Recap of my project: ScummVM includes a global fully configurable keymapper, but this requires engines to be adapted to use it. Hence the goal of this project is to integrate the customised ScummVM keymapper into the engine’s input handling system. This involves modifying the engine-driven input handling code for mapping user input to in-game actions.

Explanation: ScummVM contains multiple games due to its nature of porting several game engines, each engine has a name and the following list has the Engine name preceding it, thus each pull request is the work I did in integrating the keymap to each of the engine in separate Pull Requests in the course of my Google Summer of Code program.

1. AGOS Engine: Pull Request

2. COMPOSER Engine: Pull Request

3. MADE Engine: Pull Request

4. SAGA Engine: Pull Request

5. HUGO Engine: Pull Request

6. CINE Engine: Pull Request

7. STARK Engine: Pull Request

8. CRUISE Engine: Pull Request

9. GROOVIE Engine: Pull Request

10. LURE Engine: Pull Request

11. Introduce I18N comments and better keymapping labels: Pull Request

12. TOUCHE Engine: Pull Request

13. TRECISION Engine: Pull Request

14. BBVS Engine: Pull Request

15. FREESCAPE Engine: Pull Request

16. ILLUSIONS Engine: Pull Request

17. DRACI Engine: Pull Request

18. MADS Engine: Pull Request

19. TSAGE Engine: Pull Request

20. HOPKINS Engine: Pull Request

21. TINSEL Engine: Pull Request

22. TUCKER Engine: Pull Request

23. SWORD1 Engine: Pull Request

24. TOON Engine: Pull Request

25. Adding I18N Comments: Pull Request

Toon Engine

Toon engine in one of those point and click engines but still require keyboard inputs, it has become easier for me to identify keys and map them into the keymap, thus tackling toon engine was not that big of a hassle.
Here is the PR.

Android Build

One of the highlights of this week was that I was able to run the android build on my PC thanks to my mentors, it was tedious and strenuous process but it was important in my personal development in learning how to use docker-container and what problems could arise when building an android-build, one issue was that as my docker container was in unix environment, some executables were built for windows format so I had to remove carriage returns (windows format) for the build to be able to execute. Now I am able to access ScummVM in android mode from my PC.

TITANIC Engine

There were no demos available for titanic engine and I am running out of engines that I am capable of mapping into keymaps so this is the first game that I bought, it was pretty cheap and interesting so it was not that big of an issue. Now in terms of actually tackling the engine it seems much more complex than anticipated.

The engine is very intertwined when checking for movement, so it took some time to figure it out

it seems the above class takes input then sends it to another function which fetches the movement then this is being used for movement.
As of now the code is not complete or clean so the PR is not ready as of yet.

This week’s work mainly focused on wrapping things up as the end of the project is approaching, and I am striving to refine the details.

Quantitatively, I accomplished the following tasks:

1. I thoroughly tested my code following the correct workflow.
2. In the old version, when matching filesets in the command line, it would repeatedly create the same fileset multiple times (manifested in the logs as multiple occurrences of “Updated fileset: xxxx”). After some debugging, I found the problem. As shown in the diagram, the old version would automatically insert a new fileset after all possible match failures, which seems reasonable. However, I had already implemented the logic of “creating a fileset first before matching” in a previous version. This led to duplicate logs since the outer loop of this code was
   for matched_fileset_id, matched_count in matched_list:,
   meaning that the number of potentially matched filesets would result in the same number of duplicate logs. This issue was minor but stemmed from my lack of careful consideration.
3. I added several new features: for instance, I added checkboxes next to each file on the fileset detail page for developers to conveniently delete unnecessary files, included sorting functionality in the fileset table on the fileset page, and highlighted checksums corresponding to detection types. These features did not involve complex logic, as they were primarily frontend enhancements, and therefore were completed without difficulty.

Looking back at my entire project, I have finished nearly 3000 lines of code over 12 weeks.

I am pleased that most of what I’ve accomplished so far has met expectations. However, there are still some improvements needed before deployment, such as adaptation for MacBinary and other Mac formats, and user IP identification. My sense of responsibility drives me to continue refining this project even after GSOC ends until it can be truly put into production. I look forward to that day! 😄

Tinsel Engine

Tinsel Engine was also somewhat simple but there was an area where I had to figure out where save and load menus are opened so I can disable the keymap, it took me some time to figure out the best places to enable and disable these

This was after intense debugging but I finally figured it out.
Here is the PR.

TUCKER Engine

As like any other engine, figuring out where to disable the keymap was the hardest part but this was still relatively easy.
Here is the PR.

SWORD1 Engine

Atleast for this engine there are functions defined where savescreen is being initialized and being removed so wasy to implement keymaps, one thing weird was that this engine has 2 _keypressed, one belonging to Control class and and 1 to Engine class so had to replicate that for _customType, other than that no major issues were faced
Here is the PR.

This week, I mainly focused on testing the system I previously developed to see if there were any overlooked issues. Fortunately, the problems I discovered were minimal. Below are the issues I fixed and the new features I implemented this week:

1. Marking a fileset as full on the fileset page is straightforward; I just needed to add a button for that.
2. The set.dat file may contain checksums for “sha1” and “crc,” which I initially forgot to ignore. As a result, these two checksum types appeared in the fileset.
3. I added a “ready_for_review” page. I realized I could reuse the file_search page, so I simply created a redirect to this page.
4. Speaking of the file_search page, I finally fixed the pagination error. The issue arose because I hadn’t considered that the query page could include filter conditions, which caused the results to show the entire table (i.e., without any query conditions). This error was hidden in a small detail, but I managed to fix it.
5. I added a user_count based on different user IPs. I implemented a simple prototype, but I’m still contemplating whether there’s a better solution, such as creating a new table in the database to store user IP addresses. I’m not sure if this is necessary and I may consult my mentor for advice later.

Overall, this week has been relatively easy, but I’m still thinking about areas for improvement.