Remove clip handling from generic surface layer.

Handling clip as part of the surface state, as opposed to being part of
the operation state, is cumbersome and a hindrance to providing true proxy
surface support. For example, the clip must be copied from the surface
onto the fallback image, but this was forgotten causing undue hassle in
each backend. Another example is the contortion the meta surface
endures to ensure the clip is correctly recorded. By contrast passing the
clip along with the operation is quite simple and enables us to write
generic handlers for providing surface wrappers. (And in the future, we
should be able to write more esoteric wrappers, e.g. automatic 2x FSAA,
trivially.)

In brief, instead of the surface automatically applying the clip before
calling the backend, the backend can call into a generic helper to apply
clipping. For raster surfaces, clip regions are handled automatically as
part of the composite interface. For vector surfaces, a clip helper is
introduced to replay and callback into an intersect_clip_path() function
as necessary.

Whilst this is not primarily a performance related change (the change
should just move the computation of the clip from the moment it is applied
by the user to the moment it is required by the backend), it is important
to track any potential regression:

ppc:
Speedups
========
image-rgba         evolution-20090607-0    1026085.22 0.18% -> 672972.07 0.77%:  1.52x speedup
▌
image-rgba         evolution-20090618-0    680579.98 0.12% -> 573237.66  0.16%:  1.19x speedup
▎
image-rgba      swfdec-fill-rate-4xaa-0    460296.92 0.36% -> 407464.63  0.42%:  1.13x speedup
▏
image-rgba      swfdec-fill-rate-2xaa-0    128431.95 0.47% -> 115051.86  0.42%:  1.12x speedup
▏
Slowdowns
=========
image-rgba     firefox-periodic-table-0    56837.61 0.78% -> 66055.17    3.20%:  1.09x slowdown
▏

4 files changed, 92 insertions, 30 deletions

diff --git a/util/cairo-script/cairo-script-file.c b/util/cairo-script/cairo-script-file.c
index 18b5b862..2a7296d2 100644
--- a/util/cairo-script/cairo-script-file.c
+++ b/util/cairo-script/cairo-script-file.c
@@ -37,6 +37,7 @@
 #include <stdio.h>
 #include <limits.h> /* INT_MAX */
 #include <string.h>
+#include <zlib.h>
 
 #define CHUNK_SIZE 32768
 
@@ -148,17 +149,43 @@ csi_file_new_from_string (csi_t *ctx,
     csi_file_t *file;
 
     file = _csi_slab_alloc (ctx, sizeof (csi_file_t));
-    if (file == NULL)
+    if (_csi_unlikely (file == NULL))
 	return _csi_error (CAIRO_STATUS_NO_MEMORY);
 
     file->base.type = CSI_OBJECT_TYPE_FILE;
     file->base.ref = 1;
 
-    file->type = BYTES;
-    file->src = src; src->base.ref++;
-    file->data = src->string;
-    file->bp   = file->data;
-    file->rem  = src->len;
+    if (src->deflate) {
+	uLongf len = src->deflate;
+	csi_object_t tmp_obj;
+	csi_string_t *tmp_str;
+	csi_status_t status;
+
+	status = csi_string_new (ctx, &tmp_obj,  NULL, src->deflate);
+	if (_csi_unlikely (status))
+	    return status;
+
+	tmp_str = tmp_obj.datum.string;
+	if (uncompress ((Bytef *) tmp_str->string, &len,
+			(Bytef *) src->string, src->len) != Z_OK)
+	{
+	    csi_string_free (ctx, tmp_str);
+	    _csi_slab_free (ctx, file, sizeof (csi_file_t));
+	    return _csi_error (CAIRO_STATUS_NO_MEMORY);
+	}
+
+	file->type = BYTES;
+	file->src = tmp_str;
+	file->data = tmp_str->string;
+	file->bp   = file->data;
+	file->rem  = tmp_str->len;
+    } else {
+	file->type = BYTES;
+	file->src = src; src->base.ref++;
+	file->data = src->string;
+	file->bp   = file->data;
+	file->rem  = src->len;
+    }
 
     obj->type = CSI_OBJECT_TYPE_FILE;
     obj->datum.file = file;
diff --git a/util/cairo-script/cairo-script-objects.c b/util/cairo-script/cairo-script-objects.c
index 784376b2..9cc59c90 100644
--- a/util/cairo-script/cairo-script-objects.c
+++ b/util/cairo-script/cairo-script-objects.c
@@ -507,6 +507,7 @@ csi_string_new (csi_t *ctx,
 	string->string[len] = '\0';
     }
     string->len = len;
+    string->deflate = 0;
 
     string->base.type = CSI_OBJECT_TYPE_STRING;
     string->base.ref = 1;
@@ -518,6 +519,26 @@ csi_string_new (csi_t *ctx,
 }
 
 csi_status_t
+csi_string_deflate_new (csi_t *ctx,
+			csi_object_t *obj,
+			void *bytes,
+			int in_len,
+			int out_len)
+{
+    csi_status_t status;
+    csi_string_t *string;
+
+    status = csi_string_new (ctx, obj, bytes, in_len);
+    if (_csi_unlikely (status))
+	return status;
+
+    string = obj->datum.string;
+    string->deflate = out_len;
+
+    return CSI_STATUS_SUCCESS;
+}
+
+csi_status_t
 csi_string_new_from_bytes (csi_t *ctx,
 	                   csi_object_t *obj,
 			   char *bytes,
@@ -534,6 +555,7 @@ csi_string_new_from_bytes (csi_t *ctx,
 
     string->string = bytes;
     string->len = len;
+    string->deflate = 0;
 
     string->base.type = CSI_OBJECT_TYPE_STRING;
     string->base.ref = 1;
diff --git a/util/cairo-script/cairo-script-private.h b/util/cairo-script/cairo-script-private.h
index 996b40d7..54a840f2 100644
--- a/util/cairo-script/cairo-script-private.h
+++ b/util/cairo-script/cairo-script-private.h
@@ -386,6 +386,7 @@ struct _csi_matrix {
 struct _csi_string {
     csi_compound_object_t base;
     csi_integer_t len;
+    csi_integer_t deflate;
     char *string;
 };
 
