docs(readme): a short front page, the reference sections moved to docs/ as they were #236
No reviewers
Labels
No labels
area/accessibility
area/audio
area/input
area/install
area/menu
area/renderer
area/saves
area/upscaling
good first issue
gpu/amd
gpu/igpu
gpu/intel
gpu/nvidia
help wanted
kind
bug
kind
chore
kind
docs
kind
feature
kind
performance
kind
question
needs measurements
platform/linux
platform/macos
platform/steam-deck
platform/windows
preview
priority
critical
priority
high
priority
low
status
blocked
status
confirmed
status
duplicate
status
invalid
status
needs-info
status
wontfix
testing wanted
ci: approved
No milestone
No project
No assignees
1 participant
Notifications
Due date
No due date set.
Dependencies
No dependencies set
Reference
MagicBeanDev/Longshot!236
Loading…
Reference in a new issue
No description provided.
Delete branch "jarynclouatre/Longshot:docs/readme-for-players"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
Why
The README is 1,124 lines. Someone who only wants to install the game scrolls past the whole menu reference, the command-line switches and the build instructions to find the four steps that matter to them. This leaves a front page of 215 lines and gives each reference section its own page under
docs/.What changed
README.md: the top, Project status, Install and play (the download links and the four steps), What you need, How it works, Want to help?, License and Credits stay as they are. New: a Guides table. Build from source is one line with links.docs/install.md(On Windows, On Linux, Updating, Release channels, RecompLauncher, Where the files go, What Longshot connects to),docs/troubleshooting.md(If something goes wrong, Low FPS or stutter, AMD texture workaround),docs/menu.md(In-game menu, Controls),docs/mods.md,docs/build-from-source.md,docs/command-line.md,docs/pipeline-list.md.docs/menu.mdwhose "below" pointed at a section that is now another page (the "Save my pipeline list" row and the "ASTC as RGBA8" note): they link to it by name.CONTRIBUTING.md,docs/common-issues.md,docs/development.md,docs/getting-started.md,docs/graphics-and-controls.md,docs/saves-script.md,docs/updates-and-saves.md, and in the udmabuf warning ofhost/hos_main.c("see docs/install.md, On Linux") withtests/test_udmabuf_warning.py.docs/menu.md:tests/test_readme_menu_rows.py(file name kept),tests/test_camera_speed.py,tests/test_stick_dead_zone.py, and the website's settings page (website/src/repo-docs/pages.ts,graphics-settings.tsand their tests).website/src/content/docs/requirements.mdxcopies it) and the anchors other places link (#install-and-play,#project-status,#what-you-need,#build-from-source,#credits).Flags and settings
None.
Measurements
main:
1ef93e79(2026-11-13)this PR: docs/readme-for-players @ 3b92826a
Hardware: none used (game-free checks on a Linux x86-64 PC, Python 3.14.4, Node 24.22.2)
Scene: none, the game was not run
Method:
python3 -I -B tests/run_all.py --jobs 4 --timeout 120, two runs each, interleaved (main, PR, main, PR), both runs of a side gave the same counts; inwebsite/,npm ciandnode --test tests/*.test.tson both,npm run buildandnode scripts/check-site.tson the PR; two scripts of mine for the last two rows (below)How to reproduce: the commands above in a checkout of each commit
1ef93e79README.md, linestests/run_all.py(run 1, run 2)test_bake_portability.py,test_display_pointer.pynode --test tests/*.test.tscheck-site.ts.mdfiles outsidewebsite/andthird_party/test_bake_portability.pyfails here for a missing Python module (zstandard);test_display_pointer.pyfails to link on main as well (undefinedrx_option).Rebased since onto main
0edc1377(this PR @ 35e835f0): the rebase carried #108's reworded "HDR tone mapper" row intodocs/menu.mdand changed nothing else of this PR. Run again at 35e835f0: the six checks that read the moved pages (readme_menu_rows,camera_speed,stick_dead_zone,udmabuf_warning,forge_pr,dev_packs: 6 passed), the two scripts (2 lines, 0 of 211 links), and the website's tests: 40 of 47 on main @0edc1377and on this PR @ 35e835f0, the same seven failing on both with "names no condition for HDR tone mapper, which the menu shows only on some PCs" (see Reviewer notes).Rebased again onto main
d9648863(this PR @ f596718e) after a conflict inREADME.mdandwebsite/tests/graphics-settings.test.ts: the rebase carried the ten README lines main gained since0edc1377into the pages they belong to (the "HDR tone mapper" row of #247, the "Damage taken" row of #138 and the paragraph on the Upscaler keeping its render scale of #55 intodocs/menu.md; the Damage taken text of #138 intodocs/command-line.md), kept main's new website test beside this PR's renamed one, and changed nothing else. Run at f596718e: the suite once, 452 passed, 2 failed, 79 skipped (test_bake_portabilityandtest_menu_i18n, which fail the same way on main @d9648863); the website's tests, 53 of 53; the site build andcheck-site.ts("Site check passed: 14 pages, 788 local links and assets, their anchors and edit links"); a script found each of the ten lines in the new pages. The whole suite was not run again on main @d9648863.Rebased again onto main
