Bug 696699: Fix Text extraction mediabox information.

Since the removal of the begin_page device function, structured
text extraction has been unable to correctly establish the
mediabox for extracted pages.

Update the fz_new_stext_page call to take this mediabox
information. This is an API change, but hopefully most people
are calling fz_new_stext_page_from_page or
fz_new_stext_page_from_display_list which are updated here to
cope.

Update all the apps/tools to behave properly.

5 files changed, 15 insertions, 8 deletions

diff --git a/platform/android/viewer/jni/mupdf.c b/platform/android/viewer/jni/mupdf.c
index b70d3f1a..91a002e3 100644
--- a/platform/android/viewer/jni/mupdf.c
+++ b/platform/android/viewer/jni/mupdf.c
@@ -1251,13 +1251,15 @@ JNI_FN(MuPDFCore_searchPage)(JNIEnv * env, jobject thiz, jstring jtext)
 
 	fz_try(ctx)
 	{
+		fz_rect mediabox;
+
 		if (glo->hit_bbox == NULL)
 			glo->hit_bbox = fz_malloc_array(ctx, MAX_SEARCH_HITS, sizeof(*glo->hit_bbox));
 
 		zoom = glo->resolution / 72;
 		fz_scale(&ctm, zoom, zoom);
 		sheet = fz_new_stext_sheet(ctx);
-		text = fz_new_stext_page(ctx);
+		text = fz_new_stext_page(ctx, fz_bound_page(ctx, page, &mediabox));
 		dev = fz_new_stext_device(ctx, sheet, text);
 		fz_run_page(ctx, pc->page, dev, &ctm, NULL);
 		fz_drop_device(ctx, dev);
@@ -1342,12 +1344,13 @@ JNI_FN(MuPDFCore_text)(JNIEnv * env, jobject thiz)
 
 	fz_try(ctx)
 	{
+		fz_rect mediabox;
 		int b, l, s, c;
 
 		zoom = glo->resolution / 72;
 		fz_scale(&ctm, zoom, zoom);
 		sheet = fz_new_stext_sheet(ctx);
-		text = fz_new_stext_page(ctx);
+		text = fz_new_stext_page(ctx, fz_bound_page(ctx, page, &mediabox));
 		dev = fz_new_stext_device(ctx, sheet, text);
 		fz_run_page(ctx, pc->page, dev, &ctm, NULL);
 		fz_drop_device(ctx, dev);
@@ -1451,11 +1454,12 @@ JNI_FN(MuPDFCore_textAsHtml)(JNIEnv * env, jobject thiz)
 
 	fz_try(ctx)
 	{
+		fz_rect mediabox;
 		int b, l, s, c;
 
 		ctm = fz_identity;
 		sheet = fz_new_stext_sheet(ctx);
-		text = fz_new_stext_page(ctx);
+		text = fz_new_stext_page(ctx, fz_bound_page(ctx, page, &mediabox));
 		dev = fz_new_stext_device(ctx, sheet, text);
 		fz_run_page(ctx, pc->page, dev, &ctm, NULL);
 		fz_drop_device(ctx, dev);
diff --git a/platform/ios/Classes/MuPageViewNormal.m b/platform/ios/Classes/MuPageViewNormal.m
index b9da24e5..c56ad840 100644
--- a/platform/ios/Classes/MuPageViewNormal.m
+++ b/platform/ios/Classes/MuPageViewNormal.m
@@ -72,10 +72,11 @@ static NSArray *enumerateWords(fz_document *doc, fz_page *page)
 
 	fz_try(ctx)
 	{
+		fz_rect mediabox;
 		int b, l, c;
 
 		sheet = fz_new_stext_sheet(ctx);
-		text = fz_new_stext_page(ctx);
+		text = fz_new_stext_page(ctx, fz_bound_page(ctx, page, &mediabox));
 		dev = fz_new_stext_device(ctx, sheet, text);
 		fz_run_page(ctx, page, dev, &fz_identity, NULL);
 		fz_drop_device(ctx, dev);
diff --git a/platform/ios/Classes/MuPageViewReflow.m b/platform/ios/Classes/MuPageViewReflow.m
index 9378da61..86905b03 100644
--- a/platform/ios/Classes/MuPageViewReflow.m
+++ b/platform/ios/Classes/MuPageViewReflow.m
@@ -11,6 +11,7 @@ NSString *textAsHtml(fz_document *doc, int pageNum)
 	fz_matrix ctm;
 	fz_buffer *buf = NULL;
 	fz_output *out = NULL;
+	fz_rect mediabox;
 
 	fz_var(page);
 	fz_var(sheet);
@@ -23,7 +24,7 @@ NSString *textAsHtml(fz_document *doc, int pageNum)
 	{
 		ctm = fz_identity;
 		sheet = fz_new_stext_sheet(ctx);
-		text = fz_new_stext_page(ctx);
+		text = fz_new_stext_page(ctx, fz_bound_page(ctx, page, &mediabox));
 		dev = fz_new_stext_device(ctx, sheet, text);
 		page = fz_load_page(ctx, doc, pageNum);
 		fz_run_page(ctx, page, dev, &ctm, NULL);
diff --git a/platform/ios/common.m b/platform/ios/common.m
index 6ee3653f..80c2c738 100644
--- a/platform/ios/common.m
+++ b/platform/ios/common.m
@@ -20,9 +20,9 @@ static fz_rect hit_bbox[500];
 int search_page(fz_document *doc, int number, char *needle, fz_cookie *cookie)
 {
 	fz_page *page = fz_load_page(ctx, doc, number);
-
+	fz_rect mediabox;
 	fz_stext_sheet *sheet = fz_new_stext_sheet(ctx);
-	fz_stext_page *text = fz_new_stext_page(ctx);
+	fz_stext_page *text = fz_new_stext_page(ctx, fz_bound_page(ctx, page, &mediabox));
 	fz_device *dev = fz_new_stext_device(ctx, sheet, text);
 	fz_run_page(ctx, page, dev, &fz_identity, cookie);
 	fz_drop_device(ctx, dev);
diff --git a/platform/x11/pdfapp.c b/platform/x11/pdfapp.c
index 431c9994..e3e9ccc5 100644
--- a/platform/x11/pdfapp.c
+++ b/platform/x11/pdfapp.c
@@ -843,6 +843,7 @@ static void pdfapp_showpage(pdfapp_t *app, int loadpage, int drawpage, int repai
 
 	if (loadpage)
 	{
+		fz_rect mediabox;
 		pdfapp_loadpage(app, searching);
 
 		/* Zero search hit position */
@@ -850,7 +851,7 @@ static void pdfapp_showpage(pdfapp_t *app, int loadpage, int drawpage, int repai
 
 		/* Extract text */
 		app->page_sheet = fz_new_stext_sheet(app->ctx);
-		app->page_text = fz_new_stext_page(app->ctx);
+		app->page_text = fz_new_stext_page(app->ctx, fz_bound_page(app->ctx, app->page, &mediabox));
 
 		if (app->page_list || app->annotations_list)
 		{