It seems that you're using an outdated browser. Some things may not work as they should (or don't work at all).
We suggest you upgrade newer and better browser like: Chrome, Firefox, Internet Explorer or Opera

×
I've figured out, what caused issues described in post2115.

Ghost, indeed, had its ID changed (from ghost_1_0 to ghost_10). As for other games - it seems like gogrepo has got some problems regarding identifying franchises.

Basically, what I've done: instead of typing whole game's id, for games that feature multiple entries in series, I've used asterisk mask. And since recently, it randomly refuses to work.
For example:

python ./gogrepo.py download -os windows -lang en -ids ys* -skipgalaxy -skipextras
will download every single title from ys franchise, I own. BUT

python ./gogrepo.py download -os windows -lang en -ids metal_slug* -skipgalaxy -skipextras
wont do the same for metal slug franchise, but raise "no game with id was found" error. While mask is kinda correct, according to manifest.

Same has happend with multiple others franchises - some will download, some wont. Weird things:
- I didnt update gogrepo.py for a while. So its not like it got broken after update.
- It used to work in past - I didnt alter a thing in my autodownload script. But somehow its broken now.
- Some asterisked slugs work, while others dont. Amount of used words doesnt matter. For ones, that doesnt recognize masked slugs, typing in whole slugs/ids of games works fine, so its not like these games has suddenly missed from my library.

Welp, Im out of ideas why it randomly happens D:
Post edited October 05, 2019 by Gekko_Dekko
Could someone point me in the direction of the latest stable version of the script?
I haven't updated in almost a year so I'm guessing I'm not using the latest version of the script itself, either :)

-Edit- Nvm, found it :)
Post edited October 19, 2019 by skimmie
A few weeks ago I pretty much finished downloading almost everything by using gogrepo (except divinity original 2 mac files). Now when I try to run an update or download

https://pastebin.com/cC5FNPdZ (because the post would never post)


Is this on GOG's end? My end? . . . . .
avatar
dbzlotrfan: A few weeks ago I pretty much finished downloading almost everything by using gogrepo (except divinity original 2 mac files). Now when I try to run an update or download

https://pastebin.com/cC5FNPdZ (because the post would never post)


Is this on GOG's end? My end? . . . . .
I don't have the game but seems like there is some error in the Changelog which freaks gogrepo out, likely some misplace quotation mark or similar. Not sure why it tries to load Tyranny's Changelog when you try to update PC simulator though...
Are you using kalanyr's version or the original script?
Worst case I think you can delete the gog-manifest.dat and rescan your library
Yep . . . . deleting gog-manifest.dat worked. Actually that text would appear regardless of what I did. -updateonly, -id (insert some game here). Also i'm using eddie's original script. I think I always get errors and issues when trying kalanyr's.
Kalanyr's runs with python3. On my NAS I'm running that one and it works as a charm.
It should work with Python 2, I've maintained backwards compatibility as much as possibility (actually only like the last 3 or so updates were primarily tested on Python 3)
I wonder if there is any way to speed up library update. Coz it takes about half hour to sync 350 items via my connection. Thats... quite a lot.

Maybe download web pages with library as a whole and parse them on client side (if right now it pings gog for every item - not sure if it really works this way, just guessing), in case sync speed was reduced on purpose in order to avoid triggering anti-ddos protection?
Post edited November 13, 2019 by Gekko_Dekko
avatar
Gekko_Dekko: I wonder if there is any way to speed up library update. Coz it takes about half hour to sync 350 items via my connection. Thats... quite a lot.

Maybe download web pages with library as a whole and parse them on client side (if right now it pings gog for every item - not sure if it really works this way, just guessing), in case sync speed was reduced on purpose in order to avoid triggering anti-ddos protection?
It's to do with the structure of things your library has pages, which have games, from the games a fair bit of data is available, unfortunately the important stuff for a download manager isn't (specifically a lot of the data about the individual files). Most of the library update time is getting the real metadata for each file on a file by file basis (that's why some games are much faster than others, because they have fewer files to download).
something weird is going on with tales of maj'eyal (linux) update via gogrepopy.
Synced manifest without errors, then ran update. Everything has downloaded successfully but one file which refuses to complete with following error:

XML verification data size does not match manifest size for tales_of_maj_eyal_1_6_3x_34234.sh. manifest 946123449, received 946124208, skipping.
Again - manifest is up to date. Any ideas why it may happen?

