Auto merge of #13337 - epage:unicode, r=weihanglo

feat(tree): Control `--charset` via auto-detecting config value

### What does this PR try to resolve?

This tries to generalize `cargo tree --charset` so any part of cargo's output can use it.  For example,
- `cargo search` could use this for fancier tables
- `cargo add` could use this for fancier feature lists
- #12235 could use this for fancy rendering like miette does (CC `@Muscraft` )
- Progress bars could use fancier characters <-- this is what I'm personally working towards

This builds on the idea from #12889 that we can use fancier terminal features so long as we have good auto-detection and provide users an escape hatch until the auto-detection is improved or in case they disagree.

As a side benefit
- `cargo tree` will auto-detect when the prior default of `--charset utf8` is a good choice
- Users can control `cargo tree --charset` via `.cargo/config.toml`

### How should we test and review this PR?

This is gradually introduced through the individual commits.

### Additional information

Cargo.toml

@@ -204,6 +204,7 @@ unicase.workspace = true
 unicode-width.workspace = true
 url.workspace = true
 walkdir.workspace = true
+supports-unicode = "2.1.0"
 
 [target.'cfg(target_has_atomic = "64")'.dependencies]
 tracing-chrome.workspace = true

src/bin/cargo/commands/tree.rs

@@ -69,8 +69,7 @@ pub fn cli() -> Command {
         .arg(
             opt("charset", "Character set to use in output")
                 .value_name("CHARSET")
-                .value_parser(["utf8", "ascii"])
-                .default_value("utf8"),
+                .value_parser(["utf8", "ascii"]),
         )
         .arg(
             opt("format", "Format string used for printing dependencies")
@@ -101,6 +100,24 @@ pub fn cli() -> Command {
         ))
 }
 
+#[derive(Copy, Clone)]
+pub enum Charset {
+    Utf8,
+    Ascii,
+}
+
+impl FromStr for Charset {
+    type Err = &'static str;
+
+    fn from_str(s: &str) -> Result<Charset, &'static str> {
+        match s {
+            "utf8" => Ok(Charset::Utf8),
+            "ascii" => Ok(Charset::Ascii),
+            _ => Err("invalid charset"),
+        }
+    }
+}
+
 pub fn exec(gctx: &mut GlobalContext, args: &ArgMatches) -> CliResult {
     if args.flag("version") {
         let verbose = args.verbose() > 0;
@@ -181,8 +198,17 @@ subtree of the package given to -p.\n\
         print_available_packages(&ws)?;
     }
 
-    let charset = tree::Charset::from_str(args.get_one::<String>("charset").unwrap())
-        .map_err(|e| anyhow::anyhow!("{}", e))?;
+    let charset = args.get_one::<String>("charset");
+    if let Some(charset) = charset
+        .map(|c| Charset::from_str(c))
+        .transpose()
+        .map_err(|e| anyhow::anyhow!("{}", e))?
+    {
+        match charset {
+            Charset::Utf8 => gctx.shell().set_unicode(true)?,
+            Charset::Ascii => gctx.shell().set_unicode(false)?,
+        }
+    }
     let opts = tree::TreeOptions {
         cli_features: args.cli_features()?,
         packages,
@@ -193,7 +219,6 @@ subtree of the package given to -p.\n\
         prefix,
         no_dedupe,
         duplicates: args.flag("duplicates"),
-        charset,
         format: args.get_one::<String>("format").cloned().unwrap(),
         graph_features,
         max_display_depth: args.value_of_u32("depth")?.unwrap_or(u32::MAX),

src/cargo/core/shell.rs

@@ -53,6 +53,8 @@ impl Shell {
                 color_choice: auto_clr,
                 hyperlinks: supports_hyperlinks(),
                 stderr_tty: std::io::stderr().is_terminal(),
+                stdout_unicode: supports_unicode(&std::io::stdout()),
+                stderr_unicode: supports_unicode(&std::io::stderr()),
             },
             verbosity: Verbosity::Verbose,
             needs_clear: false,
@@ -230,11 +232,11 @@ impl Shell {
     /// Updates the color choice (always, never, or auto) from a string..
     pub fn set_color_choice(&mut self, color: Option<&str>) -> CargoResult<()> {
         if let ShellOut::Stream {
-            ref mut stdout,
-            ref mut stderr,
-            ref mut color_choice,
+            stdout,
+            stderr,
+            color_choice,
             ..
-        } = self.output
+        } = &mut self.output
         {
             let cfg = color
                 .map(|c| c.parse())
@@ -249,16 +251,40 @@ impl Shell {
         Ok(())
     }
 
