MINOR: map: Add regex matching replacement

This patch declares a new map which provides a string based on
a string with back references replaced by the content matched
by the regex.

diff --git a/doc/configuration.txt b/doc/configuration.txt
index 7dd5744d75e74f801752eed36f1324d0d0e391b8..25b94afaaf87a8bcef34227b3b0ae1f0b948c98c 100644 (file)
--- a/doc/configuration.txt
+++ b/doc/configuration.txt
@@ -11815,13 +11815,19 @@ map_<match_type>_<output_type>(<map_file>[,<default_value>])
   -----------+--------------+-----------------+-----------------+---------------
     str      | end          | map_end         | map_end_int     | map_end_ip
   -----------+--------------+-----------------+-----------------+---------------
-    str      | reg          | map_reg         | map_reg_int     | map_reg_ip
+             |              | map_reg         |                 |
+    str      | reg          +-----------------+ map_reg_int     | map_reg_ip
+             |              | map_regm        |                 |
   -----------+--------------+-----------------+-----------------+---------------
     int      | int          | map_int         | map_int_int     | map_int_ip
   -----------+--------------+-----------------+-----------------+---------------
     ip       | ip           | map_ip          | map_ip_int      | map_ip_ip
   -----------+--------------+-----------------+-----------------+---------------
 
+  The special map called "map_regm" expect matching zone in the regular
+  expression and modify the output replacing back reference (like "\1") by
+  the corresponding match text.
+
   The file contains one key + value per line. Lines which start with '#' are
   ignored, just like empty lines. Leading tabs and spaces are stripped. The key
   is then the first "word" (series of non-space/tabs characters), and the value

diff --git a/include/proto/pattern.h b/include/proto/pattern.h
index 0a132f2482890fb4d008b279b7693d4e2cf138d7..9c93db93d254dfbace89a21debabeb1e25a43e15 100644 (file)
--- a/include/proto/pattern.h
+++ b/include/proto/pattern.h
@@ -67,6 +67,7 @@ int pat_idx_list_val(struct pattern_expr *expr, struct pattern *pat, char **err)
 int pat_idx_list_ptr(struct pattern_expr *expr, struct pattern *pat, char **err);
 int pat_idx_list_str(struct pattern_expr *expr, struct pattern *pat, char **err);
 int pat_idx_list_reg(struct pattern_expr *expr, struct pattern *pat, char **err);
+int pat_idx_list_regm(struct pattern_expr *expr, struct pattern *pat, char **err);
 int pat_idx_tree_ip(struct pattern_expr *expr, struct pattern *pat, char **err);
 int pat_idx_tree_str(struct pattern_expr *expr, struct pattern *pat, char **err);
 int pat_idx_tree_pfx(struct pattern_expr *expr, struct pattern *pat, char **err);
@@ -174,6 +175,7 @@ struct pattern *pat_match_ip(struct sample *smp, struct pattern_expr *expr, int
  * and restores the previous character when leaving.
  */
 struct pattern *pat_match_reg(struct sample *smp, struct pattern_expr *expr, int fill);
+struct pattern *pat_match_regm(struct sample *smp, struct pattern_expr *expr, int fill);
 
 /*
  * pattern_ref manipulation.

diff --git a/include/types/pattern.h b/include/types/pattern.h
index a71c3434fe160d98ba942453f8d5505c850ff682..912e086b78a1505daab0fdb4a47607314ad9612d 100644 (file)
--- a/include/types/pattern.h
+++ b/include/types/pattern.h
@@ -87,6 +87,7 @@ enum {
        PAT_MATCH_DOM,   /* domain-like sub-string (str) */
        PAT_MATCH_END,   /* end of string (str) */
        PAT_MATCH_REG,   /* regex (str -> reg) */
+       PAT_MATCH_REGM,  /* regex (str -> reg) with match zones */
        /* keep this one last */
        PAT_MATCH_NUM
 };

diff --git a/src/map.c b/src/map.c
index 35feea983be25acca20600da85a7745854c5fe84..a28cedae926f69449c8c48d6ce13e3d001e4894c 100644 (file)
--- a/src/map.c
+++ b/src/map.c
@@ -163,6 +163,7 @@ static int sample_conv_map(const struct arg *arg_p, struct sample *smp, void *pr
 {
        struct map_descriptor *desc;
        struct pattern *pat;
+       struct chunk *str;
 
