ref: e733c4234c0064102f9732f23ed869660c847d31
parent: f3340de6dc30fb1c889421bdb8a3d22c1b64fa24
author: yenatch <[email protected]>
date: Thu Jun 20 20:58:35 EDT 2013
preprocess asm files individually this fixes a lot of previous hacks first off, rgbds requires that labels from includes be marked as globals. instead, 3626ddeb stuffed includes into the parent file in the preprocessor. this meant one huge file got preprocessed every time, adding an additional ten seconds to compile time. running the preprocessor once for each file would create too much overhead, so a list is fed into prequeue.py, which then makes calls to preprocessor.py. this paves the way for compiling source files separately some day. next, compiling previously required `make clean` to be executed first. f3340de6 touched main.asm to force a fresh compile instead. this behavior has been reverted. now, `make all` will only attempt to recompile if a source file has changed. preprocessor.py has some marginal changes. prequeue.py is created to keep the original functionality of preprocessor.py intact. so many files are preprocessed on first compile (1951 as of this commit) that the prequeue call has been hidden. compile time is reduced to 15-30 seconds on first compile, and 5-10 seconds subsequently. the majority of this time is spent in rgbasm.
--- a/Makefile
+++ b/Makefile
@@ -1,33 +1,23 @@
.SUFFIXES: .asm .tx .o .gbc .png .2bpp .lz
-TEXTFILES = \
- text/sweethoney.tx \
- text/phone/bill.tx \
- text/phone/elm.tx \
- text/phone/mom.tx \
- text/phone/trainers1.tx \
- text/common.tx \
- text/common_2.tx \
- text/common_3.tx \
- main.tx
+TEXTFILES := $(shell find ./ -type f -name '*.asm' | grep -v pokecrystal.asm | grep -v constants.asm | grep -v gbhw.asm | grep -v hram.asm | grep -v constants | grep -v wram.asm)
+TEXTQUEUE :=
-PNG_GFX = $(shell find gfx/ -type f -name '*.png')
-LZ_GFX = $(shell find gfx/ -type f -name '*.lz')
-TWOBPP_GFX = $(shell find gfx/ -type f -name '*.2bpp')
+PNG_GFX := $(shell find gfx/ -type f -name '*.png')
+LZ_GFX := $(shell find gfx/ -type f -name '*.lz')
+TWOBPP_GFX := $(shell find gfx/ -type f -name '*.2bpp')
all: pokecrystal.gbc
cmp baserom.gbc $<
clean:
- rm -f main.tx pokecrystal.o pokecrystal.gbc ${TEXTFILES}-pokecrystal.o: pokecrystal.asm constants.asm wram.asm ${TEXTFILES} lzs+ rm -f pokecrystal.o pokecrystal.gbc
+ @echo rm -f $$\(TEXTFILES:.asm=.tx\)
+ @rm -f $(TEXTFILES:.asm=.tx)
+pokecrystal.o: $(TEXTFILES:.asm=.tx) pokecrystal.asm constants.asm wram.asm lzs
+ python prequeue.py $(TEXTQUEUE)
rgbasm -o pokecrystal.o pokecrystal.asm
-
-pokecrystal.asm: depend
-depend:
- @touch main.asm
-
.asm.tx:
- python preprocessor.py < $< > $@
+ $(eval TEXTQUEUE := $(TEXTQUEUE) $<)
pokecrystal.gbc: pokecrystal.o
rgblink -o $@ $<
--- a/preprocessor.py
+++ b/preprocessor.py
@@ -310,25 +310,15 @@
"""
Separates asm and comments on a single line.
"""
-
- asm = ""
- comment = None
in_quotes = False
+ for i in xrange(len(l)):
+ if not in_quotes:
+ if l[i] == ";":
+ break
+ if l[i] == "\"":
+ in_quotes = not in_quotes
+ return i
- # token either belongs to the line or to the comment
- for token in l:
- if comment:
- comment += token
- else:
- if not in_quotes:
- if token == ";":
- comment = ";"
- continue
- if token == "\"":
- in_quotes = not in_quotes
- asm += token
- return asm, comment
-
def quote_translator(asm):
"""
Writes asm with quoted text translated into bytes.
