Simplify PDF resource caching table handling.

9 files changed, 78 insertions, 149 deletions

diff --git a/include/mupdf/fitz/buffer.h b/include/mupdf/fitz/buffer.h
index a616fa86..40e46e73 100644
--- a/include/mupdf/fitz/buffer.h
+++ b/include/mupdf/fitz/buffer.h
@@ -138,4 +138,9 @@ size_t fz_buffer_printf(fz_context *ctx, fz_buffer *buffer, const char *fmt, ...
 size_t fz_buffer_vprintf(fz_context *ctx, fz_buffer *buffer, const char *fmt, va_list args);
 void fz_buffer_print_pdf_string(fz_context *ctx, fz_buffer *buffer, const char *text);
 
+/*
+	fz_md5_buffer: create MD5 digest of buffer contents.
+*/
+void fz_md5_buffer(fz_context *ctx, fz_buffer *buffer, unsigned char digest[16]);
+
 #endif
diff --git a/include/mupdf/pdf/document.h b/include/mupdf/pdf/document.h
index 8a8fdd40..aabf05f9 100644
--- a/include/mupdf/pdf/document.h
+++ b/include/mupdf/pdf/document.h
@@ -12,7 +12,6 @@ typedef struct pdf_annot_s pdf_annot;
 typedef struct pdf_widget_s pdf_widget;
 typedef struct pdf_hotspot_s pdf_hotspot;
 typedef struct pdf_js_s pdf_js;
-typedef struct pdf_resource_tables_s pdf_resource_tables;
 
 enum
 {
@@ -266,7 +265,10 @@ struct pdf_document_s
 	int max_type3_fonts;
 	fz_font **type3_fonts;
 
-	pdf_resource_tables *resources;
+	struct {
+		fz_hash_table *images;
+		fz_hash_table *fonts;
+	} resources;
 };
 
 /*
diff --git a/include/mupdf/pdf/resource.h b/include/mupdf/pdf/resource.h
index e0290247..f9f99da6 100644
--- a/include/mupdf/pdf/resource.h
+++ b/include/mupdf/pdf/resource.h
@@ -16,29 +16,11 @@ void pdf_remove_item(fz_context *ctx, fz_store_drop_fn *drop, pdf_obj *key);
  * the data, enabling rapid lookup.
  */
 
-typedef struct pdf_res_table_s pdf_res_table;
-typedef pdf_obj *(pdf_res_search_fn)(fz_context *ctx, pdf_document *doc, pdf_res_table *list,
-	void *item, unsigned char md5[16]);
-
-struct pdf_res_table_s
-{
-	fz_hash_table *hash;
-	pdf_res_search_fn *search;
-};
-
-struct pdf_resource_tables_s
-{
-	pdf_res_table *image;
-	pdf_res_table *font;
-	pdf_res_table *color;
-	pdf_res_table *pattern;
-	pdf_res_table *shading;
-};
-
-pdf_obj *pdf_find_resource(fz_context *ctx, pdf_document *doc, pdf_res_table *table, void *item, unsigned char md5[16]);
-void pdf_init_resource_tables(fz_context *ctx, pdf_document *doc);
+pdf_obj *pdf_find_font_resource(fz_context *ctx, pdf_document *doc, fz_buffer *item, unsigned char md5[16]);
+pdf_obj *pdf_insert_font_resource(fz_context *ctx, pdf_document *doc, unsigned char md5[16], pdf_obj *obj);
+pdf_obj *pdf_find_image_resource(fz_context *ctx, pdf_document *doc, fz_image *item, unsigned char md5[16]);
+pdf_obj *pdf_insert_image_resource(fz_context *ctx, pdf_document *doc, unsigned char md5[16], pdf_obj *obj);
 void pdf_drop_resource_tables(fz_context *ctx, pdf_document *doc);
-pdf_obj *pdf_insert_resource(fz_context *ctx, pdf_res_table *table, void *key, pdf_obj *obj);
 
