Posts by KiddaC

I don't really use this plugin, so when I rewrote it, I just did what I presumed best. What I think best is not necessarily how other users use it.

It's first release of new code, so don't expect it to be 100% first time.

regarding h2 to history 2 - I use fuzzy logic which tries to find best matches. In reality h2 is nothing like history 2, so yes it will probably struggle to find that one.

Quote

1. Is it possible to allow go up from the top of the list (to the end of the list)? Sometimes list is too long and go up is more convenient. In addition, I think this is the reason why box do not responds after pressing 2 on top or 8 on bottom. Only restart from button helps.

Don't know why you are struggling with this. 2 and 8 work smoothly on my box for previous and next letter. It has no problem with jumping from top of list to bottom using these buttons.
Quick way to go bottom of the list is press 0 (to go top of the list then up - two button clicks)

Quote

2. Why it jumps to the top of the list after channel assigning (or unassigning)? It is very inconvenient.

It doesn't. If you assign an epg it jumps back to 2 column on the previously selected channel.
unassign stays on the same channel
Unassign all stays on the same channel

Quote

I completely agree with the case mentioned above (H2 - History2). There are many channels like this. It is a problem.

I can put the full list back in no problem. But when I was testing this yesterday, I managed to fill an entire iptv list with no issues from the short list.
I see now that there will be certain channels that will cause an issue though.

Anyone else got bouquet.tv list index crashes, like urie reported???

i don't use suls plugin ever, so I don't even know how it works.
But my references never change. So there should be no mismatch if that is the reason.

You can run the 2 plugins at the same time for the same iptv provider. We will be using different file names and refs, so there will be no clash.

Create some bouquets from suls, create the same bouquets via jedi.

That way you can determine if its the plugin messing things up, or there is something wrong with your EPG sources.

I can replicate your crash only if I change the double quotes to single quotes

#SERVICE 1:7:1:0:0:0:0:0:0:0:FROM BOUQUET 'userbouquet.jmx_iptv1.tv' ORDER BY bouquet

let me explain what that code is doing to see if you can see why yours is different.

So we reading line by line the file etc/enigma2/bouquets.tv

#NAME Bouquets (TV)

..

..

#SERVICE 1:7:1:0:0:0:0:0:0:0:FROM BOUQUET "userbouquet.favourites.tv" ORDER BY bouquet

#SERVICE 1:7:1:0:0:0:0:0:0:0:FROM BOUQUET "userbouquet.LastScanned.tv" ORDER BY bouquet

#SERVICE 1:7:1:0:0:0:0:0:0:0:FROM BOUQUET "userbouquet.jmx_iptv1.tv" ORDER BY bouquet

#SERVICE 1:7:1:0:0:0:0:0:0:0:FROM BOUQUET "userbouquet.jmx_iptv2.xyz.tv" ORDER BY bouquet

the code that is failing for you is

bouquetpath = line.split('"')[1::2][0]

which is basically splitting the line up at every quotemark " and placing the results in a list

line.split('"')

['#SERVICE 1:7:1:0:0:0:0:0:0:0:FROM BOUQUET ', 'userbouquet.jmx_iptv1.tv', ' ORDER BY bouquet']

line.split('"')[1::2] is a slice notation.

slice 0 = '#SERVICE 1:7:1:0:0:0:0:0:0:0:FROM BOUQUET '
slice 1 = 'userbouquet.jmx_iptv1.tv'
slice 2 = ORDER BY bouquet']

So its basically saying start at slice 1 and stop at slice 2
That gives us this.
['userbouquet.jmx_iptv1.tv']

As this value is in a list and I just want the value. then we add the last bit [0]
this gives us this value, and its this value I am trying to get. The name of the bouquet.

line.split('"')[1::2][0]
userbouquet.jmx_iptv1.tv

So I don't know what you have in your bouquets.tv file but it doesn't seem to be standard format.

online python 2.7 tool can be found here
https://repl.it/languages/python

test your iptv lines from your bouquets.tv file

i.e

line='#SERVICE 1:7:1:0:0:0:0:0:0:0:FROM BOUQUET "userbouquet.jmx_iptv1.tv" ORDER BY bouquet'

print(line.split('"'))
print()
print(line.split('"'))[1::2]
print()
print(line.split('"')[1::2][0])

My plugin and suls plugin use a unique custom sid to be used as a reference for the epg. Sounds like every time you are performing an update that unique ref is changing, throwing things out of sync. My Jedi plugin is based on the stream number and a few other calculations so it never changes. But suls might just be sequential. So if your provider adds or deletes a channel, then this might mess things up when an auto update is performed. Sounds plausible from what is written in that other thread

if you look in epgimport folder

etc/epgimport

you will find your source and channels files

in the channel file you will have something like this

<channel id="bbconewmid.slyk">1:0:1:029a:0296:0:0:0:0:0:http%3a//example.m3u8</channel>

