trysdyn/bcexparse
1
0
Fork 0
Code Issues Pull requests Projects Releases 1 Packages Wiki Activity Actions

Compare commits

base: trysdyn:master
Branches Tags
trysdyn:master
trysdyn:0.5.2
trysdyn:0.5.1
trysdyn:0.5.0
trysdyn:0.4.2
trysdyn:0.4.1
trysdyn:0.4.0
trysdyn:0.3.0
..
compare: trysdyn:0.3.0
Branches Tags
trysdyn:master
trysdyn:0.5.2
trysdyn:0.5.1
trysdyn:0.5.0
trysdyn:0.4.2
trysdyn:0.4.1
trysdyn:0.4.0
trysdyn:0.3.0

No commits in common. "master" and "0.3.0" have entirely different histories.
master ... 0.3.0

3 changed files with 241 additions and 675 deletions
Show stats Expand all files Collapse all files

29
LICENSE.md
View file

@ -1,29 +0,0 @@
"i'm so tired" software license 1.0
copyright (c) 2024 Trysdyn Black
this is anti-capitalist, anti-bigotry software, made by people who are tired of ill-intended organisations and individuals, and would rather not have those around their creations.
permission is granted, free of charge, to any user (be they a person or an organisation) obtaining a copy of this software, to use it for personal, commercial, or educational purposes, subject to the following conditions:
1. the above copyright notice and this permission notice shall be included in all copies or modified versions of this software.
2. the user is one of the following:
a. an individual person, labouring for themselves
b. a non-profit organisation
c. an educational institution
d. an organization that seeks shared profit for all of its members, and allows non-members to set the cost of their labor
3. if the user is an organization with owners, then all owners are workers and all workers are owners with equal equity and/or equal vote.
4. if the user is an organization, then the user is not law enforcement or military, or working for or under either.
5. the user does not use the software for ill-intentioned reasons, as determined by the authors of the software. said reasons include but are not limited to:
a. bigotry, including but not limited to racism, xenophobia, homophobia, transphobia, ableism, sexism, antisemitism, religious intolerance
b. pedophilia, zoophilia, and/or incest
c. support for cops and/or the military
d. any blockchain-related technology, including but not limited to cryptocurrencies
6. the user does not promote or engage with any of the activities listed in the previous item, and is not affiliated with any group that promotes or engages with any of such activities.
this software is provided as is, without any warranty or condition. in no event shall the authors be liable to anyone for any damages related to this software or this license, under any kind of legal claim.

11
README.md
View file

@ -6,8 +6,6 @@ FF6 BCEX Spoiler Log to JSON Parser
This tool uses only the Python standard library; no packages are needed.
Grab the latest release from the "Releases" section of the source repo, then unzip and run:
`./main.py <spoiler log filename.txt>`
It's recommended to pipe the output to `jq` or a similar json parsing tool.
@ -26,15 +24,6 @@ The parser is quite brittle since the inconsistency of the original spoiler log
Additionally the tool only parses spoiler log data needed for known use-cases, so if you plan to use it you may need to request the inclusion of spoiler log sections. The tool can be expanded by adding new functions named `parse_SECTION` where `SECTION` is the full name of a section in the log, as presented in the log.
Development targed BCEX 4.0, but BCEX 5.0 is in limited support. It works but all the bugs haven't been found yet.
BCCE (The community revival of the BCEX project) is supported, but support is geared toward taking BCCE's spoiler logs and producing identical output to BCEX. This means stats are not their own data object, but are folded into character data just like BCEX outputs it. Remonsterate is supported and inserts its data into the monsters object.
BCCE is in active development and this may break at any time; see the first paragraph in this section.
## Known Support Gaps
These sections in the BCEX/BCCE spoiler logs currently have no logic and I'm aware of it. That doesn't mean sections *not* listed here have support; they may not and I'm not aware of them.
- TREASURE CHESTS
- JUNCTIONS

546
main.py
View file

@ -2,7 +2,7 @@
"""Parse BCEX (or BCCE) logs into json objects."""
