# that will be later stripped out for consistent behavior.
if tree.tag not in lxml.html.defs.top_level_tags:
- html = b"
"
+ tree = lxml.html.fromstring(htmlb, parser=parser)
# HACK for Outlook emails, where tags like
are rendered as . We
# can generally ignore these tags so we replace them with , which
@@ -229,7 +241,7 @@ def get_html_tree(html):
return tree
-def strip_wrapping(html):
+def strip_wrapping(html: str) -> str:
"""
Remove the wrapping that might have resulted when using get_html_tree().
"""
@@ -238,7 +250,7 @@ def strip_wrapping(html):
return html.strip()
-def render_html_tree(tree):
+def render_html_tree(tree: Element) -> str:
"""
Render the given HTML tree, and strip any wrapping that was applied in
get_html_tree().
@@ -257,13 +269,15 @@ def render_html_tree(tree):
return strip_wrapping(html)
-def is_indentation_element(element):
+def is_indentation_element(element: Element) -> bool:
if isinstance(element.tag, str):
return element.tag.lower() == "blockquote"
return False
-def tree_token_generator(el, indentation_level=0):
+def tree_token_generator(
+ el: Element, indentation_level: int = 0
+) -> Iterator[None | tuple[Element, str, int] | str]:
"""
Yield tokens for the given HTML element as follows:
@@ -296,7 +310,13 @@ def tree_token_generator(el, indentation_level=0):
yield el.tail
-def tree_line_generator(el, max_lines=None):
+def tree_line_generator(
+ el: Element, max_lines: int | None = None
+) -> Iterator[
+ tuple[
+ tuple[ElementRef, str] | None, tuple[ElementRef, str] | None, int, str
+ ]
+]:
"""
Iterate through an LXML tree and yield a tuple per line.
@@ -327,7 +347,7 @@ def tree_line_generator(el, max_lines=None):
- ((, 'end'), (, 'end'), 0, 'world')
"""
- def _trim_spaces(text):
+ def _trim_spaces(text: str) -> str:
return MULTIPLE_WHITESPACE_RE.sub(" ", text).strip()
counter = 1
@@ -341,7 +361,7 @@ def _trim_spaces(text):
start_ref = None
# The indentation level at the start of the line.
- start_indentation_level = None
+ start_indentation_level = 0
for token in tree_token_generator(el):
if token is None:
@@ -393,12 +413,17 @@ def _trim_spaces(text):
else:
raise RuntimeError(f"invalid token: {token}")
+ """
+ TODO: wrong type, would trigger error if reached.
line = _trim_spaces(line)
if line:
yield line
+ """
-def indented_tree_line_generator(el, max_lines=None):
+def indented_tree_line_generator(
+ el: Element, max_lines: int | None = None
+) -> Iterator[tuple[ElementRef | None, ElementRef | None, str]]:
r"""
Like tree_line_generator, but yields tuples (start_ref, end_ref, line),
where the line already takes the indentation into account by having "> "
@@ -413,14 +438,19 @@ def indented_tree_line_generator(el, max_lines=None):
yield start_ref, end_ref, "> " * indentation_level + full_line
-def get_line_info(tree, max_lines=None):
+def get_line_info(
+ tree: Element, max_lines: int | None = None
+) -> tuple[list[ElementRef | None], list[ElementRef | None], list[str]]:
"""
Shortcut for indented_tree_line_generator() that returns an array of
start references, an array of corresponding end references (see
tree_line_generator() docs), and an array of corresponding lines.
"""
line_gen = indented_tree_line_generator(tree, max_lines=max_lines)
- line_gen_result = list(zip(*line_gen))
+ line_gen_result: (
+ tuple[list[ElementRef | None], list[ElementRef | None], list[str]]
+ | tuple[()]
+ ) = tuple(zip(*line_gen))
if line_gen_result:
return line_gen_result
return [], [], []
diff --git a/quotequail/_internal.py b/quotequail/_internal.py
index c76aa60..50c2cf8 100644
--- a/quotequail/_internal.py
+++ b/quotequail/_internal.py
@@ -12,7 +12,9 @@
"""
-def find_pattern_on_line(lines, n, max_wrap_lines):
+def find_pattern_on_line(
+ lines: list[str], n: int, max_wrap_lines: int
+) -> tuple[int, str] | None:
"""
Find a forward/reply pattern within the given lines on text on the given
line number and return a tuple with the type ('reply' or 'forward') and
@@ -20,7 +22,7 @@ def find_pattern_on_line(lines, n, max_wrap_lines):
different from the given line number in case the pattern wraps over
multiple lines.
- Returns (None, None) if no pattern was found.
+ Returns None if no pattern was found.
