Use the online Poryscript Playground to test it out, and install the VS Code Poryscript extension to make writing scripts even easier!
Poryscript is a higher-level scripting language that compiles into the scripting language used in pokeemerald, pokefirered, and pokeruby. It makes scripting faster and easier. Some advantages of using Poryscript are:
- Branching control flow with
if
,elif
,else
,while
,do...while
, andswitch
statements. - Inline text
- Auto-formatting text to fit within the in-game text box
- Better map script organization
View the Changelog to see what's new, and download the latest version from the Releases.
Table of Contents
- Usage
- Poryscript Syntax (How to Write Scripts)
- Local Development
- Versioning
- License
- Acknowledgments
Poryscript is a command-line program. It reads an input script and outputs the resulting compiled bytecode script. You can either feed it the input script from a file or from stdin
. Similarly, Poryscript can output a file or to stdout
.
> ./poryscript -h
Usage of poryscript:
-cc string
command config JSON file (default "command_config.json")
-f string
set default font id (leave empty to use default defined in font config file)
-fc string
font config JSON file (default "font_config.json")
-h show poryscript help information
-i string
input poryscript file (leave empty to read from standard input)
-l int
set default line length in pixels for formatted text (uses font config file for default)
-lm
include line markers in output (enables more helpful error messages when compiling the ROM). (To disable, use '-lm=false') (default true)
-o string
output script file (leave empty to write to standard output)
-optimize
optimize compiled script size (To disable, use '-optimize=false') (default true)
-s value
set a compile-time switch. Multiple -s options can be set. Example: -s VERSION=RUBY -s LANGUAGE=GERMAN
-v show version of poryscript
Convert a .pory
script to a compiled .inc
script, which can be directly included in a decompilation project:
./poryscript -i data/scripts/myscript.pory -o data/scripts/myscript.inc
To automatically convert your Poryscript scripts when compiling a decomp project, perform these two steps:
- Create a new
tools/poryscript/
directory, and add theporyscript
command-line executable tool to it. Also copycommand_config.json
andfont_config.json
to the same location.
# For example, on Windows, place the files here.
pokeemerald/tools/poryscript/poryscript.exe
pokeemerald/tools/poryscript/command_config.json
pokeemerald/tools/poryscript/font_config.json
It's also a good idea to add tools/poryscript
to your .gitignore
before your next commit.
- Update the Makefile with these changes (Note, don't add the
+
symbol at the start of the lines. That's just to show the line is being added.):
FIX := $(TOOLS_DIR)/gbafix/gbafix$(EXE)
MAPJSON := $(TOOLS_DIR)/mapjson/mapjson$(EXE)
JSONPROC := $(TOOLS_DIR)/jsonproc/jsonproc$(EXE)
+ SCRIPT := $(TOOLS_DIR)/poryscript/poryscript$(EXE)
include audio_rules.mk
+AUTO_GEN_TARGETS += $(patsubst %.pory,%.inc,$(shell find data/ -type f -name '*.pory'))
generated: $(AUTO_GEN_TARGETS)
%.s: ;
%.png: ;
%.pal: ;
%.aif: ;
+ %.pory: ;
%.rl: % ; $(GFX) $< $@
+ data/%.inc: data/%.pory; $(SCRIPT) -i $< -o $@ -fc tools/poryscript/font_config.json -cc tools/poryscript/command_config.json
If you're working on a large project, you may want to convert all of the existing scripts.inc
files to their scripts.pory
equivalents. Since there are a large number of script files in the Gen 3 projects, you can save yourself a lot of time by following these instructions. Again, this is completely optional, and you would only want to perform this bulk conversion if you're emabarking on large project where it would be useful to have all the existing scripts setup as Poryscript files.
Click Here to View Instructions
Convert all of your projects old map scripts.inc
files into new scripts.pory
files while maintaining the old scripts:
-
Create a file in your
pokeemerald/
directory namedconvert_inc.sh
with the following content:#!/bin/bash for directory in data/maps/* ; do pory_exists=$(find $directory -name $"scripts.pory" | wc -l) if [[ $pory_exists -eq 0 ]]; then inc_exists=$(find $directory -name $"scripts.inc" | wc -l) if [[ $inc_exists -ne 0 ]]; then echo "Converting: $directory/scripts.inc" touch "$directory/scripts.pory" echo 'raw `' >> "$directory/scripts.pory" cat "$directory/scripts.inc" >> "$directory/scripts.pory" echo '`' >> "$directory/scripts.pory" fi fi done
-
Run
chmod 777 convert_inc.sh
to ensure the script executable.
