From 054421268fff33003b892eb7bcef1a831c6c308d Mon Sep 17 00:00:00 2001
From: Ethan P <eth-p+git@hidden.email>
Date: Sun, 16 Apr 2023 20:18:40 -0700
Subject: [PATCH] Strip OSC sequences before printing

This commit strips OSC (Operating System Command) sequences before
printing lines. Eventually when time permits, I want to add back
support for printing OSC sequences (and improve it to treat hyperlinks
like an attribute).

Until then, this should help prevent garbled output :)
---
 src/printer.rs |  8 +++++---
 src/vscreen.rs | 55 +++++++++++++++++++++++++++++++++++++++++++++++++-
 2 files changed, 59 insertions(+), 4 deletions(-)

diff --git a/src/printer.rs b/src/printer.rs
index 257cc766..45fd5336 100644
--- a/src/printer.rs
+++ b/src/printer.rs
@@ -33,7 +33,7 @@ use crate::line_range::RangeCheckResult;
 use crate::preprocessor::{expand_tabs, replace_nonprintable};
 use crate::style::StyleComponent;
 use crate::terminal::{as_terminal_escaped, to_ansi_color};
-use crate::vscreen::AnsiStyle;
+use crate::vscreen::{strip_problematic_sequences, AnsiStyle};
 use crate::wrapping::WrappingMode;
 
 pub enum OutputHandle<'a> {
@@ -581,7 +581,8 @@ impl<'a> Printer for InteractivePrinter<'a> {
             let italics = self.config.use_italic_text;
 
             for &(style, region) in &regions {
-                let ansi_iterator = AnsiCodeIterator::new(region);
+                let text = strip_problematic_sequences(region);
+                let ansi_iterator = AnsiCodeIterator::new(&text);
                 for chunk in ansi_iterator {
                     match chunk {
                         // ANSI escape passthrough.
@@ -634,7 +635,8 @@ impl<'a> Printer for InteractivePrinter<'a> {
             }
         } else {
             for &(style, region) in &regions {
-                let ansi_iterator = AnsiCodeIterator::new(region);
+                let text = strip_problematic_sequences(region);
+                let ansi_iterator = AnsiCodeIterator::new(&text);
                 for chunk in ansi_iterator {
                     match chunk {
                         // ANSI escape passthrough.
diff --git a/src/vscreen.rs b/src/vscreen.rs
index ecf2bd3c..ccd0bfe8 100644
--- a/src/vscreen.rs
+++ b/src/vscreen.rs
@@ -458,9 +458,54 @@ impl<'a> Iterator for EscapeSequenceOffsetsIterator<'a> {
     }
 }
 
+/// Strips problematic ANSI escape sequences from a string.
+///
+/// Ideally, this will be replaced with something that uses [[Attributes]] to create a table of char offsets
+/// -> absolute styles and style deltas. Something like that would let us simplify the printer (and support
+/// re-printing OSC hyperlink commands).
+pub fn strip_problematic_sequences(text: &str) -> String {
+    use EscapeSequenceOffsets::*;
+
+    let mut buffer = String::with_capacity(text.len());
+    for seq in EscapeSequenceOffsetsIterator::new(text) {
+        match seq {
+            Text { start, end } => buffer.push_str(&text[start..end]),
+            Unknown { start, end } => buffer.push_str(&text[start..end]),
+
+            NF {
+                start_sequence: start,
+                start: _,
+                end,
+            } => buffer.push_str(&text[start..end]),
+
+            CSI {
+                start_sequence: start,
+                start_parameters: _,
+                start_intermediates: _,
+                start_final_byte: _,
+                end,
+            } => buffer.push_str(&text[start..end]),
+
+            OSC {
+                start_sequence: _,
+                start_command: _,
+                start_terminator: _,
+                end: _,
+            } => {
+                // TODO(eth-p): Support re-printing hyperlinks.
+                // In the meantime, strip these.
+            }
+        }
+    }
+
+    buffer
+}
+
 #[cfg(test)]
 mod tests {
-    use crate::vscreen::{EscapeSequenceOffsets, EscapeSequenceOffsetsIterator};
+    use crate::vscreen::{
+        strip_problematic_sequences, EscapeSequenceOffsets, EscapeSequenceOffsetsIterator,
+    };
 
     #[test]
     fn test_escape_sequence_offsets_iterator_parses_text() {
@@ -683,4 +728,12 @@ mod tests {
         );
         assert_eq!(iter.next(), None);
     }
+
+    #[test]
+    fn test_strip_problematic_sequences() {
+        assert_eq!(
+            strip_problematic_sequences("text\x1B[33m\x1B]OSC\x1B\\\x1B(0"),
+            "text\x1B[33m\x1B(0"
+        );
+    }
 }