[Item] Generate note from annotations (by brackets tag)/根据标注注释内自定义的标签生成笔记-支持多级标题嵌套 (2025.4.20更新/updated)

[zhikaiyici]

更新日志

2025.4.20更新： 新增解析标注翻译，翻译内容形式为🔤翻译🔤（Translate for Zotero插件默认）。

2025.3.31更新：

1. 更改样式，取消引用块，评论与标注不分段，评论加粗显示；（可更改子模板中128行myAnnotation.content += ...语句将样式更改为引用块，即注释掉现有myAnnotation.content += ...语句，并取消注释myAnnotation.content += "<blockquote>" + theContent + "</blockquote>" + myLabelsAndComment.sectionComment;）
2. 拆分模板，解决使用// @use-refresh后标题显示为空，使用时需要将主模板和子模板都导入，生成笔记时选择主模板Generate note from annotations 。
3. 将带有Table/Figure（如 Table 1/Figure 1/Figure 2）标签的标注单独列出。

2025.1.24更新： 自动标题编号，编号顺序为标题在注释中第一次出现的顺序。

2024.11.21更新：

1. 支持同一条目下多个附件标注；
2. 使用允许刷新。（使用允许刷新后将在笔记开头生成一段“代码”用于更新笔记，如下图所示。如不需要更新笔记请删除模板中// @use-refresh。）
   

Description

修改于#1194 和#1129 ，支持了嵌套多级标签（标题）。期刊标签功能需要安装Style插件。

使用时需要将主模板和子模板都导入，生成笔记时选择主模板Generate note from annotations。
Import both the main template and sub-template, and choose the main template Generate note from annotations when generating a note.

不依赖于 Zotero 标注的标签或颜色，允许在标注的评论区内以方括号“[]”自定义标签，并以其为标题自动生成笔记。例如，在多个标注中分别添加"[Heading1]example1"，”[Heading1][Heading1.1]example1.1“，”[Heading2]example2“，等评论，使用将模板生成如下笔记：
It doesn't rely on annotation tags or colors, and allows users to add a comment with square brackets (“[” and “]”) to an annotation and use the bracketed comment as titles to generate a note. For example, add comments such as "[Heading1] Example 1, [Heading1][Heading1.1] Example 1.1, [Heading2] Example 2, etc., to multiple annotations, and use the template to generate the following note:

1 Heading1

annotation1

example1

1.1 Heading1.1

annotation1.1

example1.1

2 Heading2

annotation2

example2

Screenshots

标注示例

Test on Zotero version

7.0.11

Test on Better Notes version

2.2.5

Template Type

Item

Template Share Code

主模板：

# This template is specifically for importing/sharing, using better 
# notes 'import from clipboard': copy the content and
# goto Zotero menu bar, click Tools->New Template from Clipboard.  
# Do not copy-paste this to better notes template editor directly.
name: "[item]Generate note from annotations"
zoteroVersion: "7.0.15"
pluginVersion: "2.4.8"
savedAt: "2025-04-20T13:21:28.269Z"
content: |-
  // @author zhikaiyici
  // @link [PUBLISH PAGE URL](https://github.com/windingwind/zotero-better-notes/discussions/1129)
  // @beforeloop-begin
  ${await(async() => {
      // 辅助函数：检测字符串是否包含中文字符
      const containsChinese = (text) => /[\u4e00-\u9fa5]/.test(text);
  
      // 检查标题是否为中文
      let isChineseTitle = containsChinese(items[0].getField("title"));
  
      // 获取作者显示内容
      let creators = items[0].getCreators();
      let authorDisplay = ["", ""];
      if (isChineseTitle) {
          // 中文标题：显示第一作者全名，格式为“姓名”
          if (creators.length > 0) {
              let firstAuthor = creators[0];
              let fullName = firstAuthor.lastName + firstAuthor.firstName;
              if (creators.length > 1) {
                  authorDisplay[0] = fullName + "等";
              }
              else {
                  authorDisplay[0] = fullName;
              }
              // 正文中显示前10位作者，多于10位加等
              if (creators.length > 10) {
                  authorDisplay[1] = creators.splice(0, 10).map((v) => v.firstName + v.lastName).join("; ");
                  authorDisplay[1] = authorDisplay[1] + "等";
              }
              else {
                  authorDisplay[1] = creators.map((v) => v.firstName + v.lastName).join("; ");
              }
          }
      }
      else {
          // 非中文标题：显示第一作者姓氏并附加“et al.”
          if (creators.length > 0) {
              let firstAuthorLastName = creators[0].lastName;
                  if (creators.length > 1) {
                      authorDisplay[0] = `${firstAuthorLastName} et al.`;
                  }
                  else {
                      authorDisplay[0] = creators[0].lastName + " " + creators[0].firstName;
                  }
                  // 正文中显示前10位作者，多于10位加et al.
                  if (creators.length > 10) {
                      authorDisplay[1] = creators.splice(0, 10).map((v) => v.firstName + " " + v.lastName).join("; ");
                      authorDisplay[1] = authorDisplay[1] + "; et al.";
                  }
                  else {
                      authorDisplay[1] = creators.map((v) => v.firstName + " " + v.lastName).join("; ");
                  }
              }
      }
      // 非期刊条目显示论文集、大学、出版社、仓库...
      let otherName = items[0].getField("proceedingsTitle")
          + items[0].getField("university") + items[0].getField("publisher")
          + items[0].getField("repository") + items[0].getField("institution")
          + items[0].getField("meetingName");
      let publicationName = items[0].getField("publicationTitle") + otherName;
      let sourceName = items[0].getField("journalAbbreviation") + otherName;
      
