Parsing information that is 2 pages deep?

Development Web Sources Discussion
#1

I'm going through a lot of home videos and audio files and trying to add a bunch of tags, cover art, etc. I've been able to create a web source that searches google and returns the JSON results and from there I can go into the first page of the result and pull information.

I'm hitting a new problem in that some of the information that I want to pull is on one more page deep. So basically I want to run a parsescriptindex twice. The first time to get the search results, and then a second time using a URL pulled from the selection of the first time.

For example (and this is only an example so don't get caught up on the specifics) I do a search on Elvis Costello and it returns 2 possible discography pages, one for Elvis Costello and one for Elvis Presley. After selecting the option for Elvis Costello, I would like to have the option to list out the discography and select the album and then parse all the information from this new album page.

Once again, not the exact scenario, but this describes as best as possible what I'm trying to accomplish. I think I might be out of luck and have to do some outside coding and create a txt file to import. But I'm hoping I might get lucky somehow.

Thanks in advance.

#2

Basically, this should be possible.

With the first call using [IndexUrl] you load an "overview" of possibly matching albums.
From this list, you manually pick the matching album.

Then, with a second call using [AlbumUrl], you load the specific data for the selected album.

You put both calls in the same WebSourceScript. Please have a look at the existing WebSourceScripts to see the process in detail.

From the help about "Developing Tag Sources":

image
image793×238 15.4 KB

#3

I need to take it one step further past the albumurl or make the albumurl dynamic based on the results of the indexurl.

So indexurl is searchcom (i'm removing the . because it won't let me post when there is more than 2 links), albumurl then becomes searchresult1com or searchresult2com depending which choice I make. Then a 3rd selection is what I'm going for. From my albumurl I want to choose from that page. i.e.

searchresult1com/a
searchresult1com/b
searchresult1com/c

It is on these pages that I can parse the information that I need. I don't know the albumurl that I need until the indexurl query has been run and then I would need a next step. I think that I'm trying to push this further than the program is designed to go.