-    pub fn set_hyperlinks(&mut self, yes: bool) -> CargoResult<()> {
+    pub fn set_unicode(&mut self, yes: bool) -> CargoResult<()> {
         if let ShellOut::Stream {
-            ref mut hyperlinks, ..
-        } = self.output
+            stdout_unicode,
+            stderr_unicode,
+            ..
+        } = &mut self.output
         {
+            *stdout_unicode = yes;
+            *stderr_unicode = yes;
+        }
+        Ok(())
+    }
+
+    pub fn set_hyperlinks(&mut self, yes: bool) -> CargoResult<()> {
+        if let ShellOut::Stream { hyperlinks, .. } = &mut self.output {
             *hyperlinks = yes;
         }
         Ok(())
     }
 
+    pub fn out_unicode(&self) -> bool {
+        match &self.output {
+            ShellOut::Write(_) => true,
+            ShellOut::Stream { stdout_unicode, .. } => *stdout_unicode,
+        }
+    }
+
+    pub fn err_unicode(&self) -> bool {
+        match &self.output {
+            ShellOut::Write(_) => true,
+            ShellOut::Stream { stderr_unicode, .. } => *stderr_unicode,
+        }
+    }
+
     /// Gets the current color choice.
     ///
     /// If we are not using a color stream, this will always return `Never`, even if the color
@@ -384,6 +410,8 @@ enum ShellOut {
         stderr_tty: bool,
         color_choice: ColorChoice,
         hyperlinks: bool,
+        stdout_unicode: bool,
+        stderr_unicode: bool,
     },
 }
 
@@ -416,17 +444,17 @@ impl ShellOut {
 
     /// Gets stdout as a `io::Write`.
     fn stdout(&mut self) -> &mut dyn Write {
-        match *self {
-            ShellOut::Stream { ref mut stdout, .. } => stdout,
-            ShellOut::Write(ref mut w) => w,
+        match self {
+            ShellOut::Stream { stdout, .. } => stdout,
+            ShellOut::Write(w) => w,
         }
     }
 
     /// Gets stderr as a `io::Write`.
     fn stderr(&mut self) -> &mut dyn Write {
-        match *self {
-            ShellOut::Stream { ref mut stderr, .. } => stderr,
-            ShellOut::Write(ref mut w) => w,
+        match self {
+            ShellOut::Stream { stderr, .. } => stderr,
+            ShellOut::Write(w) => w,
         }
     }
 }
@@ -519,6 +547,10 @@ fn supports_color(choice: anstream::ColorChoice) -> bool {
     }
 }
 
+fn supports_unicode(stream: &dyn IsTerminal) -> bool {
+    !stream.is_terminal() || supports_unicode::supports_unicode()
+}
+
 fn supports_hyperlinks() -> bool {
     #[allow(clippy::disallowed_methods)] // We are reading the state of the system, not config
     if std::env::var_os("TERM_PROGRAM").as_deref() == Some(std::ffi::OsStr::new("iTerm.app")) {

src/cargo/ops/tree/mod.rs

@@ -39,8 +39,6 @@ pub struct TreeOptions {
     /// appear with different versions, and report if any where found. Implies
     /// `invert`.
     pub duplicates: bool,
-    /// The style of characters to use.
-    pub charset: Charset,
     /// A format string indicating how each package should be displayed.
     pub format: String,
     /// Includes features in the tree as separate nodes.
@@ -68,23 +66,6 @@ impl Target {
     }
 }
 
-pub enum Charset {
-    Utf8,
-    Ascii,
-}
-
-impl FromStr for Charset {
-    type Err = &'static str;
-
-    fn from_str(s: &str) -> Result<Charset, &'static str> {
-        match s {
-            "utf8" => Ok(Charset::Utf8),
-            "ascii" => Ok(Charset::Ascii),
-            _ => Err("invalid charset"),
-        }
-    }
-}
-
 #[derive(Clone, Copy)]
 pub enum Prefix {
     None,
@@ -236,9 +217,10 @@ fn print(
     let format = Pattern::new(&opts.format)
         .with_context(|| format!("tree format `{}` not valid", opts.format))?;
 