      let res = `
          <h1> ${items[0].getField('title')} (${items[0].getField("year")}, ${sourceName ? sourceName + ", " : ""} ${authorDisplay[0]})</h1>
          <hr/>
          <table>
          <!--<p>-->
          <tr><td style="background-color:#f6f8fa;">
              <b><span style="color: #3c5acc">Author: </span></b> ${authorDisplay[1]}
          </td></tr>
          <!--</p>-->
  
          <!--<p>-->
          <tr><td style="background-color:#ecf3fa;"><b><span style="color: #3c5acc">Source: </span></b>
              ${(items[0].getField("volume") + items[0].getField("issue") + items[0].getField("pages")) ? 
              (publicationName + " (" + 
              (items[0].getField("volume") ? "Volume: " + items[0].getField("volume") + ", " : '') + 
              (items[0].getField("issue") ? "Issue: " + items[0].getField("issue") + ", " : '') + 
              (items[0].getField("pages") ? "Pages: " + items[0].getField("pages") : '') + ")") : publicationName}
              ${(() => {
                  if (Zotero.ZoteroStyle) {
                      let space = " ";
                      return Array.prototype.map.call(
                          Zotero.ZoteroStyle.api.renderCell(items[0], "publicationTags").childNodes,
                          e => {
                              e.innerText = space + e.innerText + space;
                              return e.outerHTML
                          }
                      ).join(space);
                  }
              })()}
          </td></tr>
          <!--</p>-->
  
          <!--<p>-->
          <tr><td style="background-color:#f6f8fa;">
              ${(() => {
                  const doi = items[0].getField("DOI");
                  if (doi) {
                      return `<b><span style="color: #3c5acc">DOI: </span></strong></b> 
                      <a href="https://doi.org/${items[0].getField('DOI')}">${items[0].getField('DOI')}</a>`;
                  } else {
                      return `<b><span style="color: #3c5acc">URL: </span></strong> </b> 
                      <a href="${items[0].getField('url')}">${items[0].getField('url')}</a>`;
                  }
             })()}
          </td></tr>
          <!--</p>-->
  
          <!--<p>-->
          <tr><td style="background-color:#ecf3fa;">
              <b><span style="color: #3c5acc">Date: </span></b> ${items[0].getField('date')}
          </td></tr>
          <!--</p>-->
  
          <!-- 本地链接 -->
          <!--<p>-->
          <tr><td style="background-color:#f6f8fa;">
              ${(() => {
                  const attachments = Zotero.Items.get(items[0].getAttachments());
                  if (attachments && attachments.length > 1) {
                      return `<b><span style="color: #3c5acc">Full Text: </span></b>
                      <a href=zotero://open-pdf/0_${Zotero.Items.get(items[0].getAttachments()).filter((i) => i.isPDFAttachment())[0].key}>
                      ${Zotero.Items.get(items[0].getAttachments()).filter((i) => i.isPDFAttachment())[0].getFilename()}</a>`;
                  } else {
                      if (attachments && attachments.length > 0) {
                          return `<b><span style="color: #3c5acc">Full Text: </span> </b> 
                          <a href="zotero://open-pdf/0_${attachments[0].key}">${attachments[0].getFilename()}</a>`;
                      }
                      else {
                          return `<b><span style="color: #3c5acc">Full Text: </span></b>`;
                      }
                  }
              })()}
          </td></tr>
          <!--</p>-->
  
          <!-- 摘要 -->
          <!--<p>-->
          <tr><td style="background-color:#ecf3fa;">
              <b><span style="color: #3c5acc">Abstract: </span></b> <i>${items[0].getField('abstractNote')}</i>
          </td></tr>
          <!--</p>-->
          </table>
  
          `;
      return res;
  })()}
  
  ${await Zotero.BetterNotes.api.template.runItemTemplate("[item]by brackets tag", {
      itemIds: items.map(item=>item.id),
      targetNoteId: targetNoteItem?.id,
      dryRun: _env.dryRun
  })}
  
