func_rename.py

from pathlib import Path
from typing import List, Tuple
from itertools import chain
import re

basemap = Path('frameworkF.map').read_text()
buildmap = Path('build/dolzel2/dolzel2.map').read_text()

buildfuncdef_re = re.compile(r'.* (?P<addr>[0-9a-f]{8}) [0-9a-f]{8}  . (?P<name>[^ ]+) .*')
basefuncdef_re = re.compile(r'.* (?P<addr>[0-9a-f]{8})  . (?P<name>[^ ]+) .*')

def get_funcname_addr(map_file: str, regex: re.Pattern):
    start = False
    addr_funcname = {}
    for line in map_file.splitlines():
        if line.startswith('.text'):
            start = True
        if not start:
            continue
        if line.startswith('.ctors'):
            break
        match = regex.match(line)
        if match:
            addr = match.group('addr')
            name = match.group('name')
            if not name.startswith('.') and not name.startswith('lbl'):
                addr_funcname[addr] = name
        # linesegs = line.strip().split()
        # if len(linesegs) >= 5:
        #     addr_str = linesegs[2]
        #     func_name = linesegs[4]
        #     if func_name == '.text' or func_name.isnumeric():
        #         continue
        #     addr_funcname[addr_str] = func_name
    return addr_funcname

basemap_map = get_funcname_addr(basemap, basefuncdef_re)
buildmap_map = get_funcname_addr(buildmap, buildfuncdef_re)

missmatch = {}

for addr, symbol in basemap_map.items():
    if addr in buildmap_map:
        bsymb = buildmap_map[addr]
        if bsymb != symbol:
            missmatch[addr] = (symbol, bsymb)

good_renames = []
for (orig_name, cur_name) in missmatch.values():
    if not '@' in orig_name and not '<' in orig_name:
        good_renames.append((orig_name, cur_name))

def do_renames(renames: List[Tuple[str, str]]):
    """
    (orig_symbol, cur_symbol)
    """
    for file in chain(Path('asm').glob('**/*.s'),
                      Path('include').glob('**/*.s')):
        is_shiftjis = False
        try:
            filedata = file.read_text()
        except UnicodeDecodeError:
            is_shiftjis = True
            filedata = file.read_text('shift-jis')
        for (orig_symbol, cur_symbol) in renames:
            filedata = filedata.replace('\n'+cur_symbol+':', '\n'+orig_symbol+':')
            filedata = filedata.replace(' '+cur_symbol+'\n', ' '+orig_symbol+'\n')
            filedata = filedata.replace(' '+cur_symbol+'@', ' '+orig_symbol+'@')
        if is_shiftjis:
            file.write_text(filedata, 'shift-jis')
        else:
            file.write_text(filedata)

    for file in chain(Path('src').glob('**/*.cpp'),
                      Path('libs').glob('**/*.cpp'),
                      Path('include').glob('**/*.h')):
        is_shiftjis = False
        try:
            filedata = file.read_text()
        except UnicodeDecodeError:
            is_shiftjis = True
            filedata = file.read_text('shift-jis')
        for (orig_symbol, cur_symbol) in renames:
            filedata = filedata.replace(' '+cur_symbol+'(', ' '+orig_symbol+'(')
        if is_shiftjis:
            file.write_text(filedata, 'shift-jis')
        else:
            file.write_text(filedata)
    
    ldscript = Path('ldscript.lcf')
    filedata = ldscript.read_text()
    for (orig_symbol, cur_symbol) in renames:
        filedata = filedata.replace('\n'+cur_symbol+'\n', '\n'+orig_symbol+'\n')
    ldscript.write_text(filedata)

# do_renames(good_renames)