From 096769917a0a461bc76bbdf10d343f1db6e93496 Mon Sep 17 00:00:00 2001
From: Zack Buhman <zack@buhman.org>
Date: Tue, 1 Aug 2023 17:04:49 +0000
Subject: [PATCH] tools/parse: add dex_text

---
 tools/parse/generic/label_data.py  | 15 +++---
 tools/parse/pokemon/dex_entries.py |  3 +-
 tools/parse/pokemon/dex_text.py    | 75 ++++++++++++++++++++++++++++++
 tools/parse/pokemon/evos_moves.py  |  3 +-
 4 files changed, 87 insertions(+), 9 deletions(-)

diff --git a/tools/parse/generic/label_data.py b/tools/parse/generic/label_data.py
index 3950e3f..a94de15 100644
--- a/tools/parse/generic/label_data.py
+++ b/tools/parse/generic/label_data.py
@@ -1,27 +1,28 @@
+from functools import partial
 from itertools import chain
 
 def is_label(token_line):
     return token_line[0].endswith(':')
 
-def is_data(token_line):
-    return token_line[0] == 'db' or token_line[0] == 'dw' \
-        or token_line[0] == 'text_far'
+def is_data(token_line, data_tokens):
+    return token_line[0] in data_tokens
 
 Label = object()
 Data = object()
 
-def _event(token_line):
+def _event(token_line, data_tokens):
     if is_label(token_line):
         label0, = token_line
         yield Label, label0.split(':')[0]
-    elif is_data(token_line):
+    elif is_data(token_line, data_tokens):
         _, args = token_line
         yield Data, args
     else:
         return
 
-def event(tokens):
-    return list(chain.from_iterable(map(_event, tokens)))
+def event(tokens, data_tokens):
+    return list(chain.from_iterable(map(
+        partial(_event, data_tokens=data_tokens), tokens)))
 
 def pointer_table(type_args, ix):
     pointer_table = []
diff --git a/tools/parse/pokemon/dex_entries.py b/tools/parse/pokemon/dex_entries.py
index 5ffba90..4504326 100644
--- a/tools/parse/pokemon/dex_entries.py
+++ b/tools/parse/pokemon/dex_entries.py
@@ -38,7 +38,8 @@ def dex_entry(type_args, ix):
     return ix+5, entry
 
 def build_tables(tokens):
-    type_args = label_data.event(tokens)
+    data_tokens = {'dw', 'db', 'text_far'}
+    type_args = label_data.event(tokens, data_tokens=data_tokens)
     ix = 0
     while ix < len(type_args):
         type, args = type_args[ix]
diff --git a/tools/parse/pokemon/dex_text.py b/tools/parse/pokemon/dex_text.py
index e69de29..f6d04bd 100644
--- a/tools/parse/pokemon/dex_text.py
+++ b/tools/parse/pokemon/dex_text.py
@@ -0,0 +1,75 @@
+import builtins
+from dataclasses import dataclass
+
+from parse.generic import tokenize
+from parse.generic import number
+from parse.generic import string
+from parse.generic import label_data
+
+def _label(type, args):
+    assert type is label_data.Label, (type, args)
+    assert builtins.type(args) == str
+    return args
+
+def _string(type, args):
+    assert type is label_data.Data, (type, args)
+    # hmm, this is a bug in label_data.event
+    # ...too lazy to fix
+    arg = ','.join(args)
+    return string.parse(arg)
+
+def type_sequence(schema, type_args):
+    assert len(schema) == len(type_args)
+    return (
+        schema[i](*type_args[i])
+        for i in range(len(schema))
+    )
+
+@dataclass
+class DexText:
+    label: str
+    content: list[tuple[str, str]]
+
+def parse_text(type_args, ix):
+    schema = [
+        _label,
+        _string, _string, _string,
+        _string, _string, _string,
+    ]
+
+    label, \
+        text1, text2, text3, \
+        text4, text5, text6 \
+        = type_sequence(schema, type_args[ix:ix+7])
+
+    dex_text = DexText(
+        label,
+        [
+            ('text', text1),
+            ('next', text2),
+            ('next', text3),
+
+            ('page', text4),
+            ('next', text5),
+            ('next', text6),
+        ]
+    )
+    return ix+7, dex_text
+
+def parse_texts(tokens):
+    data_tokens = {'text', 'next', 'page'}
+    type_args = label_data.event(tokens, data_tokens=data_tokens)
+    ix = 0
+    while ix < len(type_args):
+        ix, dex_text = parse_text(type_args, ix)
+        yield dex_text
+
+def parse(prefix):
+    path = prefix / "data/pokemon/dex_text.asm"
+    with open(path) as f:
+        return list(parse_texts(tokenize.lines(f.read().split('\n'))))
+
+
+from pathlib import Path
+from pprint import pprint
+pprint(parse(Path("pokered")))
diff --git a/tools/parse/pokemon/evos_moves.py b/tools/parse/pokemon/evos_moves.py
index 4230c8a..6c6c949 100644
--- a/tools/parse/pokemon/evos_moves.py
+++ b/tools/parse/pokemon/evos_moves.py
@@ -40,7 +40,8 @@ def parse_learnset_entry(args):
 def build_tables(tokens):
     evos_moves = EvosMoves()
 
-    type_args = label_data.event(tokens)
+    data_tokens = {'dw', 'db'}
+    type_args = label_data.event(tokens, data_tokens=data_tokens)
     ix = 0
     while ix < len(type_args):
         type, args = type_args[ix]