__version__ = "0.5.1"
__version__ = "0.3"
__author__ = "Trysdyn Black"
import json
@ -10,23 +10,7 @@ import sys
from pathlib import Path
class Parser: # noqa: PLR0904
"""
BCEX/BCCE spoiler logfile parser.
Sections missing support:
- TREASURE CHESTS
- JUNCTIONS
"""
def __init__(self, filename: str) -> None:
"""Initialize parser with filename."""
self.filename = filename
self.config_sections = {}
self.data_sections = {}
@staticmethod
def parse_MONSTERS(data: str) -> dict[str, dict]: # noqa: C901, PLR0912
def parse_MONSTERS(data: str) -> dict[str, dict]: # noqa: C901, PLR0912
"""
Parse the MONSTERS section.
@ -43,7 +27,6 @@ class Parser: # noqa: PLR0904
if "(Level " in line:
name = line.split(" (")[0]
info["stats"]["level"] = int(line.split("(Level ")[1][:-1])
info["name"] = name
# Stat chart rows
elif line.startswith("|"):
for stat in line[1:-1].split("|"):
@ -61,30 +44,16 @@ class Parser: # noqa: PLR0904
weak_text = "WEAK: "
info["nullifies"] = null_text.split(": ")[1].split(", ")
info["weak"] = weak_text.split(": ")[1].split(", ")
# Due to split oddness we can populate a blank string into weakness
# Delete it if we did.
if info["weak"] == [""]:
del info["weak"]
# Specials are name=>desc as k:v
# I *think* you can only have one special...
elif line.startswith("SPECIAL"):
content = line.split(" ", 1)[1]
if len(content) > 1:
# BCEX 5.0 changes the skill name to be single quoted instead of double
special_name = content.split('"')[1] if '"' in content else content.split("'")[1]
special_name = content.split('"')[1]
special_desc = content.split(": ")[1]
info["special"] = {special_name: special_desc}
else:
info["special"] = {}
# Morph results, with a percent chance in each one
elif line.startswith("MORPH"):
_, chance, items = line.split(" ", 2)
# BCEX 5.0 has a bug where % can be doubled sometimes
chance = int(chance[1:-3].replace("%", ""))
items = items.split(", ")
if "morph" not in info:
info["morph"] = {"percent_chance": chance, "items": items}
# Everything else is a simple k: v list where v is comma-delimited
else:
for k in ["immune", "auto", "skills", "steal", "drops", "location"]:
@ -98,8 +67,8 @@ class Parser: # noqa: PLR0904
return result
@staticmethod
def parse_REMONSTERATE(data: str) -> dict[str, dict]:
def parse_REMONSTERATE(data: str) -> dict[str, dict]:
"""
Parse the BCCE-only REMONSTERATE section.
@ -113,12 +82,13 @@ class Parser: # noqa: PLR0904
name = line.split("(")[0].strip()
originally = line.split("(", 1)[1].split(")")[0].strip()
sprite = line.split("->")[1].strip().strip(".")
result[name] = {"originally": originally, "sprite": sprite}
return result
@staticmethod
def parse_CHARACTERS(data: str) -> dict[str, dict]: # noqa: C901
def parse_CHARACTERS(data: str) -> dict[str, dict]: # noqa: C901
"""
Parse the CHARACTERS section.
@ -129,11 +99,7 @@ class Parser: # noqa: PLR0904
Regardless of flavor, core logic will snap stats back into this section
later.
"""
replacements = {
"Looks like": "looks",
"World of Ruin location": "wor_location",
"Notable equipment": "equipment",
}
replacements = {"Looks like": "looks", "World of Ruin location": "wor_location", "Notable equipment": "equipment"}
result = {}
@ -145,9 +111,9 @@ class Parser: # noqa: PLR0904
# Name
if line[0:2].isdigit():
name = line[4:]
info["name"] = name
# Stat chart rows: BCEX Version only
# Stat chart rows
# BCEX Version Only
elif line.startswith("|"):
for stat in line[1:-1].split("|"):
if ":" in stat:
@ -160,14 +126,9 @@ class Parser: # noqa: PLR0904
spell_level, spell_name = line.split("-", 1)
info["spells"][spell_name.strip()] = int(spell_level.strip().split(" ")[1])
