Issue #16203: Add re.fullmatch() function and regex.fullmatch() method,

which anchor the pattern at both ends of the string to match.

Original patch by Matthew Barnett.

3 files changed, 81 insertions, 25 deletions

diff --git a/Modules/_sre.c b/Modules/_sre.c
index 3a92db9..55a86c2 100644
--- a/Modules/_sre.c
+++ b/Modules/_sre.c
@@ -4,24 +4,25 @@
  * regular expression matching engine
  *
  * partial history:
- * 1999-10-24 fl  created (based on existing template matcher code)
- * 2000-03-06 fl  first alpha, sort of
- * 2000-08-01 fl  fixes for 1.6b1
- * 2000-08-07 fl  use PyOS_CheckStack() if available
- * 2000-09-20 fl  added expand method
- * 2001-03-20 fl  lots of fixes for 2.1b2
- * 2001-04-15 fl  export copyright as Python attribute, not global
- * 2001-04-28 fl  added __copy__ methods (work in progress)
- * 2001-05-14 fl  fixes for 1.5.2 compatibility
- * 2001-07-01 fl  added BIGCHARSET support (from Martin von Loewis)
- * 2001-10-18 fl  fixed group reset issue (from Matthew Mueller)
- * 2001-10-20 fl  added split primitive; reenable unicode for 1.6/2.0/2.1
- * 2001-10-21 fl  added sub/subn primitive
- * 2001-10-24 fl  added finditer primitive (for 2.2 only)
- * 2001-12-07 fl  fixed memory leak in sub/subn (Guido van Rossum)
- * 2002-11-09 fl  fixed empty sub/subn return type
- * 2003-04-18 mvl fully support 4-byte codes
- * 2003-10-17 gn  implemented non recursive scheme
+ * 1999-10-24 fl   created (based on existing template matcher code)
+ * 2000-03-06 fl   first alpha, sort of
+ * 2000-08-01 fl   fixes for 1.6b1
+ * 2000-08-07 fl   use PyOS_CheckStack() if available
+ * 2000-09-20 fl   added expand method
+ * 2001-03-20 fl   lots of fixes for 2.1b2
+ * 2001-04-15 fl   export copyright as Python attribute, not global
+ * 2001-04-28 fl   added __copy__ methods (work in progress)
+ * 2001-05-14 fl   fixes for 1.5.2 compatibility
+ * 2001-07-01 fl   added BIGCHARSET support (from Martin von Loewis)
+ * 2001-10-18 fl   fixed group reset issue (from Matthew Mueller)
+ * 2001-10-20 fl   added split primitive; reenable unicode for 1.6/2.0/2.1
+ * 2001-10-21 fl   added sub/subn primitive
+ * 2001-10-24 fl   added finditer primitive (for 2.2 only)
+ * 2001-12-07 fl   fixed memory leak in sub/subn (Guido van Rossum)
+ * 2002-11-09 fl   fixed empty sub/subn return type
+ * 2003-04-18 mvl  fully support 4-byte codes
+ * 2003-10-17 gn   implemented non recursive scheme
+ * 2013-02-04 mrab added fullmatch primitive
  *
  * Copyright (c) 1997-2001 by Secret Labs AB.  All rights reserved.
  *
@@ -559,6 +560,40 @@ pattern_match(PatternObject* self, PyObject* args, PyObject* kw)
 }
 
 static PyObject*
+pattern_fullmatch(PatternObject* self, PyObject* args, PyObject* kw)
+{
+    SRE_STATE state;
+    Py_ssize_t status;
+
+    PyObject* string;
+    Py_ssize_t start = 0;
+    Py_ssize_t end = PY_SSIZE_T_MAX;
+    static char* kwlist[] = { "pattern", "pos", "endpos", NULL };
+    if (!PyArg_ParseTupleAndKeywords(args, kw, "O|nn:fullmatch", kwlist,
+                                     &string, &start, &end))
+        return NULL;
+
+    string = state_init(&state, self, string, start, end);
+    if (!string)
+        return NULL;
+
+    state.match_all = 1;
+    state.ptr = state.start;
+
+    TRACE(("|%p|%p|FULLMATCH\n", PatternObject_GetCode(self), state.ptr));
+
+    status = sre_match(&state, PatternObject_GetCode(self));
+
+    TRACE(("|%p|%p|END\n", PatternObject_GetCode(self), state.ptr));
+    if (PyErr_Occurred())
+        return NULL;
+
+    state_fini(&state);
+
+    return pattern_new_match(self, &state, status);
+}
+
+static PyObject*
 pattern_search(PatternObject* self, PyObject* args, PyObject* kw)
 {
     SRE_STATE state;
@@ -1223,6 +1258,10 @@ PyDoc_STRVAR(pattern_match_doc,
 "match(string[, pos[, endpos]]) -> match object or None.\n\
     Matches zero or more characters at the beginning of the string");
 
