paulmataruso/zulip
1
0
Fork 0
mirror of https://github.com/zulip/zulip.git synced 2025-10-23 16:14:02 +00:00
Code Issues Packages Projects Releases Wiki Activity

markdown: Store ZulipMarkdown in members with the right type.

Browse Source 
Signed-off-by: Anders Kaseorg <anders@zulip.com>
(cherry picked from commit 8230324068)
This commit is contained in:
Anders Kaseorg
2022-10-06 13:58:37 -07:00
committed by Tim Abbott
parent 07e1e47db3
commit d509cd0a0f
1 changed files with 68 additions and 40 deletions
Download Patch File Download Diff File Expand all files Collapse all files

108
zerver/lib/markdown/__init__.py
View File

@@ -584,13 +584,17 @@ class InlineImageProcessor(markdown.treeprocessors.Treeprocessor):
view. view.
""" """
def __init__(self, zmd: "ZulipMarkdown") -> None:
super().__init__(zmd)
self.zmd = zmd
def run(self, root: Element) -> None: def run(self, root: Element) -> None:
# Get all URLs from the blob # Get all URLs from the blob
found_imgs = walk_tree(root, lambda e: e if e.tag == "img" else None) found_imgs = walk_tree(root, lambda e: e if e.tag == "img" else None)
for img in found_imgs: for img in found_imgs:
url = img.get("src") url = img.get("src")
assert url is not None assert url is not None
if is_static_or_current_realm_url(url, self.md.zulip_realm): if is_static_or_current_realm_url(url, self.zmd.zulip_realm):
# Don't rewrite images on our own site (e.g. emoji, user uploads). # Don't rewrite images on our own site (e.g. emoji, user uploads).
continue continue
img.set("src", get_camo_url(url)) img.set("src", get_camo_url(url))
@@ -622,6 +626,10 @@ class InlineInterestingLinkProcessor(markdown.treeprocessors.Treeprocessor):
TWITTER_MAX_TO_PREVIEW = 3 TWITTER_MAX_TO_PREVIEW = 3
INLINE_PREVIEW_LIMIT_PER_MESSAGE = 10 INLINE_PREVIEW_LIMIT_PER_MESSAGE = 10
def __init__(self, zmd: "ZulipMarkdown") -> None:
super().__init__(zmd)
self.zmd = zmd
def add_a( def add_a(
self, self,
root: Element, root: Element,
@@ -637,8 +645,8 @@ class InlineInterestingLinkProcessor(markdown.treeprocessors.Treeprocessor):
desc = desc if desc is not None else "" desc = desc if desc is not None else ""
# Update message.has_image attribute. # Update message.has_image attribute.
if "message_inline_image" in class_attr and self.md.zulip_message: if "message_inline_image" in class_attr and self.zmd.zulip_message:
self.md.zulip_message.has_image = True self.zmd.zulip_message.has_image = True
if insertion_index is not None: if insertion_index is not None:
div = Element("div") div = Element("div")
@@ -753,7 +761,7 @@ class InlineInterestingLinkProcessor(markdown.treeprocessors.Treeprocessor):
return url return url
def is_image(self, url: str) -> bool: def is_image(self, url: str) -> bool:
if not self.md.image_preview_enabled: if not self.zmd.image_preview_enabled:
return False return False
parsed_url = urllib.parse.urlparse(url) parsed_url = urllib.parse.urlparse(url)
# remove HTML URLs which end with image extensions that can not be shorted # remove HTML URLs which end with image extensions that can not be shorted
@@ -828,7 +836,7 @@ class InlineInterestingLinkProcessor(markdown.treeprocessors.Treeprocessor):
return None return None
def youtube_id(self, url: str) -> Optional[str]: def youtube_id(self, url: str) -> Optional[str]:
if not self.md.image_preview_enabled: if not self.zmd.image_preview_enabled:
return None return None
# YouTube video id extraction regular expression from https://pastebin.com/KyKAFv1s # YouTube video id extraction regular expression from https://pastebin.com/KyKAFv1s
# Slightly modified to support URLs of the forms # Slightly modified to support URLs of the forms
@@ -866,7 +874,7 @@ class InlineInterestingLinkProcessor(markdown.treeprocessors.Treeprocessor):
return None return None
def vimeo_id(self, url: str) -> Optional[str]: def vimeo_id(self, url: str) -> Optional[str]:
if not self.md.image_preview_enabled: if not self.zmd.image_preview_enabled:
return None return None
# (http|https)?:\/\/(www\.)?vimeo.com\/(?:channels\/(?:\w+\/)?|groups\/([^\/]*)\/videos\/|)(\d+)(?:|\/\?) # (http|https)?:\/\/(www\.)?vimeo.com\/(?:channels\/(?:\w+\/)?|groups\/([^\/]*)\/videos\/|)(\d+)(?:|\/\?)