  // @beforeloop-end
  // @default-begin
  // @default-end

子模板：

# This template is specifically for importing/sharing, using better 
# notes 'import from clipboard': copy the content and
# goto Zotero menu bar, click Tools->New Template from Clipboard.  
# Do not copy-paste this to better notes template editor directly.
name: "[item]by brackets tag"
zoteroVersion: "7.0.15"
pluginVersion: "2.4.8"
savedAt: "2025-04-20T13:24:19.382Z"
content: |-
  // @author zhikaiyici
  // @link [PUBLISH PAGE URL](https://github.com/windingwind/zotero-better-notes/discussions/1202)
  // @use-refresh
  ${ await(async() => {
      const getAnnotationLabelsAndComment = (_annotation) => {
          let result = { labels: [], sectionComment: "", translation: "" };
          let _comment = _annotation.annotationComment || "";
          let _tag = _annotation.getTags().map((tagObj) => tagObj.tag).join();
          let parsed = _comment.match(/^(.*?)🔤(.*?)🔤(.*?)$/s)
          if (parsed) {
              result.translation = "🔤" + parsed[2].trim() + "🔤";
              _comment = parsed[1].trim() + parsed[3].trim();;
          }
          parsed = _comment.match(/^\[(.*?)\](.*)/s);
          if (!parsed) {
              if (_tag.includes("Figure")) {
                  result.labels.push("Figures");
              }
              else if (_tag.includes("Table")) {
                  result.labels.push("Tables");
              }
              else {
                  result.labels.push("Unlabeled");
              }
              result.sectionComment = _comment;
          }
          else {
              while (parsed) {
                  result.labels.push(parsed[1].trim());
                  result.sectionComment = parsed[2].trim();
                  parsed = result.sectionComment.match(/^\[(.*?)\](.*)/);
              }
          };
          return result;
      }
  
      const colors = [
          ["#1565C0", "#E1F5FE"],
          ["#006064", "#E0F7FA"],
          ["#2E7D32", "#F1F8E9"],
          ["#E65100", "#FFF8E1"],
          ["#4A148C", "#F5F5F5"]
      ];
  
      const processAnnotations = (grouped, parentLevel) => {
          let result = "";
          let i = 0;
          // 遍历每个组
          for (const label in grouped) {
              if (label != "Unlabeled" && label != "Figures" && label != "Tables") {
                  i++;
                  const group = grouped[label];
                  let color = colors[group.level - 2] ? colors[group.level - 2]:["",""];//["#000000", "#FFFFFF"];
                  const heading = `<h${group.level} style="color:${color[0]}; background-color:${color[1]}">${(parentLevel ? parentLevel : "") + i + " " + label}</h${group.level}>`;
                  result = `${result}${heading}${group.content.filter(Boolean).join("")}`; // + "\n";
                  // 递归处理子标签
                  if (group.childLabels)
                      result = `${result}${processAnnotations(group.childLabels, (parentLevel ? parentLevel : "") + i + ".")}`;
              }
          }
          for (const label in grouped) {
              if (label === "Figures") {
                  i++;
                  const group = grouped[label];
                  const color = colors[0] ? colors[0]:["",""];
                  const heading = `<h${group.level}  style="color: ${color[0]}; background-color: ${color[1]}">${(parentLevel ? parentLevel : "") + i + " " + label}</h${group.level}>`;
                  result = `${result}${heading}${group.content.filter(Boolean).join("")}`; // + "\n";
                  break;
              }
          }
          for (const label in grouped) {
              if (label === "Tables") {
                  i++;
                  const group = grouped[label];
                  const color = colors[0] ? colors[0]:["",""];
                  const heading = `<h${group.level}  style="color: ${color[0]}; background-color: ${color[1]}">${(parentLevel ? parentLevel : "") + i + " " + label}</h${group.level}>`;
                  result = `${result}${heading}${group.content.filter(Boolean).join("")}`; // + "\n";
                  break;
              }
          }
          for (const label in grouped) {
              if (label === "Unlabeled") {
                  i++;
                  const group = grouped[label];
                  const color = colors[0] ? colors[0]:["",""];
                  const heading = `<h${group.level}  style="color: ${color[0]}; background-color: ${color[1]}">${(parentLevel ? parentLevel : "") + i + " " + label}</h${group.level}>`;
                  result = `${result}${heading}${group.content.filter(Boolean).join("")}`; // + "\n";
                  break;
              }
          }
          return result;
      }
  
      let myAnnotations = "";
      const attachments = Zotero.Items.get(topItem.getAttachments()).filter((i) => i.isPDFAttachment() || i.isSnapshotAttachment() || i.isEPUBAttachment());
      let annotatedAttachment = 0;
      for (let attachment of attachments) {
          if (attachment.getAnnotations().length > 0)
              annotatedAttachment++; 
      }
  
      let groupedAnnotations = {};
      const processedAnnotations = await Promise.all(
          attachments.map((attachment, attachmentIndex) => {
              const title = `<i>📄For Document: <a href="zotero://open-pdf/0_${attachment.key}">${attachment.getField("title")}</a></i>`;
  
              return Promise.all(
                  attachment.getAnnotations().map(async (annoItem, annotIndex) => {
                      const isNoteType = ["note", "text"].includes(annoItem.annotationType);
                      const {labels, sectionComment, translation} = getAnnotationLabelsAndComment(annoItem);
          
                      const theContent = await Zotero.BetterNotes.api.convert.annotations2html(
                          [annoItem], 
                          { noteItem: targetNoteItem, ignoreComment: !isNoteType }
                      );
  