Finally you can execute it in your pokeemerald/
directory by running ./convert_inc.sh
or bash convert_inc.sh
in the console. This script will iterate through all your data/map/
directories and convert the scripts.inc
files into scripts.pory
files by adding a raw
tag around the old scripts. convert_inc.sh
will skip over any directories that already have scripts.pory
files in them, so that it will not overwrite any maps that you have already switched over to Poryscript.
For VS Code, you can install the Poryscript extension, which provides quality-of-life improvements such as autocomplete, syntax highlighting, and error diagnostics.
For other editors with Tree-sitter support (e.g. emacs, neovim, lapce, zed, helix, etc.), some Poryscript bindings are supported here: https://github.com/Elsie19/treesitter-poryscript
A single .pory
file is composed of many top-level statements. The valid top-level statements are script
, text
, movement
, mart
, mapscripts
, and raw
.
mapscripts MyMap_MapScripts {
...
}
script MyScript {
...
}
text MyText {
"Hi, I'm some text.\n"
"I'm global and can be accessed in C code."
}
movement MyMovement {
walk_left
walk_right * 3
}
mart MyMart {
ITEM_POTION
ITEM_POKEBALL
}
raw `
MyLocalText:
.string "I'm directly included.$"
`
The script
statement creates a global script containing script commands and control flow logic. Here is an example:
script MyScript {
# Show a different message, depending on the state of different flags.
lock
faceplayer
if (flag(FLAG_RECEIVED_TOP_PRIZE)) {
msgbox("You received the best prize!")
} elif (flag(FLAG_RECEIVED_WORST_PRIZE)) {
msgbox("Ouch, you received the worst prize.")
} else {
msgbox("Hmm, you didn't receive anything.")
}
release
end
}
As you can see, using if
statements greatly simplifies writing scripts because it does not require the author to manually define new sub-labels with goto
statements everywhere.
if
statements can be nested inside each other, as you would expect.
if (flag(FLAG_TEMP) == true) {
if (var(VAR_BADGES) < 8) {
...
} else {
...
}
}
Note the special keyword elif
. This is just the way Poryscript specifies an "else if". Many elif
statements can be chained together.
Compound boolean expressions are also supported. This means you can use the AND (&&
) and OR (||
) logical operators to combine expressions. For example:
# Basic AND of two conditions.
if (!defeated(TRAINER_MISTY) && var(VAR_TIME) != DAY) {
msgbox("The Cerulean Gym's doors don't\n"
"open until morning.")
}
...
# Group nested conditions together with another set of parentheses.
if (flag(FLAG_IS_CHAMPION) && !(flag(FLAG_SYS_TOWER_GOLD) || flag(FLAG_SYS_DOME_GOLD))) {
msgbox("You should try to beat the\n"
"Battle Tower or Battle Dome!")
}
while
statements are used to do loops. They can be nested inside each or inside if
statements, as one would expect.
# Force player to answer "Yes" to NPC question.
msgbox("Do you agree to the quest?", MSGBOX_YESNO)
while (var(VAR_RESULT) != 1) {
msgbox("...How about now?", MSGBOX_YESNO)
}
setvar(VAR_QUEST_ACCEPTED, 1)
The while
statement can also be written as an infinite loop by omitting the boolean expression. This would be equivalent to while(true)
in typical programming languages. (Of course, you'll want to break
out of the infinite loop, or hard-stop the script.)
while {
msgbox("Want to see this message again?", MSGBOX_YESNO)
if (var(VAR_RESULT) != 1) {
break
}
}
do...while
statements are very similar to while
statements. The only difference is that they always execute their body once before checking the condition.
# Force player to answer "Yes" to NPC question.
do {
msgbox("Can you help me solve the puzzle?", MSGBOX_YESNO)
} while (var(VAR_RESULT) == 0)
break
can be used to break out of a loop, like many programming languages. Similary, continue
returns to the start of the loop.
