Buffers: format string check rewrite up to standards of the old one

diff --git a/Cython/Compiler/Buffer.py b/Cython/Compiler/Buffer.py
index 5ee83eb4ed471aaa340bbcf683c9a204c1d13530..f9cfb06fca3c5be8a7fc5d0beb47c87561635a8b 100644 (file)
--- a/Cython/Compiler/Buffer.py
+++ b/Cython/Compiler/Buffer.py
@@ -595,7 +595,7 @@ def get_type_information_cname(code, dtype, maxdepth=None):
                                    dtype=dtype, maxdepth=maxdepth)
     return name
 
-def type_information_code(proto, impl, name, structinfo_name, dtype):
+def type_information_code(proto, impl, name, structinfo_name, dtype, maxdepth):
     # Output the run-time type information (__Pyx_TypeInfo) for given dtype.
     # Use through get_type_information_cname
     #
@@ -754,7 +754,8 @@ static void __Pyx_BufFmt_Init(__Pyx_BufFmt_Context* ctx,
   ctx->root.offset = 0;
   ctx->head = stack;
   ctx->head->field = &ctx->root;
-  ctx->fmt_offset = ctx->head->parent_offset = 0;
+  ctx->fmt_offset = 0;
+  ctx->head->parent_offset = 0;
   ctx->packmode = '@';
   ctx->new_count = 1;
   ctx->enc_count = 0;
@@ -763,6 +764,7 @@ static void __Pyx_BufFmt_Init(__Pyx_BufFmt_Context* ctx,
   while (type->typegroup == 'S') {
     ++ctx->head;
     ctx->head->field = type->fields;
+    ctx->head->parent_offset = 0;
     type = type->fields->type;
   }
 }
@@ -920,8 +922,10 @@ static void __Pyx_BufFmt_RaiseExpected(__Pyx_BufFmt_Context* ctx) {
 }
 
 static int __Pyx_BufFmt_ProcessTypeChunk(__Pyx_BufFmt_Context* ctx) {
-  char group = __Pyx_BufFmt_TypeCharToGroup(ctx->enc_type, ctx->is_complex);
+  char group;
   size_t size, offset;
+  if (ctx->enc_type == 0) return 0;
+  group = __Pyx_BufFmt_TypeCharToGroup(ctx->enc_type, ctx->is_complex);
   do {
     __Pyx_StructField* field = ctx->head->field;
     __Pyx_TypeInfo* type = field->type;
@@ -942,8 +946,10 @@ static int __Pyx_BufFmt_ProcessTypeChunk(__Pyx_BufFmt_Context* ctx) {
     if (type->size != size || type->typegroup != group) {
       if (type->typegroup == 'C' && type->fields != NULL) {
         /* special case -- treat as struct rather than complex number */
+        size_t parent_offset = ctx->head->parent_offset + field->offset;
         ++ctx->head;
         ctx->head->field = type->fields;
+        ctx->head->parent_offset = parent_offset;
         continue;
       }
     
@@ -952,12 +958,12 @@ static int __Pyx_BufFmt_ProcessTypeChunk(__Pyx_BufFmt_Context* ctx) {
     }
 
     offset = ctx->head->parent_offset + field->offset;
-/*    if (ctx->fmt_offset != offset) {
+    if (ctx->fmt_offset != offset) {
       PyErr_Format(PyExc_ValueError,
                    "Buffer dtype mismatch; next field is at offset %"PY_FORMAT_SIZE_T"d "
                    "but %"PY_FORMAT_SIZE_T"d expected", ctx->fmt_offset, offset);
       return -1;
-    }*/
+    }
 
     ctx->fmt_offset += size;
   