                      return {
                          attachmentIndex,
                          annotIndex,
                          title,
                          labels,
                          processedContent: formatContent(theContent, labels, isNoteType, sectionComment, translation)
                      };
                  })
              );
          })
      );
  
      processedAnnotations.flat().sort((a, b) => {
          const attachOrder = a.attachmentIndex - b.attachmentIndex;
          return attachOrder !== 0 ? attachOrder : a.annotIndex - b.annotIndex;
      }).forEach(({labels, title, processedContent}) => {
          let currentLevel = groupedAnnotations;
          labels.forEach((label, index) => {
              currentLevel = currentLevel[label] ||= {
                  level: index + 2,
                  content: [],
                  childLabels: {}
              };
  
              if (index === labels.length - 1) {
                  const shouldAddTitle = annotatedAttachment > 1 && !currentLevel.content.some(t => t === title);
                  currentLevel.content.push(...(shouldAddTitle ? [title, processedContent] : [processedContent]));
              } else {
                  currentLevel = currentLevel.childLabels;
              }
          });
      });
  
      function formatContent(_content, _labels, _isNoteType, _comment, _translation) {
          // 不将标注格式化为引用块，标注和评论不分段，并加粗评论，评论在标注后
          return _isNoteType ? _content.replace(`[${_labels.join('][')}]`, "") : _content.replace("</p>", (_comment + _translation) ? `${_translation ? "<br>" + _translation : ""}${_comment ? "<br><b>" + _comment + "</b>" : ""}</p>` : "</p>");
          // 不将标注格式化为引用块，标注和评论不分段，并加粗评论，评论在标注前
          // return _isNoteType ? _content.replace(`[${_labels.join('][')}]`, "") : _content.replace("<p>", (_comment + _translation) ? `<p>${_comment ? "<b>" + _comment + "</b><br>" : ""}${_translation ? _translation + "<br>" : ""}` : "<p>");
          // 将标注格式化为引用块，标注和评论分段，评论在标注后
          // return _isNoteType ? _content.replace(`[${_labels.join('][')}]`, "") : "<blockquote>" + _content.replace("</p>", _translation ? "<br>" + _translation + "</p>" : "</p>") + "</blockquote>" + _comment;
  
      }
      myAnnotations += processAnnotations(groupedAnnotations) ? processAnnotations(groupedAnnotations) : "";
  
     	let res = `
          <!-- 生成标签笔记 -->
          ${myAnnotations}`;
      return res;        
  })()}

Anything else

No response

[yoyo2018520]

模板非常好！有个小的问题：题目和摘要能展示中英双语么？

[zhikaiyici]

在你想要显示翻译的位置添加topItem.getField('titleTranslation') 和 topItem.getField('abstractTranslation')。
比如在原标题和摘要后面显示翻译，用下面的代码替换原模板中对应部分：

<h1> ${topItem.getField('title') + topItem.getField('titleTranslation')} (${topItem.getField("year")}, ${sourceName}, ${authorDisplay[0]})</h1>

        <!-- 摘要 -->
        <p><tr>
            <td>
                ${(() => {
                    return `<b><span style="color: #3c5acc">Abstract: </span></b> <i>${topItem.getField('abstractNote') + topItem.getField('abstractTranslation')}</i>`;
                })()}
            </td>
        </tr></p>

需要先用翻译插件翻译标题和摘要。

[Susu20241010]

可以的 发自我的iPhone

…

------------------ 原始邮件 ------------------ 发件人: WANG Chao ***@***.***> 发送时间: 2024年11月21日 19:36 收件人: windingwind/zotero-better-notes ***@***.***> 抄送: Susu20241010 ***@***.***>, Mention ***@***.***> 主题: Re: [windingwind/zotero-better-notes] [Item] Generate note from annotations (by brackets tag)/根据标注注释内自定义的标签生成笔记-支持多级标题嵌套 (2024.11.21更新/updated) (Discussion #1202) 太晚了，明天吧 — Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you were mentioned.Message ID: ***@***.***>

[Susu20241010]

我在添加评论的区域输入了 [标题]xx 内容，再点击“生成笔记”，右边笔记区生成的内容显示 我的Heading在笔记的后面，咋回事
难道需要手动在笔记区打出 一级标题和二级标题吗？

[Susu20241010]

我是这样做的！

[w117247457]

请问生成笔记前面有这么一行是什么，怎么删掉

[windingwind]

去掉模板的@use-refresh行即可

[wsgwzlr]

去掉模板的@use-refresh行即可

但是我还想要保留随时更新的功能，这个能解决吗？

[Susu20241010]

请问生成笔记前面有这么一行是什么，怎么删掉

亲 使用什么插件出现 期刊标签的呀，

[w117247457]

去掉模板的@use-refresh行即可

但是我还想要保留随时更新的功能，这个能解决吗？

请问你解决了吗

[ddin13]

同问，是不是想要保留随时更新的功能就不能去掉第一行呀

[Susu20241010]

ToDesk:
设备代码:970 135 533
临时密码:4396in34

大佬能不能帮我远程一下 我今天有时间 一直开着电脑

[Susu20241010]

