Open Mardeg opened 4 months ago
The short answer is that the games.xml
file does not get its data from the detection_tables.h
files. Rather, it gets its info from this spreadsheet. Someone has to do the legwork to add that data, and we'd welcome your contribution.
Edit access is reserved for team members. You can make a copy, edit it, and then give the link back to be updated with all the games you want to see updated.
You can make a copy, edit it, and then give the link back to be updated with all the games you want to see updated.
Okay I initially added info for a dozen games, half of which are historical (pre-1990) ones for you to check. Without a google account I could only download the spreadsheet and I'm not sure which uploaded format you prefer so I'll link to both .xlsx and .ods files. Also I'm on Linux so used OpenOffice to edit it.
Please let me know if there are issues/corrections needed to the data in the files before I go off repeating any editing/formatting errors I made:
Thanks for your work. I had a look at the spreadsheet, but it's quite hard to keep track of your changes if you use a standard file instead of the cloud. If you really can't use a google account I can create a copy in my account and give you editor access via email. Would you be interested?
Thanks for your work. I had a look at the spreadsheet, but it's quite hard to keep track of your changes if you use a standard file instead of the cloud.
If you go to the "game" sheet and sort the "year" column (A to Z) the added lines are easier to track viewing both versions side by side. Since there are only 12 I'll list them here by "id" (year):
glk:adventcrowther (1976) glk:adv350_zcode (1977) glk:transylvaniav2 (1985) glk:lgop (1986) glk:beyondzork (1987) glk:questforexcalibur (1989) glk:hobbitredux_zcode (2001) glk:risorg_zcode (2003) glk:if10_gloriousrevol (2010) glk:stc18_illum (2018) glk:eehunt2000 (2020) glk:if20_vainempires (2020)
Is there more involved than importing to the cloud one of my linked files, then just copy pasting those new lines from my version to yours? It's likely what I'd do if I used a google account.
Now it's not a problem since the entries are a few of course, but if you plan to continue the work the entries to add are a lot. If you do it on the cloud we can see as you gradually add the games (and google spreadsheets also has a history feature), which would be simpler to manage.
Anyway if you really prefer to work offline, go ahead, and share a new file when you have added a significant number of entries
Okay I think to avoid being my own worst enemy here I should break this into smaller bugs with this becoming a meta tracking bug so each can be resolved sooner with data being added quicker, and I'm now convinced to make an account so I can do the work via google spreadsheets.
Would it be okay for the manner of splitting the work up to be: One bug per year of games from all comps, one bug for those pre-1990 historic games, plus extra periodic bugs as years are found for others?
There's no need to open additional bugs, you can work at your own pace, and add a comment here to give updates (e.g. when you finish a full "year"). If you want to have a means to track your work maybe you can use checkboxes as in issue #322 It'd probably be a good idea to work chronologically, starting with the historical games
Perfect! Here's my googlesheet copy with the first task done!
Nice, can you give public read access to you spreadsheet? Thanks
Nice, can you give public read access to you spreadsheet? Thanks
Oops! Sorry, try it now ðŸ¤
Ok thanks I added the historic entries
The six entries in the detection tables from IFComp 1995 added. One entry was authored in 1988 so I added that year for it.
I'm currently leaving the moby_id cell blank when the game does not have one (half of the games I added did not) as I see it is blank for other games in that sheet, though I see some games listed that include a value of -1. Which is preferred?
The six entries in the detection tables from IFComp 1995 added. One entry was authored in 1988 so I added that year for it.
Thanks, please complete more years and then I'll add those too
I'm currently leaving the moby_id cell blank when the game does not have one (half of the games I added did not) as I see it is blank for other games in that sheet, though I see some games listed that include a value of -1. Which is preferred?
Doesn't really matter, but it's preferable to put -1
When I sort my installed games by "Year" in the ScummVM UI I'm seeing far too many text adventures listed under "UNKNOWN YEAR" where we in fact know the year from our very own code. https://github.com/scummvm/scummvm/blob/master/engines/glk/glulx/detection_tables.h and https://github.com/scummvm/scummvm/blob/master/engines/glk/zcode/detection_tables.h have code comments mentioning release years.
Also historically/culturally significant ones should have an entry in games.xml such as Will Crowther's original, pre-Woods, 1976 Adventure aka Colossal Cave from https://ifarchive.org/if-archive/games/zcode/old/Advent_Crowther-r2.z5 and the 1986 Leather Goddesses of Phobos from https://web.archive.org/web/20190611060403id_/https://eblong.com/infocom/gamefiles/leathergoddesses-invclues-r4-s880405.z5
Something, maybe even this very line in the github issue description, tells you this is the right thing to do.