<channel id="bbctwohd.slyk">1:0:1:029a:0295:0:0:0:0:0:http%3a//example.m3u8</channel>

<channel id="itvhd.slyk">1:0:1:029a:0294:0:0:0:0:0:http%3a//example.m3u8</channel>

<channel id="channel4hd.slyk">1:0:1:029a:0293:0:0:0:0:0:http%3a//example.m3u8</channel>

<channel id="channel5.slyk">1:0:1:029a:0292:0:0:0:0:0:http%3a//example.m3u8</channel>

<channel id="skyonehd.slyk">1:0:1:1653:7ea:2:11a0000:0:0:0:http%3a//example.m3u8</channel>

<channel id="skywitnesshd.slyk">1:0:1:029a:0137:0:0:0:0:0:http%3a//example.m3u8</channel>

<channel id="skyatlantichd.slyk">1:0:1:029a:0278:0:0:0:0:0:http%3a//example.m3u8</channel>

<channel id="whd.slyk">1:0:1:029a:0652:0:0:0:0:0:http%3a//example.m3u8</channel>

<channel id="goldhd.slyk">1:0:1:029a:0651:0:0:0:0:0:http%3a//example.m3u8</channel>

<channel id="davehd.slyk">1:0:1:029a:0e06:0:0:0:0:0:http%3a//example.m3u8</channel>

this is normally a satellite reference, but as there is no sat ref, we make our own unique number.

I think you are running epgimporter, which is getting this ref.

Then running suls update, which might be changing this ref. So all day until epgimporter runs again, it might be out of sync.

So run suls update first, before your run epg importer update and then things should stay in sync.

I will be back in hour, If there are no bugs mentioned I will then push this new version to git so it can be used in feeds.

Sods law, when I eventually got round to adding this to my git originally, it was broken :)

Not bad for 3 days work, considering I didn't want to do much work on this plugin :)

Jedi EPG Xtream V2.00

New file on post 1

e9e7e5c258dbc42a1926ede46.png

So whats new....

Well basically I gave it some love and refactored all the code and skins.

New skin code, so should be 100% compatible with dreamboxes

New previous letter (2) / next letter (8) code that works on all 4 columns

New fuzzy logic for finding close matches for images that have fuzzywuzzy module in their feeds
Fallbacks to old logic for images that dont have fuzzywuzzy (dreamboxes)

EPG Selection column now only shows close matches for simplicity

Faster downloading and parsing of the files. Huge files are much much quicker now.

EPG sources are no longer sorted and follows the /etc/enigma2/jediepgxtream/epglist.txt file structure

I have commented out the GEN sources in this release. They are UK football blocked and are down more than they are up.

In the epglist.txt there is also now a new Rytec-ALL that can be used instead of all the individual Rytec sources. Note this ALL file is a bit laggy on load though, but more convenient.

New helper text at top, so you know what is going on, rather than just watching a loading circle unaware if anything is happening.

New Logo

New plugin icon

New font - for all languages

Has been tested on python 2.7 / python 3 and dreamboxes

394c99a365505.png

coming very shortly. Code is complete Just being tested now by buddy seagen.

Just change the picker time cron if you want to test. i.e every 5 mins instead of every 30 mins

The picker just gets images from the already downloaded list. So you can run it as frequently as you wish.

why are my file names callled "test..." :P. Did I release the latest version without tidying up the code.

c4566eb3907f2ef0.png

it still works though.

95720c3b983bc79f5.png

open atv had a job manager like this.

02f8be3d0f6bc.png

so if you got similar run the scraper, run the picker script. Then your images should have updated.

Note you will need to fully exit it out of any screens and then go back in to see the change.

you need to manual run your skyscraper.sh once so you got some data other wise you will have to wait until tomorrow until your data downloads

does obh have a script runner.

i could add it into the ipk but then you got the issue of the python folder, being python2.7, python, python3.8 etc

I got a work around. So no big deal.

don't worry about it, fuzzywuzzy is a better module for matching strings for the epg selection.

Dreambox will just have to use the old way.

yes

:quiet: I am re-writing this plugin. V2 coming soon.

mr seagen :)

Do your dreamboxes have fuzzywuzzy in their feeds

opkg install python-fuzzywuzzy

or

opkg install python3-fuzzywuzzy

Quote from seagen

No deb yet since the ascii code crash is still there

That is the code I have meant to have just fixed.

I originally had

import codecs

def getMatchList(self, name, url):

...

...

with codecs.open(temppath, 'r', encoding='utf-8') as epglist:

This crashed my python 2.7

So changed it back to a standard file open

with open(temppath, 'r') as epglist:

This now works fine on my python 2.7 and python 3 images