# If it matches, match.group('id') is the video id. # If it matches, match.group('id') is the video id.
@@ -985,7 +993,7 @@ class InlineInterestingLinkProcessor(markdown.treeprocessors.Treeprocessor):
else: else:
current_node.tail = text current_node.tail = text
db_data: Optional[DbData] = self.md.zulip_db_data db_data: Optional[DbData] = self.zmd.zulip_db_data
current_index = 0 current_index = 0
for item in to_process: for item in to_process:
# The text we want to link starts in already linked text skip it # The text we want to link starts in already linked text skip it
@@ -1226,9 +1234,9 @@ class InlineInterestingLinkProcessor(markdown.treeprocessors.Treeprocessor):
} }
# Set has_link and similar flags whenever a message is processed by Markdown # Set has_link and similar flags whenever a message is processed by Markdown
if self.md.zulip_message: if self.zmd.zulip_message:
self.md.zulip_message.has_link = len(found_urls) > 0 self.zmd.zulip_message.has_link = len(found_urls) > 0
self.md.zulip_message.has_image = False # This is updated in self.add_a self.zmd.zulip_message.has_image = False # This is updated in self.add_a
for url in unique_urls: for url in unique_urls:
# Due to rewrite_local_links_to_relative, we need to # Due to rewrite_local_links_to_relative, we need to
@@ -1239,14 +1247,16 @@ class InlineInterestingLinkProcessor(markdown.treeprocessors.Treeprocessor):
parsed_url = urllib.parse.urlsplit(urllib.parse.urljoin("/", url)) parsed_url = urllib.parse.urlsplit(urllib.parse.urljoin("/", url))
host = parsed_url.netloc host = parsed_url.netloc
if host != "" and (self.md.zulip_realm is None or host != self.md.zulip_realm.host): if host != "" and (
self.zmd.zulip_realm is None or host != self.zmd.zulip_realm.host
):
continue continue
if not parsed_url.path.startswith("/user_uploads/"): if not parsed_url.path.startswith("/user_uploads/"):
continue continue
path_id = parsed_url.path[len("/user_uploads/") :] path_id = parsed_url.path[len("/user_uploads/") :]
self.md.zulip_rendering_result.potential_attachment_path_ids.append(path_id) self.zmd.zulip_rendering_result.potential_attachment_path_ids.append(path_id)
if len(found_urls) == 0: if len(found_urls) == 0:
return return
@@ -1295,7 +1305,7 @@ class InlineInterestingLinkProcessor(markdown.treeprocessors.Treeprocessor):
netloc = urlsplit(url).netloc netloc = urlsplit(url).netloc
if netloc == "" or ( if netloc == "" or (
self.md.zulip_realm is not None and netloc == self.md.zulip_realm.host self.zmd.zulip_realm is not None and netloc == self.zmd.zulip_realm.host
): ):
# We don't have a strong use case for doing URL preview for relative links. # We don't have a strong use case for doing URL preview for relative links.
continue continue
@@ -1320,20 +1330,20 @@ class InlineInterestingLinkProcessor(markdown.treeprocessors.Treeprocessor):
# is enabled, but URL previews are a beta feature and YouTube # is enabled, but URL previews are a beta feature and YouTube
# previews are pretty stable. # previews are pretty stable.
db_data: Optional[DbData] = self.md.zulip_db_data db_data: Optional[DbData] = self.zmd.zulip_db_data
if db_data and db_data.sent_by_bot: if db_data and db_data.sent_by_bot:
continue continue
if not self.md.url_embed_preview_enabled: if not self.zmd.url_embed_preview_enabled:
continue continue
if self.md.url_embed_data is None or url not in self.md.url_embed_data: if self.zmd.url_embed_data is None or url not in self.zmd.url_embed_data:
self.md.zulip_rendering_result.links_for_preview.add(url) self.zmd.zulip_rendering_result.links_for_preview.add(url)
