From 0aa2993d55c2da8d43ccfe3fd1d968737701c541 Mon Sep 17 00:00:00 2001 From: Jin Hai Date: Fri, 10 Jul 2026 20:30:00 +0800 Subject: [PATCH] Go: format code (#16813) ### Summary Aligned to EE Signed-off-by: Jin Hai --- internal/admin/enterprise_service.go | 4 ++ internal/admin/service.go | 3 +- internal/agent/component/userfillup.go | 2 - internal/agent/tool/akshare.go | 4 +- internal/binding/cpp/darts_trie.cpp | 2 +- internal/binding/cpp/main.cpp | 2 +- internal/binding/cpp/pcre2.h | 2 +- internal/binding/cpp/rag_analyzer.cpp | 2 +- .../binding/cpp/rag_analyzer_c_api_debug.cpp | 24 +++---- internal/binding/cpp/rag_analyzer_c_test.cpp | 40 +++++------ internal/binding/cpp/term.cpp | 2 +- internal/binding/cpp/thinc_parser.cpp | 46 ++++++------- internal/binding/cpp/tokenizer.cpp | 2 +- internal/binding/cpp/wordnet_lemmatizer.cpp | 2 +- internal/dao/document.go | 6 +- internal/deepdoc/parser/pdf/parser_ocr.go | 6 +- .../parser/pdf/table/group_boxes_test.go | 3 +- .../deepdoc/parser/pdf/table/table_html.go | 32 ++++++--- .../parser/pdf/table/table_html_test.go | 1 - .../deepdoc/parser/pdf/table/table_merge.go | 2 - .../deepdoc/parser/pdf/table/table_spans.go | 66 ++++++++++++++----- .../parser/pdf/table/table_spans_test.go | 1 - internal/harness/README.md | 2 +- internal/harness/graph/pregel/engine.go | 2 +- .../compilation/extractor/dep_relation.go | 1 - internal/ingestion/task/position.go | 9 +-- internal/parser/parser/factory.go | 30 ++++----- 27 files changed, 172 insertions(+), 126 deletions(-) diff --git a/internal/admin/enterprise_service.go b/internal/admin/enterprise_service.go index f9a001a0d9..8b16427270 100644 --- a/internal/admin/enterprise_service.go +++ b/internal/admin/enterprise_service.go @@ -1169,3 +1169,7 @@ func (s *Service) RemoveIngestionTasksByCondition(tasks []string, email, status return []map[string]interface{}{element}, nil } + +func CheckLicense() (common.ErrorCode, string) { + return common.CodeLicenseValid, "" +} diff --git a/internal/admin/service.go b/internal/admin/service.go index 39638a967c..71052f63dd 100644 --- a/internal/admin/service.go +++ b/internal/admin/service.go @@ -1677,8 +1677,9 @@ func (s *Service) HandleHeartbeat(message *common.BaseMessage) (common.ErrorCode Timestamp: message.Timestamp, Ext: message.Ext, } + GlobalServerStore.UpdateServerInfo(message.ServerName, status) - return common.CodeLicenseValid, "" + return CheckLicense() } // InitDefaultAdmin initialize default admin user diff --git a/internal/agent/component/userfillup.go b/internal/agent/component/userfillup.go index 888dbc6b10..dc03ca7fa2 100644 --- a/internal/agent/component/userfillup.go +++ b/internal/agent/component/userfillup.go @@ -23,7 +23,6 @@ // (value.type starts with "file") emit a stable "" stub // so the run keeps flowing while the FileService integration // surfaces the actual bytes via the storage layer. -// package component import ( @@ -53,7 +52,6 @@ const fileStubPrefix = "%s - 新闻内容: %s -发布时间:%s + 新闻内容: %s +发布时间:%s 文章来源: %s`, item.URL, item.Title, diff --git a/internal/binding/cpp/darts_trie.cpp b/internal/binding/cpp/darts_trie.cpp index 5fc44779e0..fb508e008c 100644 --- a/internal/binding/cpp/darts_trie.cpp +++ b/internal/binding/cpp/darts_trie.cpp @@ -107,4 +107,4 @@ int DartsTrie::Get(std::string_view key) const { key = empty_null_terminated_sv; } return darts_->exactMatchSearch(key.data(), key.size()); -} \ No newline at end of file +} diff --git a/internal/binding/cpp/main.cpp b/internal/binding/cpp/main.cpp index fb8c38d6f0..d6484e9eb7 100644 --- a/internal/binding/cpp/main.cpp +++ b/internal/binding/cpp/main.cpp @@ -439,4 +439,4 @@ int main() { // test_fine_grained_tokenize_consistency_with_python(); test_tokenize_text("在本研究中，我们提出了一种novel的neural network架构，用于解决multi-modal learning问题。