Remove the ContentParser class, use functions instead
It did not help, just added to the indent.
This commit is contained in:
parent
a544453338
commit
073dd3025c
4 changed files with 394 additions and 391 deletions
|
@ -7,7 +7,7 @@ from toot import __version__
|
|||
from toot import api
|
||||
from toot.tui.utils import highlight_keys
|
||||
from toot.tui.widgets import Button, EditBox, SelectableText
|
||||
from toot.tui.richtext import ContentParser
|
||||
from toot.tui.richtext import html_to_widgets
|
||||
|
||||
|
||||
class StatusSource(urwid.Padding):
|
||||
|
@ -255,8 +255,6 @@ class Account(urwid.ListBox):
|
|||
super().__init__(walker)
|
||||
|
||||
def generate_contents(self, account, relationship=None, last_action=None):
|
||||
parser = ContentParser()
|
||||
|
||||
if self.last_action and not self.last_action.startswith("Confirm"):
|
||||
yield Button(f"Confirm {self.last_action}", on_press=take_action, user_data=self)
|
||||
yield Button("Cancel", on_press=cancel_action, user_data=self)
|
||||
|
@ -282,7 +280,7 @@ class Account(urwid.ListBox):
|
|||
if account["note"]:
|
||||
yield urwid.Divider()
|
||||
|
||||
widgetlist = parser.html_to_widgets(account["note"])
|
||||
widgetlist = html_to_widgets(account["note"])
|
||||
for line in widgetlist:
|
||||
yield (line)
|
||||
|
||||
|
@ -317,7 +315,7 @@ class Account(urwid.ListBox):
|
|||
yield urwid.Divider()
|
||||
yield urwid.Text([("bold", f"{name.rstrip(':')}"), ":"])
|
||||
|
||||
widgetlist = parser.html_to_widgets(field["value"])
|
||||
widgetlist = html_to_widgets(field["value"])
|
||||
for line in widgetlist:
|
||||
yield (line)
|
||||
|
||||
|
|
|
@ -4,7 +4,7 @@ from toot import api
|
|||
from toot.exceptions import ApiError
|
||||
from toot.utils.datetime import parse_datetime
|
||||
from .widgets import Button, CheckBox, RadioButton
|
||||
from .richtext import ContentParser
|
||||
from .richtext import html_to_widgets
|
||||
|
||||
|
||||
class Poll(urwid.ListBox):
|
||||
|
@ -86,8 +86,7 @@ class Poll(urwid.ListBox):
|
|||
def generate_contents(self, status):
|
||||
yield urwid.Divider()
|
||||
|
||||
parser = ContentParser()
|
||||
widgetlist = parser.html_to_widgets(status.data["content"])
|
||||
widgetlist = html_to_widgets(status.data["content"])
|
||||
|
||||
for line in widgetlist:
|
||||
yield (line)
|
||||
|
|
|
@ -16,10 +16,7 @@ STYLE_NAMES = [p[0] for p in PALETTE]
|
|||
BLOCK_TAGS = ["p", "pre", "li", "blockquote", "h1", "h2", "h3", "h4", "h5", "h6"]
|
||||
|
||||
|
||||
class ContentParser:
|
||||
"""Parse a limited subset of HTML and create urwid widgets."""
|
||||
|
||||
def html_to_widgets(self, html, recovery_attempt=False) -> List[urwid.Widget]:
|
||||
def html_to_widgets(html, recovery_attempt=False) -> List[urwid.Widget]:
|
||||
"""Convert html to urwid widgets"""
|
||||
widgets: List[urwid.Widget] = []
|
||||
html = unicodedata.normalize("NFKC", html)
|
||||
|
@ -33,7 +30,7 @@ class ContentParser:
