dead code: Kill off most of html_branches.

The code that was removed here was intended
for tooling that either never got built or
became obsolete.
This commit is contained in:
Steve Howell 2021-12-02 13:30:52 +00:00 committed by Tim Abbott
parent 5a5dcd6962
commit 42061359a9
2 changed files with 2 additions and 190 deletions

View File

@ -1,60 +1,10 @@
import re
from collections import defaultdict
from typing import Dict, List, Optional, Sequence, Set
from typing import Dict, List
from .template_parser import FormattedException, Token, tokenize
class HtmlTreeBranch:
"""
For <p><div id='yo'>bla<span class='bar'></span></div></p>, store a
representation of the tags all the way down to the leaf, which would
conceptually be something like "p div(#yo) span(.bar)".
"""
def __init__(self, tags: List["TagInfo"], fn: Optional[str]) -> None:
self.tags = tags
self.fn = fn
self.line = tags[-1].token.line
self.words: Set[str] = set()
for tag in tags:
for word in tag.words:
self.words.add(word)
def staircase_text(self) -> str:
"""
produces representation of a node in staircase-like format:
html
body.main-section
p#intro
"""
res = "\n"
indent = " " * 4
for t in self.tags:
res += indent + t.text() + "\n"
indent += " " * 4
return res
def text(self) -> str:
"""
produces one-line representation of branch:
html body.main-section p#intro
"""
return " ".join(t.text() for t in self.tags)
class Node:
def __init__(self, token: Optional[Token], parent: "Optional[Node]") -> None:
# FIXME parent parameter is not used!
self.token = token
self.children: List[Node] = []
self.parent: Optional[Node] = None
class TagInfo:
def __init__(self, tag: str, classes: List[str], ids: List[str], token: Token) -> None:
self.tag = tag
@ -124,50 +74,6 @@ def split_for_id_and_class(element: str) -> List[str]:
return lst
def html_branches(text: str, fn: Optional[str] = None) -> List[HtmlTreeBranch]:
tree = html_tag_tree(text, fn)
branches: List[HtmlTreeBranch] = []
def walk(node: Node, tag_info_list: Sequence[TagInfo] = []) -> None:
assert node.token is not None
info = get_tag_info(node.token)
tag_info_list = [*tag_info_list, info]
if node.children:
for child in node.children:
walk(node=child, tag_info_list=tag_info_list)
else:
tree_branch = HtmlTreeBranch(tags=tag_info_list, fn=fn)
branches.append(tree_branch)
for node in tree.children:
walk(node, [])
return branches
def html_tag_tree(text: str, fn: Optional[str] = None) -> Node:
tokens = tokenize(text)
top_level = Node(token=None, parent=None)
stack = [top_level]
for token in tokens:
# Add tokens to the Node tree first (conditionally).
if token.kind in ("html_start", "html_singleton"):
parent = stack[-1]
node = Node(token=token, parent=parent)
parent.children.append(node)
# Then update the stack to have the next node that
# we will be appending to at the top.
if token.kind == "html_start":
stack.append(node)
elif token.kind == "html_end":
stack.pop()
return top_level
def build_id_dict(templates: List[str]) -> (Dict[str, List[str]]):
template_id_dict: (Dict[str, List[str]]) = defaultdict(list)

View File

@ -2,14 +2,7 @@ import os
import unittest
import tools.lib.template_parser
from tools.lib.html_branches import (
Node,
build_id_dict,
get_tag_info,
html_branches,
html_tag_tree,
split_for_id_and_class,
)
from tools.lib.html_branches import build_id_dict, get_tag_info, split_for_id_and_class
ZULIP_PATH = os.path.dirname(os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
TEST_TEMPLATES_DIR = os.path.join(os.path.dirname(os.path.abspath(__file__)), "test_template_data")
@ -28,93 +21,6 @@ class TestHtmlBranches(unittest.TestCase):
self.assertEqual(end_tag_info.text(), "p")
self.assertEqual(text.s, "foo")
def test_html_tag_tree(self) -> None:
html = """
<!-- test -->
<!DOCTYPE html>
<html>
<!-- test -->
<head>
<title>Test</title>
<meta charset="utf-8" />
<link rel="stylesheet" href="style.css" />
</head>
<body>
<p>Hello<br>world!</p>
<p>Goodbye<!-- test -->world!</p>
</body>
</html>
<!-- test -->
"""
tree = html_tag_tree(html)
def serialize(node: Node) -> object:
return (
node.token and (node.token.kind, node.token.tag),
[serialize(child) for child in node.children],
)
expected = (
None,
[
(
("html_start", "html"),
[
(
("html_start", "head"),
[
(("html_start", "title"), []),
(("html_singleton", "meta"), []),
(("html_singleton", "link"), []),
],
),
(
("html_start", "body"),
[
(
("html_start", "p"),
[(("html_start", "br"), []), (("html_start", "p"), [])],
)
],
),
],
)
],
)
self.assertEqual(serialize(tree), expected)
def test_html_branches(self) -> None:
html = """
<!-- test -->
<!DOCTYPE html>
<html>
<!-- test -->
<head>
<title>Test</title>
<meta charset="utf-8" />
<link rel="stylesheet" href="style.css" />
</head>
<body>
<p>Hello<br>world!</p>
<p>Goodbye<!-- test -->world!</p>
</body>
</html>
<!-- test -->
"""
branches = html_branches(html)
self.assertEqual(
[(branch.text(), branch.staircase_text()) for branch in branches],
[
("html head title", "\n html\n head\n title\n"),
("html head meta", "\n html\n head\n meta\n"),
("html head link", "\n html\n head\n link\n"),
("html body p br", "\n html\n body\n p\n br\n"),
("html body p p", "\n html\n body\n p\n p\n"),
],
)
def test_build_id_dict(self) -> None:
templates = ["test_template1.html", "test_template2.html"]
templates = [os.path.join(TEST_TEMPLATES_DIR, fn) for fn in templates]