我们的方法结合了CNN(Convolutional Neural Networks)和Transformer的优势，在ImageNet数据集上达到了state-of-the-art性能。实验结果表明，在batch size为256、learning rate为0.001的条件下，我们的模型在validation set上的accuracy达到了95.7%，比baseline方法提高了3.2%。此外，我们还进行了ablation study来分析不同components的contribution。所有代码已在GitHub上开源，地址是https://github.com/example/our-project。未来工作将集中在model compression和real-time inference optimization上。"); return 0; -} \ No newline at end of file +} diff --git a/internal/binding/cpp/pcre2.h b/internal/binding/cpp/pcre2.h index 37431c7245..762355cd8b 100644 --- a/internal/binding/cpp/pcre2.h +++ b/internal/binding/cpp/pcre2.h @@ -43,7 +43,7 @@ POSSIBILITY OF SUCH DAMAGE. #define PCRE2_MAJOR 10 #define PCRE2_MINOR 47 -#define PCRE2_PRERELEASE +#define PCRE2_PRERELEASE #define PCRE2_DATE 2025-10-21 /* When an application links to a PCRE2 DLL in Windows, the symbols that are diff --git a/internal/binding/cpp/rag_analyzer.cpp b/internal/binding/cpp/rag_analyzer.cpp index 23377ff40e..a4a3bcc92d 100644 --- a/internal/binding/cpp/rag_analyzer.cpp +++ b/internal/binding/cpp/rag_analyzer.cpp @@ -2450,4 +2450,4 @@ int RAGAnalyzer::AnalyzeImpl(const Term &input, void *data, bool fine_grained, b } } return 0; -} \ No newline at end of file +} diff --git a/internal/binding/cpp/rag_analyzer_c_api_debug.cpp b/internal/binding/cpp/rag_analyzer_c_api_debug.cpp index d083382646..4601d6407e 100644 --- a/internal/binding/cpp/rag_analyzer_c_api_debug.cpp +++ b/internal/binding/cpp/rag_analyzer_c_api_debug.cpp @@ -114,28 +114,28 @@ void RAGAnalyzer_SetEnablePosition(RAGAnalyzerHandle handle, bool enable_positio int RAGAnalyzer_Analyze(RAGAnalyzerHandle handle, const char* text, RAGTokenCallback callback) { if (!handle || !text || !callback) return -1; - + fprintf(stderr, "[C_API] Analyze called with text length: %zu\n", strlen(text)); - + RAGAnalyzer* analyzer = static_cast(handle); - + Term input; input.text_ = std::string(text); - + TermList output; int ret = analyzer->Analyze(input, output); - + fprintf(stderr, "[C_API] Analyze returned: %d, tokens: %zu\n", ret, output.size()); - + if (ret != 0) { return ret; } - + // Call callback for each token for (const auto& term : output) { callback(term.text_.c_str(), term.text_.length(), term.word_offset_, term.end_offset_); } - + return 0; } @@ -144,13 +144,13 @@ char* RAGAnalyzer_Tokenize(RAGAnalyzerHandle handle, const char* text) { fprintf(stderr, "[C_API] Tokenize called with null handle or text\n"); return nullptr; } - + fprintf(stderr, "[C_API] Tokenize called with text length: %zu\n", strlen(text)); - + RAGAnalyzer* analyzer = static_cast(handle); - + std::string result = analyzer->Tokenize(std::string(text)); - + // Allocate memory for C string char* c_result = static_cast(DEBUG_MALLOC(result.size() + 1)); if (c_result) { diff --git a/internal/binding/cpp/rag_analyzer_c_test.cpp b/internal/binding/cpp/rag_analyzer_c_test.cpp index 05a8af6be4..0a166de6af 100644 --- a/internal/binding/cpp/rag_analyzer_c_test.cpp +++ b/internal/binding/cpp/rag_analyzer_c_test.cpp @@ -8,38 +8,38 @@ // Test case 1: Single thread, loop 1000 times void test_single_thread() { std::cout << "Test 1: Single thread, 1000 iterations..." << std::endl; - + // Create