|
|||
# the HTML is out of spec, doesn't start with a tag,
|
||||
# we see this in content from Pixelfed servers.
|
||||
# attempt a fix by wrapping the HTML with <p></p>
|
||||
return self.html_to_widgets(f"<p>{html}</p>", recovery_attempt=True)
|
||||
return html_to_widgets(f"<p>{html}</p>", recovery_attempt=True)
|
||||
else:
|
||||
continue
|
||||
else:
|
||||
|
@ -42,14 +39,14 @@ class ContentParser:
|
|||
# the HTML is out of spec. Attempt a fix by wrapping the
|
||||
# HTML with <p></p>
|
||||
if (first_tag and not recovery_attempt and name not in BLOCK_TAGS):
|
||||
return self.html_to_widgets(f"<p>{html}</p>", recovery_attempt=True)
|
||||
return html_to_widgets(f"<p>{html}</p>", recovery_attempt=True)
|
||||
|
||||
markup = self.render(name, e)
|
||||
markup = render(name, e)
|
||||
first_tag = False
|
||||
|
||||
if not isinstance(markup, urwid.Widget):
|
||||
# plaintext, so create a padded text widget
|
||||
txt = self.text_to_widget("", markup)
|
||||
txt = text_to_widget("", markup)
|
||||
markup = urwid.Padding(
|
||||
txt,
|
||||
align="left",
|
||||
|
@ -61,26 +58,29 @@ class ContentParser:
|
|||
widgets.append(urwid.Divider(" "))
|
||||
return widgets[:-1] # but suppress the last blank line
|
||||
|
||||
def inline_tag_to_text(self, tag) -> Tuple:
|
||||
|
||||
def inline_tag_to_text(tag) -> Tuple:
|
||||
"""Convert html tag to plain text with tag as attributes recursively"""
|
||||
markups = self.process_inline_tag_children(tag)
|
||||
markups = process_inline_tag_children(tag)
|
||||
if not markups:
|
||||
return (tag.name, "")
|
||||
return (tag.name, markups)
|
||||
|
||||
def process_inline_tag_children(self, tag) -> List:
|
||||
|
||||
def process_inline_tag_children(tag) -> List:
|
||||
"""Recursively retrieve all children
|
||||
and convert to a list of markup text"""
|
||||
markups = []
|
||||
for child in tag.children:
|
||||
if isinstance(child, Tag):
|
||||
markup = self.render(child.name, child)
|
||||
markup = render(child.name, child)
|
||||
markups.append(markup)
|
||||
else:
|
||||
markups.append(child)
|
||||
return markups
|
||||
|
||||
def text_to_widget(self, attr, markup) -> urwid.Widget:
|
||||
|
||||
def text_to_widget(attr, markup) -> urwid.Widget:
|
||||
if not has_urwidgets:
|
||||
return urwid.Text((attr, markup))
|
||||
|
||||
|
@ -99,7 +99,7 @@ class ContentParser:
|
|||
# find anchor titles with an ETX separator followed by href
|
||||
m = re.match(r"(^.+)\x03(.+$)", txt)
|
||||
if m:
|
||||
anchor_attr = self.get_best_anchor_attr(attr_list)
|
||||
anchor_attr = get_best_anchor_attr(attr_list)
|
||||
markup_list.append(
|
||||
parse_text(
|
||||
txt,
|
||||
|
@ -114,7 +114,8 @@ class ContentParser:
|
|||
|
||||
return TextEmbed(markup_list)
|
||||
|
||||
def process_block_tag_children(self, tag) -> List[urwid.Widget]:
|
||||
|
||||
def process_block_tag_children(tag) -> List[urwid.Widget]:
|
||||
"""Recursively retrieve all children
|
||||
and convert to a list of widgets
|
||||
any inline tags containing text will be
|
||||
|
@ -129,7 +130,7 @@ class ContentParser:
|
|||
if isinstance(child, Tag):
|
||||
# child is a nested tag; process using custom method
|
||||
# or default to inline_tag_to_text
|
||||
result = self.render(child.name, child)
|
||||
result = render(child.name, child)
|
||||
if isinstance(result, urwid.Widget):
|
||||
found_nested_widget = True
|
||||
child_widgets.append(result)
|
||||
|
@ -147,17 +148,18 @@ class ContentParser:
|
|||
|
||||
widget_list = []
|
||||
if len(pre_widget_markups):
|
||||
widget_list.append(self.text_to_widget(tag.name, pre_widget_markups))
|
||||
widget_list.append(text_to_widget(tag.name, pre_widget_markups))
|
||||
|
||||
if len(child_widgets):
|
||||
widget_list += child_widgets
|
||||
|
||||
if len(post_widget_markups):
|
||||
widget_list.append(self.text_to_widget(tag.name, post_widget_markups))
|
||||
widget_list.append(text_to_widget(tag.name, post_widget_markups))
|
||||
|
||||
return widget_list
|
||||
|
||||
def get_urwid_attr_name(self, tag) -> str:
|
||||
|
||||
def get_urwid_attr_name(tag) -> str:
|
||||
"""Get the class name and translate to a
|
||||
name suitable for use as an urwid
|
||||
text attribute name"""
|
||||
|
@ -174,17 +176,13 @@ class ContentParser:
|
|||
# fallback to returning the tag name
|
||||
return tag.name
|
||||
|
||||
# Tag handlers start here.
|
||||
# Tags not explicitly listed are "supported" by
|
||||
# rendering as text.
|
||||
# Inline tags return a list of marked up text for urwid.Text
|
||||
# Block tags return urwid.Widget
|
||||
|
||||
def basic_block_tag_handler(self, tag) -> urwid.Widget:
|
||||
def basic_block_tag_handler(tag) -> urwid.Widget:
|
||||
"""default for block tags that need no special treatment"""
|
||||
return urwid.Pile(self.process_block_tag_children(tag))
|
||||
return urwid.Pile(process_block_tag_children(tag))
|
||||
|
||||
def get_best_anchor_attr(self, attrib_list) -> str:
|
||||
|
||||
def get_best_anchor_attr(attrib_list) -> str:
|
||||
if not attrib_list:
|
||||
return ""
|
||||
flat_al = list(flatten(attrib_list))
|
||||
|
@ -202,46 +200,48 @@ class ContentParser:
|
|||
|
||||
return "a"
|
||||
|
||||
def render(self, attr: str, content: str):
|
||||
|
||||
def render(attr: str, content: str):
|
||||
if attr in ["a"]:
|
||||
return self.render_anchor(content)
|
||||
return render_anchor(content)
|
||||
|
||||
if attr in ["blockquote"]:
|
||||
return self.render_blockquote(content)
|
||||
return render_blockquote(content)
|
||||
|
||||
if attr in ["br"]:
|
||||
return self.render_br(content)
|
||||
return render_br(content)
|
||||
|
||||
if attr in ["em"]:
|
||||
return self.render_em(content)
|
||||
return render_em(content)
|
||||
|
||||
if attr in ["ol"]:
|
||||
return self.render_ol(content)
|
||||
return render_ol(content)
|
||||
|
||||
if attr in ["pre"]:
|
||||
return self.render_pre(content)
|
||||
return render_pre(content)
|
||||
|
||||
if attr in ["span"]:
|
||||
return self.render_span(content)
|
||||
return render_span(content)
|
||||
|
||||
if attr in ["b", "strong"]:
|
||||
return self.render_strong(content)
|
||||
return render_strong(content)
|
||||
|
||||
if attr in ["ul"]:
|
||||
return self.render_ul(content)
|
||||
return render_ul(content)
|
||||
|
||||
# Glitch-soc and Pleroma allow <H1>...<H6> in content
|
||||
# Mastodon (PR #23913) does not; header tags are converted to <P><STRONG></STRONG></P>
|
||||
if attr in ["p", "div", "li", "h1", "h2", "h3", "h4", "h5", "h6"]:
|
||||
return self.basic_block_tag_handler(content)
|
||||
return basic_block_tag_handler(content)
|
||||
|
||||
# Fall back to inline_tag_to_text handler
|
||||
return self.inline_tag_to_text(content)
|
||||
return inline_tag_to_text(content)
|
||||
|
||||
def render_anchor(self, tag) -> Tuple:
|
||||
|
||||
def render_anchor(tag) -> Tuple:
|
||||
"""anchor tag handler"""
|
||||
|
||||
markups = self.process_inline_tag_children(tag)
|
||||
markups = process_inline_tag_children(tag)
|
||||
if not markups:
|
||||
return (tag.name, "")
|
||||
|
||||
|
@ -257,14 +257,14 @@ class ContentParser:
|
|||