 /*
  * Functions, Colorspaces, Shadings and Images
diff --git a/scripts/cmapdump.c b/scripts/cmapdump.c
index 2f3493d7..4da2e329 100644
--- a/scripts/cmapdump.c
+++ b/scripts/cmapdump.c
@@ -17,6 +17,7 @@
 #include "../source/fitz/output.c"
 #include "../source/fitz/string.c"
 #include "../source/fitz/buffer.c"
+#include "../source/fitz/crypt-md5.c"
 #include "../source/fitz/stream-open.c"
 #include "../source/fitz/stream-read.c"
 #include "../source/fitz/strtod.c"
diff --git a/source/fitz/buffer.c b/source/fitz/buffer.c
index ba3451bf..10c0c055 100644
--- a/source/fitz/buffer.c
+++ b/source/fitz/buffer.c
@@ -390,6 +390,14 @@ fz_buffer_print_pdf_string(fz_context *ctx, fz_buffer *buffer, const char *text)
 	buffer->len += len;
 }
 
+void fz_md5_buffer(fz_context *ctx, fz_buffer *buffer, unsigned char digest[16])
+{
+	fz_md5 state;
+	fz_md5_init(&state);
+	fz_md5_update(&state, buffer->data, buffer->len);
+	fz_md5_final(&state, digest);
+}
+
 #ifdef TEST_BUFFER_WRITE
 
 #define TEST_LEN 1024
diff --git a/source/pdf/pdf-device.c b/source/pdf/pdf-device.c
index 2324dbd7..3b974efe 100644
--- a/source/pdf/pdf-device.c
+++ b/source/pdf/pdf-device.c
@@ -1127,11 +1127,6 @@ fz_device *pdf_new_pdf_device(fz_context *ctx, pdf_document *doc, const fz_matri
 
 		if (topctm != &fz_identity)
 			fz_buffer_printf(ctx, buf, "%M cm\n", topctm);
-
-		/* Initialise the resource tables if we have not done so
-		 * already. */
-		if (doc->resources == NULL)
-			pdf_init_resource_tables(ctx, doc);
 	}
 	fz_catch(ctx)
 	{
diff --git a/source/pdf/pdf-font.c b/source/pdf/pdf-font.c
index cc6c68ed..a0bdc5f9 100644
--- a/source/pdf/pdf-font.c
+++ b/source/pdf/pdf-font.c
@@ -2002,7 +2002,7 @@ pdf_add_cid_font(fz_context *ctx, pdf_document *doc, fz_font *font)
 		/* Before we add this font as a resource check if the same font
 		 * already exists in our resources for this doc. If yes, then
 		 * hand back that reference */
-		fref = pdf_find_resource(ctx, doc, doc->resources->font, font->buffer, digest);
+		fref = pdf_find_font_resource(ctx, doc, font->buffer, digest);
 		if (fref == NULL)
 		{
 			/* Set up desc, width, and font file */
@@ -2031,7 +2031,7 @@ pdf_add_cid_font(fz_context *ctx, pdf_document *doc, fz_font *font)
 			fref = pdf_add_object(ctx, doc, fobj);
 
 			/* Add ref to our font resource hash table. */
-			fref = pdf_insert_resource(ctx, doc->resources->font, digest, fref);
+			fref = pdf_insert_font_resource(ctx, doc, digest, fref);
 		}
 	}
 	fz_always(ctx)
@@ -2077,7 +2077,7 @@ pdf_add_simple_font(fz_context *ctx, pdf_document *doc, fz_font *font)
 		/* Before we add this font as a resource check if the same font
 		 * already exists in our resources for this doc. If yes, then
 		 * hand back that reference */
-		fref = pdf_find_resource(ctx, doc, doc->resources->font, font->buffer, digest);
+		fref = pdf_find_font_resource(ctx, doc, font->buffer, digest);
 		if (fref == NULL)
 		{
 			fobj = pdf_new_dict(ctx, doc, 10);
@@ -2119,7 +2119,7 @@ pdf_add_simple_font(fz_context *ctx, pdf_document *doc, fz_font *font)
 			fref = pdf_add_object(ctx, doc, fobj);
 
