Merge pull request #2039 from drocco007/fold-backspace-and-carriage-return

Fold backspace and carriage return

src/uu/fold/src/fold.rs

@@ -156,7 +156,7 @@ fn fold_file_bytewise<T: Read>(mut file: BufReader<T>, spaces: bool, width: usiz
             let slice = {
                 let slice = &line[i..i + width];
                 if spaces && i + width < len {
-                    match slice.rfind(char::is_whitespace) {
+                    match slice.rfind(|c: char| c.is_whitespace() && c != '\r') {
                         Some(m) => &slice[..=m],
                         None => slice,
                     }
@@ -199,7 +199,6 @@ fn fold_file<T: Read>(mut file: BufReader<T>, spaces: bool, width: usize) {
     let mut line = String::new();
     let mut output = String::new();
     let mut col_count = 0;
-    let mut char_count = 0;
     let mut last_space = None;
 
     /// Print the output line, resetting the column and character counts.
@@ -216,11 +215,10 @@ fn fold_file<T: Read>(mut file: BufReader<T>, spaces: bool, width: usize) {
 
             println!("{}", &output[..consume]);
             output.replace_range(..consume, "");
-            char_count = output.len();
 
             // we know there are no tabs left in output, so each char counts
             // as 1 column
-            col_count = char_count;
+            col_count = output.len();
 
             last_space = None;
         };
@@ -245,6 +243,7 @@ fn fold_file<T: Read>(mut file: BufReader<T>, spaces: bool, width: usize) {
             }
 
             match ch {
+                '\r' => col_count = 0,
                 '\t' => {
                     let next_tab_stop = col_count + TAB_WIDTH - col_count % TAB_WIDTH;
 
@@ -253,36 +252,24 @@ fn fold_file<T: Read>(mut file: BufReader<T>, spaces: bool, width: usize) {
                     }
 
                     col_count = next_tab_stop;
-                    last_space = if spaces { Some(char_count) } else { None };
+                    last_space = if spaces { Some(output.len()) } else { None };
                 }
                 '\x08' => {
-                    // FIXME: does not match GNU's handling of backspace
                     if col_count > 0 {
                         col_count -= 1;
-                        char_count -= 1;
-                        output.truncate(char_count);
                     }
-                    continue;
-                }
-                '\r' => {
-                    // FIXME: does not match GNU's handling of carriage return
-                    output.truncate(0);
-                    col_count = 0;
-                    char_count = 0;
-                    continue;
                 }
                 _ if spaces && ch.is_whitespace() => {
-                    last_space = Some(char_count);
-                    col_count += 1
+                    last_space = Some(output.len());
+                    col_count += 1;
                 }
                 _ => col_count += 1,
             };
 
             output.push(ch);
-            char_count += 1;
         }
 
-        if col_count > 0 {
+        if !output.is_empty() {
             print!("{}", output);
             output.truncate(0);
         }

tests/by-util/test_fold.rs

@@ -248,6 +248,71 @@ fn test_fold_at_word_boundary_only_whitespace_preserve_final_newline() {
         .stdout_is("  \n  \n");
 }
 