# delimiter between the title and the HREF
|
||||
title += f"\x03{href}"
|
||||
|
||||
attr = self.get_best_anchor_attr(attrib_list)
|
||||
attr = get_best_anchor_attr(attrib_list)
|
||||
|
||||
if attr == "a":
|
||||
# didn't find an attribute to use
|
||||
# in the child markup, so let's
|
||||
# try the anchor tag's own attributes
|
||||
|
||||
attr = self.get_urwid_attr_name(tag)
|
||||
attr = get_urwid_attr_name(tag)
|
||||
|
||||
# hashtag anchors have a class of "mention hashtag"
|
||||
# or "hashtag"
|
||||
|
@ -275,8 +275,9 @@ class ContentParser:
|
|||
|
||||
return (attr, title)
|
||||
|
||||
def render_blockquote(self, tag) -> urwid.Widget:
|
||||
widget_list = self.process_block_tag_children(tag)
|
||||
|
||||
def render_blockquote(tag) -> urwid.Widget:
|
||||
widget_list = process_block_tag_children(tag)
|
||||
blockquote_widget = urwid.LineBox(
|
||||
urwid.Padding(
|
||||
urwid.Pile(widget_list),
|
||||
|
@ -297,13 +298,15 @@ class ContentParser:
|
|||
)
|
||||
return urwid.Pile([urwid.AttrMap(blockquote_widget, "blockquote")])
|
||||
|
||||
def render_br(self, tag) -> Tuple:
|
||||
|
||||
def render_br(tag) -> Tuple:
|
||||
return ("br", "\n")
|
||||
|
||||
def render_em(self, tag) -> Tuple:
|
||||
|
||||
def render_em(tag) -> Tuple:
|
||||
# to simplify the number of palette entries
|
||||
# translate EM to I (italic)
|
||||
markups = self.process_inline_tag_children(tag)
|
||||
markups = process_inline_tag_children(tag)
|
||||
if not markups:
|
||||
return ("i", "")
|
||||
|
||||
|
@ -314,7 +317,8 @@ class ContentParser:
|
|||
|
||||
return ("i", markups)
|
||||
|
||||
def render_ol(self, tag) -> urwid.Widget:
|
||||
|
||||
def render_ol(tag) -> urwid.Widget:
|
||||
"""ordered list tag handler"""
|
||||
|
||||
widgets = []
|
||||
|
@ -329,7 +333,7 @@ class ContentParser:
|
|||
pass
|
||||
|
||||
for li in tag.find_all("li", recursive=False):
|
||||
markup = self.render("li", li)
|
||||
markup = render("li", li)
|
||||
|
||||
# li value= attribute will change the item number
|
||||
# it also overrides any ol start= attribute
|
||||
|
@ -341,11 +345,11 @@ class ContentParser:
|
|||
pass
|
||||
|
||||
if not isinstance(markup, urwid.Widget):
|
||||
txt = self.text_to_widget("li", [str(list_item_num), ". ", markup])
|
||||
txt = text_to_widget("li", [str(list_item_num), ". ", markup])
|
||||
# 1. foo, 2. bar, etc.
|
||||
widgets.append(txt)
|
||||
else:
|
||||
txt = self.text_to_widget("li", [str(list_item_num), ". "])
|
||||
txt = text_to_widget("li", [str(list_item_num), ". "])
|
||||
columns = urwid.Columns(
|
||||
[txt, ("weight", 9999, markup)], dividechars=1, min_width=3
|
||||
)
|
||||
|
@ -355,14 +359,15 @@ class ContentParser:
|
|||
|
||||
return urwid.Pile(widgets)
|
||||
|
||||
def render_pre(self, tag) -> urwid.Widget:
|
||||
|
||||
def render_pre(tag) -> urwid.Widget:
|
||||
# <PRE> tag spec says that text should not wrap,
|
||||
# but horizontal screen space is at a premium
|
||||
# and we have no horizontal scroll bar, so allow
|
||||
# wrapping.
|
||||
|
||||
widget_list = [urwid.Divider(" ")]
|
||||
widget_list += self.process_block_tag_children(tag)
|
||||
widget_list += process_block_tag_children(tag)
|
||||
|
||||
pre_widget = urwid.Padding(
|
||||
urwid.Pile(widget_list),
|
||||
|
@ -374,8 +379,9 @@ class ContentParser:
|
|||
)
|
||||
return urwid.Pile([urwid.AttrMap(pre_widget, "pre")])
|
||||
|
||||
def render_span(self, tag) -> Tuple:
|
||||
markups = self.process_inline_tag_children(tag)
|
||||
|
||||
def render_span(tag) -> Tuple:
|
||||
markups = process_inline_tag_children(tag)
|
||||
|
||||
if not markups:
|
||||
return (tag.name, "")
|
||||
|
@ -394,22 +400,23 @@ class ContentParser:
|
|||
# if "invisible" in tag.attrs["class"]:
|
||||
# return (tag.name, "")
|
||||
|
||||
style_name = self.get_urwid_attr_name(tag)
|
||||
style_name = get_urwid_attr_name(tag)
|
||||
|
||||
if style_name != "span":
|
||||
# unique class name matches an entry in our palette
|
||||
return (style_name, markups)
|
||||
|
||||
if tag.parent:
|
||||
return (self.get_urwid_attr_name(tag.parent), markups)
|
||||
return (get_urwid_attr_name(tag.parent), markups)
|
||||
else:
|
||||
# fallback
|
||||
return ("span", markups)
|
||||
|
||||
def render_strong(self, tag) -> Tuple:
|
||||
|
||||
def render_strong(tag) -> Tuple:
|
||||
# to simplify the number of palette entries
|
||||
# translate STRONG to B (bold)
|
||||
markups = self.process_inline_tag_children(tag)
|
||||
markups = process_inline_tag_children(tag)
|
||||
if not markups:
|
||||
return ("b", "")
|
||||
|
||||
|
@ -420,20 +427,21 @@ class ContentParser:
|
|||
|
||||
return ("b", markups)
|
||||
|
||||
def render_ul(self, tag) -> urwid.Widget:
|
||||
|
||||
def render_ul(tag) -> urwid.Widget:
|
||||
"""unordered list tag handler"""
|
||||
|
||||
widgets = []
|
||||
|
||||
for li in tag.find_all("li", recursive=False):
|
||||
markup = self.render("li", li)
|
||||
markup = render("li", li)
|
||||
|
||||
if not isinstance(markup, urwid.Widget):
|
||||
txt = self.text_to_widget("li", ["\N{bullet} ", markup])
|
||||
txt = text_to_widget("li", ["\N{bullet} ", markup])
|
||||
# * foo, * bar, etc.
|
||||
widgets.append(txt)
|
||||
else:
|
||||
txt = self.text_to_widget("li", ["\N{bullet} "])
|
||||
txt = text_to_widget("li", ["\N{bullet} "])
|
||||
columns = urwid.Columns(
|
||||
[txt, ("weight", 9999, markup)], dividechars=1, min_width=3
|
||||
)
|
||||
|
|
|
@ -6,6 +6,7 @@ import webbrowser
|
|||
from typing import List, Optional
|
||||
|
||||
from toot.tui import app
|
||||
from toot.tui.richtext import html_to_widgets
|
||||
from toot.utils.datetime import parse_datetime, time_ago
|
||||
from toot.utils.language import language_name
|
||||
|
||||
|
@ -13,7 +14,6 @@ from toot.entities import Status
|
|||
from toot.tui.scroll import Scrollable, ScrollBar
|
||||
from toot.tui.utils import highlight_keys
|
||||
from toot.tui.widgets import SelectableText, SelectableColumns
|
||||
from toot.tui.richtext import ContentParser
|
||||
from toot.utils import urlencode_url
|
||||
from toot.tui.stubs.urwidgets import Hyperlink, TextEmbed, parse_text, has_urwidgets
|
||||
|
||||
|
@ -356,9 +356,7 @@ class StatusDetails(urwid.Pile):
|
|||
yield ("pack", urwid.Text(("content_warning", "Marked as sensitive. Press S to view.")))
|
||||
else:
|
||||
content = status.original.translation if status.original.show_translation else status.data["content"]
|
||||
|
||||
parser = ContentParser()
|
||||
widgetlist = parser.html_to_widgets(content)
|
||||
widgetlist = html_to_widgets(content)
|
||||
|
||||
for line in widgetlist:
|
||||
yield (line)
|
||||
|
|
Loading…
Reference in a new issue