我已经解决了！！我复制的你发布的最新的代码

[Susu20241010]

哈哈哈 感激涕零 谢谢大佬的代码

[Susu20241010]

是的 发自我的iPhone

…

------------------ 原始邮件 ------------------ 发件人: WANG Chao ***@***.***> 发送时间: 2024年12月4日 15:05 收件人: windingwind/zotero-better-notes ***@***.***> 抄送: Susu20241010 ***@***.***>, Mention ***@***.***> 主题: Re: [windingwind/zotero-better-notes] [Item] Generate note from annotations (by brackets tag)/根据标注注释内自定义的标签生成笔记-支持多级标题嵌套 (2024.11.21更新/updated) (Discussion #1202) 不需要远程了是吗 — Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you were mentioned.Message ID: ***@***.***>

[Abbyecoming]

非常非常有帮助，感谢！请问，可以修改标题编号的顺序吗？如何让后出现的注释在笔记中排序靠前？

[zhikaiyici]

那又要如何决定哪个注释排在笔记前面哪个注释排在笔记后面呢？
如果依靠手动编号的话，做注释时要记住每个注释标题的顺序，而同一标题的注释是可能分散于文章各处的，这个负担还是挺重的，我觉得是偏离了依靠笔记模板减轻工作量的初衷，还不如手动做笔记。

[Grace0716]

请问自动更新是多久会自动更新一次吗？我好像更新不出来

[Grace0716]

[zhikaiyici]

请问自动更新是多久会自动更新一次吗？我好像更新不出来

没有自动更新功能，是“允许更新”，点击笔记右上角三个点-更新模板生成内容。

使用模板时，须选中条目，而不是选中条目下的附件（如PDF文件）。

[Grace0716]

是在这里建吗？我还是没看到更新模板生成内容的选项

[zhikaiyici]

更新或者重装插件或者重启zotero，你这个插件有点问题看起来。

[Grace0716]

好滴，谢谢大佬！

[hxsnow10]

我试了下，发现2个head中间的那些会被归类为unlabeled, 中间的那些是不是归类为前一个head更好呢？
我把标签改成了层级的功能
并且在windingwind/zotero-actions-tags#466 提供了一个自动更新的脚本

// @author zhikaiyici,xiahong
// @link [PUBLISH PAGE URL](https://github.com/windingwind/zotero-better-notes/discussions/1202)
${ await(async() => {
    // 辅助函数：检测字符串是否包含中文字符
    const containsChinese = (text) => /[\u4e00-\u9fa5]/.test(text);

    // 检查标题是否为中文
    let isChineseTitle = containsChinese(topItem.getField("title"));

    // 获取作者显示内容
    let creators = topItem.getCreators();
    let authorDisplay = ["", ""];
    if (isChineseTitle) {
        // 中文标题：显示第一作者全名，格式为“姓名”
        if (creators.length > 0) {
            let firstAuthor = creators[0];
            let fullName = firstAuthor.lastName + firstAuthor.firstName;
            if (creators.length > 1) {
                authorDisplay[0] = fullName + "等";
            }
            else {
                authorDisplay[0] = fullName;
            }
            // 正文中显示前10位作者，多于10位加等
            if (creators.length > 10) {
                authorDisplay[1] = creators.splice(0, 10).map((v) => v.firstName + v.lastName).join("; ");
                authorDisplay[1] = authorDisplay + "等";
            }
            else {
                authorDisplay[1] = creators.map((v) => v.firstName + v.lastName).join("; ");
            }
        }
    }
    else {
        // 非中文标题：显示第一作者姓氏并附加“et al.”
        if (creators.length > 0) {
            let firstAuthorLastName = creators[0].lastName;
                if (creators.length > 1) {
                    authorDisplay[0] = `${firstAuthorLastName} et al.`;
                }
                else {
                    authorDisplay[0] = creators[0].lastName + " " + creators[0].firstName;
                }
                // 正文中显示前10位作者，多于10位加et al.
                if (creators.length > 10) {
                    authorDisplay[1] = creators.splice(0, 10).map((v) => v.firstName + " " + v.lastName).join("; ");
                    authorDisplay[1] = authorDisplay[1] + "; et al.";
                }
                else {
                    authorDisplay[1] = creators.map((v) => v.firstName + " " + v.lastName).join("; ");
                }
            }
    }

    const getAnnotationLabelsAndComment = (_comment) => {
        let result = { labels: [], sectionComment: "" };
        let parsed = _comment.match(/^\[(.*?)\](.*)/);
        if (!parsed) {
            result.labels.push("Unlabeled");
            result.sectionComment = _comment;
        }
        else {
            while (parsed) {
                result.labels.push(parsed[1].trim());
                result.sectionComment = parsed[2].trim();
                parsed = result.sectionComment.match(/^\[(.*?)\](.*)/);
            }
        };
        return result;
    }

    const processAnnotations = (grouped, parentLevel) => {
        let result = "";
        let i = 0;
        // 遍历每个组
        for (const label in grouped) {
            if (label != "Unlabeled") {
                i++;
                const group = grouped[label];
                const heading = `<h${group.level}>${label}</h${group.level}>`;
                result += heading; // + "\n";
                // 添加内容
                group.content.forEach(content => {
                    result += content; // + "\n";
                });
                // 递归处理子标签
                result += processAnnotations(group.childLabels, parentLevel + i + ".");
            }
        }
        for (const label in grouped) {
            if (label === "Unlabeled") {
                i++;
                const group = grouped[label];
                const heading = `<h${group.level}>${label}</h${group.level}>`;
                result += heading; // + "\n";
                // 添加内容
                group.content.forEach(content => {
                    result += content; // + "\n";
                });
                break;
            }
        }
        return result;
    }

