From 95d8e7745d9b00097d50d9903d62c48268324e7a Mon Sep 17 00:00:00 2001
From: Nick Bowler <nbowler@draconx.ca>
Date: Sat, 2 Dec 2023 23:13:24 -0500
Subject: [PATCH] Rework backslash substitutions in awk scripts.

It appears that even using plain "\\" in a replacement string for sub
and gsub is still problematic.  Most awks handle this OK (returning a
literal backslash) provided the next character is not a backslash, but
on ULTRIX 4.5 nawk eats the next character regardless and interprets
it as a backslash-escape sequence, oops.

So let's try something a bit different.  We can ask awk what it does
with "\\\\", and use either "\\\\" or "\\" depending on the answer.

Hopefully this actually works everywhere.
---
 scripts/gen-options.awk | 10 +++++++---
 scripts/gen-strtab.awk  | 25 +++++++++++++++----------
 scripts/gen-tree.awk    |  6 +++++-
 3 files changed, 27 insertions(+), 14 deletions(-)

diff --git a/scripts/gen-options.awk b/scripts/gen-options.awk
index 80adcb4..c6d361c 100755
--- a/scripts/gen-options.awk
+++ b/scripts/gen-options.awk
@@ -123,6 +123,10 @@ END {
 }
 
 BEGIN {
+  # Check if "\\\\" in substitutions gives just one backslash.
+  bs = "x"; sub(/x/, "\\\\", bs);
+  bs = (length(bs) == 1 ? "\\\\" : "\\");
+
   has_actions = 0
   sopt_string = ""
   num_options = 0
@@ -238,7 +242,7 @@ END {
     lopt_strings = add_to_strtab(lopt_strings, sorted_options[i], offsets)
   }
   gsub(/[^ ]+/, "\"&", lopt_strings)
-  gsub(/ /, "\\0\"\n\t", lopt_strings)
+  gsub(/ /, bs"0\"\n\t", lopt_strings)
 
   print "static const char lopt_strings[] ="
   print "\t" lopt_strings "\";\n"
@@ -318,8 +322,8 @@ END {
       help_offsets[opt] = help_pos
       help_pos += length(help) + 1
 
-      gsub(/"/, "\\\"", help)
-      gsub(/\n/, "\\n\"\n\t    \"", help)
+      gsub(/"/, bs"\"", help)
+      gsub(/\n/, bs"n\"\n\t    \"", help)
       help = "\tPN_(\"" opt "\",\n\t    \"" help "\")"
     }
   }
diff --git a/scripts/gen-strtab.awk b/scripts/gen-strtab.awk
index 2265bb5..c5b02c6 100755
--- a/scripts/gen-strtab.awk
+++ b/scripts/gen-strtab.awk
@@ -80,6 +80,10 @@ END {
 }
 
 BEGIN {
+  # Check if "\\\\" in substitutions gives just one backslash.
+  bs = "x"; sub(/x/, "\\\\", bs);
+  bs = (length(bs) == 1 ? "\\\\" : "\\");
+
   opts["zero"] = 1
   opts["macro"] = 0
   collected = ident = ""
@@ -157,7 +161,7 @@ END {
 
   for (i = 0; i < count; i++) {
     s = sorted_strings[i]
-    gsub(/\\\\/, "\2\2", s)
+    gsub(/\\\\/, "\2", s)
     if ((n = index(strtab "\1", s "\1")) > 0) {
       offsets[sorted_strings[i]] = real_length(substr(strtab, 1, n-1));
       if (!(sorted_strings[i] in nol10n))
@@ -173,7 +177,7 @@ END {
     }
   }
 
-  gsub(/\2/, "\\", strtab);
+  gsub("\2", bs bs, strtab);
   n = split(strtab, split_strtab, "\1");
   for (i = 1; i <= n; i++) {
     printf("\t%4s ", i > !!opts["zero"] ? "\"\\0\"" : "");
@@ -206,9 +210,10 @@ END {
 # strings[ident] = val.
 function finish_string_input(strings, ident, val, n, tmpval)
 {
-  gsub(/\\\\/, "\1\1", val)
-  val = val (endline > startline ? "\n" : "")
-  gsub(/\\\n/, "", val)
+  gsub(/\\\\/, "\1", val);
+  if (endline > startline)
+    val = val "\n";
+  gsub(/\\\n/, "", val);
 
   tmpval = ""
   while ((n = match(val, /\\[^abtnvfr]/)) > 0) {
@@ -218,10 +223,10 @@ function finish_string_input(strings, ident, val, n, tmpval)
   tmpval = tmpval val
 
   # Escape special characters
-  gsub(/"/, "\\\"", tmpval)
-  gsub(/\t/, "\\t", tmpval)
-  gsub(/\n/, "\\n", tmpval)
-  gsub(/\1/, "\\", tmpval)
+  gsub(/"/,  bs"\"", tmpval)
+  gsub(/\t/, bs"t", tmpval)
+  gsub(/\n/, bs"n", tmpval)
+  gsub("\1", bs bs, tmpval)
 
   strings[ident] = tmpval
   if (!current_l10n) {
@@ -232,7 +237,7 @@ function finish_string_input(strings, ident, val, n, tmpval)
 function real_length(s, t)
 {
   t = length(s)
-  return t - gsub(/\\.|\2\2/, "&", s)
+  return t - gsub(/\\./, "&", s)
 }
 
 # bucketsort(dst, src)
diff --git a/scripts/gen-tree.awk b/scripts/gen-tree.awk
index c461e5d..55a7d5c 100755
--- a/scripts/gen-tree.awk
+++ b/scripts/gen-tree.awk
@@ -93,6 +93,10 @@ END {
 }
 
 BEGIN {
+  # Check if "\\\\" in substitutions gives just one backslash.
+  bs = "x"; sub(/x/, "\\\\", bs);
+  bs = (length(bs) == 1 ? "\\\\" : "\\");
+
   opts["strtab"] = 1;
 
   depth = max_depth = 0;
@@ -187,7 +191,7 @@ END {
       }
     }
 
-    gsub(/\1/, "\"\n\t\"\\0\" \"", entry_strtab);
+    gsub("\1", "\"\n\t\"" bs "0\" \"", entry_strtab);
     sub(/^"/, "", entry_strtab);
     sub(/\n[^\n]*$/, ";", entry_strtab);
     print "\nstatic const char tree_strtab[] =" entry_strtab
-- 
2.43.2