# Command list
# Commands: is just a CSV list of things like "fight" "magic" etc. However if command
# randomization is on, these will be exotic things like "KitMerton" that need additional
# info provided from the COMMANDS section later.
# As such, we begin with a dehydrated hash of command_name=None, then the COMMANDS section
# will provide hydration data for the values.
# Special k=v strings with comma-delimited lists
elif line.startswith("Commands:"):
info["commands"] = {command.strip(): None for command in line.split(":")[1].split(",")}
info["commands"] = [command.strip() for command in line.split(":")[1].split(",")]
elif line.startswith("Notable"):
info["equipment"] = [eq.strip() for eq in line.split(":")[1].split(",")]
@ -188,8 +149,8 @@ class Parser: # noqa: PLR0904
return result
@staticmethod
def parse_STATS(data: str) -> dict[str, dict]:
def parse_STATS(data: str) -> dict[str, dict]:
"""
Parse the BCCE-only STATS section.
@ -223,8 +184,8 @@ class Parser: # noqa: PLR0904
return result
@staticmethod
def parse_COMMANDS(data: str) -> dict[str, dict]:
def parse_COMMANDS(data: str) -> dict[str, dict]:
"""
Parse the COMMANDS section.
@ -263,29 +224,23 @@ class Parser: # noqa: PLR0904
return commands
@staticmethod
def parse_SEED(data: str) -> dict[str, bool | str]:
def parse_SEED(data: str) -> dict[str, bool | str]:
"""
Parse the injected SEED section.
This is a fake section injected by the loader code. It contains nothing
but the seed code and we derive from this if the randomizer is BCCE or
BCEX, and try to pluck out other data.
We can't do much because the format is really hard to reverse.
BCEX, and normalize the seed code to a standard format by undoing the
changes BCCE makes to it.
"""
version, mode, flags, seed_num = data.split("|") if "|" in data else data.split(".")
return {
"version": version,
"flags": flags,
"seed_num": seed_num,
"mode": mode,
"is_bcce": data.startswith("CE"),
"seed": data,
}
# Normalize seed codes to BCEX format, removing spaces and replacing pipes with dots
seed = data.replace("|", ".").replace(" ", "")
@staticmethod
def parse_SECRET_ITEMS(data: str) -> list[str]:
return {"is_bcce": data.startswith("CE"), "seed": seed}
def parse_SECRET_ITEMS(data: str) -> list[str]:
"""
Parse the BCCE-only SECRET ITEMS section.
@ -295,372 +250,12 @@ class Parser: # noqa: PLR0904
# I have no idea what this is lol, dump it to a list for now
return [line for line in data.split("\n") if not line.startswith("---")]
@staticmethod
def parse_MUSIC(data: str) -> dict[str, dict]:
"""
Parse the MUSIC section.
This differs significantly between BCEX and BCCE: Numbers and data are split
with a period in BCCE and a colon in BCEX. BCEX puts arranger on the same line
as composer, split with --, BCEX lacks the Jukebox Title section entirely.
"""
music_sections = data.split("\n\n")
replacements = {}
for section in music_sections[1:]:
if not section.strip():
continue
# BCEX and BCCE divide numbers from data differently
if ":" in section[:6]:
_, info = section.split(":", 1)
else:
_, info = section.split(".", 1)
# The name of the song being replaced preceeds a ->
old_name, info = info.split("->", 1)
old_name = old_name.strip()
replacements[old_name] = {}
# Info is, mostly, one item per line, so let's go by line
tok_info = info.split("\n")
for k in ("name", "title", "composer", "arranger", "jukebox_title"):
# Not every song has all the data
if not len(tok_info):
break
line = tok_info.pop(0).strip()
# BCCE puts arranger on its own line. BCEX puts it on the same line as
# composed, split by "--". So we have to handle both
if "-- Arranged by " in line:
line, arranger = line.split("-- Arranged by ", 1)
replacements[old_name]["arranger"] = arranger
# Jukebox Title is in prens with extra stuff to chomp
if "Jukebox title" in line:
line = line[16:-1]
# These strings are fluff but should only appear in compose/arranger lines
# so it's safe to just blindly chomp them
line = line.split("Composed by ", 1)[-1].strip()
line = line.replace("Ripped and/or arranged by ", "")
replacements[old_name][k] = line
return replacements
@staticmethod
def parse_AESTHETICS(data: str) -> dict[str, str]:
"""
Parse the BCCE-only AESTHETICS section.