-    let symbols = match opts.charset {
-        Charset::Utf8 => &UTF8_SYMBOLS,
-        Charset::Ascii => &ASCII_SYMBOLS,
+    let symbols = if gctx.shell().out_unicode() {
+        &UTF8_SYMBOLS
+    } else {
+        &ASCII_SYMBOLS
     };
 
     // The visited deps is used to display a (*) whenever a dep has

src/cargo/util/context/mod.rs

@@ -1049,6 +1049,9 @@ impl GlobalContext {
         if let Some(hyperlinks) = term.hyperlinks {
             self.shell().set_hyperlinks(hyperlinks)?;
         }
+        if let Some(unicode) = term.unicode {
+            self.shell().set_unicode(unicode)?;
+        }
 
         self.progress_config = term.progress.unwrap_or_default();
 
@@ -2646,6 +2649,7 @@ pub struct TermConfig {
     pub quiet: Option<bool>,
     pub color: Option<String>,
     pub hyperlinks: Option<bool>,
+    pub unicode: Option<bool>,
     #[serde(default)]
     #[serde(deserialize_with = "progress_or_string")]
     pub progress: Option<ProgressConfig>,

src/doc/man/cargo-tree.md

@@ -150,7 +150,7 @@ The default is the host platform. Use the value `all` to include *all* targets.
 
 {{#option "`--charset` _charset_" }}
 Chooses the character set to use for the tree. Valid values are "utf8" or
-"ascii". Default is "utf8".
+"ascii". When unspecified, cargo will auto-select a value.
 {{/option}}
 
 {{#option "`-f` _format_" "`--format` _format_" }}

src/doc/man/generated_txt/cargo-tree.txt

@@ -141,7 +141,8 @@ OPTIONS
    Tree Formatting Options
        --charset charset
            Chooses the character set to use for the tree. Valid values are
-           “utf8” or “ascii”. Default is “utf8”.
+           “utf8” or “ascii”. When unspecified, cargo will auto-select
+           a value.
 
        -f format, --format format
            Set the format string for each package. The default is “{p}”.

src/doc/src/commands/cargo-tree.md

@@ -146,7 +146,7 @@ The default is the host platform. Use the value <code>all</code> to include <em>
 
 <dt class="option-term" id="option-cargo-tree---charset"><a class="option-anchor" href="#option-cargo-tree---charset"></a><code>--charset</code> <em>charset</em></dt>
 <dd class="option-desc">Chooses the character set to use for the tree. Valid values are “utf8” or
-“ascii”. Default is “utf8”.</dd>
+“ascii”. When unspecified, cargo will auto-select a value.</dd>
 
 
 <dt class="option-term" id="option-cargo-tree--f"><a class="option-anchor" href="#option-cargo-tree--f"></a><code>-f</code> <em>format</em></dt>

src/doc/src/reference/config.md

@@ -182,6 +182,7 @@ quiet = false          # whether cargo output is quiet
 verbose = false        # whether cargo provides verbose output
 color = 'auto'         # whether cargo colorizes output
 hyperlinks = true      # whether cargo inserts links into output
+unicode = true         # whether cargo can render output using non-ASCII unicode characters
 progress.when = 'auto' # whether cargo shows progress bar
 progress.width = 80    # width of progress bar
 ```
@@ -1298,6 +1299,13 @@ Can be overridden with the `--color` command-line option.
 
 Controls whether or not hyperlinks are used in the terminal.
 
+#### `term.unicode`
+* Type: bool
+* Default: auto-detect
+* Environment: `CARGO_TERM_UNICODE`
+
+Control whether output can be rendered using non-ASCII unicode characters.
+
 #### `term.progress.when`
 * Type: string
 * Default: "auto"

src/etc/man/cargo-tree.1

@@ -175,7 +175,7 @@ The default is the host platform. Use the value \fBall\fR to include \fIall\fR t
 \fB\-\-charset\fR \fIcharset\fR
 .RS 4
 Chooses the character set to use for the tree. Valid values are \[lq]utf8\[rq] or
-\[lq]ascii\[rq]\&. Default is \[lq]utf8\[rq]\&.
+\[lq]ascii\[rq]\&. When unspecified, cargo will auto\-select a value.
 .RE
 .sp
 \fB\-f\fR \fIformat\fR,