analyzer instance RAGAnalyzerHandle handle = RAGAnalyzer_Create("./resource"); assert(handle != nullptr && "Failed to create RAGAnalyzer"); - + // Load the analyzer int result = RAGAnalyzer_Load(handle); if (result != 0) { printf("Failed to load RAGAnalyzer: %d\n", result); } assert(result == 0 && "Failed to load RAGAnalyzer"); - + const char* input = "rag"; bool all_passed = true; - + for (int i = 0; i < 1000; ++i) { char* tokens = RAGAnalyzer_Tokenize(handle, input); - + if (tokens == nullptr || strlen(tokens) == 0) { std::cerr << "Iteration " << i << ": Failed - returned empty or null string" << std::endl; all_passed = false; } - + // Free the returned string if (tokens != nullptr) { free(tokens); } } - + // Destroy analyzer instance RAGAnalyzer_Destroy(handle); - + if (all_passed) { std::cout << "Test 1: PASSED" << std::endl; } else { @@ -51,32 +51,32 @@ void test_single_thread() { // Test case 2: 16 threads, each loop 1000 times void test_multi_thread() { std::cout << "Test 2: 32 threads, each 100000 iterations..." << std::endl; - + // Create analyzer instance (shared across threads) RAGAnalyzerHandle handle = RAGAnalyzer_Create("."); assert(handle != nullptr && "Failed to create RAGAnalyzer"); - + // Load the analyzer int result = RAGAnalyzer_Load(handle); assert(result == 0 && "Failed to load RAGAnalyzer"); - + const char* input = "rag"; const int num_threads = 32; const int iterations_per_thread = 100000; - + std::vector threads; std::vector thread_results(num_threads, true); - + for (int t = 0; t < num_threads; ++t) { threads.emplace_back([&, t]() { for (int i = 0; i < iterations_per_thread; ++i) { char* tokens = RAGAnalyzer_Tokenize(handle, input); - + if (tokens == nullptr || strlen(tokens) == 0) { std::cerr << "Thread " << t << " Iteration " << i << ": Failed - returned empty or null string" << std::endl; thread_results[t] = false; } - + // Free the returned string if (tokens != nullptr) { free(tokens); @@ -84,15 +84,15 @@ void test_multi_thread() { } }); } - + // Wait for all threads to complete for (auto& t : threads) { t.join(); } - + // Destroy analyzer instance RAGAnalyzer_Destroy(handle); - + bool all_passed = true; for (int t = 0; t < num_threads; ++t) { if (!thread_results[t]) { @@ -100,7 +100,7 @@ void test_multi_thread() { break; } } - + if (all_passed) { std::cout << "Test 2: PASSED" << std::endl; } else { @@ -115,7 +115,7 @@ int main() { test_single_thread(); // test_multi_thread(); - + std::cout << "=== All tests PASSED ===" << std::endl; return 0; } diff --git a/internal/binding/cpp/term.cpp b/internal/binding/cpp/term.cpp index 8ac9e16d21..84c62afec5 100644 --- a/internal/binding/cpp/term.cpp +++ b/internal/binding/cpp/term.cpp @@ -21,4 +21,4 @@ void Term::Reset() { word_offset_ = 0; } -Term TermList::global_temporary_; \ No newline at end of file +Term TermList::global_temporary_; diff --git a/internal/binding/cpp/thinc_parser.cpp b/internal/binding/cpp/thinc_parser.cpp index c8cd195ee2..044af97e15 100644 --- a/internal/binding/cpp/thinc_parser.cpp +++ b/internal/binding/cpp/thinc_parser.cpp @@ -165,7 +165,7 @@ struct Tok2vecModel { int po_nO=96,po_nP=3,po_nI=576; struct ResBlk{bool has=false;std::vectorW,b,lnG,lnb;}; ResBlk res[4]; int n_res=0; - + bool load(const std::string& dir) { std::ifstream cf(dir+"/model.ckpt"); if(!cf)return false; std::stringstream cb;cb<& tokens, float* out) { int n=(int)tokens.size(),D=96,NE=(int)embeds.size(),EC=NE*D; @@ -241,7 +241,7 @@ struct