From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
From: Daniel Veillard <veillard@redhat.com>
Date: Sun, 29 Oct 2017 01:04:54 +0200
Subject: [PATCH] Make generate-id deterministic

Origin: upstream, https://bugzilla.gnome.org/attachment.cgi?id=306475
Bug: https://bugzilla.gnome.org/show_bug.cgi?id=751621
Bug-Debian: https://bugs.debian.org/823857
---
 libxslt/functions.c     | 91 ++++++++++++++++++++++++++++++++++++++++-
 libxslt/functions.h     |  7 ++++
 libxslt/transform.c     |  8 ++++
 libxslt/xsltInternals.h |  2 +
 4 files changed, 107 insertions(+), 1 deletion(-)

diff --git a/libxslt/functions.c b/libxslt/functions.c
index 7887dda70f67..be0d897bd471 100644
--- a/libxslt/functions.c
+++ b/libxslt/functions.c
@@ -683,6 +683,63 @@ xsltFormatNumberFunction(xmlXPathParserContextPtr ctxt, int nargs)
     xmlXPathFreeObject(decimalObj);
 }
 
+/**
+ * xsltCleanupIds:
+ * @ctxt: the transformation context
+ * @root: the root of the resulting document
+ *
+ * This clean up ids which may have been saved in Element contents
+ * by xsltGenerateIdFunction() to provide stable IDs on elements.
+ *
+ * Returns the number of items cleaned or -1 in case of error
+ */
+int
+xsltCleanupIds(xsltTransformContextPtr ctxt, xmlNodePtr root) {
+    xmlNodePtr cur;
+    int count = 0;
+
+    if ((ctxt == NULL) || (root == NULL))
+        return(-1);
+    if (root->type != XML_ELEMENT_NODE)
+        return(-1);
+
+    cur = root;
+    while (cur != NULL) {
+	if (cur->type == XML_ELEMENT_NODE) {
+	    if (cur->content != NULL) {
+	        cur->content = NULL;
+		count++;
+	    }
+	    if (cur->children != NULL) {
+		cur = cur->children;
+		continue;
+	    }
+	}
+	if (cur->next != NULL) {
+	    cur = cur->next;
+	    continue;
+	}
+	do {
+	    cur = cur->parent;
+	    if (cur == NULL)
+		break;
+	    if (cur == (xmlNodePtr) root) {
+		cur = NULL;
+		break;
+	    }
+	    if (cur->next != NULL) {
+		cur = cur->next;
+		break;
+	    }
+	} while (cur != NULL);
+    }
+
+fprintf(stderr, "Attributed %d IDs for element, cleaned up %d\n",
+        ctxt->nextid, count);
+
+    return(count);
+}
+
 /**
  * xsltGenerateIdFunction:
  * @ctxt:  the XPath Parser context
@@ -734,7 +791,39 @@ xsltGenerateIdFunction(xmlXPathParserContextPtr ctxt, int nargs){
     if (obj)
         xmlXPathFreeObject(obj);
 
-    val = (long)((char *)cur - (char *)&base_address);
+    /*
+     * Try to provide stable ID for generated document:
+     *   - usually ID are computed to be placed on elements via attributes
+     *     so using the element as the node for the ID
+     *   - the cur->content should be a correct placeholder for this, we use
+     *     it to hold element node numbers in xmlXPathOrderDocElems to
+     *     speed up XPath too
+     *   - xsltCleanupIds() clean them up before handing the XSLT output
+     *     to the API client.
+     *   - other nodes types use the node address method but that should
+     *     not end up in resulting document ID
+     *   - we can enable this by default without risk of performance issues
+     *     only the one pass xsltCleanupIds() is added
+     */
+    if (cur->type == XML_ELEMENT_NODE) {
+        if (cur->content == NULL) {
+	    xsltTransformContextPtr tctxt;
+
+	    tctxt = xsltXPathGetTransformContext(ctxt);
+	    if (tctxt == NULL) {
+		val = (long)((char *)cur - (char *)&base_address);
+	    } else {
+		tctxt->nextid++;
+		val = tctxt->nextid;
+		cur->content = (void *) (val);
+	    }
+	} else {
+	    val = (long) cur->content;
+	}
+    } else {
+	val = (long)((char *)cur - (char *)&base_address);
+    }
+
     if (val >= 0) {
       snprintf((char *)str, sizeof(str), "idp%ld", val);
     } else {
diff --git a/libxslt/functions.h b/libxslt/functions.h
index 5455b7f47802..31163613f6e9 100644
--- a/libxslt/functions.h
+++ b/libxslt/functions.h
@@ -63,6 +63,13 @@ XSLTPUBFUN void XSLTCALL
 	xsltFunctionAvailableFunction	(xmlXPathParserContextPtr ctxt,
 					 int nargs);
 
+/*
+ * Cleanup for ID generation
+ */
+XSLTPUBFUN int XSLTCALL
+	xsltCleanupIds			(xsltTransformContextPtr ctxt,
+					 xmlNodePtr root);
+
 /*
  * And the registration
  */
diff --git a/libxslt/transform.c b/libxslt/transform.c
index 57f05bf71920..9368e17197bf 100644
--- a/libxslt/transform.c
+++ b/libxslt/transform.c
@@ -706,6 +706,7 @@ xsltNewTransformContext(xsltStylesheetPtr style, xmlDocPtr doc) {
     cur->traceCode = (unsigned long*) &xsltDefaultTrace;
     cur->xinclude = xsltGetXIncludeDefault();
     cur->keyInitLevel = 0;
+    cur->nextid = 0;
 
     return(cur);
 
@@ -6038,6 +6039,13 @@ xsltApplyStylesheetInternal(xsltStylesheetPtr style, xmlDocPtr doc,
     if (root != NULL) {
         const xmlChar *doctype = NULL;
 
+        /*
+	 * cleanup ids which may have been saved in Elements content ptrs
+	 */
+	if (ctxt->nextid != 0) {
+	    xsltCleanupIds(ctxt, root);
+	}
+
         if ((root->ns != NULL) && (root->ns->prefix != NULL))
 	    doctype = xmlDictQLookup(ctxt->dict, root->ns->prefix, root->name);
 	if (doctype == NULL)
diff --git a/libxslt/xsltInternals.h b/libxslt/xsltInternals.h
index 14343d2751c8..9d3ff5fc0c06 100644
--- a/libxslt/xsltInternals.h
+++ b/libxslt/xsltInternals.h
@@ -1786,6 +1786,8 @@ struct _xsltTransformContext {
     int maxTemplateVars;
     unsigned long opLimit;
     unsigned long opCount;
+
+    unsigned long nextid;/* for generating stable ids */
 };
 
 /**