gbranden pushed a commit to branch master
in repository groff.
commit c8edb82517bbeb1e13b957bb985c5b7c80c61dd5
Author: G. Branden Robinson <[email protected]>
AuthorDate: Sat Aug 17 18:00:23 2024 -0500
[troff]: Fix buglet in escs with newline delims.
* src/roff/troff/input.cpp (do_name_test): Fix buglet in handling of
newline as delimiter for the `\A`, `\w`, and `\Z` escape sequences.
An invalid C++ conditional resulted in a newline being interpreted
both as the delimiter and as an input token, spuriously putting a
space on the output. Drop subsequent tautologous conditional: `tok !=
start_token` will always evaluate true because `tok.is_eof()` is true
and `start_token` cannot be an EOF.
Before:
$ printf 'foo\A\naz\nbar\n' | groff -a
<beginning of page>
foo1 bar
After:
$ printf 'foo\A\naz\nbar\n' | ./build/test-groff -a
<beginning of page>
foo1bar
* src/roff/groff/tests/some_escapes_accept_newline_delimiters.sh: Update
test expectations.
---
ChangeLog | 27 ++++++++++++++++++++++
.../some_escapes_accept_newline_delimiters.sh | 26 +++++++++------------
src/roff/troff/input.cpp | 26 +++++++++------------
3 files changed, 49 insertions(+), 30 deletions(-)
diff --git a/ChangeLog b/ChangeLog
index 2737c68e1..8806ebc8d 100644
--- a/ChangeLog
+++ b/ChangeLog
@@ -1,3 +1,30 @@
+2024-08-17 G. Branden Robinson <[email protected]>
+
+ * src/roff/troff/input.cpp (do_name_test): Fix buglet in
+ handling of newline as delimiter for the `\A`, `\w`, and `\Z`
+ escape sequences. An invalid C++ conditional resulted
+ in a newline being interpreted both as the delimiter and as an
+ input token, spuriously putting a space on the output. Drop
+ subsequent tautologous conditional: `tok != start_token` will
+ always evaluate true because `tok.is_eof()` is true and
+ `start_token` cannot be an EOF.
+
+ Before:
+
+ $ printf 'foo\A\naz\nbar\n' | groff -a
+ <beginning of page>
+ foo1 bar
+
+ After:
+
+ $ printf 'foo\A\naz\nbar\n' | ./build/test-groff -a
+ <beginning of page>
+ foo1bar
+
+ * src/roff/groff/tests/\
+ some_escapes_accept_newline_delimiters.sh: Update test
+ expectations.
+
2024-08-17 G. Branden Robinson <[email protected]>
* src/roff/troff/input.cpp (interpolate_register): Add assertion
diff --git a/src/roff/groff/tests/some_escapes_accept_newline_delimiters.sh
b/src/roff/groff/tests/some_escapes_accept_newline_delimiters.sh
index 595e486d9..ccc0e826e 100755
--- a/src/roff/groff/tests/some_escapes_accept_newline_delimiters.sh
+++ b/src/roff/groff/tests/some_escapes_accept_newline_delimiters.sh
@@ -43,7 +43,7 @@ test -z "$error" || wail
echo "checking correct handling of newline delimiter to 'A' escape" >&2
output=$(printf "%s\n" "$input" | "$groff" -Tascii -ww)
-test "$output" = "1 D" || wail
+test "$output" = "1D" || wail
input=".sp
\b
@@ -86,7 +86,7 @@ test -z "$error" || wail
echo "checking correct handling of newline delimiter to 'w' escape" >&2
output=$(printf "%s\n" "$input" | "$groff" -Tascii -ww)
-test "$output" = "72 D" || wail
+test "$output" = "72D" || wail
input="\X
tty: link http://example.com
@@ -101,27 +101,23 @@ echo "checking correct handling of newline delimiter to
'X' escape" >&2
output=$(printf "%s\n" "$input" | "$groff" -Tascii -ww -P -c)
test "$output" = ' D' || wail
-input="\Z
-ABC
-D
-.pl 1v>?\n(nlu"
+input='.
+.ec @
+A@Z
+BCD
+E
+.pl 1v>?@n(nlu
+.'
echo "checking that newline is accepted as delimiter to 'Z' escape" >&2
error=$(printf "%s\n" "$input" | "$groff" -Tascii -ww -z 2>&1)
test -z "$error" || wail
-# This looks really weird but is consistent. A newline used as a
-# delimiter still gets interpreted as an input line ending. What we see
-# here is: 'ABC' is formatted, the drawing position is reset to the
-# beginning of the line, a word space (from filling, overstriking 'A')
-# goes on the output, followed by 'D', so it appears as 'ADC'.
-#
-# `printf '\\Z@ABC@\nD\n'` produces the same output.
echo "checking correct handling of newline delimiter to 'Z' escape" >&2
output=$(printf "%s\n" "$input" | "$groff" -Tascii -ww \
| LC_ALL=C od -t c)
-printf "%s\n" "$output" | grep -Eqx '0000000 +A +B +\\b +D +C +\\n *' \
- || wail
+printf "%s\n" "$output" \
+ | grep -Eqx '0000000 +A +B +\\b +E +C +D +\\n *' || wail
test -z "$fail"
diff --git a/src/roff/troff/input.cpp b/src/roff/troff/input.cpp
index aa9c8ea29..84b87a60d 100644
--- a/src/roff/troff/input.cpp
+++ b/src/roff/troff/input.cpp
@@ -1647,11 +1647,10 @@ static int do_name_test()
bool got_some_char = false;
for (;;) {
tok.next();
- if (tok.is_newline() || tok.is_eof()) {
- if (tok != start_token)
- warning(WARN_DELIM, "missing closing delimiter in identifier"
- " validation escape sequence (got %1)",
- tok.description());
+ if (tok.is_eof()) {
+ warning(WARN_DELIM, "missing closing delimiter in identifier"
+ " validation escape sequence (got %1)",
+ tok.description());
// Synthesize an input line ending.
input_stack::push(make_temp_iterator("\n"));
break;
@@ -1746,11 +1745,9 @@ static node *do_zero_width()
start_token.next();
for (;;) {
tok.next();
- if (tok.is_newline() || tok.is_eof()) {
- if (tok != start_token)
- warning(WARN_DELIM, "missing closing delimiter in"
- " zero-width escape sequence (got %1)",
- tok.description());
+ if (tok.is_eof()) {
+ warning(WARN_DELIM, "missing closing delimiter in zero-width"
+ " escape sequence (got %1)", tok.description());
// Synthesize an input line ending.
input_stack::push(make_temp_iterator("\n"));
break;
@@ -5458,11 +5455,10 @@ static void do_width()
curenv = &env;
for (;;) {
tok.next();
- if (tok.is_newline() || tok.is_eof()) {
- if (tok != start_token)
- warning(WARN_DELIM, "missing closing delimiter in"
- " width computation escape sequence (got %1)",
- tok.description());
+ if (tok.is_eof()) {
+ warning(WARN_DELIM, "missing closing delimiter in width"
+ " computation escape sequence (got %1)",
+ tok.description());
// Synthesize an input line ending.
input_stack::push(make_temp_iterator("\n"));
break;
_______________________________________________
Groff-commit mailing list
[email protected]
https://lists.gnu.org/mailman/listinfo/groff-commit
