Feature: Highlight non-printable characters

Adds a new `-A`/`--show-all` option (in analogy to GNU Linux `cat`s option) that
highlights non-printable characters like space, tab or newline.

This works in two steps:
- **Preprocessing**: replace space by `•`, replace tab by `├──┤`, replace
newline by `␤`, etc.
- **Highlighting**: Use a newly written Sublime syntax to highlight
these special symbols.

Note: This feature is not technically a drop-in replacement for GNU `cat`s
`--show-all` but it has the same purpose.

assets/syntaxes/show-nonprintable.sublime-syntax

@@ -0,0 +1,25 @@
+%YAML 1.2
+---
+# http://www.sublimetext.com/docs/3/syntax.html
+name: Highlight non-printables
+file_extensions:
+    - show-nonprintable
+scope: whitespace
+contexts:
+  main:
+    - match: "•"
+      scope: support.function.show-nonprintable.space
+    - match: "├─*┤"
+      scope: constant.character.escape.show-nonprintable.tab
+    - match: "␤"
+      scope: keyword.operator.show-nonprintable.newline
+    - match: "␍"
+      scope: string.show-nonprintable.carriage-return
+    - match: "␀"
+      scope: entity.other.attribute-name.show-nonprintable.null
+    - match: "␇"
+      scope: entity.other.attribute-name.show-nonprintable.bell
+    - match: "␛"
+      scope: entity.other.attribute-name.show-nonprintable.escape
+    - match: "␈"
+      scope: entity.other.attribute-name.show-nonprintable.backspace

src/app.rs

@@ -37,6 +37,9 @@ pub struct Config<'a> {
     /// The explicitly configured language, if any
     pub language: Option<&'a str>,
 
+    /// Whether or not to show/replace non-printable characters like space, tab and newline.
+    pub show_nonprintable: bool,
+
     /// The character width of the terminal
     pub term_width: usize,
 
@@ -169,7 +172,14 @@ impl App {
 
         Ok(Config {
             true_color: is_truecolor_terminal(),
-            language: self.matches.value_of("language"),
+            language: self.matches.value_of("language").or_else(|| {
+                if self.matches.is_present("show-all") {
+                    Some("show-nonprintable")
+                } else {
+                    None
+                }
+            }),
+            show_nonprintable: self.matches.is_present("show-all"),
             output_wrap: if !self.interactive_output {
                 // We don't have the tty width when piping to another program.
                 // There's no point in wrapping when this is the case.

src/clap_app.rs

@@ -158,6 +158,18 @@ pub fn build_app(interactive_output: bool) -> ClapApp<'static, 'static> {
                      '--style=numbers'",
                 ),
         )
+        .arg(
+            Arg::with_name("show-all")
+                .long("show-all")
+                .alias("show-nonprintable")
+                .short("A")
+                .conflicts_with("language")
+                .help("Show non-printable characters (space, tab, newline, ..).")
+                .long_help(
+                    "Show non-printable characters like space, tab or newline. \
+                     Use '--tabs' to control the width of the tab-placeholders.",
+                ),
+        )
         .arg(
             Arg::with_name("line-range")
                 .long("line-range")

src/controller.rs

@@ -1,4 +1,5 @@
 use std::io::{self, Write};
+use std::mem::swap;
 
 use app::Config;
 use assets::HighlightingAssets;
@@ -6,6 +7,7 @@ use errors::*;
 use inputfile::{InputFile, InputFileReader};
 use line_range::{LineRanges, RangeCheckResult};
 use output::OutputType;
+use preprocessor::replace_nonprintable;
 use printer::{InteractivePrinter, Printer, SimplePrinter};
 
 pub struct Controller<'a> {
@@ -64,7 +66,14 @@ impl<'b> Controller<'b> {
         input_file: InputFile<'a>,
     ) -> Result<()> {
         printer.print_header(writer, input_file)?;
-        self.print_file_ranges(printer, writer, reader, &self.config.line_ranges)?;
+        self.print_file_ranges(
+            printer,
+            writer,
+            reader,
+            &self.config.line_ranges,
+            self.config.show_nonprintable,
+            self.config.tab_width,
+        )?;
         printer.print_footer(writer)?;
 
         Ok(())
@@ -76,12 +85,20 @@ impl<'b> Controller<'b> {
         writer: &mut Write,
         mut reader: InputFileReader,
         line_ranges: &LineRanges,
+        show_nonprintable: bool,
+        tab_width: usize,
     ) -> Result<()> {
         let mut line_buffer = Vec::new();
+        let mut line_buffer_processed = Vec::new();
 
         let mut line_number: usize = 1;
 
         while reader.read_line(&mut line_buffer)? {
+            if show_nonprintable {
+                replace_nonprintable(&mut line_buffer, &mut line_buffer_processed, tab_width);
+                swap(&mut line_buffer, &mut line_buffer_processed);
+            }
+
             match line_ranges.check(line_number) {
                 RangeCheckResult::OutsideRange => {
                     // Call the printer in case we need to call the syntax highlighter

src/preprocessor.rs

@@ -1,7 +1,7 @@
 use console::AnsiCodeIterator;
 
 /// Expand tabs like an ANSI-enabled expand(1).
-pub fn expand(line: &str, width: usize, cursor: &mut usize) -> String {
+pub fn expand_tabs(line: &str, width: usize, cursor: &mut usize) -> String {
     let mut buffer = String::with_capacity(line.len() * 2);
 
     for chunk in AnsiCodeIterator::new(line) {
@@ -32,3 +32,42 @@ pub fn expand(line: &str, width: usize, cursor: &mut usize) -> String {
 
     buffer
 }
+
+pub fn replace_nonprintable(input: &mut Vec<u8>, output: &mut Vec<u8>, tab_width: usize) {
+    output.clear();
+
+    let tab_width = if tab_width == 0 {
+        4
+    } else if tab_width == 1 {
+        2
+    } else {
+        tab_width
+    };
+
+    for chr in input {
+        match *chr {
+            // space
+            b' ' => output.extend_from_slice("•".as_bytes()),
+            // tab
+            b'\t' => {
+                output.extend_from_slice("├".as_bytes());
+                output.extend_from_slice("─".repeat(tab_width - 2).as_bytes());
+                output.extend_from_slice("┤".as_bytes());
+            }
+            // new line
+            b'\n' => output.extend_from_slice("␤".as_bytes()),
+            // carriage return
+            b'\r' => output.extend_from_slice("␍".as_bytes()),
+            // null
+            0x00 => output.extend_from_slice("␀".as_bytes()),
+            // bell
+            0x07 => output.extend_from_slice("␇".as_bytes()),
+            // backspace
+            0x08 => output.extend_from_slice("␈".as_bytes()),
+            // escape
+            0x1B => output.extend_from_slice("␛".as_bytes()),
+            // anything else
+            _ => output.push(*chr),
+        }
+    }
+}

src/printer.rs

@@ -22,7 +22,7 @@ use diff::get_git_diff;
 use diff::LineChanges;
 use errors::*;
 use inputfile::{InputFile, InputFileReader};
-use preprocessor::expand;
+use preprocessor::expand_tabs;
 use style::OutputWrap;
 use terminal::{as_terminal_escaped, to_ansi_color};
 
@@ -177,7 +177,7 @@ impl<'a> InteractivePrinter<'a> {
 
     fn preprocess(&self, text: &str, cursor: &mut usize) -> String {
         if self.config.tab_width > 0 {
-            expand(text, self.config.tab_width, cursor)
+            expand_tabs(text, self.config.tab_width, cursor)
         } else {
             text.to_string()
         }