Files
akbasic/docs/08-sprites.md
Andrew Kesterson 16b38c1138 Generate the documentation's figures from the listings they illustrate
Chapters 6 and 8 described what a verb draws in prose. Eight figures now show
it, and each one is produced by running the BASIC listing printed immediately
above it -- so a picture cannot drift away from the code beside it, which is the
way a screenshot goes wrong and the way nothing notices.

tools/screenshot.c is a second SDL host, much smaller than the frontend: dummy
video driver, software renderer, run to completion, read the target back, write
a PNG. It draws no text layer on purpose, so a READY in the corner is not noise
in a figure about BOX and no font has to be resolved.

tools/docs_screenshots.sh reads the new screenshot=NAME fence tag straight out
of the markdown. size=WxH is the second tag, and SCALE's figure uses it: the
point being made is a 320x200 listing filling a larger window, which cannot be
made on a 320x200 surface.

Two gates, answering different questions. docs_examples fails a tagged block
with no image, in both configurations, so a figure cannot be added and
forgotten. docs_screenshots -- a CTest, AKGL build only -- re-renders every
figure and compares byte for byte, so a listing edited without regenerating
fails. Only the second catches a stale picture.

The PNGs are checked in because a reader on the forge has no build tree, and
docs/images/README.md says loudly that they are generated. Regenerating is never
part of a build: the target is run deliberately, so a make cannot put eight
binary diffs in front of whoever ran it.

Drawing the BOX figure caught a defect in TODO.md itself. Deviation 16 claimed
in bold that BOX fills on a negative angle while its own paragraph said the fill
was filed rather than implemented. BOX cannot fill, and filled_rect is reached
by no verb as a result.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
2026-08-01 07:37:04 -04:00

