From 3f065201b18933cc436f2c541d38c2bad3bc145e Mon Sep 17 00:00:00 2001
From: Anton Kochkov <anton.kochkov@gmail.com>
Date: Fri, 3 Mar 2023 15:50:11 +0800
Subject: [PATCH] Fix string search to use proper implementation (#3414)

---
 librz/search/meson.build |  1 -
 librz/search/search.c    | 35 ++++++++++++++
 librz/search/strings.c   | 98 ----------------------------------------
 3 files changed, 35 insertions(+), 99 deletions(-)
 delete mode 100644 librz/search/strings.c

diff --git a/librz/search/meson.build b/librz/search/meson.build
index 0df10d8dfa..ab0bf39e86 100644
--- a/librz/search/meson.build
+++ b/librz/search/meson.build
@@ -5,7 +5,6 @@ rz_search_sources = [
   'regexp.c',
   'privkey-find.c',
   'search.c',
-  'strings.c'
 ]
 
 rz_search = library('rz_search', rz_search_sources,
diff --git a/librz/search/search.c b/librz/search/search.c
index 86c8eddb35..5f80b441cc 100644
--- a/librz/search/search.c
+++ b/librz/search/search.c
@@ -70,6 +70,41 @@ RZ_API int rz_search_set_string_limits(RzSearch *s, ut32 min, ut32 max) {
 	return true;
 }
 
+RZ_API int rz_search_strings_update(RzSearch *s, ut64 from, const ut8 *buf, int len) {
+	rz_return_val_if_fail(s && buf && len, -1);
+
+	RzUtilStrScanOptions scan_opt = {
+		.buf_size = len,
+		.max_uni_blocks = s->string_max,
+		.min_str_length = s->string_min,
+		.prefer_big_endian = false,
+	};
+	RzList *str_list = rz_list_new();
+	if (!str_list) {
+		return 0;
+	}
+
+	int count = rz_scan_strings_raw(buf, str_list, &scan_opt, from, from + len, RZ_STRING_ENC_GUESS);
+	if (count <= 0) {
+		rz_list_free(str_list);
+		return false;
+	}
+
+	RzListIter *iter, *iter2;
+	RzSearchKeyword *kw;
+
+	int matches = 0;
+	rz_list_foreach (s->kws, iter, kw) {
+		RzDetectedString *dstr;
+		rz_list_foreach (str_list, iter2, dstr) {
+			rz_search_hit_new(s, kw, dstr->addr);
+			matches++;
+		}
+	}
+	RZ_FREE_CUSTOM(str_list, rz_list_free);
+	return matches;
+}
+
 RZ_API int rz_search_magic_update(RzSearch *s, ut64 from, const ut8 *buf, int len) {
 	eprintf("TODO: import librz/core/cmd_search.c /m implementation into rsearch\n");
 	return false;
diff --git a/librz/search/strings.c b/librz/search/strings.c
deleted file mode 100644
index 2f1b704a41..0000000000
--- a/librz/search/strings.c
+++ /dev/null
@@ -1,98 +0,0 @@
-// SPDX-FileCopyrightText: 2006-2018 pancake <pancake@nopcode.org>
-// SPDX-License-Identifier: LGPL-3.0-only
-
-#include "rz_search.h"
-
-// TODO: this file needs some love
-enum {
-	ENCODING_ASCII = 0,
-	ENCODING_CP850 = 1
-};
-
-static char *encodings[3] = { "ascii", "cp850", NULL };
-// static int encoding = ENCODING_ASCII; // default
-// encoding = resolve_encoding(config_get("cfg.encoding"));
-
-RZ_API int rz_search_get_encoding(const char *name) {
-	int i;
-	if (!name || !*name) {
-		return ENCODING_ASCII;
-	}
-	ut32 lename = strlen(name);
-	for (i = 0; encodings[i]; i++) {
-		ut32 sz = RZ_MIN(strlen(encodings[i]), lename);
-		if (!rz_str_ncasecmp(name, encodings[i], sz)) {
-			return i;
-		}
-	}
-	return ENCODING_ASCII;
-}
-
-static bool is_encoded(int encoding, unsigned char c) {
-	switch (encoding) {
-	case ENCODING_ASCII:
-		break;
-	case ENCODING_CP850:
-		switch (c) {
-		// CP850
-		case 128: // cedilla
-		case 133: // a grave
-		case 135: // minicedilla
-		case 160: // a acute
-		case 161: // i acute
-		case 129: // u dieresi
-		case 130: // e acute
-		case 139: // i dieresi
-		case 162: // o acute
-		case 163: // u acute
-		case 164: // enye
-		case 165: // enyemay
-		case 181: // A acute
-		case 144: // E acute
-		case 214: // I acute
-		case 224: // O acute
-		case 233: // U acute
-			return true;
-		}
-		break;
-	}
-	return false;
-}
-
-RZ_API int rz_search_strings_update(RzSearch *s, ut64 from, const ut8 *buf, int len) {
-	int i = 0;
-	int matches = 0;
-	char str[4096];
-	RzListIter *iter;
-	RzSearchKeyword *kw;
-
-	rz_list_foreach (s->kws, iter, kw) {
-		for (i = 0; i < len; i++) {
-			char ch = buf[i];
-			// non-cp850 encoded
-			if (IS_PRINTABLE(ch) || IS_WHITESPACE(ch) || is_encoded(0, ch)) {
-				str[matches] = ch;
-				if (matches < sizeof(str)) {
-					matches++;
-				}
-			} else {
-				/* wide char check \x??\x00\x??\x00 */
-				if (matches && i + 2 < len && buf[i + 2] == '\0' && buf[i] == '\0' && buf[i + 1] != '\0') {
-					return 1; // widechar
-				}
-				/* check if the length fits on our request */
-				if (matches >= s->string_min && (s->string_max == 0 || matches <= s->string_max)) {
-					str[matches] = '\0';
-					int len = strlen(str);
-					if (len > 2) {
-						ut64 off = (ut64)from + i - matches;
-						rz_search_hit_new(s, kw, off);
-					}
-					fflush(stdout);
-				}
-				matches = 0;
-			}
-		}
-	}
-	return 0;
-}