ (setq body (point))
+ (when text
+ (insert text)
+ (goto-char body)
+ ;; See `nnrss-check-group', which inserts "<br /><br />".
+ (if (search-forward "<br /><br />" nil t)
+ (if (eobp)
+ (replace-match "\n")
+ (replace-match "\n\n")
+ (let ((fill-column default-fill-column)
+ (window (get-buffer-window nntp-server-buffer)))
+ (when window
+ (setq fill-column
+ (max 1 (/ (* (window-width window) 7) 8))))
+ (fill-region (point) (point-max))
+ (goto-char (point-max))
+ ;; XEmacs version of `fill-region' inserts newline.
+ (unless (bolp)
+ (insert "\n"))))
+ (goto-char (point-max))
+ (insert "\n"))
+ (when (or link enclosure)
+ (insert "\n")))
+ (when link
+ (insert link "\n"))
+ (when enclosure
+ (insert (car enclosure) " "
+ (nth 2 enclosure) " "
+ (nth 3 enclosure) "\n"))
+ (when comments
+ (insert comments "\n"))
+ (setq body (buffer-substring body (point)))))
(insert "<#/part>\n"
"<#part type=\"text/html\">\n"
"<html><head></head><body>\n")
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: Prevent nnrss from generating text/plain parts for HTML text
2006-01-02 17:41 Prevent nnrss from generating text/plain parts for HTML text Mark Plaksin
2006-01-02 17:51 ` Mark Plaksin
@ 2006-01-04 3:36 ` Katsumi Yamaoka
2006-01-05 2:38 ` Mark Plaksin
1 sibling, 1 reply; 4+ messages in thread
From: Katsumi Yamaoka @ 2006-01-04 3:36 UTC (permalink / raw)
>>>>> In <87wthiedd8.fsf@stone.tss.usg.edu>
>>>>> Mark Plaksin <happy@mcplaksin.org> wrote:
> nnrss would be better if it did not generate text/plain parts for items
> that contain HTML. As-is, when you discourage HTML and an item contains
> HTML you see the unrendered HTML first. K v will show you the HTML version
> but it's extra keystrokes.
> I use the this (from Katsumi :) to discourage HTML:
> (setq gnus-buttonized-mime-types
> '("multipart/alternative" "multipart/signed")
> mm-discouraged-alternatives
> '("text/html" "image/.*"))
> Boing Boing's RSS items (from http://boingboing.net/index.xml) are always
> HTML so I always have to hit K v to read them.
> I don't know the best way to check for HTML in elisp. I've attached a
> patch which uses a regexp to do the job. The patch works OK for me but I
> don't think it's ideal. What's a better way to do this?
How about making those variables group parameters?
(eval-after-load "gnus-sum"
'(setq gnus-newsgroup-variables
(append '((gnus-buttonized-mime-types
. '("multipart/alternative" "multipart/signed"))
(mm-discouraged-alternatives
. '("text/html" "image/.*")))
gnus-newsgroup-variables)))
(add-to-list 'gnus-parameters
'("\\`nnrss:"
(gnus-buttonized-mime-types nil)
(mm-discouraged-alternatives nil)))
I haven't fully tested them yet, though.
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: Prevent nnrss from generating text/plain parts for HTML text
2006-01-04 3:36 ` Katsumi Yamaoka
@ 2006-01-05 2:38 ` Mark Plaksin
0 siblings, 0 replies; 4+ messages in thread
From: Mark Plaksin @ 2006-01-05 2:38 UTC (permalink / raw)
Katsumi Yamaoka <yamaoka@jpl.org> writes:
>>>>>> In <87wthiedd8.fsf@stone.tss.usg.edu>
>>>>>> Mark Plaksin <happy@mcplaksin.org> wrote:
>
>> nnrss would be better if it did not generate text/plain parts for items
>> that contain HTML. As-is, when you discourage HTML and an item contains
>> HTML you see the unrendered HTML first. K v will show you the HTML version
>> but it's extra keystrokes.
>
>> I use the this (from Katsumi :) to discourage HTML:
>
>> (setq gnus-buttonized-mime-types
>> '("multipart/alternative" "multipart/signed")
>> mm-discouraged-alternatives
>> '("text/html" "image/.*"))
>
>> Boing Boing's RSS items (from http://boingboing.net/index.xml) are always
>> HTML so I always have to hit K v to read them.
>
>> I don't know the best way to check for HTML in elisp. I've attached a
>> patch which uses a regexp to do the job. The patch works OK for me but I
>> don't think it's ideal. What's a better way to do this?
>
> How about making those variables group parameters?
>
> (eval-after-load "gnus-sum"
> '(setq gnus-newsgroup-variables
> (append '((gnus-buttonized-mime-types
> . '("multipart/alternative" "multipart/signed"))
> (mm-discouraged-alternatives
> . '("text/html" "image/.*")))
> gnus-newsgroup-variables)))
>
> (add-to-list 'gnus-parameters
> '("\\`nnrss:"
> (gnus-buttonized-mime-types nil)
> (mm-discouraged-alternatives nil)))
>
> I haven't fully tested them yet, though.
This works--thanks!
I wonder whether there's any utility to having a text/plain part when
reading RSS feeds. There must be but I can't quite imagine it!
^ permalink raw reply [flat|nested] 4+ messages in thread
end of thread, other threads:[~2006-01-05 2:38 UTC | newest]
Thread overview: 4+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2006-01-02 17:41 Prevent nnrss from generating text/plain parts for HTML text Mark Plaksin
2006-01-02 17:51 ` Mark Plaksin
2006-01-04 3:36 ` Katsumi Yamaoka
2006-01-05 2:38 ` Mark Plaksin
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).