        /* get config */
        desc = arg_p[0].data.map;
@@ -172,8 +173,19 @@ static int sample_conv_map(const struct arg *arg_p, struct sample *smp, void *pr
 
        /* Match case. */
        if (pat) {
-               /* Copy sample. */
                if (pat->data) {
+                       /* In the regm case, merge the sample with the input. */
+                       if ((long)private == PAT_MATCH_REGM) {
+                               str = get_trash_chunk();
+                               str->len = exp_replace(str->str, str->size, smp->data.u.str.str,
+                                                      pat->data->u.str.str,
+                                                      (regmatch_t *)smp->ctx.a[0]);
+                               if (str->len == -1)
+                                       return 0;
+                               smp->data.u.str = *str;
+                               return 1;
+                       }
+                       /* Copy sample. */
                        smp->data = *pat->data;
                        smp->flags |= SMP_F_CONST;
                        return 1;
@@ -242,6 +254,7 @@ static struct sample_conv_kw_list sample_conv_kws = {ILH, {
        { "map_dom",     sample_conv_map, ARG2(1,STR,STR), sample_load_map, SMP_T_STR,  SMP_T_STR,  (void *)PAT_MATCH_DOM },
        { "map_end",     sample_conv_map, ARG2(1,STR,STR), sample_load_map, SMP_T_STR,  SMP_T_STR,  (void *)PAT_MATCH_END },
        { "map_reg",     sample_conv_map, ARG2(1,STR,STR), sample_load_map, SMP_T_STR,  SMP_T_STR,  (void *)PAT_MATCH_REG },
+       { "map_regm",    sample_conv_map, ARG2(1,STR,STR), sample_load_map, SMP_T_STR,  SMP_T_STR,  (void *)PAT_MATCH_REGM},
        { "map_int",     sample_conv_map, ARG2(1,STR,STR), sample_load_map, SMP_T_SINT, SMP_T_STR,  (void *)PAT_MATCH_INT },
        { "map_ip",      sample_conv_map, ARG2(1,STR,STR), sample_load_map, SMP_T_ADDR, SMP_T_STR,  (void *)PAT_MATCH_IP  },
 

diff --git a/src/pattern.c b/src/pattern.c
index 254c10650ed2967c76611357b86a736f4a1ec977..b4cb8e9ea60ba642c2a5ede069eec54ebb024898 100644 (file)
--- a/src/pattern.c
+++ b/src/pattern.c
@@ -41,6 +41,7 @@ char *pat_match_names[PAT_MATCH_NUM] = {
        [PAT_MATCH_DOM]   = "dom",
        [PAT_MATCH_END]   = "end",
        [PAT_MATCH_REG]   = "reg",
+       [PAT_MATCH_REGM]  = "regm",
 };
 
 int (*pat_parse_fcts[PAT_MATCH_NUM])(const char *, struct pattern *, int, char **) = {
@@ -57,6 +58,7 @@ int (*pat_parse_fcts[PAT_MATCH_NUM])(const char *, struct pattern *, int, char *
        [PAT_MATCH_DOM]   = pat_parse_str,
        [PAT_MATCH_END]   = pat_parse_str,
        [PAT_MATCH_REG]   = pat_parse_reg,
+       [PAT_MATCH_REGM]  = pat_parse_reg,
 };
 
 int (*pat_index_fcts[PAT_MATCH_NUM])(struct pattern_expr *, struct pattern *, char **) = {
@@ -73,6 +75,7 @@ int (*pat_index_fcts[PAT_MATCH_NUM])(struct pattern_expr *, struct pattern *, ch
        [PAT_MATCH_DOM]   = pat_idx_list_str,
        [PAT_MATCH_END]   = pat_idx_list_str,
        [PAT_MATCH_REG]   = pat_idx_list_reg,
+       [PAT_MATCH_REGM]  = pat_idx_list_regm,
 };
 
 void (*pat_delete_fcts[PAT_MATCH_NUM])(struct pattern_expr *, struct pat_ref_elt *) = {
@@ -89,6 +92,7 @@ void (*pat_delete_fcts[PAT_MATCH_NUM])(struct pattern_expr *, struct pat_ref_elt
        [PAT_MATCH_DOM]   = pat_del_list_ptr,
        [PAT_MATCH_END]   = pat_del_list_ptr,
        [PAT_MATCH_REG]   = pat_del_list_reg,
+       [PAT_MATCH_REGM]  = pat_del_list_reg,
 };
 