ParserModel { std::vector pW_pre,pb_pre,pad_pre; // preaffine std::vector pW_cls,pb_cls; // classifier std::vector move_names; - + bool load(const std::string& dir) { std::ifstream cf(dir+"/model.ckpt"); if(!cf)return false; std::stringstream cb;cb<type==JVal::ARR)for(auto& v:mn->arr)move_names.push_back(v.str);} return!pW_hid.empty() && !pW_pre.empty() && !pW_cls.empty(); } - + // Run parser forward + state machine → (heads, labels) void parse(const float* tokvecs, int n_tokens, std::vector& out_heads, std::vector& out_labels) { @@ -283,7 +283,7 @@ struct ParserModel { std::vector hidden((size_t)n_tokens*nO,0); for(int i=0;i precomp((size_t)(n_tokens+1)*nP*nO*nI,0); // Pad token (index 0) @@ -304,20 +304,20 @@ struct ParserModel { } } } - + // Helper: get feature value for a token at piece p, window w, output dim o auto feat = [&](int idx, int p, int w, int o) -> float { int ri = (idx < 0 || idx >= n_tokens) ? 0 : idx + 1; return precomp[(size_t)ri*nP*nO*nI + (size_t)p*nO*nI + (size_t)w + (size_t)o*nI]; }; - + // 3. Arc-hybrid state machine out_heads.assign(n_tokens, -1); out_labels.assign(n_tokens, ""); std::vector stack; std::vector buffer(n_tokens); for(int i=0;iint{ for(int i=0;i scores(n_actions,0); std::vector feats(nP*nI*nO,0); - + for(int step=0; stepbest_sc){best_sc=scores[a];best=a;} } if(best<0) break; - + const std::string& act=move_names[best]; if(act=="S"){stack.push_back(buffer[0]);buffer.erase(buffer.begin());} else if(act=="D"){stack.pop_back();} @@ -432,23 +432,23 @@ void ThincParser_Destroy(ThincParserHandle h) { delete (ParserState*)h; } char* ThincParser_Predict(ThincParserHandle h, const char* tokens_json) { auto* s=(ParserState*)h; if(!s||!s->loaded||!tokens_json) return strdup("[]"); - + auto j=JParser().parse(std::string(tokens_json)); if(j.type!=JVal::ARR) return strdup("[]"); std::vector tokens; for(auto& v:j.arr) tokens.push_back(v.str); int n=(int)tokens.size(); if(!n) return strdup("[]"); - + // Run tok2vec std::vector tokvecs((size_t)n*96,0); s->tok2vec.forward(tokens, tokvecs.data()); - + // Run parser std::vector heads; std::vector labels; s->parser.parse(tokvecs.data(), n, heads, labels); - + // Build JSON output std::string r="["; for(int i=0;itW.size()/96; if(!n||!n_tags)return strdup("[]"); - + // Run tok2vec to get 96-dim embeddings, then softmax + argmax std::vector tokvecs((size_t)n*96,0); s->tok2vec.forward(tokens, tokvecs.data()); - + std::vector best_tags(n, 0); for(int i=0;ibest_sc){best_sc=sc;best_tags[i]=t;} } } - + // Strip morphologizer output to just POS (e.g. "Gender=Masc|Number=Sing|POS=NOUN" → "NOUN") // For non-morphologizer models the tag string is used as-is. auto pos_only = [](const std::string& t) -> std::string { diff --git a/internal/binding/cpp/tokenizer.cpp b/internal/binding/cpp/tokenizer.cpp index edc6149173..c32941c4bf 100644 --- a/internal/binding/cpp/tokenizer.cpp +++ b/internal/binding/cpp/tokenizer.cpp @@ -312,4 +312,4 @@ bool Tokenizer::TokenizeWhite(const std::string &input_string, TermList &raw_ter } return true; -} \ No newline at end of file +} diff --git a/internal/binding/cpp/wordnet_lemmatizer.cpp b/internal/binding/cpp/wordnet_lemmatizer.cpp index d92c3463f0..f02cf37321 100644 --- a/internal/binding/cpp/wordnet_lemmatizer.cpp +++ b/internal/binding/cpp/wordnet_lemmatizer.cpp @@ -229,4 +229,4 @@ std::string WordNetLemmatizer::Lemmatize(const std::string &form, const std::str } return form; -} \ No newline at end of file +} diff --git