open urls with goto_file command (#5820)

* feat(commands): open urls with goto_file command

Add capability for `goto_file` command to open an URL under cursor.

Fixes: https://github.com/helix-editor/helix/issues/1472
Superseds: https://github.com/helix-editor/helix/pull/4398

* open files inside helix

* address code review

* bump deps

* fix based on code review comments

4 files changed, 114 insertions, 6 deletions

diff --git a/Cargo.lock b/Cargo.lock
index fda89964..544a13bb 100644
--- a/Cargo.lock
+++ b/Cargo.lock
@@ -1237,6 +1237,7 @@ dependencies = [
  "log",
  "nucleo",
  "once_cell",
+ "open",
  "pulldown-cmark",
  "serde",
  "serde_json",
@@ -1247,6 +1248,7 @@ dependencies = [
  "tokio",
  "tokio-stream",
  "toml",
+ "url",
  "which",
 ]
 
@@ -1411,6 +1413,25 @@ source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "1e186cfbae8084e513daff4240b4797e342f988cecda4fb6c939150f96315fd8"
 
 [[package]]
+name = "is-docker"
+version = "0.2.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "928bae27f42bc99b60d9ac7334e3a21d10ad8f1835a4e12ec3ec0464765ed1b3"
+dependencies = [
+ "once_cell",
+]
+
+[[package]]
+name = "is-wsl"
+version = "0.4.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "173609498df190136aa7dea1a91db051746d339e18476eed5ca40521f02d7aa5"
+dependencies = [
+ "is-docker",
+ "once_cell",
+]
+
+[[package]]
 name = "itoa"
 version = "1.0.6"
 source = "registry+https://github.com/rust-lang/crates.io-index"
@@ -1611,6 +1632,17 @@ source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "dd8b5dd2ae5ed71462c540258bedcb51965123ad7e7ccf4b9a8cafaa4a63576d"
 
 [[package]]
+name = "open"
+version = "5.0.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "cfabf1927dce4d6fdf563d63328a0a506101ced3ec780ca2135747336c98cef8"
+dependencies = [
+ "is-wsl",
+ "libc",
+ "pathdiff",
+]
+
+[[package]]
 name = "option-ext"
 version = "0.2.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
@@ -1640,6 +1672,12 @@ dependencies = [
 ]
 
 [[package]]
+name = "pathdiff"
+version = "0.2.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "8835116a5c179084a830efb3adc117ab007512b535bc1a21c991d3b32a6b44dd"
+
+[[package]]
 name = "percent-encoding"
 version = "2.3.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
diff --git a/helix-term/Cargo.toml b/helix-term/Cargo.toml
index c6374de5..378f25f8 100644
--- a/helix-term/Cargo.toml
+++ b/helix-term/Cargo.toml
@@ -57,6 +57,10 @@ pulldown-cmark = { version = "0.9", default-features = false }
 # file type detection
 content_inspector = "0.2.4"
 
+# opening URLs
+open = "5.0.0"
+url = "2.4.1"
+
 # config
 toml = "0.7"
 
diff --git a/helix-term/src/commands.rs b/helix-term/src/commands.rs
index f1d1df06..bc54abaa 100644
--- a/helix-term/src/commands.rs
+++ b/helix-term/src/commands.rs
@@ -60,8 +60,13 @@ use crate::{
 
 use crate::job::{self, Jobs};
 use futures_util::{stream::FuturesUnordered, TryStreamExt};
-use std::{collections::HashMap, fmt, future::Future};
-use std::{collections::HashSet, num::NonZeroUsize};
+use std::{
+    collections::{HashMap, HashSet},
+    fmt,
+    future::Future,
+    io::Read,
+    num::NonZeroUsize,
+};
 
 use std::{
     borrow::Cow,
@@ -70,6 +75,7 @@ use std::{
 
 use once_cell::sync::Lazy;
 use serde::de::{self, Deserialize, Deserializer};
+use url::Url;
 