c0442078(this PR @f96169f8) after a conflict inREADME.mdanddocs/development.md: the rebase carried the README lines main gained sinced9648863intodocs/menu.md(the "Audio output" row and its note with Mono of #136; the "Depends on this PC" paragraph of the bug report of #103; the Toggle paragraph, thetoggle_a=0line and thetoggle_<button>sentences of #137), took the new Discord link into the README, kept main'stoggle_<button>text indocs/development.mdbeside this PR's link tomenu.md#controls, and changed nothing else. Run atf96169f8, main @c0442078against this PR, two interleaved runs each: the suite 458 passed, 3 failed, 80 skipped on both sides in both runs (test_bake_portability,test_menu_keysandtest_pad_layout, which fail the same way on main: the last two since #137, with g++); the website's tests, 52 of 53 on both, the same one failing on both ("restart, coverage and condition drift ... stop publication"); the site build andcheck-site.tson the PR ("Site check passed: 14 pages, 788 local links and assets, their anchors and edit links"). The two scripts: the lines of main's README missing from the new README and the pages are the same set as before the rebase (heading levels, link targets and the two "below" lines); 0 of 213 links unresolved.Rebased again onto main
cf605c85(this PR @de21ed58) after a conflict inREADME.mdandwebsite/src/repo-docs/pages.ts: the rebase carried the README lines main gained sincec0442078into the places they belong to. The Navi 23, Intel and slow-movie rows of the status table, the "Below the minimum" paragraph and the catalogue links (#219, #252, #96) are in the README, whose status and requirements parts stayed. The shader preparation sentence of #252 and the "Caches and logs" row and note of #321 are indocs/install.md; thelogsfolder, the three new table rows and the "Intel laptops" section (#321, #252) indocs/troubleshooting.md; the Output resolution texts of #292 and the "Intel shader compiles" row and note of #252 indocs/menu.md; the Damage taken text of #269 and the movie and Intel compiler paragraphs of #252 indocs/command-line.md. The three links to "Intel laptops" (README,docs/menu.md,website/src/content/docs/requirements.mdx) now point atdocs/troubleshooting.md,pages.tskeeps main's new lines withdocs/menu.mdadded to its source files, and nothing else of this PR changed. Run again on main @cf605c85and this PR @de21ed58: the suite twice each, interleaved, 473 passed, 4 failed, 82 skipped on both (test_bake_portability.py,test_menu_keys.py,test_module_ids.pyandtest_pad_layout.pyfail on main too); the website's tests 71 of 72 on both, the same check failing on both ("restart, coverage and condition drift ... stop publication"); the site build (15 pages) andcheck-site.tspass on the PR (838 local links and assets); the two scripts: 2 lines, 0 of 219 links.Rebased again onto main
5e072fc8(this PR @eff187d7) after a conflict indocs/development.md: #59 added the actor motion switches to theRX_VK_DLSSrow, where this PR had changed "the readme's NVIDIA DLSS section" to "command-line.md's NVIDIA DLSS section". The row now has main's new sentence and this PR's link words; nothing else of this PR changed. Run again on main @5e072fc8and this PR @eff187d7: the suite twice each, interleaved, 474 passed, 4 failed, 83 skipped on both (the same four files fail on main too); the website's tests 71 of 72 on both; the site build (15 pages) andcheck-site.tspass on the PR (838 local links and assets).Rebased again onto main
543846eb(this PR @880a583a) after a conflict inREADME.mdanddocs/development.md: the rebase carried the README lines main gained since5e072fc8into the pages they belong to (the "Custom output size" row and its paragraph of #51, the "Skip title screen" row and paragraph of #399, and the "Walk mode" and "Walk speed" rows, their paragraph and the Walk wording under Controls of #66 intodocs/menu.md; the sentence on where the NGX SDK goes of #465 intodocs/command-line.md), kept main's Custom output size text indocs/development.mdbeside this PR's link tomenu.md#controls, and pointed two checks that are new on main and read the README for those rows (tests/test_skip_title.py,tests/test_walk_key.py) atdocs/menu.md. Nothing else changed. Run on main @543846eband this PR @880a583a, the suite twice each: main 484 passed, 3 failed, 84 skipped (test_bake_portability,test_walk_key,test_crash_record); this PR 485 passed, 2 failed, 84 skipped (test_bake_portability,test_walk_key).test_crash_recordwaits on a signalled child and failed in main's two runs on this PC, not in this PR's two, which ran after them and not in turn with main. The website's tests: 64 of 72 on both, the same eight failing with "graphics coverage mismatch at Custom output size": since #51 the settings page's reader does not take the menu's new text row. For the same reasonnpm run buildstops at that check on this PR, so the built site andcheck-site.tswere not checked on this rebase; the build was not run on main.Testing
Tested: the game-free suite, the website's tests, the site build and its link check, as in Measurements, on a Linux PC. Every line of main's README was compared by script with the new README and the seven pages. Every relative link and anchor in the tracked Markdown files outside
website/andthird_party/was resolved by script before and after.Not tested: the game was not run, and
host/hos_main.cwas not compiled: its change is the wording inside one string literal, the udmabuf warning's "see README, On Linux". The deployed website: built here, not deployed. How Forgejo renders the new pages was not looked at page by page.Needs other testers: None.