continue continue
# Existing but being None means that we did process the # Existing but being None means that we did process the
# URL, but it was not valid to preview. # URL, but it was not valid to preview.
extracted_data = self.md.url_embed_data[url] extracted_data = self.zmd.url_embed_data[url]
if extracted_data is None: if extracted_data is None:
continue continue
@@ -1356,12 +1366,13 @@ class InlineInterestingLinkProcessor(markdown.treeprocessors.Treeprocessor):
class CompiledInlineProcessor(markdown.inlinepatterns.InlineProcessor): class CompiledInlineProcessor(markdown.inlinepatterns.InlineProcessor):
def __init__(self, compiled_re: Pattern[str], md: markdown.Markdown) -> None: def __init__(self, compiled_re: Pattern[str], zmd: "ZulipMarkdown") -> None:
# This is similar to the superclass's small __init__ function, # This is similar to the superclass's small __init__ function,
# but we skip the compilation step and let the caller give us # but we skip the compilation step and let the caller give us
# a compiled regex. # a compiled regex.
self.compiled_re = compiled_re self.compiled_re = compiled_re
self.md = md self.md = zmd
self.zmd = zmd
class Timestamp(markdown.inlinepatterns.Pattern): class Timestamp(markdown.inlinepatterns.Pattern):
@@ -1476,8 +1487,12 @@ def unicode_emoji_to_codepoint(unicode_emoji: str) -> str:
class EmoticonTranslation(markdown.inlinepatterns.Pattern): class EmoticonTranslation(markdown.inlinepatterns.Pattern):
"""Translates emoticons like `:)` into emoji like `:smile:`.""" """Translates emoticons like `:)` into emoji like `:smile:`."""
def __init__(self, pattern: str, zmd: "ZulipMarkdown") -> None:
super().__init__(pattern, zmd)
self.zmd = zmd
def handleMatch(self, match: Match[str]) -> Optional[Element]: def handleMatch(self, match: Match[str]) -> Optional[Element]:
db_data: Optional[DbData] = self.md.zulip_db_data db_data: Optional[DbData] = self.zmd.zulip_db_data
if db_data is None or not db_data.translate_emoticons: if db_data is None or not db_data.translate_emoticons:
return None return None
@@ -1499,12 +1514,16 @@ class UnicodeEmoji(markdown.inlinepatterns.Pattern):
class Emoji(markdown.inlinepatterns.Pattern): class Emoji(markdown.inlinepatterns.Pattern):
def __init__(self, pattern: str, zmd: "ZulipMarkdown") -> None:
super().__init__(pattern, zmd)
self.zmd = zmd
def handleMatch(self, match: Match[str]) -> Optional[Union[str, Element]]: def handleMatch(self, match: Match[str]) -> Optional[Union[str, Element]]:
orig_syntax = match.group("syntax") orig_syntax = match.group("syntax")
name = orig_syntax[1:-1] name = orig_syntax[1:-1]
active_realm_emoji: Dict[str, EmojiInfo] = {} active_realm_emoji: Dict[str, EmojiInfo] = {}
db_data: Optional[DbData] = self.md.zulip_db_data db_data: Optional[DbData] = self.zmd.zulip_db_data
if db_data is not None: if db_data is not None:
active_realm_emoji = db_data.active_realm_emoji active_realm_emoji = db_data.active_realm_emoji
@@ -1525,7 +1544,7 @@ def content_has_emoji_syntax(content: str) -> bool:
class Tex(markdown.inlinepatterns.Pattern): class Tex(markdown.inlinepatterns.Pattern):
def handleMatch(self, match: Match[str]) -> Element: def handleMatch(self, match: Match[str]) -> Union[str, Element]:
rendered = render_tex(match.group("body"), is_inline=True) rendered = render_tex(match.group("body"), is_inline=True)
if rendered is not None: if rendered is not None:
return self.md.htmlStash.store(rendered) return self.md.htmlStash.store(rendered)
@@ -1605,18 +1624,19 @@ def url_to_a(
class CompiledPattern(markdown.inlinepatterns.Pattern): class CompiledPattern(markdown.inlinepatterns.Pattern):
def __init__(self, compiled_re: Pattern[str], md: markdown.Markdown) -> None: def __init__(self, compiled_re: Pattern[str], zmd: "ZulipMarkdown") -> None:
# This is similar to the superclass's small __init__ function, # This is similar to the superclass's small __init__ function,
# but we skip the compilation step and let the caller give us # but we skip the compilation step and let the caller give us