@@ -435,7 +436,6 @@ struct _csi_scanner {
     csi_stack_t procedure_stack;
     csi_object_t build_procedure;
 
-    int string_p;
     unsigned int accumulator;
     unsigned int accumulator_count;
 
@@ -757,6 +757,13 @@ csi_string_new (csi_t *ctx,
 		int len);
 
 csi_private csi_status_t
+csi_string_deflate_new (csi_t *ctx,
+			csi_object_t *obj,
+			void *bytes,
+			int in_len,
+			int out_len);
+
+csi_private csi_status_t
 csi_string_new_from_bytes (csi_t *ctx,
 	                   csi_object_t *obj,
 			   char *bytes,
diff --git a/util/cairo-script/cairo-script-scanner.c b/util/cairo-script/cairo-script-scanner.c
index b3217cd7..3cc39575 100644
--- a/util/cairo-script/cairo-script-scanner.c
+++ b/util/cairo-script/cairo-script-scanner.c
@@ -39,6 +39,7 @@
 #include <stdio.h> /* EOF */
 #include <string.h> /* memset */
 #include <assert.h>
+#include <zlib.h>
 
 #define DEBUG_SCAN 0
 
@@ -488,18 +489,6 @@ token_end (csi_t *ctx, csi_scanner_t *scan, csi_file_t *src)
 }
 
 static void
-string_inc_p (csi_scanner_t *scan)
-{
-    scan->string_p++;
-}
-
-static int
-string_dec_p (csi_scanner_t *scan)
-{
-    return --scan->string_p == 0;
-}
-
-static void
 string_add (csi_t *ctx, csi_scanner_t *scan, int c)
 {
     buffer_check (ctx, scan, 1);
@@ -616,7 +605,7 @@ base85_add (csi_t *ctx, csi_scanner_t *scan, int c)
 }
 
 static void
-base85_end (csi_t *ctx, csi_scanner_t *scan)
+base85_end (csi_t *ctx, csi_scanner_t *scan, cairo_bool_t deflate)
 {
     csi_object_t obj;
     cairo_status_t status;
@@ -647,12 +636,24 @@ base85_end (csi_t *ctx, csi_scanner_t *scan)
 	break;
     }
 
-    status = csi_string_new (ctx,
-			     &obj,
-			     scan->buffer.base,
-			     scan->buffer.ptr - scan->buffer.base);
-    if (_csi_unlikely (status))
-	longjmp (scan->jmpbuf, status);
+    if (deflate) {
+	uLongf len = *(uint32_t *) scan->buffer.base;
+	Bytef *source = (Bytef *) (scan->buffer.base + sizeof (uint32_t));
+
+	status = csi_string_deflate_new (ctx, &obj,
+					 source,
+					 (Bytef *) scan->buffer.ptr - source,
+					 len);
+	if (_csi_unlikely (status))
+	    longjmp (scan->jmpbuf, status);
+    } else {
+	status = csi_string_new (ctx,
+				 &obj,
+				 scan->buffer.base,
+				 scan->buffer.ptr - scan->buffer.base);
+	if (_csi_unlikely (status))
+	    longjmp (scan->jmpbuf, status);
+    }
 
     if (scan->build_procedure.type != CSI_OBJECT_TYPE_NULL)
 	status = csi_array_append (ctx,
@@ -714,6 +715,8 @@ _scan_file (csi_t *ctx, csi_file_t *src)
 	uint32_t u32;
 	float f;
     } u;
+    int deflate = 0;
+    int string_p;
 
 scan_none:
     while ((c = csi_file_getc (src)) != EOF) {
@@ -757,6 +760,8 @@ scan_none:
 		token_add_unchecked (scan, '<');
 		token_end (ctx, scan, src);
 		goto scan_none;
+	    case '|':
+		deflate = 1;
 	    case '~':
 		goto scan_base85;
 	    default:
@@ -1000,7 +1005,7 @@ scan_comment:
 
 scan_string:
     buffer_reset (&scan->buffer);
-    scan->string_p = 1;
+    string_p = 1;
     while ((c = csi_file_getc (src)) != EOF) {
 	switch (c) {
 	case '\\': /* escape */
@@ -1086,12 +1091,12 @@ scan_string:
 	    break;
 
 	case '(':
-	    string_inc_p (scan);
+	    string_p++;
 	    string_add (ctx, scan, c);
 	    break;
 
 	case ')':
-	    if (string_dec_p (scan)) {
+	    if (--string_p == 0) {
 		string_end (ctx, scan);
 		goto scan_none;
 	    }
@@ -1166,7 +1171,8 @@ scan_base85:
 		return;
 
 	    case '>':
-		base85_end (ctx, scan);
+		base85_end (ctx, scan, deflate);
+		deflate = 0;
 		goto scan_none;
 	    }
 	    csi_file_putc (src, next);