Screenshots
Nothing on screen.
Risk and rollback
An open pull request that adds a row to the README's option table, or edits one of the moved sections, will conflict; its lines belong in
docs/menu.mdor the matching page now, andtests/test_readme_menu_rows.pysays so when a menu row has no line there. Links from outside the repository to a moved README anchor (#in-game-menu,#controls,#if-something-goes-wrong,#on-windows,#on-linux,#mods) land at the top of the README, where the Guides table is. Reverting the commit puts everything back.Reviewer notes
git diff --color-moved=zebra main...HEAD -- README.md docs/.0edc1377seven website tests fail since #108 reworded the "HDR tone mapper" row (its cell no longer has the "(shown only with ...)" form the settings page reads the condition from). They pass since: 53 of 53 on this PR @ f596718e.Issues
None.
Checklist
For everyone
type(scope): description).main.python -I -B tests/run_all.pyrun and passing (any failure that also fails onmainis named above).third_party/untouched (a new folder with aPROVENANCE.mdinstead).If applicable
tests/that reads changed source updated in this PR, not weakened.WIP: docs(readme): a short front page, the reference sections moved to docs/ as they wereto docs(readme): a short front page, the reference sections moved to docs/ as they were35e835f060f596718e08Rebased onto main
d9648863after a conflict inREADME.mdandwebsite/tests/graphics-settings.test.ts(now f596718e). The rebase carried the ten README lines main gained since0edc1377into the pages they belong to: the "HDR tone mapper" row (#247), the "Damage taken" row (#138) and the Upscaler paragraph (#55) intodocs/menu.md, the Damage taken text (#138) intodocs/command-line.md. Main's new website test is kept beside this PR's renamed one. Nothing else changed. Run at f596718e: suite 452 passed, 2 failed, 79 skipped (test_bake_portability,test_menu_i18n; both fail on main @d9648863too), website tests 53 of 53, site build andcheck-site.tspass. The description has the detail.f596718e08f96169f84aRebased onto main
c0442078after a conflict inREADME.mdanddocs/development.md(nowf96169f8). The rebase carried main's new README lines intodocs/menu.md: the "Audio output" row and note with Mono (#136), the "Depends on this PC" paragraph (#103), and the Toggle paragraph,toggle_a=0and thetoggle_<button>sentences (#137). The new Discord link is in the README, anddocs/development.mdkeeps main'stoggle_<button>text. Nothing else changed. The suite and the website's tests give the same counts on main and on the branch, two runs each; the numbers are in the description.f96169f84ade21ed5857Rebased onto main
cf605c85after a conflict inREADME.mdandwebsite/src/repo-docs/pages.ts(nowde21ed58). The rebase carried main's new README lines into the places they belong to: the status rows, the "Below the minimum" paragraph and the catalogue links stay in the README (#219, #252, #96); the shader preparation sentence and "Caches and logs" are indocs/install.md(#252, #321); thelogsfolder, three table rows and the "Intel laptops" section indocs/troubleshooting.md(#321, #252); the Output resolution texts and the "Intel shader compiles" row and note indocs/menu.md(#292, #252); the Damage taken text and the movie and Intel compiler paragraphs indocs/command-line.md(#269, #252). The three links to "Intel laptops" now point atdocs/troubleshooting.md. Nothing else changed. The suite and the website's tests give the same counts on main and on the branch, two suite runs each; the numbers are in the description.de21ed5857eff187d7a8Rebased onto main
5e072fc8after a conflict indocs/development.md(noweff187d7). #59 added the actor motion switches to theRX_VK_DLSSrow; the row keeps that sentence and this PR's "command-line.md's NVIDIA DLSS section" wording. Nothing else changed. The suite (twice each) and the website's tests give the same counts on main and on the branch; the numbers are in the description.eff187d7a8880a583a4bRebased onto main
543846ebafter a conflict inREADME.mdanddocs/development.md(now880a583a). The rebase carried main's new README lines intodocs/menu.md(Custom output size of #51, Skip title screen of #399, the Walk rows and text of #66) anddocs/command-line.md(the NGX SDK folder of #465), and pointedtests/test_skip_title.pyandtests/test_walk_key.py, new on main, atdocs/menu.mdfor those rows. Nothing else changed. The suite's and the website's counts on main and on the branch are in the description; the website's tests fail the same eight cases on both since #51 (the settings page does not take the Custom output size row yet), so the site build could not be checked this time.View command line instructions
Checkout
From your project repository, check out a new branch and test the changes.