	/**
	 * 解析 Markdown 标题格式，返回#号数量和标题文本
	 * @param {string} headerStr Markdown 格式的标题字符串
	 * @returns {{level: number, title: string}} 返回对象包含级别和标题文本
	 */
	function parseMarkdownHeader(headerStr) {
		// 使用正则表达式匹配开头的#号和后面的文本
		const match = headerStr.match(/^(#{1,6})\s*(.+)$/);
		
		if (!match) {
			return null;
		}

		return {
			level: match[1].length,  // #号的数量
			title: match[2].trim()   // 标题文本，去除首尾空格
		};
	}

    let myAnnotations = "";
    const attachments = Zotero.Items.get(topItem.getAttachments()).filter((i) => i.isPDFAttachment() || i.isSnapshotAttachment() || i.isEPUBAttachment());
    let annotatedAttachment = 0;
    for (let attachment of attachments) {
        if (attachment.getAnnotations().length > 0)
            annotatedAttachment++; 
    }
	let last_label = "";
	let last_level = 1;
	let label = "";
	let level = 1;
    const groupedAnnotations = {};
    for (let attachment of attachments) {
        let annots = attachment.getAnnotations();
        let attachmentTitle = `<i>📄For Document: <a href="zotero://open-pdf/0_${attachment.key}">${attachment.getField("title")}</a></i>`;     
        for (let annoItem of annots) {
            let myAnnotation = { labels: "", content: "" };
            let myComment = annoItem.annotationComment || "";
            let myLabelsAndComment = getAnnotationLabelsAndComment(myComment);
            myAnnotation.labels = myLabelsAndComment.labels;

            if (["note", "text", "highlight", "underline"].includes(annoItem.annotationType)) {
                if(["highlight", "underline"].includes(annoItem.annotationType) && myComment){
                    annoItem.annotationComment = "";
                }
                const res = Zotero.EditorInstanceUtilities.serializeAnnotations([{
                    ...(await Zotero.Annotations.toJSON(annoItem)), ...{ attachmentItemID: annoItem.parentID }
                }]);
                if (["note", "text"].includes(annoItem.annotationType)) {
                    myAnnotation.content += res.html.replace("[" + myAnnotation.labels.join('][') + "]", "");
                }
                else {
                    myAnnotation.content += "<blockquote>" + res.html + "</blockquote>";
                    myAnnotation.content += myLabelsAndComment.sectionComment;
                }
                if (myComment){
                    annoItem.annotationComment = myComment;
                }
            }
            else {
                //if (["image", "ink"].includes(annoItem.annotationType)) {
                    myAnnotation.content += "<blockquote>" + await Zotero.BetterNotes.api.convert.annotations2html([annoItem], { noteItem: targetNoteItem, ignoreComment: true }) + "</blockquote>";
                //}
                if (myComment){
                    annoItem.annotationComment = myComment;
                }
                myAnnotation.content += myLabelsAndComment.sectionComment;
            }

            let currentLevel = groupedAnnotations;
            // 遍历每个标签，构建分组结构
			label = last_label;
			level = last_level;
            myAnnotation.labels.forEach((headStr, i) => {
				let hresult = parseMarkdownHeader(headStr);
				if (hresult) {
					label = hresult.title;
					index = hresult.level;
				}
			});
			// 如果当前标签在结果中不存在，则初始化
			if (!currentLevel[label]) {
				currentLevel[label] = {
					level: level+1, // 一级标题为h2，二级标题为h3，依此类推
					content: [],
					childLabels: {} // 存储子标签
				};
				if (annotatedAttachment > 1 && !currentLevel[label].content.includes(attachmentTitle))
					currentLevel[label].content.push(attachmentTitle);
			}
			// 添加内容
			currentLevel[label].content.push(myAnnotation.content);
			last_label = label;
			last_level = level;

        }
    }
    myAnnotations += processAnnotations(groupedAnnotations, "") ? processAnnotations(groupedAnnotations, "") : "";

    // 非期刊条目显示论文集、大学、出版社、仓库...
    let otherName = topItem.getField("proceedingsTitle")
        + topItem.getField("university") + topItem.getField("publisher")
        + topItem.getField("repository") + topItem.getField("institution")
        + topItem.getField("meetingName");
    let publicationName = topItem.getField("publicationTitle") + otherName;
    let sourceName = topItem.getField("journalAbbreviation") + otherName;
    let res = `
        <h1> Auto: ${topItem.getField('title')} (${topItem.getField("year")}, ${sourceName}, ${authorDisplay[0]})</h1>
        <!--<hr/>-->
        <p><tr><td>
            <b><span style="color: #3c5acc">Author: </span></b> ${authorDisplay[1]}
        </td></tr></p>

