input('view')) { case 'chapter-tag': /** * 分面筛选用的 tag 清单:返回每个 tag 及其命中的 pali_texts 条数。 * * 传 tags 时是**且**的关系:内层按 anchor 统计它命中了几个指定 tag, * 只留下 co = 指定 tag 个数的,也就是必须同时带上全部指定 tag; * 外层再对这批 anchor 重新按 tag 统计条数,作为下一级筛选的候选项。 * * @param string $tags 逗号分隔;不传则统计全部带 tag 的 pali_texts */ $tm = (new TagMap)->getTable(); $tg = (new Tag)->getTable(); $pt = (new PaliText)->getTable(); if ($request->input('tags') && $request->input('tags') !== '') { $tags = explode(',', $request->input('tags')); foreach ($tags as $tag) { # code... if (!empty($tag)) { $tagNames[] = $tag; } } } if (isset($tagNames)) { $where1 = " where co = " . count($tagNames); $a = implode(",", array_fill(0, count($tagNames), '?')); $in1 = "and t.name in ({$a})"; $param = $tagNames; } else { $where1 = " "; $in1 = " "; } $query = " select tags.id,tags.name,co as count from ( select tm.tag_id,count(*) as co from ( select anchor_id as id from ( select tm.anchor_id , count(*) as co from $tm as tm left join $tg as t on tm.tag_id = t.id left join $pt as pc on tm.anchor_id = pc.uid where tm.table_name = 'pali_texts' $in1 group by tm.anchor_id ) T $where1 ) CID left join $tm as tm on tm.anchor_id = CID.id group by tm.tag_id ) tid left join $tg on $tg.id = tid.tag_id order by count desc "; if (isset($param)) { $chapters = DB::select($query, $param); } else { $chapters = DB::select($query); } $all_count = count($chapters); break; case 'chapter': /** * 书目/章节清单:返回 uid、book、paragraph、level、toc、chapter_strlen 等。 * * tags 同样是**且**的关系。传了 tags 就放宽到 level < 3(书及其下一级), * 不传则只取 level = 1,即 276 本书的顶层。 * * 注意不传 tags 时这条 SQL 仍从 tag_maps 出发,所以严格说返回的是 * 「至少带一个 tag 的 level 1 记录」。目前 276 本书都带了 tag,两者恰好 * 等价;将来有书没打 tag,这里会静默漏掉它。 * * @param string $tags 逗号分隔 */ if ($request->input('tags') && $request->input('tags') !== '') { $tags = explode(',', $request->input('tags')); foreach ($tags as $tag) { # code... if (!empty($tag)) { $tagNames[] = $tag; } } } $tm = (new TagMap)->getTable(); $tg = (new Tag)->getTable(); $pt = (new PaliText)->getTable(); if (isset($tagNames)) { $where1 = " where co = " . count($tagNames); $a = implode(",", array_fill(0, count($tagNames), '?')); $in1 = "and t.name in ({$a})"; $param = $tagNames; $where2 = "where level < 3"; } else { $where1 = " "; $in1 = " "; $where2 = "where level = 1"; } $query = " select uid as id,book,paragraph,level,toc as title,chapter_strlen,parent,path from ( select anchor_id as cid from ( select tm.anchor_id , count(*) as co from $tm as tm left join $tg as t on tm.tag_id = t.id where tm.table_name = 'pali_texts' $in1 group by tm.anchor_id ) T $where1 ) CID left join $pt as pt on CID.cid = pt.uid $where2 order by book,paragraph"; if (isset($param)) { $chapters = DB::select($query, $param); } else { $chapters = DB::select($query); } $all_count = count($chapters); break; case 'chapter_children': /** * 直接下级目录:信任 pali_texts.parent 字段,只取 level < 8 的目录节点。 * * 与下面的 children 是两种取法:这里按 parent 直连,children 按段落区间 * 推算。父节点底下没有目录型子节点时,这里干脆返回空,**不会**退化成 * 列出正文段落——需要那种行为的用 children。 * * @param int $book * @param int $para 父节点的段落号 */ $table = PaliText::where('book', $request->input('book')) ->where('parent', $request->input('para')) ->where('level', '<', 8); $all_count = $table->count(); $chapters = $table->orderBy('paragraph')->get(); break; case 'children': /** * 下一层目录:不看 parent 字段,改用「段落区间 + 最浅的下一层」推算。 * * 在 [para+1, para+chapter_len-1] 区间里找 level 落在 root.level+1 到 7 * 之间、最浅的那一层,再取该层的全部节点。这样层级有跳空也取得到—— * 实测确有 level 2 直接跳到 level 4 的书(如 63:15006、143:5678)。 * * ⚠ **区间里一个目录层都没有时会退化**:改为返回区间内的全部记录, * 其中包含 level 100 的正文段落。大章节因此可能一次返回几百条, * 例如 115:2955(level 7、chapter_len 605)会返回 604 条。 * 调用方要么限制取用量,要么改用 chapter_children。 * * level >= 8 的节点(level 8 是叶子条目、level 100 是正文段落)没有 * 下级,直接返回空。 * * @param string $id pali_texts.uid;给了就优先按它定位 * @param int $book * @param int $para */ if ($request->has('id')) { $root = PaliText::where('uid', $request->input('id')) ->first(); } else { $root = PaliText::where('book', $request->input('book')) ->where('paragraph', $request->input('para')) ->first(); } if ($root->level >= 8) { $chapters = []; break; } $start = $root->paragraph + 1; $end = $root->paragraph + $root->chapter_len - 1; $nextLevelChapter = PaliText::where('book', $root->book) ->whereBetween('paragraph', [$start, $end]) ->whereBetween('level', [$root->level + 1, 7]) ->orderBy('level', 'asc') ->first(); if ($nextLevelChapter) { //存在子目录 $chapters = PaliText::where('book', $root->book) ->whereBetween('paragraph', [$start, $end]) ->where('level', $nextLevelChapter->level) ->orderBy('paragraph', 'asc') ->get(); } else { $chapters = PaliText::where('book', $root->book) ->whereBetween('paragraph', [$start, $end]) ->orderBy('paragraph', 'asc') ->get(); } $all_count = count($chapters); break; case 'paragraph': /** * 取单条 pali_texts 记录,按 (book, para) 精确匹配。 * * 这个分支**直接 return**,不走后面统一的 progress_line 补充,也不套 * rows/count 的外壳——返回的就是那条记录本身。查不到返回 error。 * * @param int $book * @param int $para */ $result = PaliText::where('book', $request->input('book')) ->where('paragraph', $request->input('para')) ->first(); if ($result) { return $this->ok($result); } else { return $this->error("no data"); } break; case 'paragraphs-info': /** * 取章节的 pali_texts 全部段落 记录,按 (book, para) 精确匹配。 * * * @param int $book * @param int $para */ $root = PaliText::where('book', $request->input('book')) ->where('paragraph', $request->input('para')) ->first(); if (!$root) { return $this->error("no paragraph"); } $chapters = PaliText::where('book', $request->input('book')) ->whereBetween('paragraph', [$root->paragraph,$root->paragraph+$root->chapter_len-1]) ->select(['book','paragraph','toc','level','lenght','chapter_len']) ->orderBy('paragraph','asc') ->get(); $all_count = count($chapters); break; case 'book-toc': /** * 获取全书目录 * 2023-1-25 改进算法 * 需求:目录显示丛书以及此丛书下面的所有书。比如,选择清净道论的一个章节。显示清净道论两本书的目录 * 算法: * 1. 查询这个目录的顶级目录 * 2. 查询book-title 获取丛书名 * 3. 根据从书名找到全部的书 * 4. 获取全部书的目录 * * 输出形状与其它分支不同,有两处:开头插一条 book=0、paragraph=0 的 * 合成记录放丛书名;其余每条的 level 都 +1,给那条丛书名让出第 1 层。 * 也正因为结构不同,末尾统一补 progress_line 时把 book-toc 排除在外。 * * @param string $series 丛书名;给了就直接按它取书目列表 * @param int $book 未给 series 时,与 para 一起定位所属丛书 * @param int $para */ if ($request->has('series')) { $book_title = $request->input('series'); //获取丛书书目列表 $books = BookTitle::where('title', $request->input('series'))->get(); } else { //查询这个目录的顶级目录 //查询书起始段落 $rootPara = app(PaliTextService::class)->getBookPara( $request->input('book'), $request->input('para') ); if (!$rootPara) { return $this->error('no book', 404, 404); } //获取丛书书名 $book_title = BookTitle::where('book', $rootPara->book) ->where('paragraph', $rootPara->paragraph) ->value('title'); //获取丛书书目列表 $books = BookTitle::where('title', $book_title)->get(); } $chapters = []; $chapters[] = ['book' => 0, 'paragraph' => 0, 'toc' => $book_title, 'level' => 1]; foreach ($books as $book) { # code... $rootPara = PaliText::where('book', $book->book) ->where('paragraph', $book->paragraph) ->first(); $table = PaliText::where('book', $rootPara->book) ->whereBetween('paragraph', [$rootPara->paragraph, ($rootPara->paragraph + $rootPara->chapter_len - 1)]) ->where('level', '<', 8); $all_count = $table->count(); $curr_chapters = $table->select(['book', 'paragraph', 'toc', 'level'])->orderBy('paragraph')->get(); foreach ($curr_chapters as $chapter) { # code... $chapters[] = ['book' => $chapter->book, 'paragraph' => $chapter->paragraph, 'toc' => $chapter->toc, 'level' => ($chapter->level + 1)]; } } break; } if ($request->input('view') !== 'book-toc' && $request->input('view') !== 'paragraphs-info') { foreach ($chapters as $key => $value) { if (is_object($value)) { //TODO $value->book 可能不存在 $progress_key = "/chapter_dynamic/{$value->book}/{$value->paragraph}/global"; $chapters[$key]->progress_line = Cache::get($progress_key); } } } return $this->ok(["rows" => $chapters, "count" => $all_count]); } /** * Store a newly created resource in storage. * * @param \Illuminate\Http\Request $request * @return \Illuminate\Http\Response */ public function store(Request $request) { // } /** * Display the specified resource. * * @param string $id * @return \Illuminate\Http\Response */ public function show(string $id) { // $para = explode('-', $id); $paragraph = PaliText::where('book', $para[0])->where('paragraph', $para[1])->first(); return $this->ok(new PaliTextResource($paragraph)); } /** * Update the specified resource in storage. * * @param \Illuminate\Http\Request $request * @param \App\Models\PaliText $paliText * @return \Illuminate\Http\Response */ public function update(Request $request, PaliText $paliText) { // } /** * Remove the specified resource from storage. * * @param \App\Models\PaliText $paliText * @return \Illuminate\Http\Response */ public function destroy(PaliText $paliText) { // } }