diff --git a/build-example.sh b/build-example.sh
index 1eaf0cc..b7a3a4b 100755
--- a/build-example.sh
+++ b/build-example.sh
@@ -39,11 +39,7 @@
     continue
   fi
 
-  if [ $f = crc32 ]; then
-    echo "Building gen/bin/example-$f"
-    # example/crc32 is unusual in that it's C++, not C.
-    $CXX -O3 example/$f/*.cc -o gen/bin/example-$f
-  elif [ $f = imageviewer ]; then
+  if [ $f = imageviewer ]; then
     # example/imageviewer is unusual in that needs additional libraries.
     echo "Building gen/bin/example-$f"
     $CC -O3 example/$f/*.c -lxcb -lxcb-image -o gen/bin/example-$f
@@ -59,6 +55,9 @@
     fi
   elif [ -e example/$f/*.c ]; then
     echo "Building gen/bin/example-$f"
-    $CC -O3 example/$f/*.c -o gen/bin/example-$f
+    $CC  -O3 example/$f/*.c  -o gen/bin/example-$f
+  elif [ -e example/$f/*.cc ]; then
+    echo "Building gen/bin/example-$f"
+    $CXX -O3 example/$f/*.cc -o gen/bin/example-$f
   fi
 done
diff --git a/doc/changelog.md b/doc/changelog.md
index 3e7b3ab..d8fb193 100644
--- a/doc/changelog.md
+++ b/doc/changelog.md
@@ -4,6 +4,7 @@
 ## Work In Progress
 
 - Added `example/imageviewer`.
+- Added `example/jsonptr`.
 - Added `std/bmp`.
 - Added `std/gif.config_decoder`.
 - Added `std/json`.
diff --git a/example/crc32/crc32.cc b/example/crc32/crc32.cc
index 2348514..09d2512 100644
--- a/example/crc32/crc32.cc
+++ b/example/crc32/crc32.cc
@@ -19,7 +19,7 @@
 similar to the standard /usr/bin/crc32 program, except that this example
 program only reads from stdin.
 
-This example program differs from the other example Wuffs programs in that it
+This example program differs from most other example Wuffs programs in that it
 is written in C++, not C.
 
 $CXX crc32.cc && ./a.out < ../../README.md; rm -f a.out
diff --git a/example/jsonptr/jsonptr.cc b/example/jsonptr/jsonptr.cc
new file mode 100644
index 0000000..bffb72e
--- /dev/null
+++ b/example/jsonptr/jsonptr.cc
@@ -0,0 +1,391 @@
+// Copyright 2020 The Wuffs Authors.
+//
+// Licensed under the Apache License, Version 2.0 (the "License");
+// you may not use this file except in compliance with the License.
+// You may obtain a copy of the License at
+//
+//    https://www.apache.org/licenses/LICENSE-2.0
+//
+// Unless required by applicable law or agreed to in writing, software
+// distributed under the License is distributed on an "AS IS" BASIS,
+// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+// See the License for the specific language governing permissions and
+// limitations under the License.
+
+// ----------------
+
+/*
+jsonptr is a JSON formatter (pretty-printer).
+
+This example program differs from most other example Wuffs programs in that it
+is written in C++, not C.
+
+$CXX jsonptr.cc && ./a.out < ../../test/data/github-tags.json; rm -f a.out
+
+for a C++ compiler $CXX, such as clang++ or g++.
+*/
+
+#include <inttypes.h>
+#include <stdio.h>
+
+// Wuffs ships as a "single file C library" or "header file library" as per
+// https://github.com/nothings/stb/blob/master/docs/stb_howto.txt
+//
+// To use that single file as a "foo.c"-like implementation, instead of a
+// "foo.h"-like header, #define WUFFS_IMPLEMENTATION before #include'ing or
+// compiling it.