        <p><tr><td><b><span style="color: #3c5acc">Source: </span></b>
            ${(topItem.getField("volume") + topItem.getField("issue") + topItem.getField("pages")) ? 
            (publicationName + " (" + 
            (topItem.getField("volume") ? "volume: " + topItem.getField("volume") + ", " : '') + 
            (topItem.getField("issue") ? "issue: " + topItem.getField("issue") + ", " : '') + 
            (topItem.getField("pages") ? "pages: " + topItem.getField("pages") : '') + ")") : publicationName}
            ${(() => {
                if (Zotero.ZoteroStyle) {
                    let space = " ";
                    return Array.prototype.map.call(
                        Zotero.ZoteroStyle.api.renderCell(topItem, "publicationTags").childNodes,
                        e => {
                            e.innerText = space + e.innerText + space;
                            return e.outerHTML
                        }
                    ).join(space);
                }
            })()}
        </td></tr></p>

        <p><tr><td>
            ${(() => {
                const doi = topItem.getField("DOI");
                if (doi) {
                    return `<b><span style="color: #3c5acc">DOI: </span></strong></b> 
                    <a href="https://doi.org/${topItem.getField('DOI')}">${topItem.getField('DOI')}</a>`;
                } else {
                    return `<b><span style="color: #3c5acc">URL: </span></strong> </b> 
                    <a href="${topItem.getField('url')}">${topItem.getField('url')}</a>`;
                }
           })()}
        </td></tr></p>

        <p><tr><td>
            <b><span style="color: #3c5acc">Date: </span></b> ${topItem.getField('date')}
        </td></tr></p>

        <!-- 本地链接 -->
        <p><tr><td>
            ${(() => {
                const attachments = Zotero.Items.get(topItem.getAttachments());
                if (attachments && attachments.length > 1) {
                    return `<b><span style="color: #3c5acc">Full Text: </span></b> 
                    <a href=zotero://open-pdf/0_${Zotero.Items.get(topItem.getAttachments()).filter((i) => i.isPDFAttachment())[0].key}>
                    ${Zotero.Items.get(topItem.getAttachments()).filter((i) => i.isPDFAttachment())[0].getFilename()}</a>`;
                } else {
                    if (attachments && attachments.length > 0) {
                        return `<b><span style="color: #3c5acc">Full Text: </span> </b> 
                        <a href="zotero://open-pdf/0_${attachments[0].key}">${attachments[0].getFilename()}</a>`;
                    }
                    else {
                        return `<b><span style="color: #3c5acc">Full Text: </span></b>`;
                    }
                }
            })()}
        </td></tr></p>

        <!-- 摘要 -->
        <p><tr><td>
            <b><span style="color: #3c5acc">Abstract: </span></b> <i>${topItem.getField('abstractNote')}</i>
        </td></tr></p>
        <!--<hr/>-->

        <!-- 生成标签笔记 -->
        ${myAnnotations}
         <br>
		build from template: [item]Generate note from annotations (by brackets tag)
    `;
    return res;        
})()}

[zhikaiyici]

👍

[lilili61]

你好，我是用你们的代码并让AI进行了一点修改，但不知道为什么现在更新内容不是替换，而是在笔记里再生成一份。问了AI也没解决，该怎么办呢？

// @author zhikaiyici
// @link PUBLISH PAGE URL
// @beforeloop-begin
// @use-refresh
name: "[item]Nested Highlights - Accurate Grouping"
zoteroVersion: "7.0.15"
pluginVersion: "2.4.8"
savedAt: "2025-04-22T00:00:00.000Z"
content: |-
${ await (async () => {

const containsChinese = (text) => /[\u4e00-\u9fa5]/.test(text);
const isChineseTitle = containsChinese(topItem.getField("title"));
const creators = topItem.getCreators();

let authorDisplay = ["", ""];
if (isChineseTitle) {
  const fullName = creators[0]?.lastName + creators[0]?.firstName;
  authorDisplay[0] = creators.length > 1 ? fullName + "等" : fullName;
  authorDisplay[1] = creators.slice(0, 10).map(c => c.firstName + c.lastName).join("; ") + (creators.length > 10 ? "等" : "");
} else {
  const last = creators[0]?.lastName;
  const first = creators[0]?.firstName;
  authorDisplay[0] = creators.length > 1 ? `${last} et al.` : `${last} ${first}`;
  authorDisplay[1] = creators.slice(0, 10).map(c => `${c.firstName} ${c.lastName}`).join("; ") + (creators.length > 10 ? "; et al." : "");
}

const colors = [
  ["#1565C0", "#E1F5FE"],
  ["#2E7D32", "#F1F8E9"],
  ["#E65100", "#FFF8E1"],
  ["#6A1B9A", "#F3E5F5"],
  ["#AD1457", "#FCE4EC"]
];

