Personal Video Database

How can i fix this?

Select Help > Check for updates to ensure your using the current plugin.

Can we PLEASE lose those captchas, they are close to impossible to read - it took me 20 minutes to register!

No. Without them, we'd be overrun by spammers. Please remove your signature.

Okay, I do see this occasionally—in the same way you mentioned originally. It happens often when changing skins, and is usually fixed by changing movies and/or expanding and collapsing all fields (I'm not sure, but maybe only collapsible fields are affected), but sometimes requires a restart. But it's not happening with your tabbed skin for me. Something else must be different...

To check if it has something to do with the configuration, rename your pvdconf.ini file, then start PVD. Test to see if the behaviour is different with the default configuration. If that makes no difference, test the behaviour with a new database. (When done, you can replace the new pvdconf.ini with the original one.)

Why do you think it's not related to the graphics card? You're probably right—I've never seen any difference after updating my driver.

A problem with Lost was discussed here. We didn't determine the cause, but the solution was to update the series using the TVDb plugin—to add all the seasons/episodes—and then update the episodes with the IMDb plugin.

I works for me now, but I don't normally log in. I believe it's only required for "adult" content. What happens if you remove the login from the plugin configuration?

As a workaround just add custom fields for each user's rating and seen values

Considering these values for one user are probably relevant to all family members, this might be called a "better solution" rather than a "workaround."

Where exactly did you say what exactly?

Take your blinders off...

The IMDb plugin has been somewhat inconsistent lately (probably due to changes in the site) in it's ability to add future episodes for which the data is incomplete (it should still get whatever is available, even if that is just an episode title and URL).

Your post was about your own self-inflicted problems which have little to do with Strat's, or this topic.

Why does it ignore the year in the filename?

It doesn't if there are two movies with the same title in the database. The case where there is only one movie with the same name is probably rare enough it can be ignored. There would probably be as many cases where the year in the filename is inadvertently not exactly the same as the year of the intended target in the database.

So, if it can be done after the fact, why not before, when it is actually needed to be done?

So, yes, that would be possible. But why do something that will just invite a different (and possibly more common) error. Why can't you just add the movie manually in the rare case you're intentionally adding a title that's the same as one already in the database?

I have been searching for any information in regards to RegEX in the help files/wiki etc...  as near as I can tell there is none, and there is no indication that changing of the RegEx search strings is even necessary. There is no indication of what flavour of RegEx is in use or what commands & operators are legal within PVD and even when someone points out an obvious problem related to RegEx search strings, no help is forthcoming. Probably explains why a NON RegEx user ignores it.

You're not looking very hard. There are a number of topics here about regex and the configuration of the file scanner. The information available and the help offered is often ignored by those who are unwilling to consider the necessity of such a system. You claim to know nothing about regex, yet suggest you're at some disadvantage because you're not told what "flavour" of regex is used. Considering the rudimentary level at which regex is used, I'd be very surprised if there where any differences between what is used and any other flavour of regex. Any simple regex reference for beginners will explain all you need to know. In most cases, you don't even need that—to make minor modifications to the regex provided so they work with your file naming practices. The Regex Tester utility is handy for testing modifications, or for creating your own expressions. Or, if you prefer not to fiddle with regex at all, just change your file naming practices to comply with the regex provided.

There's nothing wrong with the pattern matching aspect of the file scanner. The documentation leaves something to be desired. As with any such freeware application, your choices in dealing with that shortcoming are straightforward. Contribute the necessary documentation yourself, or ask the developer how much of donation would be necessary to compel him to create it.

Yes, this is exactly how it happened. The series was Tudors (tt0758790), I had data for 2 first seasons and there are 4 seasons. I deleted series record and then added it again with NMM tool and it pulled in all 4 seasons with all episodes.

You're answering my (quoted) question, so I'll assume your intent to imply you did attempt to add the missing episodes by running the plugin first. But did you Refresh before assuming the episodes were not added? It seems to me they will not appear until the list is refreshed. In any case, I'm unable to reproduce a condition where seasons/episodes are missing and running the plugin fails to add them—except in the case of Lost. I'm sure there could be more, but Lost is the only series I'm aware of that has this problem. I deleted two seasons from Tutors, and the running the plugin added them back. I deleted one episode from season 3, and the running the plugin added it back.

The same happened with the TVDB plugin.

I don't think so. It doesn't work the same way. It uses an API, and downloads a file containing the data for all episodes. So updating a series with it should produce an all or nothing result.

Is there btw any possibility to see e.g. from log files or text files what is actually retrieved?

You can record a log by starting the program with a -debug command line switch. View the log using the command on the Help menu. The log is saved as log.txt in the installation folder. You can always find the last page downloaded as page.html in the installation folder.

Also, this may be important, I am using option that deleted data is not deleted from db, it is just marked inactive I guess.

I suppose that could affect your results, but in all of my tests I permanently deleted whatever it was I wanted to try re-importing.

Related to that, episodes information is very brief with only a few data fields. However, if I select episode & import data, many more fields are actually retrieved. Maybe this is on purpose, I don't know.

In the case of the IMDb plugin, that is by design. Updating the series only adds episodes with the title, URL, year, release date and description. To add additional fields, the episodes must be selected and the plugin run again.

I have a few similar series on my db for testing purposes if you have any suggestions.

Please test those as well. Before running the IMDb plugin, make sure all episodes are visible (i.e., no filters are set). After running the plugin, refresh the list before concluding what happened. If episodes appear to be missing, verify they do exist at the source, and make a copy of the log. If you can find at least one more like Lost, that might make it easier for nostra to figure out what's going wrong.

I should add, that I know most of these problems are caused by the RegEx search strings, but I am not proficient in RegEX and I challenge anyone who does not have RegEx experience to read and understand a RegEx search string ... lol

As I said...

If the file scanner is configured correctly (so the Series title, Season and Episode numbers are correctly obtained from the file name), I would expect the episode record and file to be correctly added to the series.

I'm sorry, but I don't see the point in explaining all that goes wrong when you can't be bothered configuring the file scanner correctly. I've now tested adding new episode files for which there are no records in the database. It correctly adds new episode records, even if the series does not exist. So it's not even necessary to add any records to the database before using the scanner.

Also, if an episode title is similar to an existing movie title that is already in the PVD database, PVD will want to change the file location for the existing movie. The same can occur if two movies released in different years share the same name, irrespective of the year being included in the filename and/or location.

Since a properly configured scanner is going to recognize any episode file as an episode, it's not possible to add it as a movie—existing or otherwise. As for a movie with the same name as one in the database, but a different year (e.g., a remake), the scanner's inability to recognize the file represents a new movie is not unexpected. It's primary role is to match files with records. Finding only one record in the database with the same title, it matches it without checking the year—which is not an unreasonable thing to do. If the second movie has been added to the database, distinguished from the first by its Year, then the scanner will match the file correctly.

The code looks confusing. But it is not as complicated as it seems. There also is a newline character that might add to confusion...

Yes, that's it exactly. My brain had a hard time grasping that, with the #ConditionalAddFirst, the first part of the result came at the end of the statement—on a new line. Thanks for the clear explanation.

not web.archive.com but web.archive.org

Right. I notice many of those links are being redirected to the new site. I wonder if they all will in time. That will make it more difficult to access the old version in the archive.