Rollup merge of #121815 - nnethercote:mv-gather_comments, r=est31

Move `gather_comments`.

To the module where it is used, so it doesn't have to be `pub`.

r? ```@est31```

4 files changed, 126 insertions, 127 deletions

diff --git a/Cargo.lock b/Cargo.lock
index 1344d4152d9..a39e468e3d9 100644
--- a/Cargo.lock
+++ b/Cargo.lock
@@ -3491,6 +3491,7 @@ version = "0.0.0"
 dependencies = [
  "itertools 0.11.0",
  "rustc_ast",
+ "rustc_lexer",
  "rustc_span",
  "thin-vec",
 ]
diff --git a/compiler/rustc_ast/src/util/comments.rs b/compiler/rustc_ast/src/util/comments.rs
index bdf5143b0f7..cbc1afc6bf1 100644
--- a/compiler/rustc_ast/src/util/comments.rs
+++ b/compiler/rustc_ast/src/util/comments.rs
@@ -1,6 +1,5 @@
 use crate::token::CommentKind;
-use rustc_span::source_map::SourceMap;
-use rustc_span::{BytePos, CharPos, FileName, Pos, Symbol};
+use rustc_span::{BytePos, Symbol};
 
 #[cfg(test)]
 mod tests;
@@ -131,126 +130,3 @@ pub fn beautify_doc_string(data: Symbol, kind: CommentKind) -> Symbol {
     }
     data
 }
-
-/// Returns `None` if the first `col` chars of `s` contain a non-whitespace char.
-/// Otherwise returns `Some(k)` where `k` is first char offset after that leading
-/// whitespace. Note that `k` may be outside bounds of `s`.
-fn all_whitespace(s: &str, col: CharPos) -> Option<usize> {
-    let mut idx = 0;
-    for (i, ch) in s.char_indices().take(col.to_usize()) {
-        if !ch.is_whitespace() {
-            return None;
-        }
-        idx = i + ch.len_utf8();
-    }
-    Some(idx)
-}
-
-fn trim_whitespace_prefix(s: &str, col: CharPos) -> &str {
-    let len = s.len();
-    match all_whitespace(s, col) {
-        Some(col) => {
-            if col < len {
-                &s[col..]
-            } else {
-                ""
-            }
-        }
-        None => s,
-    }
-}
-
-fn split_block_comment_into_lines(text: &str, col: CharPos) -> Vec<String> {
-    let mut res: Vec<String> = vec![];
-    let mut lines = text.lines();
-    // just push the first line
-    res.extend(lines.next().map(|it| it.to_string()));
-    // for other lines, strip common whitespace prefix
-    for line in lines {
-        res.push(trim_whitespace_prefix(line, col).to_string())
-    }
-    res
-}
-
-// it appears this function is called only from pprust... that's
-// probably not a good thing.
-pub fn gather_comments(sm: &SourceMap, path: FileName, src: String) -> Vec<Comment> {
-    let sm = SourceMap::new(sm.path_mapping().clone());
-    let source_file = sm.new_source_file(path, src);
-    let text = (*source_file.src.as_ref().unwrap()).clone();
-
-    let text: &str = text.as_str();
-    let start_bpos = source_file.start_pos;
-    let mut pos = 0;
-    let mut comments: Vec<Comment> = Vec::new();
-    let mut code_to_the_left = false;
-
-    if let Some(shebang_len) = rustc_lexer::strip_shebang(text) {
-        comments.push(Comment {
-            style: CommentStyle::Isolated,
-            lines: vec![text[..shebang_len].to_string()],
-            pos: start_bpos,
-        });
-        pos += shebang_len;
-    }
-
-    for token in rustc_lexer::tokenize(&text[pos..]) {
-        let token_text = &text[pos..pos + token.len as usize];
-        match token.kind {
-            rustc_lexer::TokenKind::Whitespace => {
-                if let Some(mut idx) = token_text.find('\n') {
-                    code_to_the_left = false;
-                    while let Some(next_newline) = &token_text[idx + 1..].find('\n') {
-                        idx += 1 + next_newline;
-                        comments.push(Comment {
-                            style: CommentStyle::BlankLine,
-                            lines: vec![],
-                            pos: start_bpos + BytePos((pos + idx) as u32),