"""
for typ, regexes in COMPILED_PATTERN_MAP.items():
for regex in regexes:
@@ -30,25 +32,27 @@ def find_pattern_on_line(lines, n, max_wrap_lines):
match_line = match_line[1:].strip()
if regex.match(match_line.strip()):
return n + m, typ
- return None, None
+ return None
-def find_quote_position(lines, max_wrap_lines, limit=None):
+def find_quote_position(
+ lines: list[str], max_wrap_lines: int, limit: int | None = None
+) -> int | None:
"""
Return the (ending) line number of a quoting pattern. If a limit is given
and the limit is reached, the limit is returned.
"""
for n in range(len(lines)):
- end, typ = find_pattern_on_line(lines, n, max_wrap_lines)
- if typ:
- return end
+ result = find_pattern_on_line(lines, n, max_wrap_lines)
+ if result:
+ return result[0]
if limit is not None and n >= limit - 1:
return n
return None
-def join_wrapped_lines(lines):
+def join_wrapped_lines(lines: list[str]) -> str:
"""
Join one or multiple lines that wrapped. Returns the reconstructed line.
Takes into account proper spacing between the lines (see
@@ -68,7 +72,9 @@ def join_wrapped_lines(lines):
return joined
-def extract_headers(lines, max_wrap_lines):
+def extract_headers(
+ lines: list[str], max_wrap_lines: int
+) -> tuple[dict[str, str], int]:
"""
Extract email headers from the given lines. Returns a dict with the
detected headers and the amount of lines that were processed.
@@ -109,7 +115,7 @@ def extract_headers(lines, max_wrap_lines):
return hdrs, lines_processed
-def parse_reply(line):
+def parse_reply(line: str) -> dict[str, str] | None:
"""
Parse the given reply line ("On DATE, USER wrote:") and returns a
dictionary with the "Date" and "From" keys, or None, if couldn't parse.
@@ -154,8 +160,11 @@ def parse_reply(line):
def find_unwrap_start(
- lines, max_wrap_lines, min_header_lines, min_quoted_lines
-):
+ lines: list[str],
+ max_wrap_lines: int,
+ min_header_lines: int,
+ min_quoted_lines: int,
+) -> tuple[int, int, str] | None:
"""
Find the starting point of a wrapped email. Returns a tuple containing
(start_line_number, end_line_number, type), where type can be one of the
@@ -164,7 +173,7 @@ def find_unwrap_start(
* 'forward': A matching forwarding pattern was found
* 'reply': A matching reply pattern was found
* 'headers': Headers were found (usually a forwarded email)
- * 'quote': A quote was found
+ * 'quoted': A quote was found
start_line_number corresponds to the line number where the forwarding/reply
pattern starts, or where the headers/quote starts. end_line_number is only
@@ -172,7 +181,7 @@ def find_unwrap_start(
multiple lines (it does not extend to the end of the headers or of the
quoted section).
- Returns (None, None, None) if nothing was found.
+ Returns None if nothing was found.
"""
for n, line in enumerate(lines):
if not line.strip():
@@ -180,8 +189,9 @@ def find_unwrap_start(
# Find a forward / reply start pattern
- end, typ = find_pattern_on_line(lines, n, max_wrap_lines)
- if typ:
+ result = find_pattern_on_line(lines, n, max_wrap_lines)
+ if result:
+ end, typ = result
return n, end, typ
# Find a quote
@@ -210,10 +220,10 @@ def find_unwrap_start(
):
return n, n, "headers"
- return None, None, None
+ return None
-def unindent_lines(lines):
+def unindent_lines(lines: list[str]) -> list[str]:
unquoted = []
for line in lines:
if line.startswith("> "):
@@ -226,7 +236,22 @@ def unindent_lines(lines):
return unquoted
-def unwrap(lines, max_wrap_lines, min_header_lines, min_quoted_lines):
+def unwrap(
+ lines: list[str],
+ max_wrap_lines: int,
+ min_header_lines: int,
+ min_quoted_lines: int,
+) -> (
+ tuple[
+ str,
+ tuple[int | None, int | None],
+ dict[str, str] | None,
+ tuple[int | None, int | None] | None,
+ tuple[int | None, int | None] | None,
+ bool,
+ ]
+ | None
+):
"""
Return a tuple of:
- Type ('forward', 'reply', 'headers', 'quoted')
@@ -239,9 +264,13 @@ def unwrap(lines, max_wrap_lines, min_header_lines, min_quoted_lines):