The condition operators have strict rules about what conditions they accept. The operand on the left side of the condition must be a flag()
, var()
, defeated()
, or AutoVar check. They each have a different set of valid comparison operators, described below.
Type | Valid Operators |
---|---|
flag |
== |
var or AutoVar |
== , != , > , >= , < , <= |
defeated |
== |
All operators support implicit truthiness, which means you don't have to specify any of the above operators in a condition. Below are some examples of equivalent conditions:
# Check if the flag is set.
if (flag(FLAG_1))
if (flag(FLAG_1) == true)
# Check if the flag is cleared.
if (!flag(FLAG_1))
if (flag(FLAG_1) == false)
# Check if the var is not equal to 0.
if (var(VAR_1))
if (var(VAR_1) != 0)
#Check if the var is equal to 0.
if (!var(VAR_1))
if (var(VAR_1) == 0)
# Check if the trainer has been defeated.
if (defeated(TRAINER_GARY))
if (defeated(TRAINER_GARY) == true)
# Check if the trainer hasn't been defeated.
if (!defeated(TRAINER_GARY))
if (defeated(TRAINER_GARY) == false)
When not using implicit truthiness, like in the above examples, they each have different valid comparison values on the right-hand side of the condition.
Type | Valid Comparison Values |
---|---|
flag |
TRUE , true , FALSE , false |
var |
any value (e.g. 5 , VAR_TEMP_1 , VAR_FOO + BASE_OFFSET ) |
defeated |
TRUE , true , FALSE , false |
One quirk of the Gen 3 decomp scripting engine is that using the compare
scripting command with a value in the range 0x4000 <= x <= 0x40FF
or 0x8000 <= x <= 0x8015
will result in comparing against a var
, rather than the raw value. To force the comparison against a raw value, like 0x4000
, use the value()
operator. For example:
if (var(VAR_DAMAGE_DEALT) >= value(0x4000))
The resulting script use the compare_var_to_value
command, rather than the usual compare
command.
Regular non-branching commands that take arguments, such as msgbox
, must wrap their arguments in parentheses. For example:
lock
faceplayer
addvar(VAR_TALKED_COUNT, 1)
msgbox("Hello.")
release
end
Use end
or return
to early-exit out of a script.
script MyScript {
if (flag(FLAG_WON) == true) {
end
}
...
}
A switch
statement is an easy way to separate different logic for a set of concrete values. Poryscript switch
statements behave similarly to other languages. However, the cases break
implicitly. It is not possible to "fall through" to the next case by omitting a break
at the end of a case, like in C. You can use break
to break out of a case, though--it's just not required. Multiple cases can be designated by listing them immediately after another without a body. Finally, an optional default
case will take over if none of the provided case
values are met. A switch
statement's comparison value must always be a var()
operator. Of course, switch
statements can appear anywhere in the script's logic, such as inside while
loops, or even other switch
statements.
switch (var(VAR_NUM_THINGS)) {
case 0:
msgbox("You have 0 things.")
case 1:
case 2:
msgbox("You have 1 or 2 things.")
default:
msgbox("You have at least 3 things.")
}
Labels can be defined inside a script
, and they are very similar to C's goto
labels. A label isn't usually desired or needed when writing Poryscript scripts, but it can be useful and in certain situations where you might want to jump to a common part of your script from several different places. To write a label, simply add a colon (:
) after a name anywhere inside a script
. Labels are rendered as regular assembly labels, and they can be marked as local or global. By default, labels have local scope, but they can be changed to global scope using the same syntax as other statements (e.g. MyLabel(global):
).
Label Example:
// Note, this is a bad example of where a
// label would be useful.
script MyScript {
lockall
if (flag(FLAG_TEST)) {
goto(MyScript_End)
} elif (flag(FLAG_OTHER_TEST)) {
addvar(VAR_SCORE, 1)
goto(MyScript_End)
}
MyScript_End:
releaseall
}
Use text
to include text that's intended to be shared between multiple scripts or in C code. The text
statement is just a convenient way to write chunks of text, and it exports the text globally, so it is accessible in C code. Currently, there isn't much of a reason to use text
, but it will be more useful in future updates of Poryscript.
script MyScript {
msgbox(MyText)
}
text MyText {
"Hello, there.\p"
"You can refer to me in scripts or C code."