+#[test]
+fn test_backspace_should_be_preserved() {
+    new_ucmd!().pipe_in("\x08").succeeds().stdout_is("\x08");
+}
+
+#[test]
+fn test_backspaced_char_should_be_preserved() {
+    new_ucmd!().pipe_in("x\x08").succeeds().stdout_is("x\x08");
+}
+
+#[test]
+fn test_backspace_should_decrease_column_count() {
+    new_ucmd!()
+        .arg("-w2")
+        .pipe_in("1\x08345")
+        .succeeds()
+        .stdout_is("1\x0834\n5");
+}
+
+#[test]
+fn test_backspace_should_not_decrease_column_count_past_zero() {
+    new_ucmd!()
+        .arg("-w2")
+        .pipe_in("1\x08\x083456")
+        .succeeds()
+        .stdout_is("1\x08\x0834\n56");
+}
+
+#[test]
+fn test_backspace_is_not_word_boundary() {
+    new_ucmd!()
+        .args(&["-w10", "-s"])
+        .pipe_in("foobar\x086789abcdef")
+        .succeeds()
+        .stdout_is("foobar\x086789a\nbcdef");
+}
+
+#[test]
+fn test_carriage_return_should_be_preserved() {
+    new_ucmd!().pipe_in("\r").succeeds().stdout_is("\r");
+}
+
+#[test]
+fn test_carriage_return_overwrriten_char_should_be_preserved() {
+    new_ucmd!().pipe_in("x\ry").succeeds().stdout_is("x\ry");
+}
+
+#[test]
+fn test_carriage_return_should_reset_column_count() {
+    new_ucmd!()
+        .arg("-w6")
+        .pipe_in("12345\r123456789abcdef")
+        .succeeds()
+        .stdout_is("12345\r123456\n789abc\ndef");
+}
+
+#[test]
+fn test_carriage_return_is_not_word_boundary() {
+    new_ucmd!()
+        .args(&["-w6", "-s"])
+        .pipe_in("fizz\rbuzz\rfizzbuzz")
+        .succeeds()
+        .stdout_is("fizz\rbuzz\rfizzbu\nzz");
+}
+
 //
 // bytewise tests
 
@@ -398,6 +463,77 @@ fn test_bytewise_fold_at_word_boundary_only_whitespace_preserve_final_newline()
         .stdout_is("  \n  \n");
 }
 
+#[test]
+fn test_bytewise_backspace_should_be_preserved() {
+    new_ucmd!()
+        .arg("-b")
+        .pipe_in("\x08")
+        .succeeds()
+        .stdout_is("\x08");
+}
+
+#[test]
+fn test_bytewise_backspaced_char_should_be_preserved() {
+    new_ucmd!()
+        .arg("-b")
+        .pipe_in("x\x08")
+        .succeeds()
+        .stdout_is("x\x08");
+}
+
+#[test]
+fn test_bytewise_backspace_should_not_decrease_column_count() {
+    new_ucmd!()
+        .args(&["-w2", "-b"])
+        .pipe_in("1\x08345")
+        .succeeds()
+        .stdout_is("1\x08\n34\n5");
+}
+
+#[test]
+fn test_bytewise_backspace_is_not_word_boundary() {
+    new_ucmd!()
+        .args(&["-w10", "-s", "-b"])
+        .pipe_in("foobar\x0889abcdef")
+        .succeeds()
+        .stdout_is("foobar\x0889a\nbcdef");
+}
+
+#[test]
+fn test_bytewise_carriage_return_should_be_preserved() {
+    new_ucmd!()
+        .arg("-b")
+        .pipe_in("\r")
+        .succeeds()
+        .stdout_is("\r");
+}
+
+#[test]
+fn test_bytewise_carriage_return_overwrriten_char_should_be_preserved() {
+    new_ucmd!()
+        .arg("-b")
+        .pipe_in("x\ry")
+        .succeeds()
+        .stdout_is("x\ry");
+}
+
+#[test]
+fn test_bytewise_carriage_return_should_not_reset_column_count() {
+    new_ucmd!()
+        .args(&["-w6", "-b"])
+        .pipe_in("12345\r123456789abcdef")
+        .succeeds()
+        .stdout_is("12345\r\n123456\n789abc\ndef");
+}
+
+#[test]
+fn test_bytewise_carriage_return_is_not_word_boundary() {
+    new_ucmd!()
+        .args(&["-w6", "-s", "-b"])
+        .pipe_in("fizz\rbuzz\rfizzbuzz")
+        .succeeds()
+        .stdout_is("fizz\rb\nuzz\rfi\nzzbuzz");
+}
 #[test]
 fn test_obsolete_syntax() {
     new_ucmd!()
@@ -406,4 +542,4 @@ fn test_obsolete_syntax() {
         .arg("space_separated_words.txt")
         .succeeds()
         .stdout_is("test1\n \ntest2\n \ntest3\n \ntest4\n \ntest5\n \ntest6\n ");
-}
+}