feat(电子公文): 支持标引解析与原文定位

ofdgo_editor_structure.go

@@ -0,0 +1,248 @@
+// Copyright 2025-2026 肖其顿 (XIAO QI DUN)
+//
+// Licensed under the Apache License, Version 2.0 (the "License");
+// you may not use this file except in compliance with the License.
+// You may obtain a copy of the License at
+//
+// http://www.apache.org/licenses/LICENSE-2.0
+//
+// Unless required by applicable law or agreed to in writing, software
+// distributed under the License is distributed on an "AS IS" BASIS,
+// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+// See the License for the specific language governing permissions and
+// limitations under the License.
+
+package ofdgo
+
+import (
+	"bytes"
+	"context"
+	"encoding/xml"
+	"fmt"
+	"path"
+	"strings"
+)
+
+// pruneStructureReferences 清理本次编辑删除的公文图元引用，未知标引和原有悬空引用保持原样
+// 改动进入输出快照，由既有文档及版本隔离流程维护共享文件
+// 入参: parts 输出覆盖条目
+// 返回: error 读取、解析或取消错误
+func (e *Editor) pruneStructureReferences(parts map[string][]byte) error {
+	if e.source.document.CustomTags.Path == "" && len(e.source.document.CustomTags.CustomTag) == 0 {
+		return nil
+	}
+	ctx := context.Background()
+	if e.output != nil {
+		ctx = e.output.ctx
+	}
+	pages := make(map[string]bool, len(e.pages))
+	for _, page := range e.pages {
+		pages[editorResourceID(page.ID)] = true
+	}
+	removedPages := make(map[string]bool)
+	removedObjects := make(map[string]map[string]bool)
+	for id, source := range e.source.pages {
+		key := editorResourceID(id)
+		if !pages[key] {
+			removedPages[key] = true
+			continue
+		}
+		data, changed := parts[e.source.reader.ResPath(source.ref.BaseLoc)]
+		if !changed || source.original == nil {
+			continue
+		}
+		root, err := parseEditorXMLContext(ctx, data)
+		if err != nil {
+			return err
+		}
+		remaining := make(map[string]bool)
+		var scan func(*editorXML)
+		scan = func(node *editorXML) {
+			if classifyOFDNamespace(node.name.Space) == ofdXMLUnknown {
+				return
+			}
+			if id := editorResourceID(node.attr("ID")); id != "" {
+				remaining[id] = true
+			}
+			for _, child := range node.children {
+				scan(child)
+			}
+		}
+		scan(root)
+		for id := range source.nodes {
+			id = editorResourceID(id)
+			if id != "" && !remaining[id] {
+				if removedObjects[key] == nil {
+					removedObjects[key] = make(map[string]bool)
+				}
+				removedObjects[key][id] = true
+			}
+		}
+	}
+	if len(removedPages) == 0 && len(removedObjects) == 0 {
+		return nil
+	}
+	tags, err := e.source.reader.CustomTags()
+	if err != nil {
+		return err
+	}
+	for _, tag := range tags {
+		if !structureOfficial(tag, tag.NameSpace) {
+			continue
+		}
+		name := e.source.reader.ResPath(tag.FileLoc)
+		data, ok := parts[name]
+		if !ok {
+			data, err = e.source.reader.readFileViewContext(ctx, name)
+			if err != nil {
+				return err
+			}
+		}
+		updated, err := rewriteStructureReferences(ctx, data, tag, func(page string, ids []string) (string, []string) {
+			key := editorResourceID(page)
+			if removedPages[key] {
+				return page, nil
+			}
+			kept := make([]string, 0, len(ids))
+			for _, id := range ids {
+				if !removedObjects[key][editorResourceID(id)] {
+					kept = append(kept, id)
+				}
+			}
+			return page, kept
+		})
+		if err != nil {
+			return err
+		}
+		if !bytes.Equal(data, updated) {
+			parts[name] = updated
+		}
+	}
+	return nil
+}
+
+// rewriteStructureReferences 仅改写已识别公文命名空间的叶子引用，保留其他XML字节
+// 入参: ctx 取消上下文, data 原始标引, tag 标引入口, rewrite 引用映射，空标识列表表示删除引用
+// 返回: []byte 修改后的标引, error 解析或取消错误