 			/* Add ref to our font resource hash table. */
-			fref = pdf_insert_resource(ctx, doc->resources->font, digest, fref);
+			fref = pdf_insert_font_resource(ctx, doc, digest, fref);
 		}
 	}
 	fz_always(ctx)
diff --git a/source/pdf/pdf-image.c b/source/pdf/pdf-image.c
index 0b8bd6b1..c9c1fdda 100644
--- a/source/pdf/pdf-image.c
+++ b/source/pdf/pdf-image.c
@@ -310,7 +310,7 @@ pdf_add_image(fz_context *ctx, pdf_document *doc, fz_image *image, int mask)
 		/* Before we add this image as a resource check if the same image
 		 * already exists in our resources for this doc.  If yes, then
 		 * hand back that reference */
-		imref = pdf_find_resource(ctx, doc, doc->resources->image, image, digest);
+		imref = pdf_find_image_resource(ctx, doc, image, digest);
 		if (imref == NULL)
 		{
 			if (cbuffer != NULL && cbuffer->params.type != FZ_IMAGE_PNG && cbuffer->params.type != FZ_IMAGE_TIFF)
@@ -457,7 +457,7 @@ pdf_add_image(fz_context *ctx, pdf_document *doc, fz_image *image, int mask)
 			pdf_update_stream(ctx, doc, imref, buffer, 1);
 
 			/* Add ref to our image resource hash table. */
-			imref = pdf_insert_resource(ctx, doc->resources->image, digest, imref);
+			imref = pdf_insert_image_resource(ctx, doc, digest, imref);
 		}
 	}
 	fz_always(ctx)
diff --git a/source/pdf/pdf-resources.c b/source/pdf/pdf-resources.c
index 9b539fa9..0772d3a4 100644
--- a/source/pdf/pdf-resources.c
+++ b/source/pdf/pdf-resources.c
@@ -1,29 +1,7 @@
 #include "mupdf/pdf.h"
 
 static void
-res_table_free(fz_context *ctx, pdf_res_table *table)
-{
-	int i, n;
-	pdf_obj *res;
-
-	if (table == NULL)
-		return;
-	if (table->hash != NULL)
-	{
-		n = fz_hash_len(ctx, table->hash);
-		for (i = 0; i < n; i++)
-		{
-			res = fz_hash_get_val(ctx, table->hash, i);
-			if (res)
-				pdf_drop_obj(ctx, res);
-		}
-		fz_drop_hash(ctx, table->hash);
-	}
-	fz_free(ctx, table);
-}
-
-static void
-res_image_get_md5(fz_context *ctx, fz_image *image, unsigned char *digest)
+fz_md5_image(fz_context *ctx, fz_image *image, unsigned char digest[16])
 {
 	fz_pixmap *pixmap;
 	fz_md5 state;
@@ -45,7 +23,7 @@ res_image_get_md5(fz_context *ctx, fz_image *image, unsigned char *digest)
 
 /* Image specific methods */
 static void
-res_image_init(fz_context *ctx, pdf_document *doc, pdf_res_table *table)
+pdf_preload_image_resources(fz_context *ctx, pdf_document *doc)
 {
 	int len, k;
 	pdf_obj *obj;
@@ -60,7 +38,6 @@ res_image_init(fz_context *ctx, pdf_document *doc, pdf_res_table *table)
 
 	fz_try(ctx)
 	{
-		table->hash = fz_new_hash_table(ctx, 4096, 16, -1);
 		len = pdf_count_objects(ctx, doc);
 		for (k = 1; k < len; k++)
 		{
@@ -69,18 +46,15 @@ res_image_init(fz_context *ctx, pdf_document *doc, pdf_res_table *table)
 			if (pdf_name_eq(ctx, type, PDF_NAME_Image))
 			{
 				image = pdf_load_image(ctx, doc, obj);
-				res_image_get_md5(ctx, image, digest);
+				fz_md5_image(ctx, image, digest);
 				fz_drop_image(ctx, image);
 				image = NULL;
 