This is just a k=v list we split up and strip.
"""
replacements = {}
for line in data.split("\n"):
if ":" not in line:
continue
old, new = line.split(":")
old = old.strip()
new = new.strip()
replacements[old] = new
return replacements
@staticmethod
def parse_MAGITEK(data: str) -> dict[str, list]:
"""
Parse the BCCE-only MAGITEK section.
This section contains info on what Terra and the guards have in their magitek skill
lists at the start of the game. Unfortunately like the STATS block, this keys on slot
name and not randomized name, but at least it's super simple.
"""
result = {"terra": [], "others": []}
mode = None
for line in data.split("\n"):
if line.startswith("Terra Magitek"):
mode = "terra"
elif line.startswith("Other Actor"):
mode = "others"
elif line.strip() and mode:
result[mode].append(line.strip())
return result
@staticmethod
def parse_DANCES(data: str) -> dict[str, dict]:
"""
Parse the BCCE-only DANCES section.
This section's a list of dances with their effects chances. Each dance seems to
always have four effects, spaced at specific locations in the string, so we chomp
by hardcoded locations. Brittle.
"""
result = {}
dance = None
for line in data.split("\n"):
# Formatting line
if "-----" in line:
continue
# Result list
if line.startswith(" "):
# This is more brittle than I'd like but the logs space results out by
# character position and some dance results have spaces so it's hard to
# split properly.
for i in range(2, 57, 18):
chance, effect = line[i : i + 18].split(" ", 1)
result[dance][effect.strip()] = chance.strip()
# New dance section
elif line.strip():
dance = line.strip()
if dance not in result:
result[dance] = {"name": dance}
return result
@staticmethod
def parse_ESPERS(data: str) -> dict[str, dict]:
"""
Parse the ESPERS section.
This section lists espers, what they teach, their bonuses, and locations. We
assume any line with a : in it is either BONUS or LOCATION and just k=v it. Any
line after a blank line is an esper name, and anything else is a spell learn option.
"""
result = {}
esper = None
next_esper = False
for line in data.split("\n"):
# Formatting line
if "-----" in line:
continue
# Blank lines divide esper sections
if not line.strip():
next_esper = True
# The first line in a new section is the esper name
elif next_esper:
esper = line.strip()
if esper not in result:
result[esper] = {"learnset": {}, "name": esper}
next_esper = False
# Any line with ":" is a k=v we should just shove into the dict
elif ": " in line:
k, v = line.split(": ")
result[esper][k.lower()] = v.strip()
# Everything else should be spell learnset
else:
spell, mult = line.split(" x")
result[esper]["learnset"][spell.strip()] = f"x{mult}"
return result
@staticmethod
def parse_ITEM_MAGIC(data: str) -> dict[str, dict]:
"""
Parse the ITEM MAGIC section.
This section is actually three distinct subsections. Breakable items and procs
we can just store as k=v in sub-dicts. The spell-teaching items section we treat
like the esper learnset and store spell_name = learn multiplier.
"""
result = {}
section = None
for line in data.split("\n"):
# Formatting lines
if "-----" in line or not line.strip():
continue
# Anything with ":" is a k=v to insert
if ":" in line:
k, v = line.split(": ")
# Spell-teaching needs special logic to get the {spell: multiplier} format
if section == "spell-teaching":
spell, mult = v.split(" x")
result[section][k.strip()] = {spell.strip(): f"x{mult}"}
else:
result[section][k.strip()] = v.strip()
# Anything else should be a section header. Use the first word unless it's "ITEM"
else:
tok_line = line.split()
section = tok_line[1].lower() if "ITEM" in tok_line[0] else tok_line[0].lower()
result[section] = {}
return result
@staticmethod
def parse_COLOSSEUM(data: str) -> dict[str, dict]:
"""
Parse the COLOSSEUM section.