# a compiled regex. # a compiled regex.
self.compiled_re = compiled_re self.compiled_re = compiled_re
self.md = md self.md = zmd
self.zmd = zmd
class AutoLink(CompiledPattern): class AutoLink(CompiledPattern):
def handleMatch(self, match: Match[str]) -> ElementStringNone: def handleMatch(self, match: Match[str]) -> ElementStringNone:
url = match.group("url") url = match.group("url")
db_data: Optional[DbData] = self.md.zulip_db_data db_data: Optional[DbData] = self.zmd.zulip_db_data
return url_to_a(db_data, url) return url_to_a(db_data, url)
@@ -1790,7 +1810,7 @@ class LinkifierPattern(CompiledInlineProcessor):
self, self,
source_pattern: str, source_pattern: str,
format_string: str, format_string: str,
md: markdown.Markdown, zmd: "ZulipMarkdown",
) -> None: ) -> None:
# Do not write errors to stderr (this still raises exceptions) # Do not write errors to stderr (this still raises exceptions)
options = re2.Options() options = re2.Options()
@@ -1813,12 +1833,12 @@ class LinkifierPattern(CompiledInlineProcessor):
r"(?<!%)(%%)*%([a-fA-F0-9][a-fA-F0-9])", r"\1%%\2", format_string r"(?<!%)(%%)*%([a-fA-F0-9][a-fA-F0-9])", r"\1%%\2", format_string
) )
super().__init__(compiled_re2, md) super().__init__(compiled_re2, zmd)
def handleMatch( # type: ignore[override] # supertype incompatible with supersupertype def handleMatch( # type: ignore[override] # supertype incompatible with supersupertype
self, m: Match[str], data: str self, m: Match[str], data: str
) -> Union[Tuple[Element, int, int], Tuple[None, None, None]]: ) -> Union[Tuple[Element, int, int], Tuple[None, None, None]]:
db_data: Optional[DbData] = self.md.zulip_db_data db_data: Optional[DbData] = self.zmd.zulip_db_data
url = url_to_a( url = url_to_a(
db_data, db_data,
self.format_string % m.groupdict(), self.format_string % m.groupdict(),
@@ -1840,7 +1860,7 @@ class UserMentionPattern(CompiledInlineProcessor):
) -> Union[Tuple[None, None, None], Tuple[Element, int, int]]: ) -> Union[Tuple[None, None, None], Tuple[Element, int, int]]:
name = m.group("match") name = m.group("match")
silent = m.group("silent") == "_" silent = m.group("silent") == "_"
db_data: Optional[DbData] = self.md.zulip_db_data db_data: Optional[DbData] = self.zmd.zulip_db_data
if db_data is not None: if db_data is not None:
wildcard = mention.user_mention_matches_wildcard(name) wildcard = mention.user_mention_matches_wildcard(name)
@@ -1864,13 +1884,13 @@ class UserMentionPattern(CompiledInlineProcessor):
if wildcard: if wildcard:
if not silent: if not silent:
self.md.zulip_rendering_result.mentions_wildcard = True self.zmd.zulip_rendering_result.mentions_wildcard = True
user_id = "*" user_id = "*"
elif user is not None: elif user is not None:
assert isinstance(user, FullNameInfo) assert isinstance(user, FullNameInfo)
if not silent: if not silent:
self.md.zulip_rendering_result.mentions_user_ids.add(user.id) self.zmd.zulip_rendering_result.mentions_user_ids.add(user.id)
name = user.full_name name = user.full_name
user_id = str(user.id) user_id = str(user.id)
else: else:
@@ -1896,13 +1916,13 @@ class UserGroupMentionPattern(CompiledInlineProcessor):
) -> Union[Tuple[None, None, None], Tuple[Element, int, int]]: ) -> Union[Tuple[None, None, None], Tuple[Element, int, int]]:
name = m.group("match") name = m.group("match")
silent = m.group("silent") == "_" silent = m.group("silent") == "_"
db_data: Optional[DbData] = self.md.zulip_db_data db_data: Optional[DbData] = self.zmd.zulip_db_data
if db_data is not None: if db_data is not None:
user_group = db_data.mention_data.get_user_group(name) user_group = db_data.mention_data.get_user_group(name)
if user_group: if user_group:
if not silent: if not silent:
self.md.zulip_rendering_result.mentions_user_group_ids.add(user_group.id) self.zmd.zulip_rendering_result.mentions_user_group_ids.add(user_group.id)
name = user_group.name name = user_group.name
user_group_id = str(user_group.id) user_group_id = str(user_group.id)
else: else:
@@ -1925,7 +1945,7 @@ class UserGroupMentionPattern(CompiledInlineProcessor):
class StreamPattern(CompiledInlineProcessor): class StreamPattern(CompiledInlineProcessor):