-                        });
-                    }
-                }
-            }
-            rustc_lexer::TokenKind::BlockComment { doc_style, .. } => {
-                if doc_style.is_none() {
-                    let code_to_the_right = !matches!(
-                        text[pos + token.len as usize..].chars().next(),
-                        Some('\r' | '\n')
-                    );
-                    let style = match (code_to_the_left, code_to_the_right) {
-                        (_, true) => CommentStyle::Mixed,
-                        (false, false) => CommentStyle::Isolated,
-                        (true, false) => CommentStyle::Trailing,
-                    };
-
-                    // Count the number of chars since the start of the line by rescanning.
-                    let pos_in_file = start_bpos + BytePos(pos as u32);
-                    let line_begin_in_file = source_file.line_begin_pos(pos_in_file);
-                    let line_begin_pos = (line_begin_in_file - start_bpos).to_usize();
-                    let col = CharPos(text[line_begin_pos..pos].chars().count());
-
-                    let lines = split_block_comment_into_lines(token_text, col);
-                    comments.push(Comment { style, lines, pos: pos_in_file })
-                }
-            }
-            rustc_lexer::TokenKind::LineComment { doc_style } => {
-                if doc_style.is_none() {
-                    comments.push(Comment {
-                        style: if code_to_the_left {
-                            CommentStyle::Trailing
-                        } else {
-                            CommentStyle::Isolated
-                        },
-                        lines: vec![token_text.to_string()],
-                        pos: start_bpos + BytePos(pos as u32),
-                    })
-                }
-            }
-            _ => {
-                code_to_the_left = true;
-            }
-        }
-        pos += token.len as usize;
-    }
-
-    comments
-}
diff --git a/compiler/rustc_ast_pretty/Cargo.toml b/compiler/rustc_ast_pretty/Cargo.toml
index 12a08f06558..b38a2915a43 100644
--- a/compiler/rustc_ast_pretty/Cargo.toml
+++ b/compiler/rustc_ast_pretty/Cargo.toml
@@ -7,6 +7,7 @@ edition = "2021"
 # tidy-alphabetical-start
 itertools = "0.11"
 rustc_ast = { path = "../rustc_ast" }
+rustc_lexer = { path = "../rustc_lexer" }
 rustc_span = { path = "../rustc_span" }
 thin-vec = "0.2.12"
 # tidy-alphabetical-end
diff --git a/compiler/rustc_ast_pretty/src/pprust/state.rs b/compiler/rustc_ast_pretty/src/pprust/state.rs
index 6e1974f48b2..6119c6c84f8 100644
--- a/compiler/rustc_ast_pretty/src/pprust/state.rs
+++ b/compiler/rustc_ast_pretty/src/pprust/state.rs
@@ -14,7 +14,7 @@ use rustc_ast::ptr::P;
 use rustc_ast::token::{self, BinOpToken, CommentKind, Delimiter, Nonterminal, Token, TokenKind};
 use rustc_ast::tokenstream::{Spacing, TokenStream, TokenTree};
 use rustc_ast::util::classify;
-use rustc_ast::util::comments::{gather_comments, Comment, CommentStyle};
+use rustc_ast::util::comments::{Comment, CommentStyle};
 use rustc_ast::util::parser;
 use rustc_ast::{self as ast, AttrArgs, AttrArgsEq, BlockCheckMode, PatKind};
 use rustc_ast::{attr, BindingAnnotation, ByRef, DelimArgs, RangeEnd, RangeSyntax, Term};
@@ -24,7 +24,7 @@ use rustc_ast::{InlineAsmOptions, InlineAsmTemplatePiece};
 use rustc_span::edition::Edition;
 use rustc_span::source_map::{SourceMap, Spanned};
 use rustc_span::symbol::{kw, sym, Ident, IdentPrinter, Symbol};
-use rustc_span::{BytePos, FileName, Span, DUMMY_SP};
+use rustc_span::{BytePos, CharPos, FileName, Pos, Span, DUMMY_SP};
 use std::borrow::Cow;
 use thin_vec::ThinVec;
 
