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 orOpera

×
avatar
Sude: Using --platform with --list is working for me.
No problem here with Debian-packaged 2.24.
Didn’t try the latest git version.
Git version now has support for gogdownloader:// urls (https://github.com/Sude-/lgogdownloader/pull/60)
* lgogdownloader gogdownloader://GAME_NAME/FILE_ID
* lgogdownloader --download-file gogdownloader://GAME_NAME/FILE_ID
* lgogdownloader gogdownloader://GAME_NAME_1/FILE_ID_1,GAME_NAME_2/FILE_ID_2,...
* lgogdownloader --download-file gogdownloader://GAME_NAME_1/FILE_ID_1,GAME_NAME_2/FILE_ID_2,..

Git version also supports specifying output filename (-o, --output-file) when downloading single file with --download-file
https://github.com/Sude-/lgogdownloader/pull/61

This means that it's now possible to use lgogdownloader as download agent with makepkg.

DLAGENTS+=('gogdownloader::/usr/bin/lgogdownloader -o %o %u')
source=("FILENAME::gogdownloader://GAME_NAME/FILE_ID")
Useful for those who are using AUR (Arch User Repository)
Post edited August 15, 2015 by Sude
Something not specific to lgogdownloader, but when I have lots of downloads to queue, a nice CLI-only solution I've found is task-spooler (http://vicerveza.homeunix.net/~viric/soft/ts/), that allows to queue commands to run 1-by-1 or even n-by-n, keep track of their exit codes, outputs, etc.
It works quite well to run lgogdownloader, and it's probably already packaged for your favorite distro ;)
I am trying to compile LGOGDownloader from source, but I keep hitting a problem where the linker says that there are an undefined reference to `boost::filesystem::detail::status'. I am using Ubuntu 12.04 with g++-4.8 (and all dependencies installed, including libboost-filesystem-dev). Any idea why this happens?
Having had a chance to play with this, it looks like it's platform-priority that isn't working for me any more, and --platform wasn't either until I removed the platform-priority line from my config. Can anyone else check this out and confirm?
avatar
hedwards: Hi Sude, I just upgraded from 2.23 to 2.24 and I'm seeing a ton of debugging information that I wasn't seeing before.

Rather than just listing the file name, I'm getting what appears to be the contents of the XML file related to it. It does appear to be downloading anyways, but it is a tad strange.
avatar
Sude: You've probably installed debug build by accident which is really easy to do because "make install" prioritizes debug build and "make" compiles both release and debug builds. Make sure that debug build isn't compiled before "make install" (either do "make release" or just delete the debug binary before "make install").
Yep, that was it. Not sure why GOG didn't notify me sooner.

Now that I think of it, I do remember accidentally typing a command in that I don't normally use for compiling. Can't recall what it was, but that's probably it.
First of all, thanks for building a Linux downloader that has many features the official Windows GOG Downloader is missing.
One question I have, how do I set the --subdir-dlc option to the same path the offical downloader uses. I.e. for the Strike Suit Zero Raptor DLC the official donwloader uses strike_suit_zero_raptor in it's own directory, not in the directory of the main game. I've tried some setting like ../%dlcname% and ../%gamename% and stuff like that, but so far no luck. An option that showed me what path lgogdownloader uses when checking with --status would help finding the right config.
2206e98 Initial support for using platform/language strings to set options

This allows you to use language/platform codes defined in globalconstants.h to set the values for --language, --language-priority, --platform and --platform-priority
in addition to those values it also has a special case "all" which unsurprisingly enables all languages/platforms
for example:
--language en,de
--platform-priority linux,win
--platform all
avatar
Sude: 2206e98 Initial support for using platform/language strings to set options

This allows you to use language/platform codes defined in globalconstants.h to set the values for --language, --language-priority, --platform and --platform-priority
in addition to those values it also has a special case "all" which unsurprisingly enables all languages/platforms
for example:
--language en,de
--platform-priority linux,win
--platform all
Nice. So we can now use --platform linux instead of using numbers?

Also, it would be good to make it osx instead of mac for consistency, but in order not to break things may be it can be extended rather than replaced.
Post edited 3 days ago by shmerl
avatar
shmerl: Nice. So we can now use --platform linux instead of using numbers?

Also, it would be good to make it osx instead of mac for consistency, but in order not to break things may be it can be extended rather than replaced.
I'm currently working on adding regex strings to the options struct and changing Util::getOptionValue to use those regular expressions. This should make it very easy to add new values that it recognizes.
b4fd9fc Use regular expressions for getting language and platform option values

Selecting languages with --language and --language-priority
- Whatever code GOG happens to use in their API (mostly ISO 639-1)
- ISO 639-1
- ISO 639-2/T
- ISO 639-2/B
- English name

Selecting platforms with --platform and --platform-priority
- Windows: "w, win, windows"
- Linux: "l, lin, linux"
- Mac: "m, mac, osx"
avatar
Sude: (…)
In 2.24, a timeout will first delete the partially downloaded file, then start the download back from the beginning. But a manual interrupt (^C) followed by a download of the same file will resume the download from the point we stopped it at.

Could the reaction to timeouts resume the download instead of restarting it, like a manual interrupt/resume do?

(sorry for the noise if it has already been done in git)
avatar
vv221: In 2.24, a timeout will first delete the partially downloaded file, then start the download back from the beginning. But a manual interrupt (^C) followed by a download of the same file will resume the download from the point we stopped it at.

Could the reaction to timeouts resume the download instead of restarting it, like a manual interrupt/resume do?

(sorry for the noise if it has already been done in git)
I thought it worked like that already. At least it worked like that in my test when I implemented the retry on timeout feature but now that I checked the code it seems that it was missing a condition before the delete operation. Perhaps my test matched one of the other conditions and I just assumed it was working properly.

d8c6670 Fix retry on timeout
Post edited 2 days ago by Sude
avatar
Sude: b4fd9fc Use regular expressions for getting language and platform option values

Selecting languages with --language and --language-priority
- Whatever code GOG happens to use in their API (mostly ISO 639-1)
- ISO 639-1
- ISO 639-2/T
- ISO 639-2/B
- English name

Selecting platforms with --platform and --platform-priority
- Windows: "w, win, windows"
- Linux: "l, lin, linux"
- Mac: "m, mac, osx"
Great. Thanks!