@@ -335,27 +325,22 @@
"""
# split by quotes
- asms = asm.split("\"")+ asms = asm.split('"')# skip asm that actually does use ASCII in quotes
- lowasm = asms[0].lower()
+ if "SECTION" in asms[0]\
+ or "INCBIN" in asms[0]\
+ or "INCLUDE" in asms[0]:
+ return asm
- if "section" in lowasm \
- or "incbin" in lowasm:
- sys.stdout.write(asm)
- return
-
print_macro = False
if asms[0].strip() == 'print':
asms[0] = asms[0].replace('print','db 0,')print_macro = True
- output = ""
+ output = ''
even = False
- i = 0
for token in asms:
- i = i + 1
-
if even:
characters = []
# token is a string to convert to byte values
@@ -362,36 +347,28 @@
while len(token):
# read a single UTF-8 codepoint
char = token[0]
- if ord(char) >= 0xFC:
- char = char + token[1:6]
- token = token[6:]
- elif ord(char) >= 0xF8:
- char = char + token[1:5]
- token = token[5:]
- elif ord(char) >= 0xF0:
- char = char + token[1:4]
- token = token[4:]
- elif ord(char) >= 0xE0:
- char = char + token[1:3]
- token = token[3:]
- elif ord(char) >= 0xC0:
+ if ord(char) < 0xc0:
+ token = token[1:]
+ # certain apostrophe-letter pairs are considered a single character
+ if char == "'" and token:
+ if token[0] in 'dlmrstv':
+ char += token[0]
+ token = token[1:]
+ elif ord(char) < 0xe0:
char = char + token[1:2]
token = token[2:]
+ elif ord(char) < 0xf0:
+ char = char + token[1:3]
+ token = token[3:]
+ elif ord(char) < 0xf8:
+ char = char + token[1:4]
+ token = token[4:]
+ elif ord(char) < 0xfc:
+ char = char + token[1:5]
+ token = token[5:]
else:
- token = token[1:]
-
- # certain apostrophe-letter pairs are only a single byte
- if char == "'" and len(token) > 0 and \
- (token[0] == "d" or \
- token[0] == "l" or \
- token[0] == "m" or \
- token[0] == "r" or \
- token[0] == "s" or \
- token[0] == "t" or \
- token[0] == "v"):
- char = char + token[0]
- token = token[1:]
-
+ char = char + token[1:6]
+ token = token[6:]
characters += [char]
if print_macro:
@@ -421,22 +398,18 @@
output += ", ".join(["${0:02X}".format(chars[char]) for char in characters])- # if not even
else:
- output += (token)
+ output += token
even = not even
- sys.stdout.write(output)
+ return output
- return
-
def extract_token(asm):
- token = asm.split(" ")[0].replace("\t", "").replace("\n", "")- return token
+ return asm.split(" ")[0].strip()def make_macro_table():
- return dict([(macro.macro_name, macro) for macro in macros])
+ return dict(((macro.macro_name, macro) for macro in macros))
macro_table = make_macro_table()
def macro_test(asm):
@@ -443,10 +416,8 @@
"""
Returns a matching macro, or None/False.
"""
-
# macros are determined by the first symbol on the line
token = extract_token(asm)
-
# check against all names
if token in macro_table:
return (macro_table[token], token)
@@ -600,64 +571,45 @@
sys.stdout.write(output)
-def include_file(asm):
- """This is more reliable than rgbasm/rgbds including files on its own."""
-
- prefix = asm.split("INCLUDE \"")[0] + '\n'- filename = asm.split("\"")[1]- suffix = asm.split("\"")[2]-
- read_line(prefix)
-
- lines = open(filename, "r").readlines()
-
- for line in lines:
- read_line(line)
-
- read_line(suffix)
-
def read_line(l):
"""Preprocesses a given line of asm."""
- # strip and store any comment on this line
- if ";" in l:
- asm, comment = separate_comment(l)
- else:
- asm = l
- comment = None
+ # strip comments
+ asm, comment = l[:separate_comment(l)], l[separate_comment(l):]
- # handle INCLUDE as a special case
- if "INCLUDE \"" in l:
- include_file(asm)
+ # export all labels
+ if ':' in asm[:asm.find('"')]:+ sys.stdout.write('GLOBAL ' + asm.split(':')[0] + '\n')+ # expect preprocessed .asm files
+ if "INCLUDE" in asm:
+ asm = asm.replace('.asm','.tx')+ sys.stdout.write(asm)
+
# ascii string macro preserves the bytes as ascii (skip the translator)
- elif len(asm) > 6 and "\tascii " in [asm[:7], "\t" + asm[:6]]:
+ elif len(asm) > 6 and "ascii " == asm[:6] or "\tascii " == asm[:7]:
asm = asm.replace("ascii", "db", 1)sys.stdout.write(asm)
# convert text to bytes when a quote appears (not in a comment)
elif "\"" in asm:
- quote_translator(asm)
+ sys.stdout.write(quote_translator(asm))
# check against other preprocessor features
else:
macro, token = macro_test(asm)
-
if macro:
macro_translator(macro, token, asm)
else:
sys.stdout.write(asm)
+ sys.stdout.write(comment)
- # show line comment
- if comment != None:
- sys.stdout.write(comment)
-
def preprocess(lines=None):
"""Main entry point for the preprocessor."""
if not lines:
# read each line from stdin
- lines = sys.stdin
+ lines = (sys.stdin.readlines())
elif not isinstance(lines, list):
# split up the input into individual lines
lines = lines.split("\n")--- /dev/null
+++ b/prequeue.py
@@ -1,0 +1,11 @@
+import os
+import sys
+import preprocessor
+
+if __name__ == '__main__':
+ for source in sys.argv[1:]:
+ dest = os.path.splitext(source)[0] + '.tx'
+ sys.stdin = open(source, 'r')
+ sys.stdout = open(dest, 'w')
+ preprocessor.preprocess()
+