playlist/youtube.lua: Parse {title,thumb_url}

Signed-off-by: Toni Gundogdu <legatvs@gmail.com>

Merge branch 'tg/next__add_quvi/time.lua' into next

DOC: quvi-modules.7.txt: Add quvi/time

Signed-off-by: Toni Gundogdu <legatvs@gmail.com>

export/subrip.lua: Use to_timecode from quvi/time module

Now that to_timecode is available through the quvi/time module,
use it and remove the SubRip.to_timecode implementation.

Signed-off-by: Toni Gundogdu <legatvs@gmail.com>

media/arte.lua: Use quvi/time module

Use the quvi/time module now that the to_timestamp function
has been moved from quvi/util module.

Signed-off-by: Toni Gundogdu <legatvs@gmail.com>

Add quvi/time module

* Move to_timestamp from quvi/util to this module
* Add to_timecode_str and to_timecode functions

Signed-off-by: Toni Gundogdu <legatvs@gmail.com>

Merge branch 'tg/next__media_port_senat.lua' into next

tests: Add a test for media/senat.lua

Signed-off-by: Toni Gundogdu <legatvs@gmail.com>

media/senat.lua: Rewrite parse function

Rewrite `parse' function for libquvi 0.9+ series.
* Remove any now obsolete code (e.g. self.host_id)
* Force conversion from iso-8859-1 to UTF-8
* Add Senat.iter_streams function
* Parse {width,height} properties
* Update vim modeline

Signed-off-by: Toni Gundogdu <legatvs@gmail.com>

media/senat.lua: Remove query_formats function

Remove the now obsolete `query_formats' function. libquvi 0.9 no longer
supports this.

media/senat.lua: Rewrite ident function

Rewrite the `ident' function for the 0.9+ series.
* Add local Senat table and implement `can_parse_url' function
* Remove any now obsolete code

media/senat.lua: Relicense under AGPLv3+

See also:
* http://article.gmane.org/gmane.comp.web.flash.quvi/226

Move website/senat.lua to media/senat.lua

Signed-off-by: Toni Gundogdu <legatvs@gmail.com>

Merge branch 'tg/next__media_port_publicsenat.lua' into next

tests: Add a test for media/publicsenat.lua

Signed-off-by: Toni Gundogdu <legatvs@gmail.com>

media/publicsenat.lua: Rewrite parse function

Rewrite "parse" function for libquvi 0.9+ series.
* Remove any now obsolete code (e.g. self.host_id)
* Add PublicSenat.iter_streams function
* Update vim modeline

NOTE:

  This port takes a shortcut to parse the media stream URL and the
  thumbnail URL. They are now parsed from the input URL, rather than a
  secondary URL -- put together using a static URL prefix and the media
  ID. If that, really, is necessary, we can add the additional steps
  later.

Signed-off-by: Toni Gundogdu <legatvs@gmail.com>

media/publicsenat.lua: Remove query_formats function

Remove the now obsolete `query_formats' function. The libquvi 0.9+ no
longer supports this function.

media/publicsenat.lua: Rewrite ident function

Rewrite the `ident' function for the 0.9+ series.
* Add local PublicSenat table and implement `can_parse_url' function
* Remove any now obsolete code from the function

media/publicsenat.lua: Relicense under AGPLv3+

Move website/publicsenat.lua to media/publicsenat.lua

Signed-off-by: Toni Gundogdu <legatvs@gmail.com>

FIX: a typo in URL

Signed-off-by: Toni Gundogdu <legatvs@gmail.com>

Merge branch 'tg/next__media_port_cbsnews.lua' into next

tests: Add test for media/cbsnews.lua

Signed-off-by: Toni Gundogdu <legatvs@gmail.com>

media/cbsnews.lua: Rewrite parse function

Rewrite the "parse" function for libquvi 0.9+ series.
* Add CBSNews.{parse_optional,iter_streams,get_data,to_id} functions
* qargs.id would normally be optional, but required by CBSNews.get_data
* Add CBSNews.{parse_optional,iter_streams,get_data,to_id} functions
* Remove any now obsolete code (e.g. self.host_id)
* Update vim modeline

NOTES:

  quvi/lxp:find_first_tag will raise an error if something isn't found.
  This means that any optional media properties parsed using this
  function, will result in the script exiting with an error.  This is
  acceptable, although not ideal.