}
A small quality-of-life feature is that Poryscript automatically adds the $
terminator character to text, so the user doesn't need to manually type it all the time.
Text auto-formatting is also supported by Poryscript. The format()
function can be wrapped around any text, either inline or text
, and Poryscript will automatically fit the text to the size of the in-game text window by inserting automatic line breaks. A simple example:
msgbox(format("Hello, this is some long text that I want Poryscript to automatically format for me."))
Becomes:
.string "Hello, this is some long text that I\n"
.string "want Poryscript to automatically\l"
.string "format for me.$"
Like other text, formatted text can span multiple lines if you use a new set of quotes for each line. You can also manually add your own line breaks (\p
, \n
, \l
), and it will still work as expected.
text MyText {
format("Hello, are you the real-live legendary {PLAYER} that everyone talks about?\p"
"Amazing!\pSo glad to meet you!")
}
Becomes:
.string "Hello, are you the real-live legendary\n"
.string "{PLAYER} that everyone talks about?\p"
.string "Amazing!\p"
.string "So glad to meet you!$"
Additionally, format()
supports a special line break \N
, which will automatically insert the appropriate \n
or \l
line break. While this is an uncommon use case, it's useful in situations where a line break is desired for dramatic/stylistic purposes. In the following example, we want explicit line breaks for the "..."
texts, but we don't know if the first one should use \n
or \l
. Using \N
makes it easy:
text MyText {
format("You are my favorite trainer!\N...\N...\N...\NBut I'm better!")
}
The font id can optionally be specified as the second parameter to format()
.
text MyText {
format("Hello, are you the real-live legendary {PLAYER} that everyone talks about?\pAmazing!\pSo glad to meet you!", "1_latin_rse")
}
Becomes:
.string "Hello, are you the real-live legendary\n"
.string "{PLAYER} that everyone talks about?\p"
.string "Amazing!\p"
.string "So glad to meet you!$"
The font configuration JSON file informs Poryscript how many pixels wide each character in the message is, as well as setting a default maximum line length. Fonts have different character widths, and games have different text box sizes. For convenience, Poryscript comes with font_config.json
, which contains the configuration for pokeemerald's 1_latin
font as 1_latin_rse
, as well as pokefirered's equivalent as 1_latin_frlg
. More fonts can be added to this file by simply creating anothing font id node under the fonts
key in font_config.json
.
cursorOverlapWidth
can be used to ensure there is always enough room for the cursor icon to be displayed in the text box. (This "cursor icon" is the small icon that's shown when the player needs to press A to advance the text box.)
numLines
is the number of lines displayed within a single message box. If editing text for a taller space, this can be adjusted in font_config.json
.
The length of a line can optionally be specified as the third parameter to format()
if a font id was specified as the second parameter.
text MyText {
format("Hello, are you the real-live legendary {PLAYER} that everyone talks about?\pAmazing!\pSo glad to meet you!", "1_latin_rse", 100)
}
Becomes:
.string "Hello, are you the\n"
.string "real-live\l"
.string "legendary\l"
.string "{PLAYER} that\l"
.string "everyone talks\l"
.string "about?\p"
.string "Amazing!\p"
.string "So glad to meet\n"
.string "you!$"
Finally, format()
takes the following optional named parameters, which override settings from the font config:
fontId
maxLineLength
numLines
cursorOverlapWidth
text MyText {
format("This is an example of named parameters!", numLines=3, maxLineLength=100)
}
When Poryscript compiles text, the resulting text content is rendered using the .string
assembler directive. The decomp projects' build process then processes those .string
directives and substituted the string characters with the game-specific text representation. It can be useful to specify different types of strings, though. For example, implementing print-debugging commands might make use of ASCII text. Poryscript allows you to specify which assembler directive to use for text. Simply add the directive as a prefix to the string content like this:
ascii"My ASCII string."
custom"My Custom string."
// compiles to...
