From 65b57a281390310f0097cbdb9d7967d97f94679c Mon Sep 17 00:00:00 2001
From: Joshua Haberman <joshua@reverberate.org>
Date: Fri, 25 Feb 2011 19:01:58 -0800
Subject: Added escaping for text output.

---
 lang_ext/lua/test.lua |  2 --
 src/upb_textprinter.c | 55 +++++++++++++++++++++++++++++++++++++++++++++++++--
 2 files changed, 53 insertions(+), 4 deletions(-)

diff --git a/lang_ext/lua/test.lua b/lang_ext/lua/test.lua
index b7f980e..d7f71df 100644
--- a/lang_ext/lua/test.lua
+++ b/lang_ext/lua/test.lua
@@ -31,8 +31,6 @@ print(msg:ToText())
 msg = SpeedMessage2()
 f = assert(io.open("../../benchmarks/google_message2.dat"))
 msg:Parse(f:read("*all"))
-msg.field2 = ""
-print(msg.field2)
 print(msg:ToText())
 --msg:Serialize()
 --msg:FromText(str)
diff --git a/src/upb_textprinter.c b/src/upb_textprinter.c
index 7e99ebd..076e1df 100644
--- a/src/upb_textprinter.c
+++ b/src/upb_textprinter.c
@@ -8,6 +8,7 @@
 
 #include <inttypes.h>
 #include <stdlib.h>
+#include <ctype.h>
 #include "upb_def.h"
 #include "upb_string.h"
 
@@ -20,6 +21,56 @@ struct _upb_textprinter {
 
 #define CHECK(x) if ((x) < 0) goto err;
 
+static int upb_textprinter_putescaped(upb_textprinter *p, upb_string *str,
+                                      bool preserve_utf8) {
+  // Based on CEscapeInternal() from Google's protobuf release.
+  // TODO; we could write directly into a bytesink's buffer instead.
+  char dstbuf[4096], *dst = dstbuf, *dstend = dstbuf + sizeof(dstbuf);
+  const char *src = upb_string_getrobuf(str), *end = src + upb_string_len(str);
+
+  // I think hex is prettier and more useful, but proto2 uses octal; should
+  // investigate whether it can parse hex also.
+  bool use_hex = false;
+  bool last_hex_escape = false; // true if last output char was \xNN
+
+  for (; src < end; src++) {
+    if (dstend - dst < 4) {
+      upb_string str = UPB_STACK_STRING_LEN(dstbuf, dst - dstbuf);
+      CHECK(upb_bytesink_putstr(p->bytesink, &str, &p->status));
+      dst = dstbuf;
+    }
+
+    bool is_hex_escape = false;
+    switch (*src) {
+      case '\n': *(dst++) = '\\'; *(dst++) = 'n';  break;
+      case '\r': *(dst++) = '\\'; *(dst++) = 'r';  break;
+      case '\t': *(dst++) = '\\'; *(dst++) = 't';  break;
+      case '\"': *(dst++) = '\\'; *(dst++) = '\"'; break;
+      case '\'': *(dst++) = '\\'; *(dst++) = '\''; break;
+      case '\\': *(dst++) = '\\'; *(dst++) = '\\'; break;
+      default:
+        // Note that if we emit \xNN and the src character after that is a hex
+        // digit then that digit must be escaped too to prevent it being
+        // interpreted as part of the character code by C.
+        if ((!preserve_utf8 || (uint8_t)*src < 0x80) &&
+            (!isprint(*src) || (last_hex_escape && isxdigit(*src)))) {
+          sprintf(dst, (use_hex ? "\\x%02x" : "\\%03o"), (uint8_t)*src);
+          is_hex_escape = use_hex;
+          dst += 4;
+        } else {
+          *(dst++) = *src; break;
+        }
+    }
+    last_hex_escape = is_hex_escape;
+  }
+  // Flush remaining data.
+  upb_string outstr = UPB_STACK_STRING_LEN(dstbuf, dst - dstbuf);
+  CHECK(upb_bytesink_putstr(p->bytesink, &outstr, &p->status));
+  return 0;
+err:
+  return -1;
+}
+
 static int upb_textprinter_indent(upb_textprinter *p) {
   if(!p->single_line)
     for(int i = 0; i < p->indent_depth; i++)
@@ -81,9 +132,9 @@ static upb_flow_t upb_textprinter_value(void *_p, upb_fielddef *f,
       CASE("%hhu", bool);
     case UPB_TYPE(STRING):
     case UPB_TYPE(BYTES):
-      // TODO: escaping.
       CHECK(upb_bytesink_putstr(p->bytesink, UPB_STRLIT("\""), &p->status));
-      CHECK(upb_bytesink_putstr(p->bytesink, upb_value_getstr(val), &p->status))
+      CHECK(upb_textprinter_putescaped(p, upb_value_getstr(val),
+                                       f->type == UPB_TYPE(STRING)));
       CHECK(upb_bytesink_putstr(p->bytesink, UPB_STRLIT("\""), &p->status));
       break;
   }
-- 
cgit v1.2.3