def find_stream_id(self, name: str) -> Optional[int]: def find_stream_id(self, name: str) -> Optional[int]:
db_data: Optional[DbData] = self.md.zulip_db_data db_data: Optional[DbData] = self.zmd.zulip_db_data
if db_data is None: if db_data is None:
return None return None
stream_id = db_data.stream_names.get(name) stream_id = db_data.stream_names.get(name)
@@ -1956,7 +1976,7 @@ class StreamPattern(CompiledInlineProcessor):
class StreamTopicPattern(CompiledInlineProcessor): class StreamTopicPattern(CompiledInlineProcessor):
def find_stream_id(self, name: str) -> Optional[int]: def find_stream_id(self, name: str) -> Optional[int]:
db_data: Optional[DbData] = self.md.zulip_db_data db_data: Optional[DbData] = self.zmd.zulip_db_data
if db_data is None: if db_data is None:
return None return None
stream_id = db_data.stream_names.get(name) stream_id = db_data.stream_names.get(name)
@@ -2010,6 +2030,10 @@ class AlertWordNotificationProcessor(markdown.preprocessors.Preprocessor):
"`", "`",
} }
def __init__(self, zmd: "ZulipMarkdown") -> None:
super().__init__(zmd)
self.zmd = zmd
def check_valid_start_position(self, content: str, index: int) -> bool: def check_valid_start_position(self, content: str, index: int) -> bool:
if index <= 0 or content[index] in self.allowed_before_punctuation: if index <= 0 or content[index] in self.allowed_before_punctuation:
return True return True
@@ -2021,14 +2045,14 @@ class AlertWordNotificationProcessor(markdown.preprocessors.Preprocessor):
return False return False
def run(self, lines: List[str]) -> List[str]: def run(self, lines: List[str]) -> List[str]:
db_data: Optional[DbData] = self.md.zulip_db_data db_data: Optional[DbData] = self.zmd.zulip_db_data
if db_data is not None: if db_data is not None:
# We check for alert words here, the set of which are # We check for alert words here, the set of which are
# dependent on which users may see this message. # dependent on which users may see this message.
# #
# Our caller passes in the list of possible_words. We # Our caller passes in the list of possible_words. We
# don't do any special rendering; we just append the alert words # don't do any special rendering; we just append the alert words
# we find to the set self.md.zulip_rendering_result.user_ids_with_alert_words. # we find to the set self.zmd.zulip_rendering_result.user_ids_with_alert_words.
realm_alert_words_automaton = db_data.realm_alert_words_automaton realm_alert_words_automaton = db_data.realm_alert_words_automaton
@@ -2040,11 +2064,15 @@ class AlertWordNotificationProcessor(markdown.preprocessors.Preprocessor):
if self.check_valid_start_position( if self.check_valid_start_position(
content, end_index - len(original_value) content, end_index - len(original_value)
) and self.check_valid_end_position(content, end_index + 1): ) and self.check_valid_end_position(content, end_index + 1):
self.md.zulip_rendering_result.user_ids_with_alert_words.update(user_ids) self.zmd.zulip_rendering_result.user_ids_with_alert_words.update(user_ids)
return lines return lines
class LinkInlineProcessor(markdown.inlinepatterns.LinkInlineProcessor): class LinkInlineProcessor(markdown.inlinepatterns.LinkInlineProcessor):
def __init__(self, pattern: str, zmd: "ZulipMarkdown") -> None:
super().__init__(pattern, zmd)
self.zmd = zmd
def zulip_specific_link_changes(self, el: Element) -> Union[None, Element]: def zulip_specific_link_changes(self, el: Element) -> Union[None, Element]:
href = el.get("href") href = el.get("href")
assert href is not None assert href is not None
@@ -2055,7 +2083,7 @@ class LinkInlineProcessor(markdown.inlinepatterns.LinkInlineProcessor):
return None # no-op; the link is not processed. return None # no-op; the link is not processed.
# Rewrite local links to be relative # Rewrite local links to be relative
db_data: Optional[DbData] = self.md.zulip_db_data db_data: Optional[DbData] = self.zmd.zulip_db_data
href = rewrite_local_links_to_relative(db_data, href) href = rewrite_local_links_to_relative(db_data, href)
# Make changes to <a> tag attributes # Make changes to <a> tag attributes