.ascii "My ASCII string.\0"
.custom "My Custom string."
Note that Poryscript will automatically add the \0
suffix character to ASCII strings. It will not add suffix to any other directives.
Use movement
statements to conveniently define movement data that is typically used with the applymovement
command. *
can be used as a shortcut to repeat a single command many times. Data defined with movement
is created with local scope, not global.
script MyScript {
lock
applymovement(2, MyMovement)
waitmovement(0)
release
}
movement MyMovement {
walk_left
walk_up * 5
face_down
}
Becomes:
MyScript::
lock
applymovement 2, MyMovement
waitmovement 0
release
return
MyMovement:
walk_left
walk_up
walk_up
walk_up
walk_up
walk_up
face_down
step_end
However, movement can also be inlined inside commands similar to text, using the moves()
operator. This is often much more convenient, and it can help simplify your scripts. Anything that can be used in a movement
statement can also be used inside moves()
.
Looking at the previous example, the movement can be inlined like this:
script MyScript {
lock
applymovement(2, moves(
walk_left
walk_up * 5
face_down
))
waitmovement(0)
release
}
Note, whitespace doesn't matter. This can also be written all on a single line:
applymovement(2, moves(walk_left walk_up * 5 face_down))
// You can even use commas to separate each movement command, since
// that may be easier to read.
applymovement(2, moves(walk_left, walk_up * 5, face_down))
Use mart
statements to define a list of items for use with the pokemart
command. Data defined with the mart
statement is created with local scope by default. It is not neccesary to add ITEM_NONE
to the end of the list, but if Poryscript encounters it, any items after it will be ignored.
script ScriptWithPokemart {
lock
message("Welcome to my store.")
waitmessage
pokemart(MyMartItems)
msgbox("Come again soon.")
release
}
mart MyMartItems {
ITEM_LAVA_COOKIE
ITEM_MOOMOO_MILK
ITEM_RARE_CANDY
ITEM_LEMONADE
ITEM_BERRY_JUICE
}
Becomes:
ScriptWithPokemart::
lock
message ScriptWithPokemart_Text_0
waitmessage
pokemart MyMartItems
msgbox ScriptWithPokemart_Text_1
release
return
.align 2
MyMartItems:
.2byte ITEM_LAVA_COOKIE
.2byte ITEM_MOOMOO_MILK
.2byte ITEM_RARE_CANDY
.2byte ITEM_LEMONADE
.2byte ITEM_BERRY_JUICE
.2byte ITEM_NONE
ScriptWithPokemart_Text_0:
.string "Welcome to my store.$"
ScriptWithPokemart_Text_1:
.string "Come again soon.$"
Use mapscripts
to define a set of map script definitions. Scripts can be inlined for convenience, or a label to another script can simply be specified. Some map script types, like MAP_SCRIPT_ON_FRAME_TABLE
, require a list of comparison variables and scripts to execute when the variable's value is equal to some value. In these cases, you use brackets []
to specify that list of scripts. Below is a full example showing map script definitions for a new map called MyNewCity
:
mapscripts MyNewCity_MapScripts {
MAP_SCRIPT_ON_RESUME: MyNewCity_OnResume
MAP_SCRIPT_ON_TRANSITION {
random(2)
switch (var(VAR_RESULT)) {
case 0: setweather(WEATHER_ASH)
case 1: setweather(WEATHER_RAIN_HEAVY)
}
}
MAP_SCRIPT_ON_FRAME_TABLE [
VAR_TEMP_0, 0: MyNewCity_OnFrame_0
VAR_TEMP_0, 1 {
lock
msgbox("This script is inlined.")
setvar(VAR_TEMP_0, 2)
release
}
]
}
script MyNewCity_OnResume {
...
}
script MyNewCity_OnFrame_0 {
...
