Major change to how `deftype` shows up in our code:
- the decompiler will no longer emit the `offset-assert`,
`method-count-assert`, `size-assert` and `flag-assert` parameters. There
are extremely few cases where having this in the decompiled code is
helpful, as the types there come from `all-types` which already has
those parameters. This also doesn't break type consistency because:
- the asserts aren't compared.
- the first step of the test uses `all-types`, which has the asserts,
which will throw an error if they're bad.
- the decompiler won't emit the `heap-base` parameter unless necessary
now.
- the decompiler will try its hardest to turn a fixed-offset field into
an `overlay-at` field. It falls back to the old offset if all else
fails.
- `overlay-at` now supports field "dereferencing" to specify the offset
that's within a field that's a structure, e.g.:
```lisp
(deftype foobar (structure)
((vec vector :inline)
(flags int32 :overlay-at (-> vec w))
)
)
```
in this structure, the offset of `flags` will be 12 because that is the
final offset of `vec`'s `w` field within this structure.
- **removed ID from all method declarations.** IDs are only ever
automatically assigned now. Fixes#3068.
- added an `:overlay` parameter to method declarations, in order to
declare a new method that goes on top of a previously-defined method.
Syntax is `:overlay <method-name>`. Please do not ever use this.
- added `state-methods` list parameter. This lets you quickly specify a
list of states to be put in the method table. Same syntax as the
`states` list parameter. The decompiler will try to put as many states
in this as it can without messing with the method ID order.
Also changes `defmethod` to make the first type definition (before the
arguments) optional. The type can now be inferred from the first
argument. Fixes#3093.
---------
Co-authored-by: Hat Kid <6624576+Hat-Kid@users.noreply.github.com>
Started at 349,880,038 allocations and 42s
- Switched to making `Symbol` in GOOS be a "fixed type", just a wrapper
around a `const char*` pointing to the string in the symbol table. This
is a step toward making a lot of things better, but by itself not a huge
improvement. Some things may be worse due to more temp `std::string`
allocations, but one day all these can be removed. On linux it saved
allocations (347,685,429), and saved a second or two (41 s).
- cache `#t` and `#f` in interpreter, better lookup for special
forms/builtins (hashtable of pointers instead of strings, vector for the
small special form list). Dropped time to 38s.
- special-case in quasiquote when splicing is the last thing in a list.
Allocation dropped to 340,603,082
- custom hash table for environment lookups (lexical vars). Dropped to
36s and 314,637,194
- less allocation in `read_list` 311,613,616. Time about the same.
- `let` and `let*` in Interpreter.cpp 191,988,083, time down to 28s.
This change adds a few new features:
- Decompiler automatically knows the type of `find-parent-method` use in
jak 1 and jak2 when used in a method or virtual state handler.
- Decompiler inserts a call to `call-parent-method` or
`find-parent-state`
- Removed most casts related to these functions
There are still a few minor issues around this:
- There are still some casts needed when using `post` methods, as `post`
is just a `function`, and needs a cast to `(function none)` or similar.
It didn't seem easy to change the type of `post`, so I'm not going to
worry about it for this PR. It only shows up in like 3 places in jak 2.
(and 0 in jak 1)
- If "call the handler if it's not #f" logic should probably be another
macro.
Fixes#805
- state handlers that are not inlined lambdas have smarter type
checking, getting rid of 99.9% of the casts emitted (they were not
useful)
- art groups were not being properly linked to their "master" groups.
- `max` in `ja` in Jak 2 was not being detected.
Another huge PR...
This will create a folder like `decompiler_out/jak1/entities` and save a
JSON file per level with all the actors.
Also, it should hopefully make custom level building a little faster.
Fixes empty boxed arrays of strings breaking some decomp
(`ctywide-speech` and `race-info`).
Adds `decomp-as` tag to decompiler types so that the static data
decompiler can use macros like `meters` and `seconds` on fields that
aren't of type `meters` or `time-frame`.
Adds `override` tag to decompiler types which overrides the type of
field with that name. The type must be a child type of the original
field's type (or the same type, but why would you do this?).
Fixes the camera being offset for `drillmtn` after loading `palout`
once.
This is a huge refactor sadly.