a/internal/dao/document.go b/internal/dao/document.go index d279be8a82..b0b2e31984 100644 --- a/internal/dao/document.go +++ b/internal/dao/document.go @@ -75,9 +75,9 @@ func (dao *DocumentDAO) IncrementCounts(id string, kbID string, chunkNum int64, return DB.Model(&entity.Document{}). Where("id = ? AND kb_id = ?", id, kbID). Updates(map[string]interface{}{ - "chunk_num": gorm.Expr("chunk_num + ?", chunkNum), - "token_num": gorm.Expr("token_num + ?", tokenNum), - "process_duration": gorm.Expr("process_duration + ?", duration), + "chunk_num": gorm.Expr("chunk_num + ?", chunkNum), + "token_num": gorm.Expr("token_num + ?", tokenNum), + "process_duration": gorm.Expr("process_duration + ?", duration), }).Error } diff --git a/internal/deepdoc/parser/pdf/parser_ocr.go b/internal/deepdoc/parser/pdf/parser_ocr.go index d4fb6347e0..4c2114d516 100644 --- a/internal/deepdoc/parser/pdf/parser_ocr.go +++ b/internal/deepdoc/parser/pdf/parser_ocr.go @@ -66,8 +66,10 @@ func (p *Parser) ocrDetectAndRecognize(ctx context.Context, pageImg image.Image, for _, t := range texts { if strings.TrimSpace(t.Text) != "" { result = append(result, pdf.TextBox{ - X0: px0, X1: px1, - Top: py0, Bottom: py1, + X0: px0, + X1: px1, + Top: py0, + Bottom: py1, Text: t.Text, PageNumber: pageNum, }) diff --git a/internal/deepdoc/parser/pdf/table/group_boxes_test.go b/internal/deepdoc/parser/pdf/table/group_boxes_test.go index 6ff203e967..70cf3d47c2 100644 --- a/internal/deepdoc/parser/pdf/table/group_boxes_test.go +++ b/internal/deepdoc/parser/pdf/table/group_boxes_test.go @@ -1,4 +1,3 @@ - package table import ( @@ -17,7 +16,7 @@ func TestGroupBoxesByRC_RDiffSplitsRows(t *testing.T) { {X0: 210, X1: 290, Top: 0, Bottom: 30, Text: "C", R: 2, C: 2}, {X0: 10, X1: 90, Top: 35, Bottom: 65, Text: "D", R: 3, C: 0}, {X0: 110, X1: 190, Top: 35, Bottom: 65, Text: "E", R: 4, C: 1}, - {X0: 210, X1: 290, Top: 35, Bottom: 65, Text: "F", R: 5, C: 2}, + {X0: 210, X1: 290, Top: 35, Bottom: 65, Text: "F", R: 5, C: 2}, } rows := GroupBoxesByRC(boxes) // R=0,1,2,3,4,5 → 6 rows (Python: R differs → new row). diff --git a/internal/deepdoc/parser/pdf/table/table_html.go b/internal/deepdoc/parser/pdf/table/table_html.go index 8b3b320c3b..947ff18338 100644 --- a/internal/deepdoc/parser/pdf/table/table_html.go +++ b/internal/deepdoc/parser/pdf/table/table_html.go @@ -19,9 +19,13 @@ func RowsToHTML(rows [][]pdf.TSRCell, caption string, headerRows map[int]bool, s for ri, row := range rows { b.WriteString("") for ci, cell := range row { - if covered[[2]int{ri, ci}] { continue } + if covered[[2]int{ri, ci}] { + continue + } tag := "td" - if headerRows[ri] { tag = "th" } + if headerRows[ri] { + tag = "th" + } b.WriteString("<") b.WriteString(tag) sp := "" @@ -30,7 +34,9 @@ func RowsToHTML(rows [][]pdf.TSRCell, caption string, headerRows map[int]bool, s sp = fmt.Sprintf("colspan=%d", s[0]) } if s[1] > 1 { - if sp != "" { sp += " " } + if sp != "" { + sp += " " + } sp += fmt.Sprintf("rowspan=%d", s[1]) } } @@ -59,17 +65,23 @@ func SimpleRowsToHTML(rows [][]string) string { } nCols := 0 for _, row := range rows { - if len(row) > nCols { nCols = len(row) } + if len(row) > nCols { + nCols = len(row) + } } var b strings.Builder b.WriteString("") for ri, row := range rows { b.WriteString("") tag := "td" - if ri == 0 { tag = "th" } + if ri == 0 { + tag = "th" + } for ci := 0; ci < nCols; ci++ { text := "" - if ci < len(row) { text = row[ci] } + if ci < len(row) { + text = row[ci] + } b.WriteString("<") b.WriteString(tag) b.WriteString(" >") @@ -98,7 +110,9 @@ func RowsToStrings(rows [][]pdf.TSRCell) [][]string { func HasText(rows [][]pdf.TSRCell) bool { for _, row := range rows { for _, c := range row { - if strings.TrimSpace(c.Text) != "" { return true } + if strings.TrimSpace(c.Text) != "" { + return true + } } } return false @@ -106,7 +120,9 @@ func HasText(rows [][]pdf.TSRCell) bool { func HasAnyText(cells []pdf.TSRCell) bool { for _, c := range cells { - if strings.TrimSpace(c.Text) != "" { return true } + if strings.TrimSpace(c.Text) != "" { + return true + } } return false } diff --git a/internal/deepdoc/parser/pdf/table/table_html_test.go b/internal/deepdoc/parser/pdf/table/table_html_test.go index 4c96706848..d5b4e2375e 100644 --- a/internal/deepdoc/parser/pdf/table/table_html_test.go +++ b/internal/deepdoc/parser/pdf/table/table_html_test.go @@ -1,4 +1,3 @@ - package table import ( diff --git a/internal/deepdoc/parser/pdf/table/table_merge.go b/internal/deepdoc/parser/pdf/table/table_merge.go index a31fdccdae..4164ae22a0 100644 --- a/internal/deepdoc/parser/pdf/table/table_merge.go +++ b/internal/deepdoc/parser/pdf/table/table_merge.go @@ -1,4 +1,3 @@ - package table import ( @@ -125,4 +124,3 @@ func MergeTablesAcrossPages(tables []pdf.TableItem, medianHeights map[int]float6 } return result } - diff --git a/internal/deepdoc/parser/pdf/table/table_spans.go b/internal/deepdoc/parser/pdf/table/table_spans.go index 9fc79f7712..d7860199cb 100644 --- a/internal/deepdoc/parser/pdf/table/table_spans.go +++ b/internal/deepdoc/parser/pdf/table/table_spans.go @@ -12,7 +12,9 @@ import ( func CalSpans(rows [][]pdf.TSRCell) (map[[2]int][2]int, map[[2]int]bool) { spanInfo := make(map[[2]int][2]int) covered := make(map[[2]int]bool) - if len(rows) == 0 || len(rows[0]) == 0 { return spanInfo, covered } + if len(rows) == 0 || len(rows[0]) == 0 { + return spanInfo, covered + } // Compute column center positions. nCols := len(rows[0]) @@ -32,40 +34,64 @@ func CalSpans(rows [][]pdf.TSRCell) (map[[2]int][2]int, map[[2]int]bool) { for i, row := range rows { for j, cell := range row { - if j >= nCols { continue } + if j >= nCols { + continue + } // Exclude spanning cells from column/row boundary calculations. // Use label-based detection (O(1), no dependency on column midpoints). - if strings.Contains(cell.Label, "spanning") { continue } - if cell.X0 < colLeft[j] { colLeft[j] = cell.X0 } - if cell.X1 > colRight[j] { colRight[j] = cell.X1 } - if cell.Y0 < rowTop[i] { rowTop[i] = cell.Y0 } - if cell.Y1 > rowBott[i] { rowBott[i] = cell.Y1 } + if strings.Contains(cell.Label, "spanning") { + continue + } + if cell.X0 < colLeft[j] { + colLeft[j] = cell.X0 + } + if cell.X1 > colRight[j] { + colRight[j] = cell.X1 + } + if cell.Y0 < rowTop[i] { + rowTop[i] = cell.Y0 + } + if cell.Y1 > rowBott[i] { + rowBott[i] = cell.Y1 + } } } // For each spanning cell, compute how many cols/rows it covers. for i, row := range rows { for j, cell := range row { - if j >= nCols || covered[[2]int{i,j}] { continue } + if j >= nCols || covered[[2]int{i, j}] { + continue + } // Skip cells without position data (they can't span). - if cell.X0 == 0 && cell.X1 == 0 && cell.Y0 == 0 && cell.Y1 == 0 { continue } + if cell.X0 == 0 && cell.X1 == 0 && cell.Y0 == 0 && cell.Y1 == 0 { + continue + } cs, rs := 1, 1 // Count columns whose center is inside this cell's X range. - for k := j+1; k < nCols; k++ { + for k := j + 1; k < nCols; k++ { // Skip columns with no