Pomax / node-flickrapi

A node.js (and client-library) implementation of the Flickr API with oauth API key authentication and API method proxying
177 stars 52 forks source link

Can't get original photos, not authenticated #110

Open DougCollinge opened 5 years ago

DougCollinge commented 5 years ago

I was trying to rescue my original photos before Flickr deletes them but I was not getting them in the download. I am posting this issue for anyone else who is in the same situation.

It turned out to be just because I was not getting authenticated for the getSizes, getInfo, and getPhoto calls. It was necessary to add the following lines, after which everything worked well. Same solution as issue #60.

handlers/downsync/sets.js:38: user_id: flickr.options.user_id, // suppresses an error complaining about no user_id handlers/downsync/sets.js:42: authenticated: true handlers/downsync/photos.js:49: authenticated: true handlers/downsync/photos.js:66: authenticated: true,

Pomax commented 5 years ago

Then for anyone else in this situation: don't use this, just ask Flickr to give you your data back.

Go to https://www.flickr.com/account, and in the lower right on the main account page, simply request all your data (metadata, photos, your flickr mail, comments, everything) with a convenient click. It'll take their system about a day to roll up your entire history as zip files for you to download, after which those'll be available for a generously long time.

In my case, that was one zip file for the metadata, and six zip files for the photographs.

DougCollinge commented 5 years ago

Well, who knew? (You, evidently). I'll do it too and see what the format looks like. Yours is quite nice...

Pomax commented 5 years ago

Everyone who had an account (even a free one) with an email address they actually read rather than a spam account: Flickr's been mailing out the fact that you can do this for the last few weeks several times over =)

As for the format, it's all amazingly conherent: you get .json files that all neatly link together and point to photos by flickr name, with the photo datasets using those same names, so you can really easily write a thing on top for auto generating your own frontend. I'm not a fan of their decision to start deleting images from free accounts, nor their sudden doubling of the pro account price (it used to be $45 per 2 years, now it'll be $50 for one) but at least the Smugmug folks who now own Flickr understand that the data is ours, and that they should help people get their data if they actually care, instead of pulling a Yahoo! and making things even harder to get to.

And as of January, we can finally log in without Yahoo! logins again, at which point I'll probably pick work on this back up. They already took the film strip idea I added to FlickrMirror back in 2013, if they go back to being a normal service I can probably start playing around with new front end concepts that they can then incorporate into Flickr proper again.

DougCollinge commented 5 years ago

All it says is that we can download our content. We have always been able to download our content but the facilities provided were pretty much unusable and didn't provide any metadata. So, yeah, if Smugmug added this new download system then great, good for them. Yahoo was awful, for sure.

Pomax commented 5 years ago

We have absolutely never had a way to download all our content from Flickr itself. Not under original Flickr ownership, and not under Yahoo! ownership. We've only ever been able to use the API for that.