const getAnnotationLabelsAndComment = (_comment) => {
  let result = { labels: [], sectionComment: "", translation: "" };
  let comment = _comment || "";

  const transMatch = comment.match(/^(.*?)🔤(.*?)🔤(.*?)$/s);
  if (transMatch) {
    result.translation = "🔤" + transMatch[2].trim() + "🔤";
    comment = transMatch[1].trim() + transMatch[3].trim();
  }

  let parsed = comment.match(/^\[(.*?)\](.*)/s);
  if (!parsed) {
    result.labels = [];
  } else {
    while (parsed) {
      result.labels.push(parsed[1].trim());
      comment = parsed[2].trim();
      parsed = comment.match(/^\[(.*?)\](.*)/s);
    }
  }

  result.sectionComment = comment;
  return result;
};

const formatContent = (_content, _comment, _translation) => {
  return _content.replace("</p>", 
    (_translation || _comment) 
      ? `${_translation ? "<br><i>" + _translation + "</i>" : ""}${_comment ? "<br><b>" + _comment + "</b>" : ""}</p>` 
      : "</p>");
};

const insertIntoGroup = (root, pathLabels, content) => {
  let current = root;
  pathLabels.forEach((label, idx) => {
    current[label] = current[label] || {
      level: idx + 2,
      content: [],
      childLabels: {}
    };
    if (idx === pathLabels.length - 1) {
      current[label].content.push(content);
    }
    current = current[label].childLabels;
  });
};

const processAnnotations = (grouped, prefix = "") => {
  let html = "";
  let i = 0;
  for (const label in grouped) {
    i++;
    const group = grouped[label];
    const color = colors[group.level - 2] || ["#000", "#EEE"];
    const heading = `<h${group.level} style="color:${color[0]};background-color:${color[1]}">${prefix}${i} ${label}</h${group.level}>`;
    html += heading + group.content.join("");
    if (Object.keys(group.childLabels).length > 0) {
      html += processAnnotations(group.childLabels, `${prefix}${i}.`);
    }
  }
  return html;
};

const groupedAnnotations = {};
const attachments = Zotero.Items.get(topItem.getAttachments()).filter(att => att.isPDFAttachment());

for (const attachment of attachments) {
  let lastLabelPath = ["默认分组"]; // 每个附件重置lastLabelPath
  const annots = attachment.getAnnotations();

  for (const anno of annots) {
    const rawComment = anno.annotationComment || "";
    const { labels, sectionComment, translation } = getAnnotationLabelsAndComment(rawComment);
    const html = await Zotero.BetterNotes.api.convert.annotations2html([anno], { ignoreComment: true }); // 移除noteItem参数
    const content = formatContent(html, sectionComment, translation);

    const currentLabelPath = labels.length > 0 ? labels : lastLabelPath;
    insertIntoGroup(groupedAnnotations, currentLabelPath, content);

    if (labels.length > 0) lastLabelPath = labels;
  }
}

let sourceTitle = topItem.getField("publicationTitle") + topItem.getField("proceedingsTitle") + topItem.getField("publisher");
let doiOrUrl = topItem.getField("DOI") 
  ? `<a href="https://doi.org/${topItem.getField("DOI")}">${topItem.getField("DOI")}</a>` 
  : `<a href="${topItem.getField("url")}">${topItem.getField("url")}</a>`;

return `
  <h1>${topItem.getField("title")} (${topItem.getField("year")}, ${authorDisplay[0]})</h1>
  <hr/>
  <p><b>Author: </b>${authorDisplay[1]}</p>
  <p><b>Source: </b>${sourceTitle}</p>
  <p><b>DOI / URL: </b>${doiOrUrl}</p>
  <p><b>Date: </b>${topItem.getField("date")}</p>
  <p><b>Abstract: </b><i>${topItem.getField("abstractNote")}</i></p>
  <hr/>
  ${processAnnotations(groupedAnnotations)}
`;

})()}

[zhikaiyici]

去掉分割线语法<hr/>

[lilili61]

非常感谢

[Tingtingyang1234]

你好我在使用代码的时候发现根据标注颜色生成笔记，刷新的同时会导致自己做的文字笔记丢失，这个有什么办法吗

[zhikaiyici]

没有，看这里的回复#1129 (comment)
另一个思路是把你自己的文字笔记放在相应批注的评论里，而不是在笔记中添加。

[evagao2001]

大佬请问，为什么我导入模版后，显示“items is not defined”呢

[zhikaiyici]

推荐使用“从剪切板导入笔记模板”功能。如果手动新建模板然后粘贴的话，注意删除// @ 行首多余空格。

[woyeyao]

大佬，我读代码里面Figure和tables标签是什么啊，才开始用zotero

[zhikaiyici]

将带有Table/Figure（如 Table 1/Figure 1/Figure 2）标签的标注单独列出。

哪里没说清楚吗？可以动手测试看看效果。

[woyeyao]

对不起，我把Figure和table误以为是BN的特殊标签了