+PyDoc_STRVAR(pattern_fullmatch_doc,
+"fullmatch(string[, pos[, endpos]]) -> match object or None.\n\
+    Matches against all of the string");
+
 PyDoc_STRVAR(pattern_search_doc,
 "search(string[, pos[, endpos]]) -> match object or None.\n\
     Scan through string looking for a match, and return a corresponding\n\
@@ -1258,6 +1297,8 @@ PyDoc_STRVAR(pattern_doc, "Compiled regular expression objects");
 static PyMethodDef pattern_methods[] = {
     {"match", (PyCFunction) pattern_match, METH_VARARGS|METH_KEYWORDS,
         pattern_match_doc},
+    {"fullmatch", (PyCFunction) pattern_fullmatch, METH_VARARGS|METH_KEYWORDS,
+        pattern_fullmatch_doc},
     {"search", (PyCFunction) pattern_search, METH_VARARGS|METH_KEYWORDS,
         pattern_search_doc},
     {"sub", (PyCFunction) pattern_sub, METH_VARARGS|METH_KEYWORDS,
diff --git a/Modules/sre.h b/Modules/sre.h
index 42fe28d..621e2d8 100644
--- a/Modules/sre.h
+++ b/Modules/sre.h
@@ -86,6 +86,7 @@ typedef struct {
     SRE_REPEAT *repeat;
     /* hooks */
     SRE_TOLOWER_HOOK lower;
+    int match_all;
 } SRE_STATE;
 
 typedef struct {
diff --git a/Modules/sre_lib.h b/Modules/sre_lib.h
index 214c22a..df86697 100644
--- a/Modules/sre_lib.h
+++ b/Modules/sre_lib.h
@@ -454,17 +454,24 @@ do { \
 #define JUMP_ASSERT          12
 #define JUMP_ASSERT_NOT      13
 
-#define DO_JUMP(jumpvalue, jumplabel, nextpattern) \
+#define DO_JUMPX(jumpvalue, jumplabel, nextpattern, matchall) \
     DATA_ALLOC(SRE(match_context), nextctx); \
     nextctx->last_ctx_pos = ctx_pos; \
     nextctx->jump = jumpvalue; \
     nextctx->pattern = nextpattern; \
+    nextctx->match_all = matchall; \
     ctx_pos = alloc_pos; \
     ctx = nextctx; \
     goto entrance; \
     jumplabel: \
     while (0) /* gcc doesn't like labels at end of scopes */ \
 
+#define DO_JUMP(jumpvalue, jumplabel, nextpattern) \
+    DO_JUMPX(jumpvalue, jumplabel, nextpattern, ctx->match_all)
+
+#define DO_JUMP0(jumpvalue, jumplabel, nextpattern) \
+    DO_JUMPX(jumpvalue, jumplabel, nextpattern, 0)
+
 typedef struct {
     Py_ssize_t last_ctx_pos;
     Py_ssize_t jump;
@@ -477,6 +484,7 @@ typedef struct {
         SRE_CODE chr;
         SRE_REPEAT* rep;
     } u;
+    int match_all;
 } SRE(match_context);
 
 /* check if string matches the given pattern.  returns <0 for
@@ -499,6 +507,7 @@ SRE(match)(SRE_STATE* state, SRE_CODE* pattern)
     ctx->last_ctx_pos = -1;
     ctx->jump = JUMP_NONE;
     ctx->pattern = pattern;
+    ctx->match_all = state->match_all;
     ctx_pos = alloc_pos;
 
 entrance:
@@ -571,8 +580,11 @@ entrance:
         case SRE_OP_SUCCESS:
             /* end of pattern */
             TRACE(("|%p|%p|SUCCESS\n", ctx->pattern, ctx->ptr));
-            state->ptr = ctx->ptr;
-            RETURN_SUCCESS;
+            if (!ctx->match_all || ctx->ptr == state->end) {
+                state->ptr = ctx->ptr;
+                RETURN_SUCCESS;
+            }
+            RETURN_FAILURE;
 
         case SRE_OP_AT:
             /* match at given position */
@@ -726,7 +738,8 @@ entrance:
             if (ctx->count < (Py_ssize_t) ctx->pattern[1])
                 RETURN_FAILURE;
 
-            if (ctx->pattern[ctx->pattern[0]] == SRE_OP_SUCCESS) {
+            if (ctx->pattern[ctx->pattern[0]] == SRE_OP_SUCCESS &&
+                (!ctx->match_all || ctx->ptr == state->end)) {
                 /* tail is empty.  we're finished */
                 state->ptr = ctx->ptr;
                 RETURN_SUCCESS;
@@ -810,7 +823,8 @@ entrance:
                 ctx->ptr += ctx->count;
             }
 
-            if (ctx->pattern[ctx->pattern[0]] == SRE_OP_SUCCESS) {
+            if (ctx->pattern[ctx->pattern[0]] == SRE_OP_SUCCESS &&
+                (!ctx->match_all || ctx->ptr == state->end)) {
                 /* tail is empty.  we're finished */
                 state->ptr = ctx->ptr;
                 RETURN_SUCCESS;
@@ -1082,7 +1096,7 @@ entrance:
             state->ptr = ctx->ptr - ctx->pattern[1];
             if (state->ptr < state->beginning)
                 RETURN_FAILURE;
-            DO_JUMP(JUMP_ASSERT, jump_assert, ctx->pattern+2);
+            DO_JUMP0(JUMP_ASSERT, jump_assert, ctx->pattern+2);
             RETURN_ON_FAILURE(ret);
             ctx->pattern += ctx->pattern[0];
             break;
@@ -1094,7 +1108,7 @@ entrance:
                    ctx->ptr, ctx->pattern[1]));
             state->ptr = ctx->ptr - ctx->pattern[1];
             if (state->ptr >= state->beginning) {
-                DO_JUMP(JUMP_ASSERT_NOT, jump_assert_not, ctx->pattern+2);
+                DO_JUMP0(JUMP_ASSERT_NOT, jump_assert_not, ctx->pattern+2);
                 if (ret) {
                     RETURN_ON_ERROR(ret);
                     RETURN_FAILURE;