SPDX: license tags (mostly by guesswork)

[exim.git] / src / src / regex.c

diff --git a/src/src/regex.c b/src/src/regex.c
index 35af5e1b5b20578f6b9db05fbfa06819695d397c..eefba8ecf93e258557d167e361c084de092f5f5b 100644 (file)
--- a/src/src/regex.c
+++ b/src/src/regex.c
@@ -2,9 +2,11 @@
 *     Exim - an Internet mail transport agent    *
 *************************************************/
 
-/* Copyright (c) Tom Kistner <tom@duncanthrax.net> 2003-2015
+/*
+ * Copyright (c) The Exim Maintainers 2016 - 2022
+ * Copyright (c) Tom Kistner <tom@duncanthrax.net> 2003-2015
  * License: GPL
- * Copyright (c) The Exim Maintainers 2016 - 2018
+ * SPDX-License-Identifier: GPL-2.0-only
  */
 
 /* Code for matching regular expressions against headers and body.
@@ -17,9 +19,9 @@
 
 /* Structure to hold a list of Regular expressions */
 typedef struct pcre_list {
-  pcre2_code *re;
-  uschar *pcre_text;
-  struct pcre_list *next;
+  const pcre2_code *   re;
+  uschar *             pcre_text;
+  struct pcre_list *   next;
 } pcre_list;
 
 uschar regex_match_string_buffer[1024];
@@ -27,35 +29,31 @@ uschar regex_match_string_buffer[1024];
 extern FILE *mime_stream;
 extern uschar *mime_current_boundary;
 
+
 static pcre_list *
-compile(const uschar * list)
+compile(const uschar * list, BOOL cacheable)
 {
 int sep = 0;
-uschar *regex_string;
-pcre_list *re_list_head = NULL;
-pcre_list *ri;
+uschar * regex_string;
+pcre_list * re_list_head = NULL;
+pcre_list * ri;
 
 /* precompile our regexes */
 while ((regex_string = string_nextinlist(&list, &sep, NULL, 0)))
   if (strcmpic(regex_string, US"false") != 0 && Ustrcmp(regex_string, "0") != 0)
     {
-    pcre2_code * re;
-    int err;
-    PCRE2_SIZE pcre_erroffset;
-
     /* compile our regular expression */
-    if (!(re = pcre2_compile( (PCRE2_SPTR) regex_string, PCRE2_ZERO_TERMINATED,
-                 0, &err, &pcre_erroffset, pcre_cmp_ctx)))
+    uschar * errstr;
+    const pcre2_code * re = regex_compile(regex_string,
+      cacheable ? MCS_CACHEABLE : MCS_NOFLAGS, &errstr, pcre_gen_cmp_ctx);
+
+    if (!re)
       {
-      uschar errbuf[128];
-      pcre2_get_error_message(err, errbuf, sizeof(errbuf));
-      log_write(0, LOG_MAIN,
-          "regex acl condition warning - error in regex '%s': %s at offset %l, skipped.",
-          regex_string, errbuf, (long)pcre_erroffset);
+      log_write(0, LOG_MAIN, "regex acl condition warning - %s, skipped", errstr);
       continue;
       }
 
-    ri = store_get(sizeof(pcre_list), FALSE);
+    ri = store_get(sizeof(pcre_list), GET_UNTAINTED);
     ri->re = re;
     ri->pcre_text = regex_string;
     ri->next = re_list_head;
@@ -74,7 +72,7 @@ for (pcre_list * ri = re_list_head; ri; ri = ri->next)
   int n;
 
   /* try matcher on the line */
-  if ((n = pcre2_match(ri->re, (PCRE2_SPTR)linebuffer, len, 0, 0, md, pcre_mtc_ctx)) > 0)
+  if ((n = pcre2_match(ri->re, (PCRE2_SPTR)linebuffer, len, 0, 0, md, pcre_gen_mtc_ctx)) > 0)
     {
     Ustrncpy(regex_match_string_buffer, ri->pcre_text,
              sizeof(regex_match_string_buffer)-1);
@@ -82,31 +80,41 @@ for (pcre_list * ri = re_list_head; ri; ri = ri->next)
 
     for (int nn = 1; nn < n; nn++)
       {
-      PCRE2_UCHAR * cstr;
-      PCRE2_SIZE cslen;
-      pcre2_substring_get_bynumber(md, nn, &cstr, &cslen);
-      regex_vars[nn-1] = US cstr;
+      PCRE2_SIZE * ovec = pcre2_get_ovector_pointer(md);
+      int off = nn * 2;
+      int len = ovec[off + 1] - ovec[off];
+      regex_vars[nn-1] = string_copyn(linebuffer + ovec[off], len);
       }
 
     return OK;
     }
   }
-pcre2_match_data_free(md);
+/* pcre2_match_data_free(md);  gen ctx needs no free */
 return FAIL;
 }
 
+
+/* reset expansion variables */
+void
+regex_vars_clear(void)
+{
+regex_match_string = NULL;
+for (int i = 0; i < REGEX_VARS; i++) regex_vars[i] = NULL;
+}
+
+
+
 int
-regex(const uschar **listptr)
+regex(const uschar ** listptr, BOOL cacheable)
 {
 unsigned long mbox_size;
-FILE *mbox_file;
-pcre_list *re_list_head;
-uschar *linebuffer;
+FILE * mbox_file;
+pcre_list * re_list_head;
+uschar * linebuffer;
 long f_pos = 0;
 int ret = FAIL;
 
-/* reset expansion variable */
-regex_match_string = NULL;
+regex_vars_clear();
 
 if (!mime_stream)                              /* We are in the DATA ACL */
   {
@@ -129,11 +137,11 @@ else
   }
 
 /* precompile our regexes */
-if (!(re_list_head = compile(*listptr)))
+if (!(re_list_head = compile(*listptr, cacheable)))
   return FAIL;                 /* no regexes -> nothing to do */
 
 /* match each line against all regexes */
-linebuffer = store_get(32767, TRUE);   /* tainted */
+linebuffer = store_get(32767, GET_TAINTED);
 while (fgets(CS linebuffer, 32767, mbox_file))
   {
   if (  mime_stream && mime_current_boundary           /* check boundary */
@@ -166,19 +174,18 @@ return ret;
 
 
 int
-mime_regex(const uschar **listptr)
+mime_regex(const uschar **listptr, BOOL cacheable)
 {
-pcre_list *re_list_head = NULL;
-FILE *f;
-uschar *mime_subject = NULL;
+pcre_list * re_list_head = NULL;
+FILE * f;
+uschar * mime_subject = NULL;
 int mime_subject_len = 0;
 int ret;
 
-/* reset expansion variable */
-regex_match_string = NULL;
+regex_vars_clear();
 
 /* precompile our regexes */
-if (!(re_list_head = compile(*listptr)))
+if (!(re_list_head = compile(*listptr, cacheable)))
   return FAIL;                 /* no regexes -> nothing to do */
 
 /* check if the file is already decoded */
@@ -204,7 +211,7 @@ if (!(f = fopen(CS mime_decoded_filename, "rb")))
   }
 
 /* get 32k memory, tainted */
-mime_subject = store_get(32767, TRUE);
+mime_subject = store_get(32767, GET_TAINTED);
 
 mime_subject_len = fread(mime_subject, 1, 32766, f);