More parsing

5 files changed, 176 insertions, 15 deletions

diff --git a/examples/conf.pom b/examples/conf.pom
index 82e1abd..36a6dcf 100644
--- a/examples/conf.pom
+++ b/examples/conf.pom
@@ -1,7 +1,7 @@
 indentation-type = tabs
 show-line-numbers = yes
 tab-size = 4
-font-size = "18"
+font-size = 18
 
 [file-extensions]
 C = .c, .h
diff --git a/examples/read_conf.py b/examples/read_conf.py
index e49fd4f..f8ce8b9 100644
--- a/examples/read_conf.py
+++ b/examples/read_conf.py
@@ -6,5 +6,10 @@ import sys
 sys.path.append(str(Path(__file__).parent.parent))
 
 import pom_parser
-filename = 'examples/conf.pom' if len(sys.argv) < 2 else sys.argv[1]
-print(pom_parser.load_path(filename))
+try:
+	filename = 'examples/conf.pom' if len(sys.argv) < 2 else sys.argv[1]
+	conf = pom_parser.load_path(filename)
+	print(conf.get('indentation-type','hafkjd'))
+	print(conf.section('plug-in'))
+except pom_parser.Error as e:
+	print('Parse error:', str(e), sep = '\n')
diff --git a/mypy.ini b/mypy.ini
index 37da880..be0671c 100644
--- a/mypy.ini
+++ b/mypy.ini
@@ -1,3 +1,2 @@
 [mypy]
 strict = true
-ignore_missing_imports = True
diff --git a/pom_parser/__init__.py b/pom_parser/__init__.py
index 7a53e2d..7e6401c 100644
--- a/pom_parser/__init__.py
+++ b/pom_parser/__init__.py
@@ -1,17 +1,167 @@
 import io
+from typing import Optional, Any, Iterable
+
+class Error(ValueError):
+	next: Optional['Error']
+	message: str
+	file: str
+	line: int
+	def __init__(self, file: str, line_number: int, message: str) -> None:
+		self.file = file
+		self.line_number = line_number
+		self.message = message
+		self.next = None
+
+	def __str__(self) -> str:
+		err: Optional['Error'] = self
+		messages = []
+		while err:
+			messages.append(f'{err.file}:{err.line_number}: {err.message}')
+			err = err.next
+		return '\n'.join(messages)
+
+	@staticmethod
+	def _from_list(l: list['Error']) -> 'Error':
+		for (i, e) in enumerate(l[:-1]):
+			e.next = l[i+1]
+		return l[0]
+
+class Item:
+	key: str
+	value: str
+	file: str
+	line: int
+	def __repr__(self) -> str:
+		return f'<Item {self.key} at {self.file}:{self.line}>'
 
 class Configuration:
-	pass
+	_items: dict[str, Item]
+	def __repr__(self) -> str:
+		result = []
+		for item in self._items.values():
+			result.append(f'{item.key}: {repr(item.value)}')
+		return '\n'.join(result)
+
+	def get(self, key: str, default: Optional[str] = None) -> Optional[str]:
+		item = self._items.get(key)
+		if item is None:
+			return default
+		return item.value
+
+	def items(self) -> Iterable[Item]:
+		import copy
+		return map(copy.copy, self._items.values())
+
+	def section(self, name: str) -> 'Configuration':
+		import copy
+		section_items = {}
+		name_dot = name + '.'
+		for item in self.items():
+			if item.key.startswith(name_dot):
+				item_copy = copy.copy(item)
+				section_items[item.key[len(name_dot):]] = item_copy
+		conf = Configuration()
+		conf._items = section_items
+		return conf
+
+class _Parser:
+	line_number: int
+	filename: str
+	current_section: str
+	errors: list[Error]
+	file: io.BufferedIOBase
+	items: dict[str, Item]
+
+	def __init__(self, filename: str, file: io.BufferedIOBase):
+		self.errors = []
+		self.filename = filename
+		self.file = file
+		self.line_number = 0
+		self.current_section = ''
+		self.items = {}
+
+	def _error(self, message: str) -> None:
+		self.errors.append(Error(self.filename, self.line_number, message))
+
+	def _check_key(self, key: str) -> None:
+		if not key:
+			self._error('Empty key (expected something before =)')
+			return
+		if '..' in key:
+			self._error(f"Key {key} shouldn't contain ..")
+			return
+		if key.startswith('.'):
+			self._error(f"Key {key} shouldn't start with .")
+			return
+		if key.endswith('.'):
+			self._error(f"Key {key} shouldn't end with .")
+			return
+		# TODO
+
+	def _parse_quoted_value(self, value_start: str) -> str:
+		raise NotImplementedError('TODO: quoted value')
 