}
For maps with no map scripts, simply make an empty mapscripts
statement:
mapscripts MyNewCity_MapScripts {}
Use raw
to include raw bytecode script. Anything in a raw
statement will be directly included into the compiled script. This is useful for defining custom data, or data types not supported in regular Poryscript.
raw `
TestMap_MapScripts::
.byte 0
`
script MyScript {
lock
faceplayer
# Text can span multiple lines. Use a new set of quotes for each line.
msgbox("This is shorter text,\n"
"but we can still put it\l"
"on multiple lines.")
applymovement(OBJ_EVENT_ID_PLAYER, MyScript_Movement)
waitmovement(0)
msgbox(MyScript_LongText)
release
end
}
raw `
MyScript_Movement:
walk_left
walk_down
step_end
MyScript_LongText:
.string "Hi, there.\p"
.string "This text is too long\n"
.string "to inline above.\p"
.string "We'll put it down here\n"
.string "instead, so it's out of\l"
.string "the way.$"
`
Use single-line comments with #
or //
. Everything after the #
or //
will be ignored. Comments cannot be placed in a raw
statement. (Users who wish to run the C preprocessor on Poryscript files should use //
comments to avoid conflict with C preprocessor directives that use the #
character.)
# This script does some cool things.
script MyScript {
// This is also a valid comment.
...
}
Use const
to define constants that can be used in the current script. This is especially useful for giving human-friendly names to event object ids, or temporary flags. Constants must be defined before they are used. Constants can also be composed of previously-defined constants.
const PROF_BIRCH_ID = 3
const ASSISTANT_ID = PROF_BIRCH_ID + 1
const FLAG_GREETED_BIRCH = FLAG_TEMP_2
script ProfBirchScript {
applymovement(PROF_BIRCH_ID, moves(walk_left * 4, face_down))
showobject(ASSISTANT_ID)
setflag(FLAG_GREETED_BIRCH)
}
Note that these constants are not a general macro system. They can only be used in certain places in Poryscript syntax. Below is an example of all possible places where constants can be substituted into the script:
const CONSTANT = 1
mapscripts MyMapScripts {
MAP_SCRIPT_ON_FRAME_TABLE [
// The operand and comparison values can both use constants in a
// table-based map script.
CONSTANT, CONSTANT: MyOnFrameScript_0
]
}
script MyScript {
// Any parameter of any command can use constants.
somecommand(CONSTANT)
// Any comparison operator can use constants, as well as their comparison values.
if (flag(CONSTANT)) {}
if (var(CONSTANT) == CONSTANT) {}
if (defeated(CONSTANT)) {}
// A switch var value can be a constant, as well as the individual cases.
switch (var(CONSTANT)) {
case CONSTANT: break
}
}
To control whether a script should be global or local, a scope modifier can be specified. This is supported for script
, text
, movement
, and mapscripts
. In this context, "global" means that the label will be defined with two colons ::
. Local scopes means one colon :
.
script(global) MyGlobalScript {
...
}
script(local) MyLocalScript {
...
}
Becomes:
MyGlobalScript::
...
MyLocalScript:
...
The top-level statements have different default scopes. They are as follows:
Type | Default Scope |
---|---|
script |
Global |
text |
Global |
movement |
Local |
mart |
Local |
mapscripts |
Global |
Some scripting commands always store their result in the same variable. For example, checkitem
always stores its result in VAR_RESULT
. Poryscript can simplify working with these commands with a concept called "AutoVar" commands.
Without using an AutoVar, a script would be written like this:
checkitem(ITEM_ROOT_FOSSIL)
if (var(VAR_RESULT) == TRUE) {
// player has the Root Fossil
}
However, AutoVars can be used inside the condition, which helps streamline the script:
if (checkitem(ITEM_ROOT_FOSSIL) == TRUE) {
// player has the Root Fossil
}
AutoVars can be used anywhere a var()
operator can be used. e.g. if
conditions, switch
statements--any boolean expression!
AutoVar commands are fully configurable with the command_config.json
file. Use the -cc
command line parameter to specifying the location of that config.
There are two types of AutoVar commands:
- Implicit
- The stored var is defined in the config file, and is not present in the authored script.
- Examples:
checkitem
,getpartysize
,random
- Explicit
- The stored var is provided as part of the command, and the config file stores the 0-based index of the command that specifies the stored var.
- Examples:
specialvar
,checkcoins
Let's take a look at the example config file:
// command_config.json
{
"autovar_commands": {
"specialvar": {
"var_name_arg_position": 0
},
"checkitem": {
"var_name": "VAR_RESULT"
},
...
