Nemcio/ChanSort
1
0
Fork 0
mirror of https://github.com/PredatH0r/ChanSort.git synced 2026-05-07 05:06:31 +02:00
Code Issues Packages Projects Releases Wiki Activity

- added support for Orsay .zip files which contain a Clone/map-AirD and other files similar to Samsung .scm

Browse Source 
- Reference list dialog is now resizable
- Grundig dvb\*_config.xml lists are now separated into TV and Radio lists with individual ordering
- ability to load Grundig dvb\*\_config.xml files containing invalid XML characters (e.g. 0x10)
This commit is contained in:
Horst Beham
2022-04-19 10:36:52 +02:00
parent 0ea5e0ff0c
commit 76d88c63cc
10 changed files with 214 additions and 103 deletions
Download Patch File Download Diff File Expand all files Collapse all files

17
source/ChanSort.Loader.Panasonic/XmlSerializer.cs
View File

@@ -1,10 +1,8 @@
using System;
using System.Collections.Generic;
using System.Collections.Generic;
using System.IO;
using System.Linq;
using System.Reflection;
using System.Text;
using System.Text.RegularExpressions;
using System.Xml;
using System.Xml.Schema;
using ChanSort.Api;
@@ -13,11 +11,20 @@ namespace ChanSort.Loader.Panasonic
{
/*
Panasonic Android TVs (2020 and later) use the same unstandardized compressed .bin binary format as many Philips lists and some Sharp TVs.
Additionally it exports a .xml file with very limited information that only includes channel numbers and channel names truncated at 8 characters.
MediaTek based Android TVs (e.g. Panasonic 2020 and later, Nokia, ...) use the same unstandardized compressed .bin binary format as many
Philips lists and some Sharp TVs.
Additionally it exports a .xml file with very limited information that only includes channel numbers and channel names truncated at 8 bytes.
This truncation makes it impossible for a user to distinguish between channels that have longer names like "Sky Bundesliga ...", therefore
this loader adds the "SvlId" as the ShortName. This SvlId is probably a "service list id" and refers to a a data record inside the .bin file.
The truncation can also happen in the middle of a multi-type UTF-8 character sequence. Non-latin characters, including German umlauts or all
cyrillic characters require 2 bytes/character, effectively reducing the channel name length to 4-8 characters.
Another severe issue with these files is that XML special characters in channel names are not escaped properly. Some preprocessing is required
in order to guess if a "&" is meant as an &amp; data value or an XML attribute. It's likely that < and > inside channel names have the same problem.
When the TV has channels from various sources, it is not possible to determine to which internal source a channel belongs, making sorting of
sub-lists more or less impossible.
*/
class XmlSerializer : SerializerBase