Updates the decompiler for the new format and there's new macros. This
new format should be easier to read/parse.
Also rewrote `sp-init-fields!` (both jak 1 and 2) from assembly to GOAL.
Hopefully I did not miss any regressions in Jak 1/2 while updating the
files, it's a lot.
- decompile `neon-baron-part`, which also has the hideout door for some
reason
- improve a few error messages in static data decompilation
- fix bug with disabling fog in merc
Work in progress minimap. Known issues:
- "path finding" doesn't appear to work - it gets stuck forever in many
cases
- some nasty patches around timer-based code
- jak arrow blending issues
- would be nice to make it higher resolution
if the search is forced to terminate due to iteration/time limits, the
icon is not in the right place
Increase level heaps and borrow heaps. The level heap increase was
likely not needed, but better safe than sorry. We allocate the 128 MB
main heap anyway so there's no harm.
Also fix the crash when using `-boot`. As I thought it was just a
one-line typo in the kernel.
Manual patches:
- `drill-turret`: The static data for `*turret-13-path*`,
`*turret-14-path*` and `*turret-15-path*` was decompiled by hand and the
integers in the `set-speed-mult` events have been replaced with boxed
integer arrays that contain only that integer in order to make the
compiler happy. To that effect, the event handler in `target-turret` was
changed to access that array instead of just accessing the int.
- `hover-nav-control`: In `hover-nav-control::10`, `arg2` is usually a
`vector`, but there are some places where it is called with `#t` as
`arg2` and, subsequently, crashes the game because it tries to access
the `quad` of `arg2` if `arg2` is truthy. To mitigate this, the
condition `arg2` has been replaced with `(and (!= arg2 #t) arg2)` (in
this case, it would jump to the `else` that just resets the `dest-vel`
and `transv` `quad`s)
- `drill-baron`: The static data for `*drill-ship-turret-speed-event*`
has been decompiled by hand.
TODOs:
- Jellyfish crash the game
- Destroying the metalhead eggs that are on the breakable wall crashes
the game (already happened with the Peacemaker before)
- Figure out why static data of type `turret-path-event` doesn't
decompile
The docs for all the hover-nav and nav-network code could use some love
in the future, I'm not smart enough to figure out what any of that code
actually means, but it seems to work...
Also threw in the fix for the ▲ that was accidentally left commented
out.
Adding support for better child-type method docstrings. This is a
problem unique to methods.
Usually, a child-type will have the same signature and a common name
will apply, but the implementation is different. This means, you
probably want a different docstring to describe what is happening.
Currently this is possible to do via `:replace`. The problem with
replace is two fold:
- a replaced method ends up in the generated `deftype`...because you
usually change the signature!
- we don't put docstrings in the `deftype` in normal GOAL, this is just
something we do for the `all-types` file (they go in the `defmethod`
instead)
- more importantly, this means anytime you now want to change the
parent's name/args/return type -- you have to apply that change
everywhere.
So this is a better design you can now just declare the method like so:
```clj
(:override-doc "my new docstring" <method_id>)
```
And internally a pseudo-replaced method will be added, but it will
inherit everything from the parent (except the docstring of course)
Unrelated - I also made all the keyword args for declaring methods not
depend on ordering
This also adds support for documenting virtual and non-virtual state
handlers. For example:
```clj
(:states
(part-tester-idle (:event "test") symbol))
```
or
```clj
(idle () _type_ :state (:event "test") 20)
```
I will probably add the ability to give some sort of over-view docstring
at a later date.
Co-authored-by: water <awaterford111445@gmail.com>
- started documenting the files I glossed over, some are totally done,
others are just partially done
- I changed the decompiler to automatically initialize the
art-group-info from the json file. This makes updating gsrc, even a
single file at a time, have consistent naming
- Though I disabled this functionality for jak 1, as I have no idea if
using the ntsc art groups will cause a regression for different versions
- fix indentation for docstrings -- it still doesn't look great, but
this is now a formatting concern, rather than the docstring having a
bunch of happen-stance leading whitespace.
`whack` is missing a single function (`birth-func-whack-score`) that I
couldn't figure out the right args for.
I was not able to test any of this code because I cannot get `atoll` to
extract.
