[gstreamer-bugs] [Bug 319986] [RFE][PATCH] GStreamer annodex decoding support

GStreamer (bugzilla.gnome.org) bugzilla-daemon at bugzilla.gnome.org
Tue Nov 29 06:39:31 PST 2005


Do not reply to this via email (we are currently unable to handle email
responses and they get discarded).  You can add comments to this bug at
http://bugzilla.gnome.org/show_bug.cgi?id=319986
 GStreamer | gst-plugins-good | Ver: HEAD CVS





------- Additional Comments From Michael Smith  2005-11-29 14:39 -------
I took a quick first look at the new elements, comments follow:
 - generally looks excellent. Good to see tests, etc!
 - in gst_annodex_parse_headers, the 2nd g_strstr_len() call might pass a
negative length on malformed input; should be checked for.
 - the to/from npt functions disagree on the signedness of what they're dealing
with, which might not matter in practice. But the parser doesn't validate that
the  fields are in-range; should be fixed.
 - bunch of (VL_UNRESOLVED) in the header comment chunks.
 - I have no idea from the description what the 'extract-mode' property does.
The description should be sufficient to decribe what it does.
 - gst_cmml_dec_parse_ident_header() (and probably the other parsing functions,
I didn't check them all) aren't checking that the buffer is large enough for the
data they're trying to read, so they can overrun their buffer.
 - "CmmlEnc", "Coenc/Encoder",  <-- note typo in Codec.
 - You use libxml, but don't link against it? This will only work on some OSes.
 - lines should always be under 80 columns: see the one that starts with
preamble = g_strdup_printf ("<?xml  (and maybe others)
 - "utc" seems like an overly short and undescriptive property name. Also, the
description doesn't mention what format it is (this latter applies to 'timebase'
also).
 - "base" is also undescriptive; no idea what it's meant to do.
 - I don't understand this in skeldec:
+    /* this should be application/x-ogg-skeleton but that would break
+     * autoplugging */
+    GST_STATIC_CAPS ("application/octet-stream")
   Shouldn't you be adding a typefind function for skeleton so you can typefind
this stream?
 - Use of g_strsplit doesn't seem to check that there's at least one element
(though it checks if there's a 2nd element before using it). Also doesn't look
like it frees the tokens array?
 - fisbone->granulerate_* seem to be inconsistently signed.

Well, there's a first pass. Also, see thomas's
gstreamer/docs/random/moving-plugins document; I didn't notice any gtkdoc
documentation for the plugins.

Mike

------- You are receiving this mail because: -------
You are the assignee for the bug.
You are the QA contact for the bug.




More information about the Gstreamer-bugs mailing list