Each line is one item offered, which has a resulting item, a monster level, and a
a monster name. This results in basic string splitting giving us all the data we
need.
"""
result = {}
for line in data.split("\n"):
if "-----" in line or not line.strip():
continue
item, tok_line = line.split("->")
new_item, tok_line = tok_line.split(": LV ")
level, name = tok_line.split(" ", 1)
result[item.strip()] = {
"becomes": new_item.strip(),
"battle": {"monster": name.strip(), "level": int(level.strip())},
}
return result
@staticmethod
def parse_ITEM_EFFECTS(data: str) -> dict[str, dict]: # noqa: C901, PLR0912
"""
Parse the BCCE-only ITEM EFFECTS section.
This is a weird chimera section with multiple sub-sections. Most of these
sub-sections can just be k:v or k:v, v, v parsed. Elemental properties is
a special indented list and special features can contain command changers
so we need a lot of special parsing.
"""
result = {}
mode = None
item = None
for line in data.split("\n"):
if "-----" in line or not line.strip():
continue
if line == line.upper():
mode = line.lower().strip().replace(" ", "_")
result[mode] = {}
# Everything here is k:v splitable except the elemental properties
# section which is k:v but with a return and indent and multiple values.
elif mode == "elemental_properties":
if line.startswith(" "):
tok_line = line.split()
operator = "+" if "Gained" in tok_line else "-" if "Lost" in tok_line else ""
element = tok_line[-1]
effect = tok_line[2].strip(":")
if item not in result[mode]:
result[mode][item] = []
result[mode][item].append(f"{operator}{element} {effect}")
else:
item = line.split(":")[0]
result[mode][item] = []
else:
item, effect = line.split(":")
item = item.strip()
effect = effect.strip()
for effect_token in effect.split(", "):
# This is a command change. This can appear in command_changers *or* features
# In either case the result goes in command changers for consistency
if "->" in effect_token:
old, new = effect_token.split("->")
# We don't have a guarantee COMMAND CHANGERS is in yet...
if "command_changers" not in result:
result["command_changers"] = {}
if item not in result["command_changers"]:
result["command_changers"][item] = {}
result["command_changers"][item][old.strip()] = new.strip()
# Everything else should hopefully just be a basic effect list
else:
if item not in result[mode]:
result[mode][item] = []
result[mode][item].append(effect_token.strip())
return result
@staticmethod
def parse_SHOPS(data: str) -> dict[str, dict]:
"""
Parse SHOPS section.
For now this is just a dict of shop name => {item => cost}. I would like to split
this up better so you have, for example narshe["wob"]["after_kefka"]["weapons"] but
that's probably more lifting than is necessary. Most people will just be searching
for buyable items period.
"""
result = {}
shop = None
for line in data.split("\n"):
if "-----" in line or not line.strip():
continue
if line == line.upper():
shop = line.strip()
result[shop] = {"stock": {}, "female_discount": False}
elif line.startswith("Discounts for female characters"):
result[shop]["female_discount"] = True
else:
tok_line = line.split()
item = " ".join(tok_line[:-1])
cost = tok_line[-1]
result[shop]["stock"][item.strip()] = int(cost.strip())
return result
@staticmethod
def cleanup_STATS(data: dict) -> bool:
"""
Fold BCCE-only STATS section back into CHARACTERS data.
This returns BCCE logs back to how they were laid out in BCEX: where stat blocks
were simply part of the CHARACTERS data.
The BCCE STATS section keys on character slot (Terra, Locke, etc) and not the
new randomized character name, so some hunting has to happen here.
