[REL] IMDB

[antp]

IMDb est peut-être encore en train de changer quelque chose, ils font parfois des essais au hasard.
Mais tu avais le même genre de problème aléatoire avec Allociné ? Est-ce vraiment lié au site ou au script ou bien est-ce un truc bizarre chez toi ?
C'est avec quelles options pour "AllActors" et "ActorsLayout" ?

[lobest1]

j'ai fait beaucoup de tests, au début je croyais que c'était allociné qui posait problème car je n'avais pas bien compris comment fonctionnait les acteurs avec ce script.
Dans mon cas j'utilise allociné pour les in infos et images du films et imdb uniquement pour les acteurs (noms, rôles, photo) pour faire apparaitre le nom des acteurs et leur rôles en dessous des leurs photos en utilisant les champs extras, car à priori allociné n'importe pas les photos des acteurs ni les noms et rôles dans des champs extras mais uniquement dans le champ acteurs.

Tu peux tester avec le film From Hell, ça me fait le problème à tous les coups.

Voici mes options (Allactors =2, ActorLayauot =3)


Merci pour ton aide.

[Dorohedoro]

Happy holidays and happy new year, thank you all for the hard work put into this project.

[antp]

Dans mon cas j'utilise allociné pour les in infos et images du films et imdb uniquement pour les acteurs (noms, rôles, photo) pour faire apparaitre le nom des acteurs et leur rôles en dessous des leurs photos en utilisant les champs extras, car à priori allociné n'importe pas les photos des acteurs ni les noms et rôles dans des champs extras mais uniquement dans le champ acteurs.

Tu peux tester avec le film From Hell, ça me fait le problème à tous les coups.

Voici mes options (Allactors =2, ActorLayauot =3)

Ah c'est dans les extras, je pensais que c'était dans la liste des acteurs de la fiche principale.
Je vais essayer de voir ça prochainement

[antp]

Version 5.5.10 to fix the actor names in extras (as described in posts above in French)

[humungus]

It appears IMDb has changed something on their end again. I updated a couple of movies today and in both cases the script failed to retrieve the director.

The movies in question:
https://www.imdb.com/title/tt1315962
https://www.imdb.com/title/tt0829482

Same seems to be true for TV episodes. Example: https://www.imdb.com/title/tt1154812

I also re-checked a movie and an episode I updated two days ago when the director was retrieved, and they, too, failed to do so today: https://www.imdb.com/title/tt0111008
https://www.imdb.com/title/tt1154810

Note that I use script 5.5.09 as I don't need the actors fix in 5.5.10.

[antp]

Fixed as version 5.5.11
I kept the old line in comment in case they change back in the next days, we never know with them

[humungus]

Thank you! I just updated three titles and all seems good.

[lobest1]

Merci beaucoup pour les maj. Je vais tester tout ça

[Dorohedoro]

Is not working at all for me, and when I try to see the source code in the browser it shows some kind of bot detection java script.

Is just me?

[Lerxst]

Not working for me either. I get the message "Movie number could not be extracted from the page (script outdated?)". Script version is 5.5.11.

[lobest1]

Pareil pour moi le script ne fonctionne plus du tout, même pas en collant l'url.
ça renvoie l'erreur : Movie number could not be extracted from the page (sc ipt outdated?)

[Dorohedoro]

Pareil pour moi le script ne fonctionne plus du tout, même pas en collant l'url.
ça renvoie l'erreur : Movie number could not be extracted from the page (sc ipt outdated?)

Thank God I'm not the only one.

I think imdb is trying to block scraping. The API works though.

Maybe it's the time to fully embrace the API? It's there any way to extract actor images using that method?

[antp]

I think imdb is trying to block scraping. The API works though.

Maybe it's the time to fully embrace the API? It's there any way to extract actor images using that method?

Or they just made a tiny change in their page, on a tag that was used by the script to start the analysis.
I'll take a look at it tomorrow, hoping it is just a small change.

[Dorohedoro]

I think imdb is trying to block scraping. The API works though.

Maybe it's the time to fully embrace the API? It's there any way to extract actor images using that method?

Or they just made a tiny change in their page, on a tag that was used by the script to start the analysis.
I'll take a look at it tomorrow, hoping it is just a small change.

Thank you so much.

[antp]

Unfortunately it won't be so easy
It seems that indeed AMC is blocked, but it is not targeted explicitly to AMC via user agent: they use some filtering that detects bots (Firefox using AMC's user agent works fine, for example).
I tried to use what was created for Filmaffinity, via cURL, but it returns then a blank page
In the meantime, until we find a solution for using IMDb itself, the fallback would be the IMDb API script, even if it has less options.

[azasap]

Doesn't works as expected - using IMDb API script returns correct data but somehow saves it into three slots - current prior and next by replacing old data there - it looks we need wait for correct solution

[antp]

What do you mean by "three slots" ? It should just save for the current movie that you import

[azasap]

yes - it should - but I got same data in the selected slot - and in the prior one and in the next one too - it just replaced old data - actually I got three slots in order with same data - found stupid solution by creating three empty slots and using the middle one as target -
UPD. doesn't works at all now

[Dorohedoro]

Unfortunately it won't be so easy
It seems that indeed AMC is blocked, but it is not targeted explicitly to AMC via user agent: they use some filtering that detects bots (Firefox using AMC's user agent works fine, for example).
I tried to use what was created for Filmaffinity, via cURL, but it returns then a blank page
In the meantime, until we find a solution for using IMDb itself, the fallback would be the IMDb API script, even if it has less options.

Yeah, I think those non-stop changes on their end are to stop bot scraping.

Let's hope the free api last, for now that's the only way to get data from there. If not we all should buy the official api from them (:hihi: just $150k)

Thank you for you hard work.