path: root/gajim/common/styling.py

# This file is part of Gajim.
#
# Gajim is free software; you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published
# by the Free Software Foundation; version 3 only.
#
# Gajim is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with Gajim. If not, see <http://www.gnu.org/licenses/>.

from __future__ import annotations

from typing import Any
from typing import Match

import re
import string
from dataclasses import dataclass
from dataclasses import field

from gi.repository import GLib

from gajim.common import regex
from gajim.common.const import URIType
from gajim.common.helpers import parse_uri as analyze_uri
from gajim.common.helpers import validate_jid
from gajim.common.util.text import escape_iri_query

PRE = '`'
STRONG = '*'
STRIKE = '~'
EMPH = '_'

WHITESPACE = set(string.whitespace)
SPAN_DIRS = {PRE, STRONG, STRIKE, EMPH}
VALID_SPAN_START = WHITESPACE | SPAN_DIRS

PRE_RX = r'(?P<pre>^```.+?(^```$(.|\Z)))'
PRE_NESTED_RX = r'(?P<pre>^```.+?((^```$(.|\Z))|\Z))'
QUOTE_RX = r'(?P<quote>^(?=>).*?(^|\Z)(?!>))'

BLOCK_RX = re.compile(PRE_RX + '|' + QUOTE_RX, re.S | re.M)
BLOCK_NESTED_RX = re.compile(PRE_NESTED_RX + '|' + QUOTE_RX, re.S | re.M)
UNQUOTE_RX = re.compile(r'^> |^>', re.M)

URI_OR_JID_RX = re.compile(
    fr'(?P<uri>(?<![\w+.-]){regex.IRI})|(?P<jid>{regex.XMPP.jid})')

SD = 0
SD_POS = 1
MAX_QUOTE_LEVEL = 20


@dataclass
class StyleObject:
    start: int
    end: int
    text: str


@dataclass
class BaseHyperlink(StyleObject):
    name: str
    uri: str
    start_byte: int
    end_byte: int

    def get_markup_string(self) -> str:
        href = GLib.markup_escape_text(self.uri)
        text = GLib.markup_escape_text(self.text)
        return f'<a href="{href}">{text}</a>'


@dataclass
class Hyperlink(BaseHyperlink):
    name: str = field(default='uri', init=False)


@dataclass
class Address(BaseHyperlink):
    name: str = field(default='address', init=False)


@dataclass
class XMPPAddress(BaseHyperlink):
    name: str = field(default='xmppadr', init=False)


@dataclass
class MailAddress(BaseHyperlink):
    name: str = field(default='mailadr', init=False)


@dataclass
class Block(StyleObject):

    @classmethod
    def from_match(cls: Any, match: Match[str]) -> Block:
        return cls(start=match.start(),
                   end=match.end(),
                   text=match.group(cls.name))


@dataclass
class PlainBlock(Block):
    name: str = field(default='plain', init=False)
    spans: list[Span] = field(default_factory=list)
    uris: list[BaseHyperlink] = field(default_factory=list)

    @classmethod
    def from_quote_match(cls, match: Match[str]) -> PlainBlock:
        return cls(start=match.start(),
                   end=match.end(),
                   text=match.group('quote'))


@dataclass
class PreBlock(Block):
    name: str = field(default='pre', init=False)


@dataclass
class QuoteBlock(Block):
    name: str = field(default='quote', init=False)
    blocks: list[Block] = field(default_factory=list)

    def unquote(self) -> str:
        return UNQUOTE_RX.sub('', self.text)


@dataclass
class Span(StyleObject):
    start_byte: int
    end_byte: int
    name: str


@dataclass
class PlainSpan(Span):
    name: str = field(default='plain', init=False)


@dataclass
class StrongSpan(Span):
    name: str = field(default='strong', init=False)


@dataclass
class EmphasisSpan(Span):
    name: str = field(default='emphasis', init=False)


@dataclass
class PreTextSpan(Span):
    name: str = field(default='pre', init=False)


@dataclass
class StrikeSpan(Span):
    name: str = field(default='strike', init=False)