"""
for slot, stats in data.get("STATS", {}).items():
for c_data in data.get("CHARACTERS", {}).values():
if c_data["originally"].lower() == slot.lower():
c_data["stats"] = stats
return True
@staticmethod
def cleanup_COMMANDS(data: dict) -> bool:
"""Fold COMMANDS expanded descriptions into CHARACTERS command data."""
# If our COMMANDS section is missing or somehow missing a given command, we just
# repeat the command's name as its description. This should only be simple things
# like "fight" and "magic" unless something goes wrong.
command_info = data.get("COMMANDS", {})
for c_data in data.get("CHARACTERS", {}).values():
for command in c_data.get("commands", {}):
c_data["commands"][command] = command_info.get(command, command)
return False
@staticmethod
def cleanup_REMONSTERATE(data: dict) -> bool:
"""Fold REMONSTERATE section into MONSTERS section data."""
for name, info in data.get("REMONSTERATE", {}).items():
for m_name, m_info in data.get("MONSTERS", {}).items():
if name == m_name:
m_info["originally"] = info["originally"]
m_info["sprite"] = info["sprite"]
return True
def get_sections(self, data: str) -> dict[str, str]:
"""Split logfile text and return a dict of sections for parsing."""
tok_data = data.split("============================================================\n")
def load(filename: str) -> dict[str, str]:
"""Load file and tokenize into sections."""
# Load our file, tokenize by section header (starting with ====)
with Path(filename).open(encoding="utf-8") as infile:
tok_data = infile.read().split("============================================================\n")
sections = {}
@ -676,43 +271,54 @@ class Parser: # noqa: PLR0904
section_header, section_data = s.split("\n", 1)
sections[section_header[5:]] = section_data
self.config_sections = sections
return sections
def parse(self) -> dict:
"""Fully parse the logfile and return the full data object."""
# Get individual sections to work on
with Path(self.filename).open(encoding="utf-8") as infile:
sections = self.get_sections(infile.read())
data = {}
# For each section attempt to run a parser function for it
for k, v in sections.items():
section_func = f"parse_{k.replace(' ', '_')}"
if hasattr(self, section_func):
data[k] = getattr(self, section_func)(v)
# Do post-parse cleanup. We need all sections parsed to do these
# Any cleanup function that returns true has its respective section deleted
section_dels = set()
for k in data:
section_func = f"cleanup_{k.replace(' ', '_')}"
if hasattr(self, section_func) and getattr(self, section_func)(data):
section_dels.add(k)
# Any section cleanup that returns true means delete that section
for k in section_dels:
if k in data:
del data[k]
self.data_sections = data
return data
if __name__ == "__main__":
p = Parser(sys.argv[1])
data = p.parse()
sections = load(sys.argv[1])
data = {}
# This mess tries to run a function named parse_SECTION for each section,
# and just continues to the next section if one doesn't exist.
for k, v in sections.items():
try:
section_func = f"parse_{k.replace(' ', '_')}"
data[k] = globals()[section_func](v)
except KeyError:
continue
# Subkey CHARACTERS commands with COMMANDS data
# This turns lists of commands each character has into hashes where
# Command name => Textual desc of command
# Certain flags don't shuffle commands like this so we have to check
if "COMMANDS" in data:
for c_data in data["CHARACTERS"].values():
new_commands = {}
for command in c_data["commands"]:
new_commands[command] = data["COMMANDS"].get(command, command)
c_data["commands"] = new_commands
# If we have a STATS block, snap it into CHARACTER data
# BCCE broke this out into its own section
# Worse, it keys on slot name, not randomized character name
if "STATS" in data:
for slot, stats in data["STATS"].items():
for c_data in data["CHARACTERS"].values():
if c_data["originally"].lower() == slot.lower():
c_data["stats"] = stats
del data["STATS"]
# If we ran BCCE Remonsterate, fold sprite data into monster block
if "REMONSTERATE" in data:
for name, info in data["REMONSTERATE"].items():
for m_name, m_info in data["MONSTERS"].items():
if name == m_name:
m_info["originally"] = info["originally"]
m_info["sprite"] = info["sprite"]
del data["REMONSTERATE"]
# Barf this pile of trash out
print(json.dumps(data))