Thanks. when do you get around to re-architecting things, it'd probably be a good idea to mimic various other Unixy tools and set up the option parser to make every boolean option have a no- counterpart.

Ideally, allowing them to appear multiple times on the command line with later occurrences overriding earlier ones. That simplifies scripts and frontends because they don't have to reinvent the understanding of which options override which other ones.

For example:

There is no DLC counter on the game shelf. (first attachment)

There is a data-gameindex attribute.
I've expanded the main download and the DLC download from the HTML DOM. I hope this helps (second attachment)

One other thing:
I can't download the new witcher 2 19GB Linux file with lgogdownloader, it just doesn't do it.

I'm using:

lgogdownloader --check-orphans --download --repair --create-xml automatic --no-subdirectories --language=524287 --platform=7 --game witcher

HTTP: Login successful
API: Login failed

:(

This is where is fails. The downloader expects every game that has DLC to have the DLC counter so it doesn't need to spend lots of time getting info about every game from the account page.
https://github.com/Sude-/lgogdownloader/blob/master/src/downloader.cpp#L1982
This could be fixed by getting info for all games from the account page but it would make Downloader::getGames() much slower

Does the API response (json from debug info of API::getGameDetails) for Surgeon Simulator include info about the DLC?

I could make an exceptions list of games that have DLC but don't have the DLC counter but that would need constant updates.
It would be best if GOG added DLC counter to it.
This is caused by the --repair option which fails because it can't get the XML data
The API response says that there is XML data available but the data doesn't exists and thus it fails with 404 error
I reported this issue here and also sent a message to GOG support about it.

edit: this isn't an ideal solution but as a workaround before GOG adds the XML data you can download
https://sites.google.com/site/gogdownloader/gog_the_witcher_2_assassins_of_kings_1.0.0.3.tar.gz.xml
to $XDG_CACHE_HOME/lgogdownloader/xml/the_witcher_2/ and then use --no-remote-xml

I installed lgogdownloader following the instruction in the OP, thanks! I got a couple of test downloads with success and now I need to work my way through downloading the rest of my library. What I'd like to know is: does the lgogdownloader automatically verify the files for integrity once downloading is finished or do I need to check them manually?

For the test downloads I used the line: and it states 'OK' for each of the files so it all looks good. Should I do this check for every game (with extras) downloaded?

It doesn't verify the files after downloading. The downloader assumes that everything is OK if download finishes with CURLE_OK.

--status checks the hash from remote XML against hash from locally saved XML data
If no locally saved XML data is found then it calculates md5 of the file and compares it to the remote XML
This makes --status fairly fast but it assumes that nothing went wrong during the download or that nothing has corrupted after the download.

--repair is much better for making sure everything is OK as it always hashes the file and compares the chunk hashes to chunk hashes from remote XML data. This also allows it to repair only corrupted chunks.
However --repair takes much longer time because it requires hashing all the files.

--repair is better for checking integrity
--status is more useful for checking if you're missing anything or if something has updated
Both depend on XML data being available. This isn't a problem for installers and patches as GOG provides XML data for them. However extras require you to have saved the XML data with --create-xml because GOG doesn't provide XML data for extras.

Thanks! I'll check again using the --repair command

I didn't change anything when I started downloading, rather than redownload the extras can I make the XML now or is it only possible to create the XML when initiating a download? EDIT: my attempts to guess a command to make the XML failed. Please excuse my lack of certainty on this, I rarely use the Terminal so I feel like I'm guessing most of the time...

On an unrelated note: when I started the download for Fallout it fetched the info for 3 games, does this mean it's about to download Fallout 2 and Tactics as well? EDIT: indeed, --download fallout does get the series, so if one wants to download only the first game is there a way to do that e.g. download "fallout" with the quotation marks?

You could use "--create-xml automatic" but it only works if the download finishes with CURLE_OK so it won't work for resume attempt of already finished downloads as libcurl returns CURLE_RANGE_ERROR

You can also use "--create-xml /path/to/file" to create XML data for specific file.
However the downloader doesn't know what game the XML data belongs to so you'll need to move the XML data to proper subdirectory in ~/.cache/lgogdownloader/xml/ or specify the directory with "--xml-directory ~/.cache/lgogdownloader/xml/gamename/ --create-xml /path/to/file"

There's some code changes here that allow the downloader to create XML even on resume attempts. I'll probably add something similar to git version later

The downloader uses partial matching so it downloads everything that contains "fallout" in it (it's basically the same as regular expression ".*fallout.*"). You want to use exact matching. You can use exact matching by putting it between '^' and '$'
--game "^fallout$"

Thank you once again, I'd add +10 to your posts if I was able to!

I have the next game downloading now with the --create-xml automatic command added. I will go back to the others at a later time and see if I specify the XMLs manually.

^...$ is ideal, I think it will be useful on a few more so that I can make sure I don't go over my bandwidth limits. On the subject on bandwidth limits, is it possible to download a few specific games together? e.g. I want Syberia + Arx Fatalis + Sanitarium. Would the command be: or something else?

lgogdownloader --download --game "^(syberia|arx_fatalis|sanitarium)$"
or
lgogdownloader --download --game "^syberia$|^arx_fatalis$|^sanitarium$"
both should give the same result

edit: you can see which games will be downloaded by using it on the list
lgogdownloader --list --game "^(syberia|arx_fatalis|sanitarium)$"

I can't log in any more, all I get is:

Failed to get login token
HTTP: Login failed

I take it that this is on their end as my log in details are definitely correct and it didn't used to be a problem. In fact I haven't tried downloading things in a couple weeks, so I shouldn't be blocked or anything like that.

I wonder if this means that Galaxy is imminent. I would assume they'd like to have it out in time for TW3 and WL2.

I've been downloading almost non stop for the last 2-3 days, I've got one game 45 min from finishing and another queued up to start immediately after, so we'll see if the next one is halted... EDIT: downloads continued as they should, I'll see if I am still logged in when it's all done.

I notice my library is still sorted by purchase date after I checked stuff earlier, before now lgogdownloader forces the sort order to alphabetical frequently (I assume every time it starts a new file or something)

I'm currently putting this through its paces by getting it to download my library. Seems to work alright so far.

I'm curious though.. does the API give you file sizes? It would be helpful/interesting to be able to be able to do a --du before downloading to make sure I have enough room for --download.

cba5fb9 Make Downloader::HTTP_Login print some more info

Try debug build of the current git version.
It will print a little more info during login.

The reason that --update-check works while --list fails is that --update-check only uses the API. Your API login details are still valid but the website cookies are not valid or there's some other error preventing the downloader getting list of games.

It does give file sizes. Extras don't currently have filesizes because at one point GOG removed extras from the API response for some reason so extras are now parsed from account page. I haven't made code to parse the filesize from the account page yet.