}
With the above config, a script could be written like so:
if (checkitem(ITEM_POKEBLOCK_CASE)) {
if (specialvar(VAR_RESULT, GetFirstFreePokeblockSlot) != -1 &&
specialvar(VAR_RESULT, PlayerHasBerries)
) {
msgbox("Great! You can use the Berry Blender!)
}
} else {
msgbox("You don't have a Pokeblock case!")
}
Use the poryswitch
statement to change compiler behavior depending on custom switches. This makes it easy to make scripts behave different depending on, say, the GAME_VERSION
or LANGUAGE
. Any content that does not match the compile-time switch will not be included in the final output. To define custom switches, use the -s
option when running poryscript
. You can specify multiple switches, and each key/value pair must be separated by an equals sign. For example:
./poryscript -i script.pory -o script.inc -s GAME_VERSION=RUBY -s LANGUAGE=GERMAN
The poryswitch
statement can be embedded into any script section, including text
, movement
, and mart
statements. The underscore _
case is used as the fallback, if none of the other cases match. Cases that only contain a single statement or command can be started with a colon :
. Otherwise, use curly braces to define the case's block.
Here are some examples of compile-time switches. This assumes that two compile-time switches are defined, GAME_VERSION
and LANGUAGE
.
script MyScript {
lock
faceplayer
poryswitch(GAME_VERSION) {
RUBY {
msgbox("Here, take this Ruby Orb.")
giveitem(ITEM_RUBY_ORB)
}
SAPPHIRE {
msgbox("Here, take this Sapphire Orb.")
giveitem(ITEM_SAPPHIRE_ORB)
}
_: msgbox(format("This case is used when GAME_VERSION doesn't match either of the above."))
}
release
}
text MyText {
poryswitch(LANGUAGE) {
GERMAN: "Hallo. Ich spreche Deutsch."
ENGLISH: "Hello. I speak English."
}
}
movement MyMovement {
face_player
walk_down
poryswitch(GAME_VERSION) {
RUBY: walk_left * 2
SAPPHIRE {
walk_right * 2
walk_left * 4
}
}
}
mart MyMart {
ITEM_POTION
ITEM_POKEBALL
poryswitch(GAME_VERSION) {
RUBY {
ITEM_LAVA_COOKIE
ITEM_RED_SCARF
}
SAPPHIRE {
ITEM_FRESH_WATER
ITEM_BLUE_SCARF
}
}
}
Note, poryswitch
can also be embedded inside inlined mapscripts
scripts.
By default, Poryscript produces optimized output. It attempts to minimize the number of goto
commands and unnecessary script labels. To disable optimizations, pass the -optimize=false
option to poryscript
.
By default, Poryscript includes C Preprocessor line markers in the compiled output. This improves error messages. To disable line markers, specify -lm=false
when invoking Poryscript.
These instructions will get you setup and working with Poryscript's code. You can either build the Poryscript tool from source, or simply download the latest release from the Releases tab on GitHub.
First, install Go. Poryscript has no additional dependencies. It uses Go modules, so you shouldn't need to be located in a Go workspace.
Navigate to the Poryscript working directory, and build it:
cd your/path/to/poryscript
go build
This will create a poryscript
executable binary in the same directory. Then you can simply install it into your project by running ./install.sh ../yourprojectname
instead of manually copying the files over, similarly to how agbcc is installed into projects.
Poryscript has automated tests for its emitter
, parser
, and lexer
packages. To run all of the tests from the base directory:
> go test ./...
? github.com/huderlem/poryscript [no test files]
? github.com/huderlem/poryscript/ast [no test files]
ok github.com/huderlem/poryscript/emitter 0.523s
ok github.com/huderlem/poryscript/lexer 0.273s
ok github.com/huderlem/poryscript/parser 0.779s
? github.com/huderlem/poryscript/token [no test files]
Poryscript uses Semantic Versioning. For the available versions, see the tags on this repository.
This project is licensed under the MIT License - see the LICENSE.md file for details
- Thorsten Ball's Writing An Interpreter In Go helped bootstrap the lexer, AST, and parser for this project. A chunk of that code was derived and/or copied from that book, as I had never written something of this nature before.