+#define WUFFS_IMPLEMENTATION
+
+// Defining the WUFFS_CONFIG__MODULE* macros are optional, but it lets users of
+// release/c/etc.c whitelist which parts of Wuffs to build. That file contains
+// the entire Wuffs standard library, implementing a variety of codecs and file
+// formats. Without this macro definition, an optimizing compiler or linker may
+// very well discard Wuffs code for unused codecs, but listing the Wuffs
+// modules we use makes that process explicit. Preprocessing means that such
+// code simply isn't compiled.
+#define WUFFS_CONFIG__MODULES
+#define WUFFS_CONFIG__MODULE__BASE
+#define WUFFS_CONFIG__MODULE__JSON
+
+// If building this program in an environment that doesn't easily accommodate
+// relative includes, you can use the script/inline-c-relative-includes.go
+// program to generate a stand-alone C++ file.
+#include "../../release/c/wuffs-unsupported-snapshot.c"
+
+#ifndef DST_BUFFER_SIZE
+#define DST_BUFFER_SIZE (32 * 1024)
+#endif
+#ifndef SRC_BUFFER_SIZE
+#define SRC_BUFFER_SIZE (32 * 1024)
+#endif
+#ifndef TOKEN_BUFFER_SIZE
+#define TOKEN_BUFFER_SIZE (4 * 1024)
+#endif
+
+uint8_t dst_buffer[DST_BUFFER_SIZE];
+uint8_t src_buffer[SRC_BUFFER_SIZE];
+wuffs_base__token tok_buffer[TOKEN_BUFFER_SIZE];
+
+wuffs_base__io_buffer dst;
+wuffs_base__io_buffer src;
+wuffs_base__token_buffer tok;
+
+wuffs_json__decoder dec;
+wuffs_base__status dec_status;
+
+// dec_current_token_end_src_index is the src.data.ptr index of the end of the
+// current token. An invariant is that (dec_current_token_end_src_index <=
+// src.meta.ri).
+size_t dec_current_token_end_src_index;
+
+#define MAX_INDENT 8
+#define INDENT_STRING "        "
+size_t indent;
+
+#define TRY(error_msg)         \
+  do {                         \
+    const char* z = error_msg; \
+    if (z) {                   \
+      return z;                \
+    }                          \
+  } while (false)
+
+// ----
+
+const char* read_src() {
+  src.compact();
+  if (src.meta.wi >= src.data.len) {
+    return "main: src buffer is full";
+  }
+  size_t n = fread(src.data.ptr + src.meta.wi, sizeof(uint8_t),
+                   src.data.len - src.meta.wi, stdin);
+  src.meta.wi += n;
+  if (n == 0) {
+    if (feof(stdin)) {
+      return "main: read error: unexpected EOF";
+    } else {
+      return "main: read error";
+    }
+  }
+  return nullptr;
+}
+
+const char* flush_dst() {
+  size_t n = dst.meta.wi - dst.meta.ri;
+  if (n > 0) {
+    size_t i = fwrite(dst.data.ptr + dst.meta.ri, sizeof(uint8_t), n, stdout);
+    dst.meta.ri += i;
+    if (i != n) {
+      return "main: write error";
+    }
+    dst.compact();
+  }
+  return nullptr;
+}
+
+const char* write_dst(const void* s, size_t n) {
+  const uint8_t* p = static_cast<const uint8_t*>(s);
+  while (n > 0) {
+    size_t i = dst.writer_available();
+    if (i == 0) {
+      const char* z = flush_dst();
+      if (z) {
+        return z;
+      }
+      i = dst.writer_available();
+      if (i == 0) {
+        return "main: dst buffer is full";
+      }
+    }
+
+    if (i > n) {
+      i = n;
+    }
+    memcpy(dst.data.ptr + dst.meta.wi, p, i);
+    dst.meta.wi += i;
+    p += i;
+    n -= i;
+  }
+  return nullptr;
+}
+
+// ----
+
+enum class context {
+  none,
+  in_list_after_bracket,
+  in_list_after_value,
+  in_dict_after_brace,
+  in_dict_after_key,
+  in_dict_after_value,
+};
+
+// parsed_token is a result type, combining a wuffs_base_token and an error.
