From d06ae2f5c345741192a0e3f9086765382690e37b Mon Sep 17 00:00:00 2001
From: Simon Tatham <anakin@pobox.com>
Date: Sun, 1 May 2022 11:19:10 +0100
Subject: [PATCH] New utility function base64_valid().

For when you want to tell the difference between a base64-encoded
string and some other kind of string that might replace it.
---
 misc.h               |  1 +
 utils/CMakeLists.txt |  1 +
 utils/base64_valid.c | 54 ++++++++++++++++++++++++++++++++++++++++++++
 3 files changed, 56 insertions(+)
 create mode 100644 utils/base64_valid.c

diff --git a/misc.h b/misc.h
index eeed80c1..fe8eb06a 100644
--- a/misc.h
+++ b/misc.h
@@ -114,6 +114,7 @@ strbuf *base64_decode_sb(ptrlen data);
 void base64_encode_bs(BinarySink *bs, ptrlen data, int cpl);
 void base64_encode_fp(FILE *fp, ptrlen data, int cpl);
 strbuf *base64_encode_sb(ptrlen data, int cpl);
+bool base64_valid(ptrlen data);
 
 struct bufchain_granule;
 struct bufchain_tag {
diff --git a/utils/CMakeLists.txt b/utils/CMakeLists.txt
index 949166f5..9d079ecb 100644
--- a/utils/CMakeLists.txt
+++ b/utils/CMakeLists.txt
@@ -5,6 +5,7 @@ add_sources_from_current_dir(utils
   base64_decode.c
   base64_encode_atom.c
   base64_encode.c
+  base64_valid.c
   bufchain.c
   buildinfo.c
   burnstr.c
diff --git a/utils/base64_valid.c b/utils/base64_valid.c
new file mode 100644
index 00000000..8eb1f3a0
--- /dev/null
+++ b/utils/base64_valid.c
@@ -0,0 +1,54 @@
+/*
+ * Determine whether a string looks like valid base64-encoded data.
+ */
+
+#include "misc.h"
+
+static inline bool valid_char_main(char c)
+{
+    return ((c >= 'A' && c <= 'Z') ||
+            (c >= 'a' && c <= 'z') ||
+            (c >= '0' && c <= '9') ||
+            c == '+' || c == '/');
+}
+
+bool base64_valid(ptrlen data)
+{
+    size_t blocklen = 0, nequals = 0;
+
+    for (size_t i = 0; i < data.len; i++) {
+        char c = ((const char *)data.ptr)[i];
+
+        if (c == '\n' || c == '\r')
+            continue;
+
+        if (valid_char_main(c)) {
+            if (nequals)               /* can't go back to data after = */
+                return false;
+            blocklen++;
+            if (blocklen == 4)
+                blocklen = 0;
+            continue;
+        }
+
+        if (c == '=') {
+            if (blocklen == 0 && nequals) /* started a fresh block */
+                return false;
+
+            nequals++;
+            blocklen++;
+            if (blocklen == 4) {
+                if (nequals > 2)
+                    return false;      /* nonsensical final block */
+                blocklen = 0;
+            }
+            continue;
+        }
+
+        return false;                  /* bad character */
+    }
+
+    if (blocklen == 0 || blocklen == 2 || blocklen == 3)
+        return true;                   /* permit eliding the trailing = */
+    return false;
+}