@@ -59,6 +59,127 @@ pub struct Comments<'a> {
     current: usize,
 }
 
+/// Returns `None` if the first `col` chars of `s` contain a non-whitespace char.
+/// Otherwise returns `Some(k)` where `k` is first char offset after that leading
+/// whitespace. Note that `k` may be outside bounds of `s`.
+fn all_whitespace(s: &str, col: CharPos) -> Option<usize> {
+    let mut idx = 0;
+    for (i, ch) in s.char_indices().take(col.to_usize()) {
+        if !ch.is_whitespace() {
+            return None;
+        }
+        idx = i + ch.len_utf8();
+    }
+    Some(idx)
+}
+
+fn trim_whitespace_prefix(s: &str, col: CharPos) -> &str {
+    let len = s.len();
+    match all_whitespace(s, col) {
+        Some(col) => {
+            if col < len {
+                &s[col..]
+            } else {
+                ""
+            }
+        }
+        None => s,
+    }
+}
+
+fn split_block_comment_into_lines(text: &str, col: CharPos) -> Vec<String> {
+    let mut res: Vec<String> = vec![];
+    let mut lines = text.lines();
+    // just push the first line
+    res.extend(lines.next().map(|it| it.to_string()));
+    // for other lines, strip common whitespace prefix
+    for line in lines {
+        res.push(trim_whitespace_prefix(line, col).to_string())
+    }
+    res
+}
+
+fn gather_comments(sm: &SourceMap, path: FileName, src: String) -> Vec<Comment> {
+    let sm = SourceMap::new(sm.path_mapping().clone());
+    let source_file = sm.new_source_file(path, src);
+    let text = (*source_file.src.as_ref().unwrap()).clone();
+
+    let text: &str = text.as_str();
+    let start_bpos = source_file.start_pos;
+    let mut pos = 0;
+    let mut comments: Vec<Comment> = Vec::new();
+    let mut code_to_the_left = false;
+
+    if let Some(shebang_len) = rustc_lexer::strip_shebang(text) {
+        comments.push(Comment {
+            style: CommentStyle::Isolated,
+            lines: vec![text[..shebang_len].to_string()],
+            pos: start_bpos,
+        });
+        pos += shebang_len;
+    }
+
+    for token in rustc_lexer::tokenize(&text[pos..]) {
+        let token_text = &text[pos..pos + token.len as usize];
+        match token.kind {
+            rustc_lexer::TokenKind::Whitespace => {
+                if let Some(mut idx) = token_text.find('\n') {
+                    code_to_the_left = false;
+                    while let Some(next_newline) = &token_text[idx + 1..].find('\n') {
+                        idx += 1 + next_newline;
+                        comments.push(Comment {
+                            style: CommentStyle::BlankLine,
+                            lines: vec![],
+                            pos: start_bpos + BytePos((pos + idx) as u32),
+                        });
+                    }
+                }
+            }
+            rustc_lexer::TokenKind::BlockComment { doc_style, .. } => {
+                if doc_style.is_none() {
+                    let code_to_the_right = !matches!(
+                        text[pos + token.len as usize..].chars().next(),
+                        Some('\r' | '\n')
+                    );
+                    let style = match (code_to_the_left, code_to_the_right) {
+                        (_, true) => CommentStyle::Mixed,
+                        (false, false) => CommentStyle::Isolated,
+                        (true, false) => CommentStyle::Trailing,
+                    };
+
+                    // Count the number of chars since the start of the line by rescanning.
+                    let pos_in_file = start_bpos + BytePos(pos as u32);
+                    let line_begin_in_file = source_file.line_begin_pos(pos_in_file);
+                    let line_begin_pos = (line_begin_in_file - start_bpos).to_usize();
+                    let col = CharPos(text[line_begin_pos..pos].chars().count());
+
+                    let lines = split_block_comment_into_lines(token_text, col);
+                    comments.push(Comment { style, lines, pos: pos_in_file })
+                }
+            }
+            rustc_lexer::TokenKind::LineComment { doc_style } => {
+                if doc_style.is_none() {
+                    comments.push(Comment {
+                        style: if code_to_the_left {
+                            CommentStyle::Trailing
+                        } else {
+                            CommentStyle::Isolated
+                        },
+                        lines: vec![token_text.to_string()],
+                        pos: start_bpos + BytePos(pos as u32),
+                    })
+                }
+            }
+            _ => {
+                code_to_the_left = true;
+            }
+        }
+        pos += token.len as usize;
+    }
+
+    comments
+}
+
 impl<'a> Comments<'a> {
     pub fn new(sm: &'a SourceMap, filename: FileName, input: String) -> Comments<'a> {
         let comments = gather_comments(sm, filename, input);