Signed-off-by: Toni Gundogdu <legatvs@gmail.com>

media/cbscnews.lua: Remove query_formats function

Remove the `query_formats' function which is no longer used by the
libquvi 0.9+ series.

media/cbsnews.lua: Rewrite ident function

Rewrite the `ident' function for the 0.9+ series.
* Remove any now obsolete code from the function
* Add CBSNews.can_parse_url

media/cbsnews.lua: Relicense under AGPLv3+

Move website/cbsnews.lua to media/cbsnews.lua

Signed-off-by: Toni Gundogdu <legatvs@gmail.com>

FIX: quvi/stream: stream_new: audio.encoding: empty

Set the initial value of audio.encoding to an empty string, rather than
a zero.

Signed-off-by: Toni Gundogdu <legatvs@gmail.com>

gen-news.sh: Do not print commits with >1 one parent

Signed-off-by: Toni Gundogdu <legatvs@gmail.com>

Merge branch 'tg/next__media_port_audioboo.lua' into next

tests: Add media_audioboo

Add a test for the media/audioboo.lua script.

Signed-off-by: Toni Gundogdu <legatvs@gmail.com>

media/audioboo.lua: Rewrite parse function

Rewrite the `parse' function for libquvi 0.9+ series.
* Remove now obsolete code (e.g. self.host_id)
* Make qargs.{title,id} optional
* Add Audioboo.iter_streams
* Update vim modeline

Signed-off-by: Toni Gundogdu <legatvs@gmail.com>

media/audioboo.lua: Remove obsolete query_formats

The `query_formats' is no longer used by libquvi 0.9+, remove the
function from the script altogether.

media/audioboo.lua: Rewrite ident function

Rewrite the `ident' function for libquvi 0.9+ series.
* Remove now obsolete code from the function
* Add Audioboo.can_parse_url

media/audioboo.lua: Relicense under AGPLv3+

Signed-off-by: Toni Gundogdu <legatvs@gmail.com>

Move website/audioboo.lua to media/audioboo.lua

Signed-off-by: Toni Gundogdu <legatvs@gmail.com>

dist: Include playlist/soundcloud.lua

Signed-off-by: Toni Gundogdu <legatvs@gmail.com>

Merge branch 'tg/next__relicense_soundcloud' into next

tests: Enable media_soundcloud

Signed-off-by: Toni Gundogdu <legatvs@gmail.com>

media/soundcloud.lua: Relicense under AGPLv3+

See also:
* http://article.gmane.org/gmane.comp.web.flash.quvi/230

Signed-off-by: Toni Gundogdu <legatvs@gmail.com>

test_media_arte: Rewrite using the capture_*

Besides using the new capture_* function set, pick a random media URL
from the page, instead of the the first one.

Some attempts to access the video pages results in a random binary blob
that the server returns instead of the page HTML. Geoblocked in some
bizarre new way?

Signed-off-by: Toni Gundogdu <legatvs@gmail.com>

Merge branch 'tg/next__revise_test_lib' into next

tests: lib: Revise re.c transl. unit

Add a more flexible set of capture_* functions, replacing the old
`capture' function.

Signed-off-by: Toni Gundogdu <legatvs@gmail.com>

tests: lib: Add slist.c transl. unit

Add slist_free_full to the library. This function provides a convenient
way (for those who still use an earlier version of GLib) to release an
entire singly-linked list. Normally we'd just use g_slist_free_full
(since GLib 2.28).