UPD: problem solved. Looks like I ran update in the wrong time - right when they replaces old files with new - so gogdownloader tried to get different files than already existed. Running update & download few hours later fixed everything
Post edited November 27, 2019 by Gekko_Dekko
avatar
Gekko_Dekko: I've figured out, what caused issues described in post2115.

Ghost, indeed, had its ID changed (from ghost_1_0 to ghost_10). As for other games - it seems like gogrepo has got some problems regarding identifying franchises.

Basically, what I've done: instead of typing whole game's id, for games that feature multiple entries in series, I've used asterisk mask. And since recently, it randomly refuses to work.
For example:

python ./gogrepo.py download -os windows -lang en -ids ys* -skipgalaxy -skipextras
avatar
Gekko_Dekko: will download every single title from ys franchise, I own. BUT

python ./gogrepo.py download -os windows -lang en -ids metal_slug* -skipgalaxy -skipextras
avatar
Gekko_Dekko: wont do the same for metal slug franchise, but raise "no game with id was found" error. While mask is kinda correct, according to manifest.

Same has happend with multiple others franchises - some will download, some wont. Weird things:
- I didnt update gogrepo.py for a while. So its not like it got broken after update.
- It used to work in past - I didnt alter a thing in my autodownload script. But somehow its broken now.
- Some asterisked slugs work, while others dont. Amount of used words doesnt matter. For ones, that doesnt recognize masked slugs, typing in whole slugs/ids of games works fine, so its not like these games has suddenly missed from my library.

Welp, Im out of ideas why it randomly happens D:
I am very late for this (and it may have been answered already) but you need to put those id into quotes (simple or double quotes) to avoid the shell trying to expand them. Ie, maybe you had a subdir named "metal_slugsomething" and instead of receiving "metal_slug*" your python received that expanded name instead. So: use:

python ./gogrepo.py download -os windows -lang en -ids "metal_slug*" -skipgalaxy -skipextras
to avoid the shell interpretation of the "*" and have the shell pass it directly to the python script
I was just recreating the gog-cookies.dat file with login credentials, and received an e-mail from GOG with two step authentication code. Which worked, so everything seems to be in order.

But the e-mail was for some reason in some Asian language (I can't tell one alphabet from the other), fortunately the code was still in regular numbers.

Any idea why, and if the script has anything to do with that?
avatar
piranha1: I was just recreating the gog-cookies.dat file with login credentials, and received an e-mail from GOG with two step authentication code. Which worked, so everything seems to be in order.

But the e-mail was for some reason in some Asian language (I can't tell one alphabet from the other), fortunately the code was still in regular numbers.

Any idea why, and if the script has anything to do with that?
Don't think it has anything to do with it, GOG's geolocation just seems to be hit and miss on the regular.

I've randomly been served German, Polish and Russian front pages and emails for no apparent reason, and sometimes trying to change it back doesn't work while it happens. Why GOG doesn't just use whatever language we set our account to is a mystery, but then they're hardly alone in doing it (see Google, Youtube, etc.). Just seems to be more error prone on GOG's end than other sites.
Post edited December 02, 2019 by erephine
avatar
EdhelDil: I am very late for this (and it may have been answered already) but you need to put those id into quotes (simple or double quotes) to avoid the shell trying to expand them. Ie, maybe you had a subdir named "metal_slugsomething" and instead of receiving "metal_slug*" your python received that expanded name instead. So: use: to avoid the shell interpretation of the "*" and have the shell pass it directly to the python script
Actually, that's kind of the point. They want to use the * as a wild card to receive anything that starts with "metal_slug"
EDIT: NVM: I see what you meant now. I missed the last part about passing it to the script.
Post edited December 02, 2019 by paladin181