protected static void removeDefinedCustomXmlParts( WordprocessingMLPackage wmlPackage, String xpathStorageItemId) { List<PartName> partsToRemove = new ArrayList<PartName>(); RelationshipsPart relationshipsPart = wmlPackage.getMainDocumentPart().getRelationshipsPart(); List<Relationship> relationshipsList = ((relationshipsPart != null) && (relationshipsPart.getRelationships() != null) ? relationshipsPart.getRelationships().getRelationship() : null); Part part = null; CustomXmlDataStoragePart dataPart = null; if (relationshipsList != null) { for (Relationship relationship : relationshipsList) { if (Namespaces.CUSTOM_XML_DATA_STORAGE.equals(relationship.getType())) { part = relationshipsPart.getPart(relationship); if (IsPartToRemove(part, xpathStorageItemId)) { partsToRemove.add(part.getPartName()); } } } } if (!partsToRemove.isEmpty()) { for (int i = 0; i < partsToRemove.size(); i++) { relationshipsPart.removePart(partsToRemove.get(i)); } } }
public static void visit( WordprocessingMLPackage wmlPackage, boolean bodyOnly, Callback callback) { MainDocumentPart mainDocument = null; RelationshipsPart relPart = null; List<Relationship> relList = null; List<Object> elementList = null; if ((wmlPackage != null) && (callback != null)) { mainDocument = wmlPackage.getMainDocumentPart(); callback.walkJAXBElements(mainDocument.getJaxbElement().getBody()); if (!bodyOnly) { relPart = mainDocument.getRelationshipsPart(); relList = relPart.getRelationships().getRelationship(); for (Relationship rs : relList) { elementList = null; if (Namespaces.HEADER.equals(rs.getType())) { elementList = ((HeaderPart) relPart.getPart(rs)).getJaxbElement().getContent(); } else if (Namespaces.FOOTER.equals(rs.getType())) { elementList = ((FooterPart) relPart.getPart(rs)).getJaxbElement().getContent(); } else if (Namespaces.ENDNOTES.equals(rs.getType())) { // elementList = ((EndnotesPart) relPart.getPart(rs)).getContent(); elementList = new ArrayList(); elementList.addAll(((EndnotesPart) relPart.getPart(rs)).getJaxbElement().getEndnote()); } else if (Namespaces.FOOTNOTES.equals(rs.getType())) { // elementList = ((FootnotesPart) relPart.getPart(rs)).getContent(); elementList = new ArrayList(); elementList.addAll( ((FootnotesPart) relPart.getPart(rs)).getJaxbElement().getFootnote()); } else if (Namespaces.COMMENTS.equals(rs.getType())) { elementList = new ArrayList(); for (Comment comment : ((CommentsPart) relPart.getPart(rs)).getJaxbElement().getComment()) { elementList.addAll(comment.getEGBlockLevelElts()); } } if ((elementList != null) && (!elementList.isEmpty())) { log.debug("Processing target: " + rs.getTarget() + ", type: " + rs.getType()); callback.walkJAXBElements(elementList); } } } } }
protected static Set<ContentAccessor> getParts(WordprocessingMLPackage srcPackage) { Set<ContentAccessor> partList = new HashSet<ContentAccessor>(); partList.add(srcPackage.getMainDocumentPart()); // Add headers/footers RelationshipsPart rp = srcPackage.getMainDocumentPart().getRelationshipsPart(); for (Relationship r : rp.getRelationships().getRelationship()) { if (r.getType().equals(Namespaces.HEADER)) { partList.add((HeaderPart) rp.getPart(r)); } else if (r.getType().equals(Namespaces.FOOTER)) { partList.add((FooterPart) rp.getPart(r)); } } return partList; }
/** * Traverse the document, and return a map of all styles which are used directly in the document. * (IE this does not include styles on which others are just BasedOn). * * @return */ public Set<String> getStylesInUse() { org.docx4j.wml.Document wmlDocumentEl = (org.docx4j.wml.Document) this.getJaxbElement(); Body body = wmlDocumentEl.getBody(); List<Object> bodyChildren = body.getContent(); Set<String> stylesInUse = new HashSet<String>(); FontAndStyleFinder finder = new FontAndStyleFinder(null, null, stylesInUse); finder.defaultCharacterStyle = this.getStyleDefinitionsPart().getDefaultCharacterStyle(); finder.defaultParagraphStyle = this.getStyleDefinitionsPart().getDefaultParagraphStyle(); new TraversalUtil(bodyChildren, finder); finder.finish(); // Styles in headers, footers? RelationshipsPart rp = this.getRelationshipsPart(); if (rp != null) { for (Relationship r : rp.getRelationships().getRelationship()) { Part part = rp.getPart(r); if (part instanceof FooterPart) { Ftr ftr = ((FooterPart) part).getJaxbElement(); finder.walkJAXBElements(ftr); } else if (part instanceof HeaderPart) { Hdr hdr = ((HeaderPart) part).getJaxbElement(); finder.walkJAXBElements(hdr); } } } // Styles in endnotes, footnotes? if (this.getEndNotesPart() != null) { log.debug("Looking at endnotes"); CTEndnotes endnotes = this.getEndNotesPart().getJaxbElement(); finder.walkJAXBElements(endnotes); } if (this.getFootnotesPart() != null) { log.debug("Looking at footnotes"); CTFootnotes footnotes = this.getFootnotesPart().getJaxbElement(); finder.walkJAXBElements(footnotes); } // Comments if (this.getCommentsPart() != null) { log.debug("Looking at comments"); Comments comments = this.getCommentsPart().getJaxbElement(); finder.walkJAXBElements(comments); } return stylesInUse; }
protected static boolean IsPartToRemove(Part part, String xpathStorageItemId) { boolean ret = false; RelationshipsPart relationshipsPart = part.getRelationshipsPart(); List<Relationship> relationshipsList = ((relationshipsPart != null) && (relationshipsPart.getRelationships() != null) ? relationshipsPart.getRelationships().getRelationship() : null); CustomXmlDataStoragePropertiesPart propertiesPart = null; DatastoreItem datastoreItem = null; if ((relationshipsList != null) && (!relationshipsList.isEmpty())) { for (Relationship relationship : relationshipsList) { if (Namespaces.CUSTOM_XML_DATA_STORAGE_PROPERTIES.equals(relationship.getType())) { propertiesPart = (CustomXmlDataStoragePropertiesPart) relationshipsPart.getPart(relationship); break; } } } if (propertiesPart != null) { datastoreItem = propertiesPart.getJaxbElement(); } if (datastoreItem != null) { if ((datastoreItem.getItemID() != null) && (datastoreItem.getItemID().length() > 0)) { ret = datastoreItem.getItemID().equals(xpathStorageItemId); } if ((!ret) && (datastoreItem.getSchemaRefs() != null) && (datastoreItem.getSchemaRefs().getSchemaRef() != null) && (!datastoreItem.getSchemaRefs().getSchemaRef().isEmpty())) { for (SchemaRef ref : datastoreItem.getSchemaRefs().getSchemaRef()) { if (PART_TO_REMOVE_SCHEMA_TYPES.contains(ref.getUri())) { ret = true; break; } } } } return ret; }
/** * Traverse the document, looking for fonts which have been applied, either directly, or via a * style. * * @return */ public Set<String> fontsInUse() { log.info("fontsInUse.."); getPropertyResolver(); // this inits our virtual DocDefaults style // Setup Set<String> fontsDiscovered = new java.util.HashSet<String>(); // // Keep track of styles we encounter, so we can // // inspect these for fonts // Set<String> stylesInUse = new java.util.HashSet<String>(); // // org.docx4j.wml.Styles styles = null; // if (this.getStyleDefinitionsPart()!=null) { // styles = (org.docx4j.wml.Styles)this.getStyleDefinitionsPart().getJaxbElement(); // } // // It is convenient to have a HashMap of styles // Map<String, Style> stylesDefined = new java.util.HashMap<String, Style>(); // if (styles!=null) { // for (Iterator<Style> iter = styles.getStyle().iterator(); iter.hasNext();) { // Style s = iter.next(); // stylesDefined.put(s.getStyleId(), s); // } // } // // We need to know what fonts and styles are used in the document org.docx4j.wml.Document wmlDocumentEl = (org.docx4j.wml.Document) this.getJaxbElement(); Body body = wmlDocumentEl.getBody(); List<Object> bodyChildren = body.getContent(); FontDiscoveryCharacterVisitor visitor = new FontDiscoveryCharacterVisitor(fontsDiscovered); RunFontSelector runFontSelector = new RunFontSelector( (WordprocessingMLPackage) this.pack, visitor, RunFontActionType.DISCOVERY); FontAndStyleFinder finder = new FontAndStyleFinder(runFontSelector, fontsDiscovered, null); finder.defaultCharacterStyle = this.getStyleDefinitionsPart().getDefaultCharacterStyle(); finder.defaultParagraphStyle = this.getStyleDefinitionsPart().getDefaultParagraphStyle(); new TraversalUtil(bodyChildren, finder); // finder.finish(); fontsDiscovered.add(runFontSelector.getDefaultFont()); // fonts in headers, footers? RelationshipsPart rp = this.getRelationshipsPart(); if (rp != null) { for (Relationship r : rp.getRelationships().getRelationship()) { Part part = rp.getPart(r); if (part instanceof FooterPart) { Ftr ftr = ((FooterPart) part).getJaxbElement(); finder.walkJAXBElements(ftr); } else if (part instanceof HeaderPart) { Hdr hdr = ((HeaderPart) part).getJaxbElement(); finder.walkJAXBElements(hdr); } } } // Styles in endnotes, footnotes? if (this.getEndNotesPart() != null) { log.debug("Looking at endnotes"); CTEndnotes endnotes = this.getEndNotesPart().getJaxbElement(); finder.walkJAXBElements(endnotes); } if (this.getFootnotesPart() != null) { log.debug("Looking at footnotes"); CTFootnotes footnotes = this.getFootnotesPart().getJaxbElement(); finder.walkJAXBElements(footnotes); } // Comments if (this.getCommentsPart() != null) { log.debug("Looking at comments"); Comments comments = this.getCommentsPart().getJaxbElement(); finder.walkJAXBElements(comments); } // Add fonts used in the styles we discovered // .. 2013 03 10: no longer necessary // Fonts can also be used in the numbering part // For now, treat any font mentioned in that part as in use. // Ideally, we'd only register fonts used in numbering levels // that were actually used in the document if (getNumberingDefinitionsPart() != null) { Numbering numbering = getNumberingDefinitionsPart().getJaxbElement(); for (Numbering.AbstractNum abstractNumNode : numbering.getAbstractNum()) { for (Lvl lvl : abstractNumNode.getLvl()) { if (lvl.getRPr() != null && lvl.getRPr().getRFonts() != null) { String fontName = lvl.getRPr().getRFonts().getAscii(); if (fontName != null) { fontsDiscovered.add(fontName); log.debug( "Registered " + fontName + " for abstract list " + abstractNumNode.getAbstractNumId() + " lvl " + lvl.getIlvl()); } } } } } return fontsDiscovered; }
/* recursively (i) get each Part listed in the relationships (ii) add the Part to the zip file (iii) traverse its relationship */ public void addPartsFromRelationships(ZipOutputStream out, RelationshipsPart rp) throws Docx4JException { // for (Iterator it = rp.iterator(); it.hasNext(); ) { // Relationship r = (Relationship)it.next(); // log.info("For Relationship Id=" + r.getId() + " Source is " + r.getSource().getPartName() + // ", Target is " + r.getTargetURI() ); for (Relationship r : rp.getRelationships().getRelationship()) { log.debug( "For Relationship Id=" + r.getId() + " Source is " + rp.getSourceP().getPartName() + ", Target is " + r.getTarget()); if (r.getType().equals(Namespaces.HYPERLINK)) { continue; // whether internal or external } if (r.getTargetMode() != null && r.getTargetMode().equals("External")) { // ie its EXTERNAL // As at 1 May 2008, we don't have a Part for these; // there is just the relationship. log.warn("Encountered external resource " + r.getTarget() + " of type " + r.getType()); // So continue; } try { // String resolvedPartUri = URIHelper.resolvePartUri(r.getSourceURI(), r.getTargetURI() // ).toString(); String resolvedPartUri = URIHelper.resolvePartUri(rp.getSourceURI(), new URI(r.getTarget())).toString(); // Now drop leading "/' resolvedPartUri = resolvedPartUri.substring(1); // Now normalise it .. ie abc/def/../ghi // becomes abc/ghi // Maybe this isn't necessary with a zip file, // - ZipFile class may be smart enough to do it. // But it is certainly necessary in the JCR case. // target = (new java.net.URI(target)).normalize().toString(); // log.info("Normalised, it is " + target ); // Document contents = getDocumentFromZippedPart( zf, target); if (!false) { log.debug("Getting part /" + resolvedPartUri); // Part part = p.getParts().get(new PartName("/" + resolvedPartUri)); Part part = rp.getPart(r); // 2012 09 26: If the part is actually attached to // a different package, using this you can still get it. // Use this 'feature' at your own risk! if (part == null) { log.error("Part " + resolvedPartUri + " not found!"); } else { log.debug(part.getClass().getName()); } if (!part.getPackage().equals(p)) { log.warn("Part " + resolvedPartUri + " is attached to some other package"); } savePart(out, part); } } catch (Exception e) { throw new Docx4JException( "Failed to add parts from relationships of " + rp.getSourceP().getPartName(), e); } } }