httpd-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Rob Hartill <r...@imdb.com>
Subject [Near] Fix for multiple content encodings bug (fwd)
Date Sun, 07 Apr 1996 20:14:44 GMT

not acked..

MIME-Version: 1.0 (NeXT Mail 3.3risc v118.3)
Content-Type: multipart/alternative; boundary=NeXT-Mail-1515183957-1
Content-Transfer-Encoding: 7bit
X-Nextstep-Mailer: Mail 3.3 (Enhance X)
From: Bill Bumgarner <bbum@friday.com>
Date: Sun,  7 Apr 96 14:29:40 -0500
To: apache-bugs@mail.apache.org
Subject: [Near] Fix for multiple content encodings bug
cc: jeff@netsurfer.com, tom@pdh.com
Reply-To: bbum@friday.com
X-A: Budded on earth....
X-B: ...to bloom in heaven.

--NeXT-Mail-1515183957-1
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: quoted-printable


Enclosed is a slightly modified version of find_ct()-- assuming you have a =
mime-capable mail reader, the lines that are changed should be bold [if =
not, the changes are minor and should be easy to figure out].

With the URL and configuration from the original bug report =
[http://www.friday.com/~bbum/Urlifier-1_7.bundle.tar.gz], apache now =
yields the headers:

	HTTP/1.0 200 OK
	Date: Sun, 07 Apr 1996 ? GMT
	Server: Apache/1.0.3
	Content-type: application/bundle-MAB
	Content-encoding: x-gzip
	Content-encoding: x-tar
	Content-length: 17428
	Last-modified: Sun, 07 Apr 1996 ? GMT

Unfortunately, this [likely] cannot be counted as a *complete* fix-- if =
the targeted URL contains THREE or more encodings -- IE; =
http://www.friday.com/~bbum/Urlifier-1_7.bundle.tar.gz.Z -- the resultant =
header will have the last content-encoding by itself, but all the rest =
will appear as a comma seperated list.  IE;

	HTTP/1.0 200 OK
	Date: Sun, 07 Apr 1996 ? GMT
	Server: Apache/1.0.3
	Content-type: application/bundle-MAB
	Content-encoding: x-compress
	Content-encoding: x-gzip, x-tar
	Content-length: 17428
	Last-modified: Sun, 07 Apr 1996 ? GMT

Unless the Content-encoding: field CAN contain a comma seperated list of =
encodings [notice that the order is preserved, left to right, top to =
bottom...], the enclosed patch is not totally correct.  Unfortunately, to =
break the list out into individual headers would require a much more =
significant patch.

[Jeff:  I noticed that the NetKit DOES correctly handle multiple encoding =
headers with some being comma separated lists.  Most Excellent!  I will =
have a quad-fat compile of Apache to you in short order.]

thanks,
b.bum

int find_ct(request_rec *r)
{
    int i;
    char *fn =3D pstrdup (r->pool, r->filename);
    mime_dir_config *conf =3D
    (mime_dir_config *)get_module_config(r->per_dir_config, &mime_module);
    char *type;

    if (S_ISDIR(r->finfo.st_mode)) {
        r->content_type =3D DIR_MAGIC_TYPE;
        return OK;
    }

    if((i=3Drind(fn,'.')) < 0) return DECLINED;
    ++i;

    while ((type =3D table_get (conf->encoding_types, &fn[i])))
        {
            if (r->content_encoding =3D=3D NULL) {
                r->content_encoding =3D type;
            } else {
                table_merge(r->headers_out, "Content-encoding", type);
            }

            /* go back to previous extension to try to use it as a =
language */

            fn[i-1] =3D '\0';
            if((i=3Drind(fn,'.')) < 0) return OK;
            ++i;
        }

    if ((type =3D table_get (conf->language_types, &fn[i])))
        {
            r->content_language =3D type;

            /* go back to previous extension to try to use it as a type */

            fn[i-1] =3D '\0';
            if((i=3Drind(fn,'.')) < 0) return OK;
            ++i;
        }

    if ((type =3D table_get (conf->forced_types, &fn[i]))
        || (type =3D table_get (hash_buckets[hash(fn[i])], &fn[i])))
        {
            r->content_type =3D type;
        }

    return OK;
}


--NeXT-Mail-1515183957-1
Content-Type: text/enriched; charset=us-ascii
Content-Transfer-Encoding: quoted-printable


Enclosed is a slightly modified version of find_ct()-- assuming you have a =
mime-capable mail reader, the lines that are changed should be=20
<bold>bold
</bold> [if not, the changes are minor and should be easy to figure out].