 use grep_regex::RegexMatcherBuilder;
 use grep_searcher::{sinks, BinaryDetection, SearcherBuilder};
@@ -331,7 +337,7 @@ impl MappableCommand {
         goto_implementation, "Goto implementation",
         goto_file_start, "Goto line number <n> else file start",
         goto_file_end, "Goto file end",
-        goto_file, "Goto files in selection",
+        goto_file, "Goto files/URLs in selection",
         goto_file_hsplit, "Goto files in selection (hsplit)",
         goto_file_vsplit, "Goto files in selection (vsplit)",
         goto_reference, "Goto references",
@@ -1190,10 +1196,53 @@ fn goto_file_impl(cx: &mut Context, action: Action) {
                 .to_string(),
         );
     }
+
     for sel in paths {
         let p = sel.trim();
-        if !p.is_empty() {
-            let path = &rel_path.join(p);
+        if p.is_empty() {
+            continue;
+        }
+
+        if let Ok(url) = Url::parse(p) {
+            return open_url(cx, url, action);
+        }
+
+        let path = &rel_path.join(p);
+        if path.is_dir() {
+            let picker = ui::file_picker(path.into(), &cx.editor.config());
+            cx.push_layer(Box::new(overlaid(picker)));
+        } else if let Err(e) = cx.editor.open(path, action) {
+            cx.editor.set_error(format!("Open file failed: {:?}", e));
+        }
+    }
+}
+
+/// Opens the given url. If the URL points to a valid textual file it is open in helix.
+//  Otherwise, the file is open using external program.
+fn open_url(cx: &mut Context, url: Url, action: Action) {
+    let doc = doc!(cx.editor);
+    let rel_path = doc
+        .relative_path()
+        .map(|path| path.parent().unwrap().to_path_buf())
+        .unwrap_or_default();
+
+    if url.scheme() != "file" {
+        return open_external_url(cx, url);
+    }
+
+    let content_type = std::fs::File::open(url.path()).and_then(|file| {
+        // Read up to 1kb to detect the content type
+        let mut read_buffer = Vec::new();
+        let n = file.take(1024).read_to_end(&mut read_buffer)?;
+        Ok(content_inspector::inspect(&read_buffer[..n]))
+    });
+
+    // we attempt to open binary files - files that can't be open in helix - using external
+    // program as well, e.g. pdf files or images
+    match content_type {
+        Ok(content_inspector::ContentType::BINARY) => open_external_url(cx, url),
+        Ok(_) | Err(_) => {
+            let path = &rel_path.join(url.path());
             if path.is_dir() {
                 let picker = ui::file_picker(path.into(), &cx.editor.config());
                 cx.push_layer(Box::new(overlaid(picker)));
@@ -1204,6 +1253,23 @@ fn goto_file_impl(cx: &mut Context, action: Action) {
     }
 }
 
+/// Opens URL in external program.
+fn open_external_url(cx: &mut Context, url: Url) {
+    let commands = open::commands(url.as_str());
+    cx.jobs.callback(async {
+        for cmd in commands {
+            let mut command = tokio::process::Command::new(cmd.get_program());
+            command.args(cmd.get_args());
+            if command.output().await.is_ok() {
+                return Ok(job::Callback::Editor(Box::new(|_| {})));
+            }
+        }
+        Ok(job::Callback::Editor(Box::new(move |editor| {
+            editor.set_error("Opening URL in external program failed")
+        })))
+    });
+}
+
 fn extend_word_impl<F>(cx: &mut Context, extend_fn: F)
 where
     F: Fn(RopeSlice, Range, usize) -> Range,
diff --git a/helix-view/Cargo.toml b/helix-view/Cargo.toml
index 6218e9e4..a5218b9a 100644
--- a/helix-view/Cargo.toml
+++ b/helix-view/Cargo.toml
@@ -26,7 +26,7 @@ helix-vcs = { version = "0.6", path = "../helix-vcs" }
 
 # Conversion traits
 once_cell = "1.18"
-url = "2"
+url = "2.4.1"
 
 arc-swap = { version = "1.6.0" }