From e33c4b3c11f80a1252fbb17e0a3147d74888b61d Mon Sep 17 00:00:00 2001
From: Willy Tarreau <w@1wt.eu>
Date: Fri, 12 Mar 2021 18:59:31 +0100
Subject: [PATCH] MINOR: tools: add the ability to update a word fingerprint

Instead of making a new one from scratch, let's support not wiping the
existing fingerprint and updating it, and to do the same char by char.
The word-by-word one will still result in multiple beginnings and ends,
but that will accurately translate word boundaries. The char-based one
has more flexibility and requires that the caller maintains the previous
char to indicate the transition, which also allows to insert delimiters
for example.
---
 include/haproxy/tools.h | 29 +++++++++++++++++++++++++++++
 src/tools.c             | 16 +++++++++++++---
 2 files changed, 42 insertions(+), 3 deletions(-)

diff --git a/include/haproxy/tools.h b/include/haproxy/tools.h
index 56948ea6c..901dca0bb 100644
--- a/include/haproxy/tools.h
+++ b/include/haproxy/tools.h
@@ -865,6 +865,7 @@ int my_unsetenv(const char *name);
 char *env_expand(char *in);
 uint32_t parse_line(char *in, char *out, size_t *outlen, char **args, int *nbargs, uint32_t opts, char **errptr);
 size_t sanitize_for_printing(char *line, size_t pos, size_t width);
+void update_word_fingerprint(uint8_t *fp, const char *word);
 void make_word_fingerprint(uint8_t *fp, const char *word);
 int word_fingerprint_distance(const uint8_t *fp1, const uint8_t *fp2);
 
@@ -1072,5 +1073,33 @@ static inline unsigned int statistical_prng()
 	return statistical_prng_state = x;
 }
 
+/* Update array <fp> with the character transition <prev> to <curr>. If <prev>
+ * is zero, it's assumed that <curr> is the first character. If <curr> is zero
+ * its assumed to mark the end. Both may be zero. <fp> is a 1024-entries array
+ * indexed as 32*from+to. Positions for 'from' and 'to' are:
+ *   0..25=letter, 26=digit, 27=other, 28=begin, 29=end, others unused.
+ */
+static inline void update_char_fingerprint(uint8_t *fp, char prev, char curr)
+{
+	int from, to;
+
+	switch (prev) {
+	case 0:         from = 26; break; // begin
+	case 'a'...'z': from = prev - 'a'; break;
+	case 'A'...'Z': from = tolower(prev) - 'a'; break;
+	case '0'...'9': from = 26; break;
+	default:        from = 27; break;
+	}
+
+	switch (curr) {
+	case 0:         to = 28; break; // end
+	case 'a'...'z': to = curr - 'a'; break;
+	case 'A'...'Z': to = tolower(curr) - 'a'; break;
+	case '0'...'9': to = 26; break;
+	default:        to = 27; break;
+	}
+
+	fp[32 * from + to]++;
+}
 
 #endif /* _HAPROXY_TOOLS_H */
diff --git a/src/tools.c b/src/tools.c
index 0fd3edead..1255e748b 100644
--- a/src/tools.c
+++ b/src/tools.c
@@ -5369,18 +5369,17 @@ size_t sanitize_for_printing(char *line, size_t pos, size_t width)
 	return pos - shift;
 }
 
-/* Initialize array <fp> with the fingerprint of word <word> by counting the
+/* Update array <fp> with the fingerprint of word <word> by counting the
  * transitions between characters. <fp> is a 1024-entries array indexed as
  * 32*from+to. Positions for 'from' and 'to' are:
  *   0..25=letter, 26=digit, 27=other, 28=begin, 29=end, others unused.
  */
-void make_word_fingerprint(uint8_t *fp, const char *word)
+void update_word_fingerprint(uint8_t *fp, const char *word)
 {
 	const char *p;
 	int from, to;
 	int c;
 
-	memset(fp, 0, 1024);
 	from = 28; // begin
 	for (p = word; *p; p++) {
 		c = tolower(*p);
@@ -5397,6 +5396,17 @@ void make_word_fingerprint(uint8_t *fp, const char *word)
 	fp[32 * from + to]++;
 }
 
+/* Initialize array <fp> with the fingerprint of word <word> by counting the
+ * transitions between characters. <fp> is a 1024-entries array indexed as
+ * 32*from+to. Positions for 'from' and 'to' are:
+ *   0..25=letter, 26=digit, 27=other, 28=begin, 29=end, others unused.
+ */
+void make_word_fingerprint(uint8_t *fp, const char *word)
+{
+	memset(fp, 0, 1024);
+	update_word_fingerprint(fp, word);
+}
+
 /* Return the distance between two word fingerprints created by function
  * make_word_fingerprint(). It's a positive integer calculated as the sum of
  * the squares of the differences between each location.