 void (*pat_prune_fcts[PAT_MATCH_NUM])(struct pattern_expr *) = {
@@ -105,6 +109,7 @@ void (*pat_prune_fcts[PAT_MATCH_NUM])(struct pattern_expr *) = {
        [PAT_MATCH_DOM]   = pat_prune_ptr,
        [PAT_MATCH_END]   = pat_prune_ptr,
        [PAT_MATCH_REG]   = pat_prune_reg,
+       [PAT_MATCH_REGM]  = pat_prune_reg,
 };
 
 struct pattern *(*pat_match_fcts[PAT_MATCH_NUM])(struct sample *, struct pattern_expr *, int) = {
@@ -121,6 +126,7 @@ struct pattern *(*pat_match_fcts[PAT_MATCH_NUM])(struct sample *, struct pattern
        [PAT_MATCH_DOM]   = pat_match_dom,
        [PAT_MATCH_END]   = pat_match_end,
        [PAT_MATCH_REG]   = pat_match_reg,
+       [PAT_MATCH_REGM]  = pat_match_regm,
 };
 
 /* Just used for checking configuration compatibility */
@@ -138,6 +144,7 @@ int pat_match_types[PAT_MATCH_NUM] = {
        [PAT_MATCH_DOM]   = SMP_T_STR,
        [PAT_MATCH_END]   = SMP_T_STR,
        [PAT_MATCH_REG]   = SMP_T_STR,
+       [PAT_MATCH_REGM]  = SMP_T_STR,
 };
 
 /* this struct is used to return information */
@@ -539,6 +546,30 @@ struct pattern *pat_match_bin(struct sample *smp, struct pattern_expr *expr, int
        return ret;
 }
 
+/* Executes a regex. It temporarily changes the data to add a trailing zero,
+ * and restores the previous character when leaving. This function fills
+ * a matching array.
+ */
+struct pattern *pat_match_regm(struct sample *smp, struct pattern_expr *expr, int fill)
+{
+       struct pattern_list *lst;
+       struct pattern *pattern;
+       struct pattern *ret = NULL;
+
+       list_for_each_entry(lst, &expr->patterns, list) {
+               pattern = &lst->pat;
+
+               if (regex_exec_match2(pattern->ptr.reg, smp->data.u.str.str, smp->data.u.str.len,
+                                     MAX_MATCH, pmatch, 0)) {
+                       ret = pattern;
+                       smp->ctx.a[0] = pmatch;
+                       break;
+               }
+       }
+
+       return ret;
+}
+
 /* Executes a regex. It temporarily changes the data to add a trailing zero,
  * and restores the previous character when leaving.
  */
@@ -1146,7 +1177,7 @@ int pat_idx_list_str(struct pattern_expr *expr, struct pattern *pat, char **err)
        return 1;
 }
 
-int pat_idx_list_reg(struct pattern_expr *expr, struct pattern *pat, char **err)
+int pat_idx_list_reg_cap(struct pattern_expr *expr, struct pattern *pat, int cap, char **err)
 {
        struct pattern_list *patl;
 
@@ -1169,7 +1200,8 @@ int pat_idx_list_reg(struct pattern_expr *expr, struct pattern *pat, char **err)
        }
 
        /* compile regex */
-       if (!regex_comp(pat->ptr.str, patl->pat.ptr.reg, !(expr->mflags & PAT_MF_IGNORE_CASE), 0, err)) {
+       if (!regex_comp(pat->ptr.str, patl->pat.ptr.reg,
+                       !(expr->mflags & PAT_MF_IGNORE_CASE), cap, err)) {
                free(patl->pat.ptr.reg);
                free(patl);
                return 0;
@@ -1183,6 +1215,16 @@ int pat_idx_list_reg(struct pattern_expr *expr, struct pattern *pat, char **err)
        return 1;
 }
 
+int pat_idx_list_reg(struct pattern_expr *expr, struct pattern *pat, char **err)
+{
+       return pat_idx_list_reg_cap(expr, pat, 0, err);
+}
+
+int pat_idx_list_regm(struct pattern_expr *expr, struct pattern *pat, char **err)
+{
+       return pat_idx_list_reg_cap(expr, pat, 1, err);
+}
+
 int pat_idx_tree_ip(struct pattern_expr *expr, struct pattern *pat, char **err)
 {
        unsigned int mask;