176 lines
5.7 KiB
Markdown

# 8. Sprites
Eight sprites, numbered 1 to 8, as on a C128. They need the SDL build.
A sprite here is a real object in the host's graphics library, registered alongside
whatever the host's own game is drawing — so a game that embeds this interpreter can
see and manipulate a script's sprites.
## Giving a sprite a picture
`SPRSAV` does it, and it takes three kinds of source.
### From an image file
```basic requires=akgl setup=ship
10 SPRSAV "ship.png", 1
```
The most useful form, and not one a C128 has. The path is tried against the working
directory first and then against the directory the running program was loaded from — so
a `.bas` stored beside its artwork works wherever you launch it from. Anything the
image library can decode will do.
A sprite loaded this way takes **the image's own size**. It is not forced to 24 by 21.
### From a saved region
```basic requires=akgl
10 BOX 1, 0, 0, 24, 21
20 SSHAPE A$, 0, 0, 24, 21
30 SPRSAV A$, 1
```
Draw it with the graphics verbs, capture it, install it. The C128 documents `SPRSAV`'s
string as the `SSHAPE` data format at a fixed 24 by 21, so sharing the mechanism is
faithful rather than a shortcut.
### From data
```basic norun
10 DIM P#(63)
20 FOR I# = 0 TO 62
30 READ P#(I#)
40 NEXT I#
50 SPRSAV P#, 1
60 DATA 255, 129, 129, ...
```
63 bytes: three per row, twenty-one rows, most significant bit leftmost. This is the
form a type-in listing uses.
**It takes an integer array, not a string.** A C128 puts the raw bytes in a string; a
string here is NUL-terminated, so it cannot hold a zero byte and therefore cannot hold
a sprite. `DIM P#(63)` is exactly the right size.
Writing a sprite back out — `SPRSAV 1, A$` — is refused. That would be a disk
operation.
## Showing and moving
```basic requires=akgl screenshot=sprites
10 COLOR 1, 2
20 CIRCLE 1, 20, 20, 18, 18
30 PAINT 1, 20, 20
40 SSHAPE A$, 0, 0, 40, 40
50 GRAPHIC 1, 1
60 FOR N# = 1 TO 3
70 SPRSAV A$, N#
80 NEXT N#
90 SPRITE 1, 1
100 SPRITE 2, 1, 6
110 SPRITE 3, 1, 8, 0, 1, 1
120 MOVSPR 1, 30, 80
130 MOVSPR 2, 110, 80
140 MOVSPR 3, 190, 60
```
![](images/sprites.png)
Three sprites from one drawing: the first in the default colour, the second in colour
6, and the third in colour 8 with both expansion bits set, which is what makes it twice
the size. `GRAPHIC 1, 1` on line 50 wipes the drawing the sprites were captured from —
the picture would otherwise still show the original disc in the top-left corner.
**A sprite's colour multiplies the artwork rather than replacing it**, so a white disc
takes the colour cleanly and a coloured one comes out darker than you asked for.
`SPRITE n [,on] [,colour] [,priority] [,xexpand] [,yexpand] [,multicolour]`. Only the
number is required and **an argument you leave out is left alone**, so `SPRITE 1, 1`
turns sprite 1 on without disturbing its colour.
`MOVSPR` has four forms, and the punctuation is what tells them apart:
| Form | What it does |
|---|---|
| `MOVSPR 1, 100, 50` | put it at (100, 50) |
| `MOVSPR 1, +10, -20` | move it *by* that much |
| `MOVSPR 1, 10 ; 90` | move it 10 pixels along bearing 90 |
| `MOVSPR 1, 45 # 8` | set it moving along bearing 45 at speed 8 |
A bearing is degrees **clockwise from straight up**, so 0 is north and 90 is east. In
the polar form the distance comes first and the angle second — the opposite order from
the continuous form, which is a trap worth remembering.
The continuous form does not move anything on the statement itself. The sprite moves as
the program runs, paced by the host's clock. Speed 0 stops it.
Coordinates are the same window pixels the drawing verbs use, not the VIC-II's raster
coordinates — see Chapter 6, and `RGR(1)` and `RGR(2)` for how big the window is.
`SCALE` does not apply to them: a sprite is positioned in device pixels whatever the
drawing verbs are doing.
## Collision
```basic norun
10 COLLISION 1, BUMPED
20 REM ... main loop ...
90 GOTO 20
100 LABEL BUMPED
110 PRINT "HIT: " + BUMP(1)
120 RETURN
```
`COLLISION 1, target` calls a subroutine when two sprites overlap. The handler is
entered between lines and must end in `RETURN`, exactly like a `GOSUB` body. Omitting
the target disarms it.
`BUMP(1)` returns a bitmask of which sprites have collided — bit 0 is sprite 1 — and
**reading it clears it**, which is what makes "has anything hit me since I last looked"
answerable.
Only type 1, sprite-to-sprite, is implemented. Types 2 and 3 are refused by name.
Collision is by bounding box, not by pixel: two sprites whose boxes overlap but whose
artwork does not are reported as colliding.
```basic requires=akgl screenshot=sprite-collision
10 COLOR 1, 2
20 CIRCLE 1, 20, 20, 14, 14
30 PAINT 1, 20, 20
40 BOX 1, 0, 0, 40, 40
50 SSHAPE A$, 0, 0, 40, 40
60 GRAPHIC 1, 1
70 SPRSAV A$, 1
80 SPRSAV A$, 2
90 SPRITE 1, 1, 6
100 SPRITE 2, 1, 3
110 MOVSPR 1, 110, 55
120 MOVSPR 2, 145, 90
```
![](images/sprite-collision.png)
The artwork here includes a border around the whole 40 by 40 sprite, so each sprite
draws its own bounding box. Those boxes overlap at one corner and the two discs are
nowhere near each other — and `BUMP(1)` reports a collision.
## Reading state back
| Function | Gives |
|---|---|
| `RSPPOS(n, 0)` | x |
| `RSPPOS(n, 1)` | y |
| `RSPPOS(n, 2)` | speed |
| `RSPRITE(n, f)` | one of `SPRITE`'s settings, in `SPRITE`'s own argument order |
| `RSPCOLOR(1)` or `RSPCOLOR(2)` | one of the shared multicolour registers |
These read the interpreter's own state rather than asking the device, so they work
even with no device attached.
## SPRDEF
Not implemented, and deliberately. It is an interactive full-screen sprite editor
driven by single keystrokes, not something a program can call — and this interpreter
does not own the screen it would take over. The three `SPRSAV` forms replace it.