non-spanning cells (initial values unchanged). - if colLeft[k] == 1e9 && colRight[k] == -1e9 { continue } + if colLeft[k] == 1e9 && colRight[k] == -1e9 { + continue + } colCenter := (colLeft[k] + colRight[k]) / 2 - if colCenter >= cell.X0 && colCenter <= cell.X1 { cs++ } + if colCenter >= cell.X0 && colCenter <= cell.X1 { + cs++ + } } // Count rows whose center is inside this cell's Y range. - for k := i+1; k < nRows; k++ { + for k := i + 1; k < nRows; k++ { // Skip rows with no non-spanning cells. - if rowTop[k] == 1e9 && rowBott[k] == -1e9 { continue } + if rowTop[k] == 1e9 && rowBott[k] == -1e9 { + continue + } rowCenter := (rowTop[k] + rowBott[k]) / 2 - if rowCenter >= cell.Y0 && rowCenter <= cell.Y1 { rs++ } + if rowCenter >= cell.Y0 && rowCenter <= cell.Y1 { + rs++ + } } if cs > 1 || rs > 1 { - spanInfo[[2]int{i,j}] = [2]int{cs, rs} + spanInfo[[2]int{i, j}] = [2]int{cs, rs} // Mark covered cells. for ri := i; ri < i+rs && ri < nRows; ri++ { for cj := j; cj < j+cs && cj < nCols; cj++ { @@ -83,8 +109,12 @@ func CalSpans(rows [][]pdf.TSRCell) (map[[2]int][2]int, map[[2]int]bool) { // flattenGrid flattens a 2D grid into a 1D slice for fillCellTextFromBoxes. func FlattenGrid(grid [][]pdf.TSRCell) []pdf.TSRCell { n := 0 - for _, row := range grid { n += len(row) } + for _, row := range grid { + n += len(row) + } flat := make([]pdf.TSRCell, 0, n) - for _, row := range grid { flat = append(flat, row...) } + for _, row := range grid { + flat = append(flat, row...) + } return flat } diff --git a/internal/deepdoc/parser/pdf/table/table_spans_test.go b/internal/deepdoc/parser/pdf/table/table_spans_test.go index 389b13dc98..29db5251d6 100644 --- a/internal/deepdoc/parser/pdf/table/table_spans_test.go +++ b/internal/deepdoc/parser/pdf/table/table_spans_test.go @@ -1,4 +1,3 @@ - package table import ( diff --git a/internal/harness/README.md b/internal/harness/README.md index ecffc3d68f..179585116c 100644 --- a/internal/harness/README.md +++ b/internal/harness/README.md @@ -725,7 +725,7 @@ idle ──beginPlanningTurn──▶ planning ──beginActiveTurn──▶ ac - **TurnContext** — per-turn Preempted/Stopped channels, StopCause - **Callbacks**: `GenInput`, `GenResume`, `PrepareAgent`, `OnAgentEvents` -**Push options:** `WithPreempt`, `WithPreemptTimeout`, `WithPreemptDelay` +**Push options:** `WithPreempt`, `WithPreemptTimeout`, `WithPreemptDelay` **Stop options:** `WithGraceful`, `WithImmediate`, `WithGracefulTimeout`, `UntilIdleFor`, `WithSkipCheckpoint`, `WithStopCause` --- diff --git a/internal/harness/graph/pregel/engine.go b/internal/harness/graph/pregel/engine.go index c1c788f7c7..4a71c8d413 100644 --- a/internal/harness/graph/pregel/engine.go +++ b/internal/harness/graph/pregel/engine.go @@ -48,7 +48,7 @@ type Engine struct { versionsSeen map[string]map[string]int cache Cache backgroundExec *BackgroundExecutor - callbacks *CallbackManager // lifecycle callbacks (event recording, metrics) + callbacks *CallbackManager // lifecycle callbacks (event recording, metrics) deferredCheckpoints []deferredCheckpoint // for DurabilityExit mode } diff --git a/internal/ingestion/compilation/extractor/dep_relation.go b/internal/ingestion/compilation/extractor/dep_relation.go index aacb9b8a97..17db44fc21 100644 --- a/internal/ingestion/compilation/extractor/dep_relation.go +++ b/internal/ingestion/compilation/extractor/dep_relation.go @@ -9,7 +9,6 @@ import ( "strings" ) - // Verb lemmatization — multi-language var verbLemma = map[string]string{ // English diff --git a/internal/ingestion/task/position.go b/internal/ingestion/task/position.go index 8c42a162e8..cc9b1dc766 100644 --- a/internal/ingestion/task/position.go +++ b/internal/ingestion/task/position.go @@ -21,10 +21,11 @@ package task // every 5 elements. pn is 0-indexed; output is 1-indexed. // // Mirrors Python: rag.nlp.add_positions() -// for pn, left, right, top, bottom in poss: -// page_num_int.append(int(pn + 1)) -// top_int.append(int(top)) -// position_int.append((int(pn + 1), int(left), int(right), int(top), int(bottom))) +// +// for pn, left, right, top, bottom in poss: +// page_num_int.append(int(pn + 1)) +// top_int.append(int(top)) +// position_int.append((int(pn + 1), int(left), int(right), int(top), int(bottom))) func AddPositions(chunk map[string]any, positions []float64) { if len(positions) == 0 || len(positions)%5 != 0 { return diff --git a/internal/parser/parser/factory.go b/internal/parser/parser/factory.go index b267921eba..98fc440f4f 100644 --- a/internal/parser/parser/factory.go +++ b/internal/parser/parser/factory.go @@ -69,18 +69,18 @@ func GetParserByID(parserID string) (ParseResultProducer, error) { } // Stub constructors for each parser type. -func NewNaivePDFParser() *stubParser { return &stubParser{name: "naive"} } -func NewPaperPDFParser() *stubParser { return &stubParser{name: "paper"} } -func NewBookPDFParser() *stubParser { return &stubParser{name: "book"} } -func NewPresentationParser() *stubParser { return &stubParser{name: "presentation"} } -func NewManualPDFParser() *stubParser { return &stubParser{name: "manual"} } -func NewLawsPDFParser() *stubParser { return &stubParser{name: "laws"} } -func NewQAPDFParser() *stubParser { return &stubParser{name: "qa"} } -func NewTableParser() *stubParser { return &stubParser{name: "table"} } -func NewResumePDFParser() *stubParser { return &stubParser{name: "resume"} } -func NewPicturePDFParser() *stubParser { return &stubParser{name: "picture"} } -func NewOnePDFParser() *stubParser { return &stubParser{name: "one"} } -func NewAudioParser() *stubParser { return &stubParser{name: "audio"} } -func NewEmailParser() *stubParser { return &stubParser{name: "email"} } -func NewTagPDFParser() *stubParser { return &stubParser{name: "tag"} } -func NewKGPDFParser() *stubParser { return &stubParser{name: "knowledge_graph"} } +func NewNaivePDFParser() *stubParser { return &stubParser{name: "naive"} } +func NewPaperPDFParser() *stubParser { return &stubParser{name: "paper"} } +func NewBookPDFParser() *stubParser { return &stubParser{name: "book"} } +func NewPresentationParser() *stubParser { return &stubParser{name: "presentation"} } +func NewManualPDFParser() *stubParser { return &stubParser{name: "manual"} } +func NewLawsPDFParser() *stubParser { return &stubParser{name: "laws"} } +func NewQAPDFParser() *stubParser { return &stubParser{name: "qa"} } +func NewTableParser() *stubParser { return &stubParser{name: "table"} } +func NewResumePDFParser() *stubParser { return &stubParser{name: "resume"} } +func NewPicturePDFParser() *stubParser { return &stubParser{name: "picture"} } +func NewOnePDFParser() *stubParser { return &stubParser{name: "one"} } +func NewAudioParser() *stubParser { return &stubParser{name: "audio"} } +func NewEmailParser() *stubParser { return &stubParser{name: "email"} } +func NewTagPDFParser() *stubParser { return &stubParser{name: "tag"} } +func NewKGPDFParser() *stubParser { return &stubParser{name: "knowledge_graph"} }