With the URL and configuration from the original bug report =
[http://www.friday.com/~bbum/Urlifier-1_7.bundle.tar.gz], apache now =
yields the headers:


	HTTP/1.0 200 OK

	Date: Sun, 07 Apr 1996 ? GMT

	Server: Apache/1.0.3

	Content-type: application/bundle-MAB

	Content-encoding: x-gzip

	Content-encoding: x-tar

	Content-length: 17428

	Last-modified: Sun, 07 Apr 1996 ? GMT


Unfortunately, this [likely] cannot be counted as a *complete* fix-- if =
the targeted URL contains THREE or more encodings -- IE; =
http://www.friday.com/~bbum/Urlifier-1_7.bundle.tar.gz.Z -- the resultant =
header will have the last content-encoding by itself, but all the rest =
will appear as a comma seperated list.  IE;


	HTTP/1.0 200 OK

	Date: Sun, 07 Apr 1996 ? GMT

	Server: Apache/1.0.3

	Content-type: application/bundle-MAB

	Content-encoding: x-compress

	Content-encoding: x-gzip, x-tar

	Content-length: 17428

	Last-modified: Sun, 07 Apr 1996 ? GMT


Unless the Content-encoding: field CAN contain a comma seperated list of =
encodings [notice that the order is preserved, left to right, top to =
bottom...], the enclosed patch is not totally correct.  Unfortunately, to =
break the list out into individual headers would require a much more =
significant patch.


[Jeff:  I noticed that the NetKit DOES correctly handle multiple encoding =
headers with some being comma separated lists.  Most Excellent!  I will =
have a quad-fat compile of Apache to you in short order.]


thanks,

b.bum


int find_ct(request_rec *r)


    int i;

    char *fn =3D pstrdup (r->pool, r->filename);

    mime_dir_config *conf =3D

    (mime_dir_config *)get_module_config(r->per_dir_config, &mime_module);

    char *type;


    if (S_ISDIR(r->finfo.st_mode))=20

        r->content_type =3D DIR_MAGIC_TYPE;

        return OK;

   =20


    if((i=3Drind(fn,'.')) << 0) return DECLINED;

    ++i;


    while ((type =3D table_get (conf->encoding_types, &fn[i])))

       =20

<bold>            if (r->content_encoding =3D=3D NULL)=20

                r->content_encoding =3D type;

             else=20

                table_merge(r->headers_out, "Content-encoding", type);

           =20


</bold>            /* go back to previous extension to try to use it as a =
language */


            fn[i-1] =3D '';

            if((i=3Drind(fn,'.')) << 0) return OK;

            ++i;

       =20


    if ((type =3D table_get (conf->language_types, &fn[i])))

       =20

            r->content_language =3D type;


            /* go back to previous extension to try to use it as a type */


            fn[i-1] =3D '';

            if((i=3Drind(fn,'.')) << 0) return OK;

            ++i;

       =20


    if ((type =3D table_get (conf->forced_types, &fn[i]))

        || (type =3D table_get (hash_buckets[hash(fn[i])], &fn[i])))

       =20

            r->content_type =3D type;

       =20


    return OK;




--NeXT-Mail-1515183957-1--

----- End of forwarded message from Bill Bumgarner -----

-- 
Rob Hartill (robh@imdb.com)
The Internet Movie Database (IMDb)  http://www.imdb.com/
           ...more movie info than you can poke a stick at.

Mime
View raw message