Signed-off-by: Toni Gundogdu <legatvs@gmail.com>

Merge branch 'tg/next__relicense_arte' into next

tests: media: Enable media_arte

Signed-off-by: Toni Gundogdu <legatvs@gmail.com>

media/arte.lua: Relicense under AGPLv3+

See also:
* http://article.gmane.org/gmane.comp.web.flash.quvi/226

Signed-off-by: Toni Gundogdu <legatvs@gmail.com>

FIX: playlist/soundcloud.lua: item pattern

Signed-off-by: Toni Gundogdu <legatvs@gmail.com>

Merge branch 'tg/next/1.0' into next

Conflicts:
m4/version.m4
share/Makefile.am
share/lua/website/break.lua
share/lua/website/dailymotion.lua
share/lua/website/gaskrank.lua
share/lua/website/quvi/util.lua
share/lua/website/soundcloud.lua
share/lua/website/youtube.lua
tests/data/format/default/audioboo.json
tests/data/format/default/break.json
tests/data/format/default/fixme/bloomberg.json
tests/data/format/default/fixme/collegehumor.json
tests/data/format/default/fixme/mgnetwork.json
tests/data/format/default/funnyordie.json
tests/data/format/default/gaskrank.json
tests/data/format/default/metacafe.json
tests/data/format/default/nsfw/empflix.json
tests/data/format/default/nsfw/fixme/pornhub.json
tests/data/format/default/nsfw/xvideos.json
tests/data/format/default/nsfw/youjizz.json
tests/data/format/default/soundcloud-2.json
tests/data/format/default/soundcloud.json
tests/data/format/default/spiegel.json
tests/data/format/default/theonion.json
tests/data/format/default/theonion_2.json
tests/data/format/default/tvlux.json
tests/data/format/default/videobash.json
tests/data/format/default/vimeo.json
tests/t/expire.t

share/README: Revise

Add missing directories, revise descriptions.

Signed-off-by: Toni Gundogdu <legatvs@gmail.com>

FIX: tests/lib/fetch.c: include "tests.h"

Signed-off-by: Toni Gundogdu <legatvs@gmail.com>

Merge branch 'tg/next/1.0__asciidoc.conf' into tg/next/1.0

man7: Add footer.txt

Signed-off-by: Toni Gundogdu <legatvs@gmail.com>

man7: quvi-modules-3rdparty.7.txt: Revise for footer.txt

* Use linkman from asciidoc.conf
* Include footer.txt

Signed-off-by: Toni Gundogdu <legatvs@gmail.com>

man7: quvi-modules.7.txt: Revise for footer.txt

* Use linkman from asciidoc.conf
* Include footer.txt

Signed-off-by: Toni Gundogdu <legatvs@gmail.com>

man7: libquvi-scripts.7.txt: Revise for footer.txt

* Use linkman from asciidoc.conf
* Include footer.txt

Signed-off-by: Toni Gundogdu <legatvs@gmail.com>

man7: Revise Makefile.am for asciidoc.conf

Add ASCIIDOC_OPTS, call a2x with --asciidoc-opts, and add footer.txt to
EXTRA_DIST.

Signed-off-by: Toni Gundogdu <legatvs@gmail.com>

doc: Add asciidoc.conf

This file adds the linkman macro. See the comment field for the
instructions. This is a modified version from the pacman (of Arch Linux)
source tree uses to produce the manpages: this macro does not insert
<date>.

Signed-off-by: Toni Gundogdu <legatvs@gmail.com>

AC_CHECK_HEADERS: Add string.h

Signed-off-by: Toni Gundogdu <legatvs@gmail.com>

AC_CHECK_FUNCS: Add memset

Signed-off-by: Toni Gundogdu <legatvs@gmail.com>

Workaround: AM_PROG_AR

Signed-off-by: Toni Gundogdu <legatvs@gmail.com>

Merge branch 'tg/next/1.0__cookie' into tg/next/1.0

Bump version to 0.4.14

Signed-off-by: Toni Gundogdu <legatvs@gmail.com>

Update NEWS for v0.4.14

Signed-off-by: Toni Gundogdu <legatvs@gmail.com>

Change soundcloud test URL

Some of the content may be 'protected' in which case the server will
return HTTP/401 (unauthorized) when the stream URL is being accessed.

Signed-off-by: Toni Gundogdu <legatvs@gmail.com>

quvi/const: Remove qoo_http_user_agent

quvi.http.header provides a way to set the UA.

Signed-off-by: Toni Gundogdu <legatvs@gmail.com>

media/dailymotion.lua: Set cookie using quvi.http.header

Signed-off-by: Toni Gundogdu <legatvs@gmail.com>

quvi-modules(7): Document quvi/http/cookie module

Signed-off-by: Toni Gundogdu <legatvs@gmail.com>

Add quvi/http/cookie.lua

This module provides a set of shorthand functions that wrap the use of
quvi.http.cookie function.