+// For the parsed_token returned by make_parsed_token, it also contains the src
+// data bytes for the token. This slice is just a view into the src_buffer
+// array, and its contents may change on the next call to parse_next_token.
+//
+// An invariant is that (token.length() == data.len).
+typedef struct {
+  const char* error_msg;
+  wuffs_base__token token;
+  wuffs_base__slice_u8 data;
+} parsed_token;
+
+parsed_token make_pt_error(const char* error_msg) {
+  parsed_token p;
+  p.error_msg = error_msg;
+  p.token = wuffs_base__make_token(0);
+  p.data = wuffs_base__make_slice_u8(nullptr, 0);
+  return p;
+}
+
+parsed_token make_pt_token(uint64_t token_repr,
+                           uint8_t* data_ptr,
+                           size_t data_len) {
+  parsed_token p;
+  p.error_msg = nullptr;
+  p.token = wuffs_base__make_token(token_repr);
+  p.data = wuffs_base__make_slice_u8(data_ptr, data_len);
+  return p;
+}
+
+parsed_token parse_next_token() {
+  while (true) {
+    // Return a previously produced token, if one exists.
+    //
+    // We do this before checking dec_status. This is analogous to Go's
+    // io.Reader's documented idiom, when processing io.Reader.Read's returned
+    // (n int, err error), to "process the n > 0 bytes returned before
+    // considering the error err. Doing so correctly handles I/O errors that
+    // happen after reading some bytes".
+    if (tok.meta.ri < tok.meta.wi) {
+      wuffs_base__token t = tok.data.ptr[tok.meta.ri++];
+
+      uint64_t n = t.length();
+      if ((src.meta.ri - dec_current_token_end_src_index) < n) {
+        return make_pt_error("main: internal error: inconsistent src indexes");
+      }
+      dec_current_token_end_src_index += n;
+
+      // Filter out any filler tokens (e.g. whitespace).
+      if (t.value_base_category() == 0) {
+        continue;
+      }
+
+      return make_pt_token(
+          t.repr, src.data.ptr + dec_current_token_end_src_index - n, n);
+    }
+
+    // Now consider dec_status.
+    if (dec_status.repr == nullptr) {
+      return make_pt_error("main: internal error: parser stopped");
+
+    } else if (dec_status.repr == wuffs_base__suspension__short_read) {
+      if (dec_current_token_end_src_index != src.meta.ri) {
+        return make_pt_error("main: internal error: inconsistent src indexes");
+      }
+      const char* z = read_src();
+      if (z) {
+        return make_pt_error(z);
+      }
+      dec_current_token_end_src_index = src.meta.ri;
+
+    } else if (dec_status.repr == wuffs_base__suspension__short_write) {
+      tok.compact();
+
+    } else {
+      return make_pt_error(dec_status.message());
+    }
+
+    // Retry a "short read" or "short write" suspension.
+    dec_status = dec.decode_tokens(&tok, &src);
+  }
+}
+
+// ----
+
+const char* handle_string(parsed_token pt) {
+  while (true) {
+    TRY(write_dst("\"", 1));
+    TRY(write_dst(pt.data.ptr, pt.data.len));
+    if ((pt.token.value_base_detail() & 1) == 0) {
+      break;
+    }
+    pt = parse_next_token();
+    if (pt.error_msg) {
+      return pt.error_msg;
+    }
+  }
+  TRY(write_dst("\"", 1));
+  return nullptr;
+}
+
+const char* main2() {
+  dec_status = dec.initialize(sizeof__wuffs_json__decoder(), WUFFS_VERSION, 0);
+  if (!dec_status.is_ok()) {
+    return dec_status.message();
+  }
+  dec_status = dec.decode_tokens(&tok, &src);
+  dec_current_token_end_src_index = 0;
+
+  uint64_t depth = 0;
+  context ctx = context::none;
+
+continue_loop:
+  while (true) {
+    parsed_token pt = parse_next_token();
+    if (pt.error_msg) {
+      return pt.error_msg;
+    }
+    uint64_t vbc = pt.token.value_base_category();
+    uint64_t vbd = pt.token.value_base_detail();
+
+    // Handle ']' or '}'.