Manual patches in:
- `tomb-boulder`: a few `ppointer->handle`s (I think) were not being
decompiled properly (all used in combination with `clone-anim-once`)
- `ruins-obs`, `pillar-collapse`: `art-joint-anim` casts
- `tomb-beetle`: commented out a call to `shadow-control-method-14` that
was crashing the game when spawning the beetles
- `grunt-mech`: commented out `(.mula.s)` instruction
Notes:
- `enemy-info`'s `idle-anim-script` is most likely a `(pointer
idle-control-frame)`, however, some `nav-enemy-info` labels set it to
`#f` (first encountered in `tomb-beetle`, but also present in `hal`,
`roboguard` and `metalkor-setup`), which crashes the decompiler. This
may become a problem in the future when we eventually get to these
files. For this PR, I made `tomb-beetle` decompile with
`idle-anim-script` set to `#f` and have not noticed any issues/crashes
with that.
- `tomb-boulder` compiles and doesn't crash, but when trying to play the
Daxter chase sequence, the boulder sometimes either spawns at the origin
or spawns in the correct place, but doesn't move.
Co-authored-by: water <awaterford111445@gmail.com>
Adds the following files:
- `amphibian`
- `centurion`
- `ginsu`
- `grenadier`
- `hopper`
- `metalmonk`
- `monster-frog`
- `predator-graph`
- `predator-h`
- `predator`
- `rapid-gunner`
- `rhino`
- `rhino-wall`
- `tomb-baby-spider`
Also adds the DGOs for the following levels:
- Mar's Tomb
- Mountain Temple
- Drill Platform
- Sacred Site (Sage Hut)
Manual patches:
- The decompiler emits `(b! #t cfg-17)` in `(trans hostile hopper)`
without putting a `(label cfg-17)` anywhere
- Added cast to `art-joint-anim` in `(code broken rhino-wall)` and
`(code hit rhino-wall)`
Other notes:
- `amphibian` seems to occasionally crash when using its tongue attack.
Haven't investigated this yet
- `ginsu` crashes after being killed somewhere in the `deactivate`
method, possibly because of its `part-spawner`
- Predators aren't spawning in the forest hunt mission, not sure if
`forest-obs` might be needed for that or if it's something else
- The `rhino-wall` STR animation seems to load (albeit a bit broken due
to missing VAG stream playback), but causes the viewport to shrink as
soon as it's played
- I added `pegasus::74` to the `event_handler_hack` bool in
`variable_naming.cpp` because I got a `none` cast after changing the
return value of `enemy::74`
The things that aren't working right now:
- returning the board (maybe cutscene related?)
- draft until i see if this can be sorted out
- seems to need streaming audio as well, the `scene-player` gets
properly initialized from what i can tell
- had to hack around the streaming audio parts.
I did some manual modifications in a few places to work around some
truly mysterious control flow, and some unsupported stack array stuff.
Also fixes a bug in decompiling static improper lists.
As far as I can tell, guns work, other than some graphical issues and
the crazy particle spawning issue, but I strongly suspect these are
problems with the sparticle/graphics side.
- Split up DGOs between threads in the multithreaded offline test
- fix some random warnings
- make the sig paths decompile a bit nicer to make some files smaller
This solves two main problems:
- the looming threat of running out of memory since every thread would
consume duplicate (and probably not needed) resources
- though I will point out, jak 2's offline tests seem to hardly use any
memory even with 400+ files, duplicated across many threads. Where as
jak 1 does indeed use tons more memory. So I think there is something
going on besides just the source files
- condense the output so it's much easier to see what is happening / how
close the test is to completing.
- one annoying thing about the multiple thread change was errors were
typically buried far in the middle of the output, this fixes that
- refactors the offline test code in general to be a lot more modular
The pretty printing is not enabled by default, run with `-p` or
`--pretty-print` if you want to use it
https://user-images.githubusercontent.com/13153231/205513212-a65c20d4-ce36-44f6-826a-cd475505dbf9.mp4
- fix issue described in
https://github.com/open-goal/jak-project/issues/1939
- fix `text`, which was manually patched with the wrong offset (was
reading the symbol value off by one byte)
- clean up some random useless prints
- make the offline tests keep trying if there's a comparison error,
clean up the output a bit so the diffs are all at the end.
ManDude
