Quoting DJ Doena:

Quote:
Quoting AiAustria:

Quote:
First of all, here is a downloadable copy of goodguys former website.

But exactly the CLT file seems to be broken.

Just downloaded the file with your link, extracted the hta file from the zip file from the 7z file on my local drive... -> it work on my Windows 7 machine...

Two remarks:
Far in the back of my head, I can remember there was some security issue, which forces the file to be locally stored on C: (I extracted it to the desktop).
Since I use the export feature, I have installed the commdlg32.ocx as described in the readme...

I am currently using this tool for:
- getting an overview
-- sorting and Grouping by original title
-- selecting by cast or crew
-- drilling into collapsed groups to see profiles and localities
- export to XML (real destination is Excel)

What would be nice to have (wish list for christmas):
- editing of the original title on screen (no write back into the data base for sure)
-- this would make the export smoother and ...
-- the manual counting much easier)
-- preferable by drag and drop (dragging one title onto another syncs the origianl titles
- excluding titles from counting
-- ability to mark features which do not belong to the person currently under investigation
- entering all known name variants at once to get all relevant titles on one screen
-- this would allow to show an exact CLT result for the person under investigation,
-- which could be recalulated on the fly when original titles are altered or features are excluded
- formatting the export
-- usually I only need Original title, year of production and credited as for setting up common name threads...
-- I would need only one export line for the above combination (until now Excel does this job...)

Quoting GSyren:

Quote:
Quoting mediadogg:

Quote:
there is at least one tools developer that likes to build such tools.

I hope you mean Karsten, because I'm not interested. Oh no, not me!

I have way too many unwatched movies to waste time on a project like that, so not me!

It would need screen scraping, and I've never done that, so not me!

Of course there are some interesting screen scraping packages that might be interesting to investigate ... but oh no, not me!

I won't ... I can't ... I shouldn't ... I mustn't ... Oh dagnabbit you mediadogg you! You've planted an idea! I'm not sure anything will come of it, and if something does, it probably won't be anything like goodguy's program. 

   
Can't wait!!!!  Seems like a bunch of loyal members of the community would love it, and it seems tailor-made for your skills. Seriously, I wish you the best if you decide to take a stab at it.

Quoting mediadogg:

Quote:
Seems like a bunch of loyal members of the community would love it, and it seems tailor-made for your skills. Seriously, I wish you the best if you decide to take a stab at it.

Well, I'm sorry to disappoint you. I have looked at it, and I don't understand what his program does, and even if I did I wouldn't have a frickin' clue how to go about it. 

Looks to me that you'd need a Master's degree in screenscraping to pull this one off. If anyone can do it, it would probably be Karsten. I doubt that he's interested, but who knows ...

As I see it there are three basic steps:
- Collect the informatiom
- Collate the information
- Present the information

There are several difficulties that I can see, but the biggest one is just to collect the information.
Just think of how you would do this manually.


Go to the CLT page and enter the name.
Click the first profile and collect the info from the detail window.
Repeat for every profile on that page.
Repeat for every page for that name.

For Clint Eastwood that's 120 pages, and you can see how many profiles.
Now do this with screen scraping ...

I wouldn't hesitate to collate and present the information, but collecting it, well that's just too big a task for a total newbie at screen scraping.

I downloaded goodguy's program and looked at it. It is not a plugin. It is in fact just HTML with embedded Javascript. One big mother of a Javascript, and from what I could tell from a cursory glance, it does screen scraping exactly the way I described. So it does in fact run inside the browser.

If AiAustria can still run it on Win 7, maybe he can tell us how long it takes to gather the CLT info for Clint Eastwood? 

So, if it still works, and still has value, the only issue is that the author does not want it to be used or distributed. That means, as a professional software author, that my hands are tied without his permission. I seem to recall that somebody is able to get in touch with him, and I remember having a decent relationship with him. Is it possible that he would authorize a "clone" project that would allow us to extract / adapt the javascript logic and repackage as an HTML5 or windows or Android app? Would he want to be paid? If so, maybe pool together and buy the rights? I don't mind contributing a few funds and time to the project.

On the other hand, if the community would tolerate a plugin, the developer API has evolved since that program was written. The entire Online is available to a plugin. All profiles, all data, all cast / crew / credits - everything - in real time. I have been messing with it for a different purpose recently, and it works.

It would be fairly simple to create a thin server plugin that would return a data set (say XML) populated with the results of a specific query made on the Online database. Then some other program could handle the UI, query formation, etc. Just give me the query - I return the result - do with it what you will. EZ PZ. (I or somebody else - I don't care if another developer wants to do the thin server.) There would have to be some reasonable limitations. A query that says "return the entire profile XML for all profiles in locality x," would clearly be disallowed.

Any ideas anyone?

Quoting AiAustria:

Quote:
Quoting mediadogg:

Quote:
It would be fairly simple to create a thin server plugin that would return a data set (say XML) populated with the results of a specific query made on the Online database. Then some other program could handle the UI, query formation, etc. Just give me the query - I return the result - do with it what you will. EZ PZ. (I or somebody else - I don't care if another developer wants to do the thin server.) There would have to be some reasonable limitations. A query that says "return the entire profile XML for all profiles in locality x," would clearly be disallowed.

I can't speak for others around here. But for my purpose a simple XML would be enough; Excel is a usable front end for XML data...

What would I need?
All CREDITASSIGN (?) where "credited as" is equal to the prepared name (or better any one out of a small list of names) with the following information:
- name (first/middle/last)
- BY
- credited as

- profile title
- profile production year
- profile original title
- profile UPC/EAN/Disc ID
- profile locality

- role name
- function (if crew member)
if it is not too expensive (the CLT lists it): group/episode the name is listed in

Closing in on it.

Is it possible to model the concept by first looking at the local database, and maybe using ProfilerQuery as the UI model? If the concept works on the local, then it will work with the Online. Then we package the query up in XML format based on the Profiler filter format maybe. Once happy with that, we test the idea of blasting that query into a plugin that executes on the Online instead of the local, and returns exactly the same formatted result that the standalone program has already perfected using the local database. Then the standalone program just does its UI thing, not knowing or caring where the result has come from?

Thoughts?
(Meanwhile I think I will mess around and see if I can replicate the Clint Eastwood query and how much time it would take in a plugin)