-class Settings:
-	pass
+	def _read_line(self) -> bool:
+		line_bytes = self.file.readline()
+		if not line_bytes:
+			return False
+		self.line_number += 1
+		try:
+			line = line_bytes.decode()
+		except UnicodeDecodeError:
+			self._error('Bad UTF-8')
+			return True
+		if line.endswith('\r\n'):
+			line = line[:-2]
+		elif line.endswith('\n'):
+			line = line[:-1]
+		for c in line:
+			if ord(c) < 32 and c != '\t':
+				self._error(f'Invalid character in file: ASCII control character {ord(c)}')
+				return True
+		line = line.lstrip(' \t')
+		if not line or line.startswith('#'):
+			return True
+		if line.startswith('['):
+			line = line.rstrip(' \t')
+			if not line.endswith(']'):
+				self._error('[ with no matching ]')
+				return True
+			self.current_section = line[1:-1]
+			if self.current_section:
+				self._check_key(self.current_section)
+			return True
+		equals_idx = line.find('=')
+		if equals_idx == -1:
+			self._error('Invalid line — should either start with [ or contain =')
+			return True
+		relative_key = line[:equals_idx].rstrip(' \t')
+		self._check_key(relative_key)
+		value = line[equals_idx+1:].lstrip(' \t')
+		if value.startswith('"') or value.startswith('`'):
+			value = self._parse_quoted_value(value)
+		key = f'{self.current_section}.{relative_key}' if self.current_section else relative_key
+		item = Item()
+		item.key = key
+		item.value = value
+		item.file = self.filename
+		item.line = self.line_number
+		self.items[key] = item
+		return True
 
-def load_file(filename: str, file: io.IOBase, settings: Settings = Settings()) -> Configuration:
-	raise NotImplementedError('not implemented')
+def load_file(filename: str, file: io.BufferedIOBase) -> Configuration:
+	parser = _Parser(filename, file)
+	while parser._read_line():
+		pass
+	if parser.errors:
+		raise Error._from_list(parser.errors)
+	conf = Configuration()
+	conf._items = parser.items
+	return conf
 
-def load_string(filename: str, string: str, settings: Settings = Settings()) -> Configuration:
-	return load_file(filename, io.BytesIO(string.encode()), settings)
+def load_string(filename: str, string: str) -> Configuration:
+	return load_file(filename, io.BytesIO(string.encode()))
 
-def load_path(path: str, settings: Settings = Settings()) -> Configuration:
+def load_path(path: str) -> Configuration:
 	with open(path, 'rb') as file:
-		return load_file(path, file, settings)
+		return load_file(path, file)
diff --git a/pylintrc.toml b/pylintrc.toml
index d7f2d64..ad744ac 100644
--- a/pylintrc.toml
+++ b/pylintrc.toml
@@ -318,7 +318,7 @@ max-module-lines = 1000
 
 [tool.pylint.imports]
 # List of modules that can be imported at any level, not just the top level one.
-allow-any-import-level = ["boto3", "botocore", "pydo"]
+allow-any-import-level = ["copy"]
 
 # Allow explicit reexports by alias from a package __init__.
 # allow-reexport-from-package =
@@ -374,7 +374,14 @@ confidence = ["HIGH", "CONTROL_FLOW", "INFERENCE", "INFERENCE_FAILURE", "UNDEFIN
 # --enable=similarities". If you want to run only the classes checker, but have
 # no Warning level messages displayed, use "--disable=all --enable=classes
 # --disable=W".
-disable = ["raw-checker-failed", "bad-inline-option", "locally-disabled", "file-ignored", "suppressed-message", "useless-suppression", "deprecated-pragma", "use-implicit-booleaness-not-comparison-to-string", "use-implicit-booleaness-not-comparison-to-zero", "use-symbolic-message-instead"]
+disable = ["raw-checker-failed", "bad-inline-option", "locally-disabled", "file-ignored",
+	"suppressed-message", "useless-suppression", "deprecated-pragma",
+	"use-implicit-booleaness-not-comparison-to-string", "use-implicit-booleaness-not-comparison-to-zero",
+	"use-symbolic-message-instead",
+	"protected-access", # way too strict to be useful
+	"too-many-return-statements", # who cares?
+	"too-few-public-methods", # who cares?
+]
 
 # Enable the message, report, category or checker with the given id(s). You can
 # either give multiple identifier separated by comma (,) or put this option