SPAN_CLS_DICT = {
    STRONG: StrongSpan,
    EMPH: EmphasisSpan,
    PRE: PreTextSpan,
    STRIKE: StrikeSpan
}


@dataclass
class ParsingResult:
    text: str
    blocks: list[Block]


def find_byte_index(text: str, index: int):
    byte_index = -1
    for index_, c in enumerate(text):
        byte_index += len(c.encode())
        if index == index_:
            return byte_index

    raise ValueError('index not in string: %s, %s' % (text, index))


def process(text: str | bytes, level: int = 0) -> ParsingResult:
    if isinstance(text, bytes):
        text = text.decode()

    blocks = _parse_blocks(text, level)
    for block in blocks:
        if isinstance(block, PlainBlock):
            offset = 0
            offset_bytes = 0
            for line in block.text.splitlines(keepends=True):
                block.spans += _parse_line(line, offset, offset_bytes)
                block.uris += _parse_uris(line, offset, offset_bytes)

                offset += len(line)
                offset_bytes += len(line.encode())

        if isinstance(block, QuoteBlock):
            result = process(block.unquote(), level=level + 1)
            block.blocks = result.blocks

    return ParsingResult(text, blocks)


def process_uris(text: str | bytes) -> list[BaseHyperlink]:
    if isinstance(text, bytes):
        text = text.decode()

    uris: list[BaseHyperlink] = []
    offset = 0
    offset_bytes = 0
    for line in text.splitlines(keepends=True):
        uris += _parse_uris(line, offset, offset_bytes)
        offset += len(line)
        offset_bytes += len(line.encode())

    return uris


def _parse_blocks(text: str, level: int) -> list[Block]:
    blocks: list[Block] = []
    text_len = len(text)
    last_end_pos = 0

    rx = BLOCK_NESTED_RX if level > 0 else BLOCK_RX

    for match in rx.finditer(text):
        if match.start() != last_end_pos:
            blocks.append(PlainBlock(
                start=last_end_pos,
                end=match.start(),
                text=text[last_end_pos:match.start()]))

        last_end_pos = match.end()
        group_dict = match.groupdict()

        if group_dict.get('quote') is not None:
            if level > MAX_QUOTE_LEVEL:
                blocks.append(PlainBlock.from_quote_match(match))
            else:
                blocks.append(QuoteBlock.from_match(match))

        if group_dict.get('pre') is not None:
            blocks.append(PreBlock.from_match(match))

    if last_end_pos != text_len:
        blocks.append(PlainBlock(
            start=last_end_pos,
            end=text_len,
            text=text[last_end_pos:]))

    return blocks


def _parse_line(line: str, offset: int, offset_bytes: int) -> list[Span]:
    index: int = 0
    length = len(line)
    stack: list[tuple[str, int]] = []
    spans: list[Span] = []

    while index < length:
        sd = line[index]
        if sd not in SPAN_DIRS:
            index += 1
            continue

        is_valid_start = _is_valid_span_start(line, index)
        is_valid_end = _is_valid_span_end(line, index)

        if is_valid_start and is_valid_end:
            # Favor end over new start, this means parsing is done non-greedy
            if sd in [open_sd for open_sd, _ in stack]:
                is_valid_start = False

        if is_valid_start:
            if sd == PRE:
                index = _handle_pre_span(line,
                                         index,
                                         offset,
                                         offset_bytes,
                                         spans)
                continue

            stack.append((sd, index))
            index += 1
            continue

        if is_valid_end:
            if sd not in [open_sd for open_sd, _ in stack]:
                index += 1
                continue

            if _is_span_empty(sd, index, stack):
                stack.pop()
                index += 1
                continue

            start_pos = _find_span_start_position(sd, stack)
            spans.append(_make_span(line,
                                    sd,
                                    start_pos,
                                    index,
                                    offset,
                                    offset_bytes))

        index += 1

    return spans


def _parse_uris(line: str,
                offset: int,
                offset_bytes: int) -> list[BaseHyperlink]:
    uris: list[BaseHyperlink] = []