 				/* Don't allow overwrites. */
-				if (fz_hash_find(ctx, table->hash, digest) == NULL)
-					fz_hash_insert(ctx, table->hash, digest, obj);
-			}
-			else
-			{
-				pdf_drop_obj(ctx, obj);
+				if (!fz_hash_find(ctx, doc->resources.images, digest))
+					fz_hash_insert(ctx, doc->resources.images, digest, pdf_keep_obj(ctx, obj));
 			}
+			pdf_drop_obj(ctx, obj);
 			obj = NULL;
 		}
 	}
@@ -91,126 +65,88 @@ res_image_init(fz_context *ctx, pdf_document *doc, pdf_res_table *table)
 	}
 	fz_catch(ctx)
 	{
-		res_table_free(ctx, table);
 		fz_rethrow(ctx);
 	}
 }
 
-static pdf_obj *
-res_image_search(fz_context *ctx, pdf_document *doc, pdf_res_table *table, void *item, unsigned char *digest)
+pdf_obj *
+pdf_find_image_resource(fz_context *ctx, pdf_document *doc, fz_image *item, unsigned char digest[16])
 {
-	fz_image *image = item;
-	fz_hash_table *hash = table->hash;
 	pdf_obj *res;
 
-	if (hash == NULL)
-		res_image_init(ctx, doc, doc->resources->image);
-	hash = doc->resources->image->hash;
+	if (!doc->resources.images)
+	{
+		doc->resources.images = fz_new_hash_table(ctx, 4096, 16, -1);
+		pdf_preload_image_resources(ctx, doc);
+	}
 
 	/* Create md5 and see if we have the item in our table */
-	res_image_get_md5(ctx, image, digest);
-	res = fz_hash_find(ctx, hash, digest);
+	fz_md5_image(ctx, item, digest);
+	res = fz_hash_find(ctx, doc->resources.images, digest);
 	if (res)
 		pdf_keep_obj(ctx, res);
 	return res;
 }
 
-/* Font specific methods */
-
-/* We do need to come up with an effective way to see what is already in the
- * file to avoid adding to what is already there. This is avoided for pdfwrite
- * as we check as we add each font.  For adding text to an existing file though
- * it may be more problematic */
-static void
-res_font_init(fz_context *ctx, pdf_document *doc, pdf_res_table *table)
+pdf_obj *
+pdf_insert_image_resource(fz_context *ctx, pdf_document *doc, unsigned char digest[15], pdf_obj *obj)
 {
-	table->hash = fz_new_hash_table(ctx, 4096, 16, -1);
+	pdf_obj *res = fz_hash_insert(ctx, doc->resources.images, digest, obj);
+	if (res)
+		fz_warn(ctx, "warning: image resource already present");
+	else
+		res = pdf_keep_obj(ctx, obj);
+	return res;
 }
 
-static void
-res_font_get_md5(fz_context *ctx, fz_buffer *buffer, unsigned char *digest)
-{
-	fz_md5 state;
-
-	fz_md5_init(&state);
-	fz_md5_update(&state, buffer->data, buffer->len);
-	fz_md5_final(&state, digest);
-}
+/* We do need to come up with an effective way to see what is already in the
+ * file to avoid adding to what is already there. This is avoided for pdfwrite
+ * as we check as we add each font. For adding text to an existing file though
+ * it may be more problematic. */
 