headers = {}
# Get line number and wrapping type.
- start, end, typ = find_unwrap_start(
+ result = find_unwrap_start(
lines, max_wrap_lines, min_header_lines, min_quoted_lines
)
+ if not result:
+ return None
+
+ start, end, typ = result
# We found a line indicating that it's a forward/reply.
if typ in ("forward", "reply"):
@@ -256,19 +285,23 @@ def unwrap(lines, max_wrap_lines, min_header_lines, min_quoted_lines):
# Find where the headers or the quoted section starts.
# We can set min_quoted_lines to 1 because we expect a quoted section.
- start2, end2, typ = find_unwrap_start(
+ result = find_unwrap_start(
lines[end + 1 :], max_wrap_lines, min_header_lines, 1
)
+ start2 = result[0] if result else 0
+ typ2 = result[2] if result else None
- if typ == "quoted":
+ if typ2 == "quoted":
# Quoted section starts. Unindent and check if there are headers.
quoted_start = end + 1 + start2
unquoted = unindent_lines(lines[quoted_start:])
rest_start = quoted_start + len(unquoted)
- start3, end3, typ = find_unwrap_start(
+ result = find_unwrap_start(
unquoted, max_wrap_lines, min_header_lines, min_quoted_lines
)
- if typ == "headers":
+ start3 = result[0] if result else 0
+ typ3 = result[2] if result else None
+ if typ3 == "headers":
hdrs, hdrs_length = extract_headers(
unquoted[start3:], max_wrap_lines
)
@@ -330,10 +363,12 @@ def unwrap(lines, max_wrap_lines, min_header_lines, min_quoted_lines):
if typ == "quoted":
unquoted = unindent_lines(lines[start:])
rest_start = start + len(unquoted)
- start2, end2, typ = find_unwrap_start(
+ result = find_unwrap_start(
unquoted, max_wrap_lines, min_header_lines, min_quoted_lines
)
- if typ == "headers":
+ start2 = result[0] if result else 0
+ typ2 = result[2] if result else None
+ if typ2 == "headers":
main_type = "forward"
hdrs, hdrs_length = extract_headers(
unquoted[start2:], max_wrap_lines
@@ -358,4 +393,4 @@ def unwrap(lines, max_wrap_lines, min_header_lines, min_quoted_lines):
True,
)
- return None
+ raise RuntimeError(f"invalid type: {typ}")
diff --git a/quotequail/types.py b/quotequail/types.py
new file mode 100644
index 0000000..af1278b
--- /dev/null
+++ b/quotequail/types.py
@@ -0,0 +1,8 @@
+from typing import TYPE_CHECKING, TypeAlias
+
+if TYPE_CHECKING:
+ from lxml.html import HtmlElement
+
+ Element: TypeAlias = HtmlElement
+
+ElementRef = tuple["Element", str]
diff --git a/setup.cfg b/setup.cfg
deleted file mode 100644
index 5332636..0000000
--- a/setup.cfg
+++ /dev/null
@@ -1,64 +0,0 @@
-[flake8]
-ignore=
- # !!! make sure you have a comma at the end of each line EXCEPT the LAST one
- # line length, already enforced by black
- E501,
- # https://pypi.org/project/flake8-future-import/
- FI1
- # Missing docstrings
- D1,
- # One-line docstring should fit on one line with quotes.
- # We ignore this because it's OK to buy yourself a few extra characters
- # for the summary line even if the summary line is *the only* line.
- D200,
- # 1 blank line required between summary line and description
- D205,
- # Multi-line docstring summary should start at the first line.
- # We ignore this because we agreed in #20553 that we we want to put the
- # summary line below """ for multi-line docstrings.
- D212,
- # First line should end with a period
- D400,
- # This is not PEP8-compliant and conflicts with black
- W503,
- W504,
- # This is not PEP8-compliant and conflicts with black
- E203,
- # Too intrusive, sometimes makes code less readable
- SIM106
- # Allow f-strings
- SFS301,
- # Allow .format
- SFS201
-exclude=venv
-#max-complexity=2
-banned-modules=
- typing.Text = use str
-require-code=True
-
-[isort]
-skip=venv,src
-known_first_party=quotequail
-known_tests=tests
-sections=FUTURE,STDLIB,THIRDPARTY,FIRSTPARTY,TESTS,LOCALFOLDER
-default_section=THIRDPARTY
-use_parentheses=true
-multi_line_output=3
-include_trailing_comma=True
-force_grid_wrap=0
-combine_as_imports=True
-line_length=87
-
-[mypy]
-python_version = 3.7
-ignore_missing_imports = True
-no_implicit_optional = True
-strict_equality = True
-follow_imports = normal
-warn_unreachable = True
-show_error_context = True
-pretty = True
-files = quotequail
-
-[tool:pytest]
-norecursedirs=venv