+func rewriteStructureReferences(ctx context.Context, data []byte, tag CustomTag, rewrite func(string, []string) (string, []string)) ([]byte, error) {
+	root, err := parseEditorXMLContext(ctx, data)
+	if err != nil {
+		return nil, err
+	}
+	if !structureOfficial(tag, root.name.Space) {
+		return data, nil
+	}
+	var patches []editorXMLPatch
+	count := 0
+	var walk func(*editorXML, int) error
+	walk = func(node *editorXML, depth int) error {
+		if err := ctx.Err(); err != nil {
+			return err
+		}
+		count++
+		if depth > 128 || count > 100000 {
+			return fmt.Errorf("custom tag structure limit exceeded")
+		}
+		if node.name.Space != root.name.Space {
+			return nil
+		}
+		if node.name.Local == "ObjectRef" && len(node.children) == 0 {
+			value := strings.TrimSpace(editorImportText(data, node))
+			ids := structureObjectIDs(value)
+			if len(ids) == 0 {
+				return nil
+			}
+			page, next := rewrite(node.attr("PageRef"), ids)
+			if len(next) == 0 {
+				patches = append(patches, editorXMLPatch{start: node.start, end: node.end})
+				return nil
+			}
+			if strings.Join(next, ",") != strings.Join(ids, ",") {
+				var text bytes.Buffer
+				_ = xml.EscapeText(&text, []byte(strings.Join(next, ",")))
+				patches = append(patches, editorXMLContent(data, node, text.Bytes()))
+			}
+			if page != node.attr("PageRef") {
+				patch, err := editorXMLAttributePatch(data, node, "PageRef", page)
+				if err != nil {
+					return err
+				}
+				patches = append(patches, patch)
+			}
+			return nil
+		}
+		for _, child := range node.children {
+			if err := walk(child, depth+1); err != nil {
+				return err
+			}
+		}
+		return nil
+	}
+	if err := walk(root, 1); err != nil {
+		return nil, err
+	}
+	if len(patches) == 0 {
+		return data, nil
+	}
+	return editorPatchXML(data, patches), nil
+}
+
+// structures 迁移所选页面的公文标引，按已迁移图元映射引用，不猜测其他行业的引用规则
+// 返回: []byte 标引索引条目, error 读取、解析或取消错误
+func (m *editorPageImport) structures() ([]byte, error) {
+	tags, err := m.reader.CustomTags()
+	if err != nil {
+		return nil, err
+	}
+	var entries []byte
+	for i, tag := range tags {
+		if !structureOfficial(tag, tag.NameSpace) {
+			continue
+		}
+		data, err := m.readFile(m.reader.ResPath(tag.FileLoc))
+		if err != nil {
+			return nil, err
+		}
+		count := 0
+		updated, err := rewriteStructureReferences(m.ctx, data, tag, func(page string, ids []string) (string, []string) {
+			key := editorImportKey(page)
+			if !m.pages[key] {
+				return page, nil
+			}
+			var mapped []string
+			for _, id := range ids {
+				if value := m.ids[editorImportKey(id)]; value != "" {
+					mapped = append(mapped, value)
+				}
+			}
+			count += len(mapped)
+			return m.ids[key], mapped
+		})
+		if err != nil {
+			return nil, err
+		}
+		if count == 0 {
+			continue
+		}
+		name := packageAvailableName(m.target, m.files, path.Join(m.directory, "Tags", fmt.Sprintf("Tag_%d.xml", i)))
+		m.files[name] = updated
+		var content []byte
+		if tag.SchemaLoc != "" {
+			schema, err := m.copyData(m.reader.ResPath(tag.SchemaLoc), path.Join(m.directory, "Tags", path.Base(tag.SchemaLoc)), false)
+			if err != nil {
+				return nil, err
+			}
+			content = append(content, editorXMLText("SchemaLoc", "/"+strings.TrimPrefix(schema, "/"))...)
+		}
+		content = append(content, editorXMLText("FileLoc", "/"+name)...)
+		var attrs ofdAttrs
+		attrs.add("TypeID", tag.TypeID)
+		attrs.add("NameSpace", tag.NameSpace)
+		entry, err := editorXMLContainer("CustomTag", attrs, content)
+		if err != nil {
+			return nil, err
+		}
+		entries = append(entries, entry...)
+	}
+	return entries, nil
+}