-static pdf_obj *
-res_font_search(fz_context *ctx, pdf_document *doc, pdf_res_table *table, void *item, unsigned char digest[16])
+pdf_obj *
+pdf_find_font_resource(fz_context *ctx, pdf_document *doc, fz_buffer *item, unsigned char digest[16])
 {
-	fz_buffer *buffer = item;
-	fz_hash_table *hash = table->hash;
 	pdf_obj *res;
 
-	if (hash == NULL)
-		res_font_init(ctx, doc, doc->resources->font);
-	hash = doc->resources->font->hash;
+	if (!doc->resources.fonts)
+		doc->resources.fonts = fz_new_hash_table(ctx, 4096, 16, -1);
 
 	/* Create md5 and see if we have the item in our table */
-	res_font_get_md5(ctx, buffer, digest);
-	res = fz_hash_find(ctx, hash, digest);
+	fz_md5_buffer(ctx, item, digest);
+	res = fz_hash_find(ctx, doc->resources.fonts, digest);
 	if (res)
 		pdf_keep_obj(ctx, res);
 	return res;
 }
 
-/* Accessible methods */
 pdf_obj *
-pdf_find_resource(fz_context *ctx, pdf_document *doc, pdf_res_table *table, void *item, unsigned char md5[16])
+pdf_insert_font_resource(fz_context *ctx, pdf_document *doc, unsigned char digest[15], pdf_obj *obj)
 {
-	return table->search(ctx, doc, table, item, md5);
+	pdf_obj *res = fz_hash_insert(ctx, doc->resources.fonts, digest, obj);
+	if (res)
+		fz_warn(ctx, "warning: font resource already present");
+	else
+		res = pdf_keep_obj(ctx, obj);
+	return res;
 }
 
-pdf_obj *
-pdf_insert_resource(fz_context *ctx, pdf_res_table *table, void *key, pdf_obj *obj)
+static void
+res_table_free(fz_context *ctx, fz_hash_table *hash)
 {
-	pdf_obj *res;
-
-	fz_try(ctx)
-	{
-		res = fz_hash_insert(ctx, table->hash, key, obj);
-		if (res != NULL)
-			fz_warn(ctx, "warning: resource already present");
-		else
-			res = pdf_keep_obj(ctx, obj);
-	}
-	fz_catch(ctx)
+	int i, n;
+	if (hash)
 	{
-		fz_rethrow(ctx);
+		n = fz_hash_len(ctx, hash);
+		for (i = 0; i < n; i++)
+			pdf_drop_obj(ctx, fz_hash_get_val(ctx, hash, i));
+		fz_drop_hash(ctx, hash);
 	}
-	return res;
 }
 
 void
 pdf_drop_resource_tables(fz_context *ctx, pdf_document *doc)
 {
-	if (doc->resources == NULL)
-		return;
-	res_table_free(ctx, doc->resources->color);
-	res_table_free(ctx, doc->resources->font);
-	res_table_free(ctx, doc->resources->image);
-	res_table_free(ctx, doc->resources->pattern);
-	res_table_free(ctx, doc->resources->shading);
-	fz_free(ctx, doc->resources);
-	doc->resources = NULL;
-}
-
-void
-pdf_init_resource_tables(fz_context *ctx, pdf_document *doc)
-{
-	fz_try(ctx)
-	{
-		doc->resources = fz_malloc_struct(ctx, pdf_resource_tables);
-		doc->resources->image = fz_malloc_struct(ctx, pdf_res_table);
-		doc->resources->image->search = res_image_search;
-		doc->resources->font = fz_malloc_struct(ctx, pdf_res_table);
-		doc->resources->font->search = res_font_search;
-	}
-	fz_catch(ctx)
-	{
-		pdf_drop_resource_tables(ctx, doc);
-		fz_rethrow(ctx);
-	}
+	res_table_free(ctx, doc->resources.fonts);
+	res_table_free(ctx, doc->resources.images);
 }