1"""Join raw text tokens with the rest of the text
2
3This is set as a separate rule to provide an opportunity for plugins
4to run text replacements after text join, but before escape join.
5
6For example, `\\:)` shouldn't be replaced with an emoji.
7"""
8
9from __future__ import annotations
10
11from ..token import Token
12from .state_core import StateCore
13
14
15def text_join(state: StateCore) -> None:
16 """Join raw text for escape sequences (`text_special`) tokens with the rest of the text"""
17
18 for inline_token in state.tokens[:]:
19 if inline_token.type != "inline":
20 continue
21
22 children = inline_token.children or []
23 for child_token in children:
24 # image `alt` is parsed into its own token tree
25 if child_token.children:
26 child_token.children = _join_text_tokens(child_token.children)
27 inline_token.children = _join_text_tokens(children)
28
29
30def _join_text_tokens(tokens: list[Token]) -> list[Token]:
31 """Convert `text_special` to `text` and join all adjacent text tokens"""
32 new_tokens: list[Token] = []
33 i = 0
34 while i < len(tokens):
35 token = tokens[i]
36 if token.type == "text_special":
37 token.type = "text"
38 if token.type == "text" and new_tokens and new_tokens[-1].type == "text":
39 # Collapse a run of adjacent text nodes in a single join, instead
40 # of pairwise `a + b` concatenation. The pairwise form is O(L*k)
41 # in the size of the run because each step rebuilds the growing
42 # prefix; "".join is O(L).
43 parts = [new_tokens[-1].content, token.content]
44 i += 1
45 while i < len(tokens):
46 next_token = tokens[i]
47 if next_token.type == "text_special":
48 next_token.type = "text"
49 if next_token.type != "text":
50 break
51 parts.append(next_token.content)
52 i += 1
53 new_tokens[-1].content = "".join(parts)
54 else:
55 new_tokens.append(token)
56 i += 1
57 return new_tokens