    def make(start: int, end: int, is_jid: bool) -> BaseHyperlink | None:
        if line[end - 1] == ',':
            # Trim one trailing comma
            end -= 1
        if '(' in line[:start] and line[end - 1] == ')':
            # Trim one trailing closing parenthesis if the match is preceded
            # by an opening one somewhere on the line
            end -= 1
        if re.fullmatch('[^:]+:', line[start:end]):
            # URIs that consist only of a scheme are thusly excluded
            return None
        return _make_hyperlink(line,
                               start,
                               end - 1,
                               offset,
                               offset_bytes,
                               is_jid)

    for match in URI_OR_JID_RX.finditer(line):
        start, end = match.span()
        hyperlink = make(start, end, bool(match.group('jid')))
        if hyperlink:
            uris.append(hyperlink)

    return uris


def _handle_pre_span(line: str,
                     index: int,
                     offset: int,
                     offset_bytes: int,
                     spans: list[Span]) -> int:

    # Scan ahead for the end
    end = line.find(PRE, index + 1)
    if end == -1:
        return index + 1

    if end - index == 1:
        # empty span
        return index + 1

    spans.append(_make_span(line, PRE, index, end, offset, offset_bytes))
    return end + 1


def _make_span(line: str,
               sd: str,
               start: int,
               end: int,
               offset: int,
               offset_bytes: int) -> Span:

    text = line[start:end + 1]

    start_byte = find_byte_index(line, start) + offset_bytes
    end_byte = find_byte_index(line, end) + offset_bytes + 1

    start += offset
    end += offset + 1
    span_class = SPAN_CLS_DICT.get(sd)
    assert span_class is not None

    return span_class(start=start,
                      start_byte=start_byte,
                      end=end,
                      end_byte=end_byte,
                      text=text)


def _make_hyperlink(line: str,
                    start: int,
                    end: int,
                    offset: int,
                    offset_bytes: int,
                    is_jid: bool) -> BaseHyperlink | None:

    text = line[start:end + 1]

    start_byte = find_byte_index(line, start) + offset_bytes
    end_byte = find_byte_index(line, end) + offset_bytes + 1

    start += offset
    end += offset + 1

    uri = text
    if is_jid:
        cls_ = Address
        try:
            validate_jid(text)
        except ValueError:
            return None
        uri = 'about:ambiguous-address?' + escape_iri_query(text)

    else:
        uri = text
        auri = analyze_uri(uri)
        if auri.type == URIType.XMPP:
            cls_ = XMPPAddress
        elif auri.type == URIType.MAIL:
            cls_ = MailAddress
        elif auri.type == URIType.INVALID:
            return None
        else:
            cls_ = Hyperlink

    return cls_(start=start,
                start_byte=start_byte,
                end=end,
                end_byte=end_byte,
                uri=uri,
                text=text)


def _is_span_empty(sd: str, index: int, stack: list[tuple[str, int]]) -> bool:
    start_sd = stack[-1][SD]
    if start_sd != sd:
        return False

    pos = stack[-1][SD_POS]
    return pos == index - 1


def _find_span_start_position(sd: str, stack: list[tuple[str, int]]) -> int:
    while stack:
        start_sd, pos = stack.pop()
        if start_sd == sd:
            return pos

    raise ValueError('Unable to find opening span')


def _is_valid_span_start(line: str, index: int) -> bool:
    '''
    https://xmpp.org/extensions/xep-0393.html#span

    ... The opening styling directive MUST be located at the beginning
    of the line, after a whitespace character, or after a different opening
    styling directive. The opening styling directive MUST NOT be followed
    by a whitespace character ...
    '''

    try:
        char = line[index + 1]
    except IndexError:
        return False

    if char in WHITESPACE:
        return False

    if index == 0:
        return True

    char = line[index - 1]
    return char in VALID_SPAN_START


def _is_valid_span_end(line: str, index: int) -> bool:
    '''
    https://xmpp.org/extensions/xep-0393.html#span

    ... and the closing styling directive MUST NOT be preceded
    by a whitespace character ...
    '''

    if index == 0:
        return False

    char = line[index - 1]
    return char not in WHITESPACE