wp_pl grabber (1 Viewer)

[amilewski]

Hi everybody,

I'm constructing new grabber for tv.wp.pl site and have a problem with template text. When I try my grabber with webepg designer v4 everything seems to be OK, but when I run it with webepg.exe some program positions (with no description) are ommited and lost. Can anybody help with this?

Siteurl=http://tv.wp.pl/index.html?T%5Bdate%5D=[YYYY]-[MM]-[DD]&T%5Bstation%5D=[ID]&T%5Bcategory%5D=ALL&T%5Btime%5D=0" post="" external="false" encoding=""

start="Program na"

end="Legenda"

TemplateText=<tr><td><b><#START></b></td><td><b><z(><a></z)?><#TITLE><z(> </a></z)?></b><br><span><z(><#DESCRIPTION><a>więcej</a><b></b><br></z)?></span></td></tr>

tags="TBAS"

You may test it with channelid="1"

Thanks in advance

 

[James]

I have had a report that things work in Designer and not in the latest SVN. I may need to update the design.

Are you using the latest SVN?

I would suggest not using the Z tags for the <a> instead just leave them out with TBS

<tr><td><b><#START></b></td><td><b><a><#TITLE></a></b><br><span><z(><#DESCRIPTION><a>więcej</a><b></b><br></z)?></span></td></tr>

 

[amilewski]

Nice to see you here James, before I started this thread I had been reading other posts and found your suggestions very usefull.
Unfortunatelly, work with this site led me into so many troubles that I' ve decided to abandon this work and start another one with tv.panoramainternetu.pl which seems to provide much more information and channels.
And here I've got another problem to solve which seems to be very common - optional fields.
Let's concider following example of source html:
<tr><td><b>str1</b><a>str2</a><b>str3</b></td></tr>
we parse it to three variables (#v1,#v2,#v3), if all three strings are present in source line variables get correct values, but if line does not contain string2 #v2 gets value of str3 and #v3 becomes empty, in this example #v2 should be empty and #v3 should be of str3.
Does anybody have clue how to achive that?

 

[James]

If there are parsable tags between them it shouldn't be a problem

So the html example should work, but if the you have the source

<td>(#v1,#v2,#v3)</td>

with a template

<td>(<#STR1>,<#STR2>,<#STR3>)</td>

then if the #2, is missing it can be a bit of a problem.

Can you provide more details about the case you are working on. (URL, etc)

 

[amilewski]

Site url: http://tv.panoramainternetu.pl/index.php?strt_hr=-1&cat=-1&day=1176674400&chnl=1

Start: drukuj
End: poprzedni dzie

Tags: T
TemplateText: <TR><TD><FONT><B><#START></B></FONT><BR><BR></TD><TD><FONT><A><#TITLE></A> (<#SUBTITLE>) odcinek <#EPISODE><BR><I><#GENRE></I></FONT><FONT><BR><#SUBTITLES></FONT></TD></TR>

So far only <#START> parses right.

 

[amilewski]

If you add "I" to tags you will get more accurate results but you will lose some broadcasts (not conaining <IMG>)....

 

[amilewski]

again, litle progress but still in trouble

my template text:

<TR><TD><FONT><B><#START></B></FONT><BR><BR></TD>
<TD><FONT><A><B><#TITLE></B></A> (<#SUBTITLE>) odcinek <#EPISODE><BR>
<I><#GENRE></I></FONT>
<FONT><BR><#OPIS><BR><#SUBTITLES></FONT></TD></TR>

my tags:

TF

<#GENRE> never recognized and its value always added to previous variable, it seems that I tag is completely ignored. Why? What can I do with that?

 

[James]

You have told webepg to ignore it.

If you want to use the I tags then you need to add an I to the tags "TFI".

 

[amilewski]

If I add "I" to tags as you suggested I loose some lines because not all lines contain <I> tag and <Z> tag does not help. Is this stalemate?

 

[amilewski]

After many days of tests with number of sites I must say it is very difficult to prepare grabber which is able to grab something more than just start time, title and description (maybe this is due only to sites available in my country). To get more, the parsing engine needs some improvements. The one I want to suggest is to add new option in "SectionTemplate" ie "SearchTags". If "SearchTags" is empty, parser will work the way as it does now to get full compatibility with current grabbers. If it contains anything else, the tags enumerated in "Tags" will be used to find html source lines (as now) and tags from "SearchTags" will be used for finding and matching values.

What, James, do you think about it?