+    if ((vbc == 1) && ((vbd & 0x2) != 0)) {
+      if (depth <= 0) {
+        return "main: internal error: inconsistent depth";
+      }
+      depth--;
+
+      // Write preceding whitespace.
+      if ((ctx != context::in_list_after_bracket) &&
+          (ctx != context::in_dict_after_brace)) {
+        TRY(write_dst("\n", 1));
+        for (size_t i = 0; i < depth; i++) {
+          TRY(write_dst(INDENT_STRING, indent));
+        }
+      }
+
+      TRY(write_dst((vbd & 0x10) ? "]" : "}", 1));
+      ctx = (vbd & 0x1000) ? context::in_list_after_value
+                           : context::in_dict_after_key;
+      goto after_value;
+    }
+
+    // Write preceding whitespace and punctuation, if it wasn't ']' or '}'.
+    if (ctx == context::in_dict_after_key) {
+      TRY(write_dst(": ", 2));
+    } else if (ctx != context::none) {
+      if ((ctx != context::in_list_after_bracket) &&
+          (ctx != context::in_dict_after_brace)) {
+        TRY(write_dst(",", 1));
+      }
+      TRY(write_dst("\n", 1));
+      for (size_t i = 0; i < depth; i++) {
+        TRY(write_dst(INDENT_STRING, indent));
+      }
+    }
+
+    // Handle the token itself: either a container ('[' or '{') or a simple
+    // value (number, string or literal).
+    switch (vbc) {
+      case 1:
+        TRY(write_dst((vbd & 0x10) ? "[" : "{", 1));
+        depth++;
+        ctx = (vbd & 0x10) ? context::in_list_after_bracket
+                           : context::in_dict_after_brace;
+        goto continue_loop;
+
+      case 3:
+        TRY(handle_string(pt));
+        goto after_value;
+    }
+
+    // Return an error if we didn't match the (vbc, vbd) pair.
+    return "main: unexpected token";
+
+    // Book-keeping after completing a value (whether a container value or a
+    // simple value). Empty parent containers are no longer empty. If the
+    // parent container is a "{...}" object, toggle between keys and values.
+  after_value:
+    if (depth <= 0) {
+      return nullptr;
+    }
+    switch (ctx) {
+      case context::in_list_after_bracket:
+        ctx = context::in_list_after_value;
+        break;
+      case context::in_dict_after_brace:
+        ctx = context::in_dict_after_key;
+        break;
+      case context::in_dict_after_key:
+        ctx = context::in_dict_after_value;
+        break;
+      case context::in_dict_after_value:
+        ctx = context::in_dict_after_key;
+        break;
+    }
+  }
+}
+
+const char* main1(int argc, char** argv) {
+  dst = wuffs_base__make_io_buffer(
+      wuffs_base__make_slice_u8(dst_buffer, DST_BUFFER_SIZE),
+      wuffs_base__empty_io_buffer_meta());
+
+  src = wuffs_base__make_io_buffer(
+      wuffs_base__make_slice_u8(src_buffer, SRC_BUFFER_SIZE),
+      wuffs_base__empty_io_buffer_meta());
+
+  tok = wuffs_base__make_token_buffer(
+      wuffs_base__make_slice_token(tok_buffer, TOKEN_BUFFER_SIZE),
+      wuffs_base__empty_token_buffer_meta());
+
+  indent = 4;
+
+  TRY(main2());
+  TRY(write_dst("\n", 1));
+  return nullptr;
+}
+
+int main(int argc, char** argv) {
+  const char* z0 = main1(argc, argv);
+  const char* z1 = flush_dst();
+  const char* z = z0 ? z0 : z1;
+  if (z) {
+    fprintf(stderr, "%s\n", z);
+    return 1;
+  }
+  return 0;
+}