Signed-off-by: Toni Gundogdu <legatvs@gmail.com>

quvi/const: Add quvi-object http cookie options

Signed-off-by: Toni Gundogdu <legatvs@gmail.com>

share/Makefile.am: Add wimp.lua

Signed-off-by: Toni Gundogdu <legatvs@gmail.com>

tests: default: add wimp.json

Signed-off-by: Toni Gundogdu <legatvs@gmail.com>

Add website script for wimp.com

Signed-off-by: Martin Herkt <lachs0r@srsfckn.biz>

website utils: add base64 decoder

Signed-off-by: Martin Herkt <lachs0r@srsfckn.biz>

Bump version to 0.4.13

Signed-off-by: Toni Gundogdu <legatvs@gmail.com>

Update NEWS for v0.4.13

Signed-off-by: Toni Gundogdu <legatvs@gmail.com>

tests: default: Change xvideos test URL

Signed-off-by: Toni Gundogdu <legatvs@gmail.com>

quvi/const: Remove qoo_http_cookie

Rename ard.lua to ardmediathek.lua

Signed-off-by: Toni Gundogdu <legatvs@gmail.com>

FIX: ard.lua: Assign self.script_dir to package.path

Assign the value of the self.script_dir to the package.path so that
the 'quvi/*' modules can be found.

Reported-by: Fabian Homborg <fhomborg@gmail.com>
Signed-off-by: Toni Gundogdu <legatvs@gmail.com>

Update NEWS for v0.4.12

Signed-off-by: Toni Gundogdu <legatvs@gmail.com>

Bump version to 0.4.12

Signed-off-by: Toni Gundogdu <legatvs@gmail.com>

Move mgnetwork.lua under FIXME

Signed-off-by: Toni Gundogdu <legatvs@gmail.com>

tests: default: xvideos.json: Update test URL

Signed-off-by: Toni Gundogdu <legatvs@gmail.com>

Move fastjizz.lua under FIXME

Signed-off-by: Toni Gundogdu <legatvs@gmail.com>

tests: default: videobash.json: Update page_title

Signed-off-by: Toni Gundogdu <legatvs@gmail.com>

tests: default: tvlux.json: Update

Signed-off-by: Toni Gundogdu <legatvs@gmail.com>

FIX: tvlux.lua: Update patterns, parse thumb URL

Signed-off-by: Toni Gundogdu <legatvs@gmail.com>

soundcloud-2.json: Update test URL

Signed-off-by: Toni Gundogdu <legatvs@gmail.com>

ard.lua: cleanup

* remove superfluous fsk table from test_availability()
* s/films/videos/ in test_availability()
* move the call to test_availability() into iter_formats(), so it
  gets always executed
* remove superfluous match in container_from()

Merge branch 'tg/next/0.4__ard.lua_misc_improvements' into next

ard.lua: Ard.to_s: Use service internal stream ID

Since the service provides an ID (1,2,3,...), use it in
the media stream ID to make them unique. Otherwise we'll
end up with identical media stream IDs in some cases.

Use 'i' for the lack of a better prefix. This is similar to
that of youtube.lua (0.9) and the use of the "itag" value in
YouTube the media stream IDs.

Signed-off-by: Toni Gundogdu <legatvs@gmail.com>

ard.lua: Cleanup Ard.iter_formats

Realign, remove webX extraction which is now done in Ard.quality_from,
make the for loop more compact.

Signed-off-by: Toni Gundogdu <legatvs@gmail.com>

ard.lua: Add Ard.container_from function

ard.lua: Add Ard.height_from function

ard.lua: Add Ard.quality_from function

Have the Ard.quality function return a string containing one of the
strings 'ld', 'md', 'sd', 'hd'. Other websites.

Plenty of guessing going on here, these strings are roughly based on
the findings posted here:
  http://article.gmane.org/gmane.comp.web.flash.quvi/170

  0 = '256x144' -- ld
  1 = '512x288' -- md
  2 = '960x540' -- sd
  3 ='1280x720' -- hd

ard.lua: Rewrite Ard.to_s for 0.9 compatibility

The 0.9 series of libquvi-scripts will attempt to return the media
stream IDs as "$quality_$container_$encoding_$height(p)" whenever
possible.  Prepare for the foreseeable future.

Add ard.lua

This adds support for the 'ARD mediathek', the movie archive of the
primary national television broadcaster of Germany. It's mediathek also
servers videos from many smaller german broadcasters.

Note: films are only available online for two weeks after airing.
Certain films are only available at certain daytimes. In this case an
error is raised.

Signed-off-by: Thomas Weißschuh <thomas@t-8ch.de>
Tested-by: Toni Gundogdu <legatvs@gmail.com>

FIX: wdrmaus.lua:101: invalid escape sequence

Fix the "wdrmaus.lua:101: invalid escape sequence near \/" by
removing the backslash.

Reported-by: Thomas Weißschuh <thomas@t-8ch.de>
Signed-off-by: Toni Gundogdu <legatvs@gmail.com>