@@ -1001,13 +1007,11 @@ static const char* __Pyx_BufFmt_CheckString(__Pyx_BufFmt_Context* ctx, const cha
   while (1) {
     switch(*ts) {
       case 0:
-        if (ctx->enc_type != 0) {
-          if (ctx->head == NULL) {
-            __Pyx_BufFmt_RaiseExpected(ctx);
-            return NULL;
-          }
-          if (__Pyx_BufFmt_ProcessTypeChunk(ctx) == -1) return NULL;
+        if (ctx->enc_type != 0 && ctx->head == NULL) {
+          __Pyx_BufFmt_RaiseExpected(ctx);
+          return NULL;
         }
+        if (__Pyx_BufFmt_ProcessTypeChunk(ctx) == -1) return NULL;
         if (ctx->head != NULL) {
           __Pyx_BufFmt_RaiseExpected(ctx);
           return NULL;
@@ -1063,7 +1067,14 @@ static const char* __Pyx_BufFmt_CheckString(__Pyx_BufFmt_Context* ctx, const cha
       case '}': /* end of substruct; either repeat or move on */
         ++ts;
         return ts;
-
+      case 'x':
+        if (__Pyx_BufFmt_ProcessTypeChunk(ctx) == -1) return NULL;
+        ctx->fmt_offset += ctx->new_count;
+        ctx->new_count = 1;
+        ctx->enc_count = 0;
+        ctx->enc_type = 0;
+        ++ts;
+        break;
       case 'Z':
         got_Z = 1;
         ++ts;
@@ -1080,11 +1091,7 @@ static const char* __Pyx_BufFmt_CheckString(__Pyx_BufFmt_Context* ctx, const cha
           ctx->enc_count += ctx->new_count;
         } else {
           /* New type */
-          if (ctx->enc_type != 0) {
-            if (__Pyx_BufFmt_ProcessTypeChunk(ctx) == -1) {
-              return NULL;
-            }
-          }
+          if (__Pyx_BufFmt_ProcessTypeChunk(ctx) == -1) return NULL;
           ctx->enc_count = ctx->new_count;
           ctx->enc_type = *ts;
           ctx->is_complex = got_Z;

diff --git a/Cython/Includes/numpy.pxd b/Cython/Includes/numpy.pxd
index 3aa2220515836e852bfa892c6726d953a824bffb..d5b122748c333bf0bcd9f4f0ae26a3bddfe57d05 100644 (file)
--- a/Cython/Includes/numpy.pxd
+++ b/Cython/Includes/numpy.pxd
@@ -272,18 +272,18 @@ cdef inline char* _util_dtypestring(dtype descr, char* f, char* end, int* offset
         if (end - f) - (new_offset - offset[0]) < 15: # this should leave room for "T{" and "}" as well
             raise RuntimeError("Format string allocated too short, see comment in numpy.pxd")
 
-        new_byteorder = child.byteorder
-        if new_byteorder == '|': new_byteorder = '='
+#        new_byteorder = child.byteorder
+#        if new_byteorder == '|': new_byteorder = '='
 #        if byteorder[0] != new_byteorder:
 #            f[0] = new_byteorder
 #            f += 1
 #            byteorder[0] = new_byteorder
 
         # Output padding bytes
-        while offset[0] < new_offset:
-            f[0] = 120 # "x"; pad byte
-            f += 1
-            offset[0] += 1
+#        while offset[0] < new_offset:
+#            f[0] = 120 # "x"; pad byte
+#            f += 1
+#            offset[0] += 1
 
         offset[0] += child.itemsize
             

diff --git a/tests/run/buffmt.pyx b/tests/run/buffmt.pyx
index 891c1535665af9b33a2d472126db5e8e2fc73eef..245e41f94596748371601d056cc580e320b77741 100644 (file)
--- a/tests/run/buffmt.pyx
+++ b/tests/run/buffmt.pyx
@@ -154,6 +154,20 @@ def char3int(fmt):
     >>> char3int("ci2i")
     >>> char3int("c@i@2i")
 
+    Extra pad bytes (assuming int size is 4 or more)
+    >>> char3int("cxiii")
+    >>> char3int("c3xiii")
+    >>> char3int("cxxxiii")
+
+    Standard alignment (assming int size is 4)
+    >>> char3int("=c3xiii")
+    >>> char3int("=cxxx@iii")
+    >>> char3int("=ciii")
+    Traceback (most recent call last):
+        ...
+    ValueError: Buffer dtype mismatch; next field is at offset 1 but 4 expected
+    
+    Error:
     >>> char3int("cii")
     Traceback (most recent call last):
        ...