Merge pull request #6 from oolong-sh/graph-serialize

Add graph serialization

.gitignore

@@ -23,3 +23,5 @@ go.work.sum
 
 # env file
 .env
+
+oolong

internal/state/state.go

@@ -7,6 +7,7 @@ import (
 
 	"github.com/oolong-sh/oolong/internal/documents"
 	"github.com/oolong-sh/oolong/internal/linking/ngrams"
+	"github.com/oolong-sh/oolong/pkg/graph"
 	"github.com/oolong-sh/oolong/pkg/keywords"
 	"github.com/oolong-sh/oolong/pkg/notes"
 )
@@ -87,5 +88,16 @@ func UpdateState(docs []*documents.Document) error {
 		panic(err)
 	}
 
+	kw := keywords.NGramsToKeywordsMap(state.NGrams)
+	n := notes.DocumentsToNotes(state.Documents)
+
+	dat, err := graph.SerializeGraph(kw, n, 0.1, 80)
+	if err != nil {
+		return err
+	}
+	if err := os.WriteFile("graph.json", dat, 0644); err != nil {
+		return err
+	}
+
 	return nil
 }

pkg/graph/graph.go

@@ -0,0 +1,88 @@
+package graph
+
+import (
+	"encoding/json"
+	"path/filepath"
+
+	"github.com/oolong-sh/oolong/pkg/keywords"
+	"github.com/oolong-sh/oolong/pkg/notes"
+)
+
+type NodeJSON struct {
+	ID   string  `json:"id"`
+	Name string  `json:"name"`
+	Val  float64 `json:"val"`
+}
+
+type LinkJSON struct {
+	Source string `json:"source"`
+	Target string `json:"target"`
+}
+
+type Graph struct {
+	Nodes []NodeJSON `json:"nodes"`
+	Links []LinkJSON `json:"links"`
+}
+
+func clamp(value, min, max float64) float64 {
+	if value < min {
+		return min
+	}
+	if value > max {
+		return max
+	}
+	return value
+}
+
+const NOTE_NODE_VAL = 50
+
+func SerializeGraph(keywordMap map[string]keywords.Keyword, notes []notes.Note, lowerBound, upperBound float64) ([]byte, error) {
+	nodes := []NodeJSON{}
+	links := []LinkJSON{}
+
+	for _, keyword := range keywordMap {
+		// Only add nodes above the minimum threshold
+		if keyword.Weight >= lowerBound {
+			clampedWeight := clamp(keyword.Weight, lowerBound, upperBound)
+			nodes = append(nodes, NodeJSON{
+				ID:   keyword.Keyword,
+				Name: keyword.Keyword,
+				Val:  clampedWeight,
+			})
+		}
+	}
+
+	for _, note := range notes {
+		// Add Note node with a fixed value
+		noteID := note.Path
+		noteName := filepath.Base(note.Path) // /home/patrick/notes/home/blogs/bayes.md -> bayes.md
+		nodes = append(nodes, NodeJSON{
+			ID:   noteID,
+			Name: noteName,
+			Val:  NOTE_NODE_VAL,
+		})
+
+		// Link notes to keywords
+		for keywordID := range note.Weights {
+			keyword, exists := keywordMap[keywordID]
+			if exists && keyword.Weight >= lowerBound {
+				links = append(links, LinkJSON{
+					Source: noteID,
+					Target: keyword.Keyword,
+				})
+			}
+		}
+	}
+
+	graph := Graph{
+		Nodes: nodes,
+		Links: links,
+	}
+
+	jsonData, err := json.Marshal(graph)
+	if err != nil {
+		return nil, err
+	}
+
+	return jsonData, nil
+}

pkg/keywords/keywords.go

@@ -59,3 +59,21 @@ func NGramsToKeywords(ngmap map[string]*ngrams.NGram) []Keyword {
 
 	return keywords
 }
+
+func NGramsToKeywordsMap(ngmap map[string]*ngrams.NGram) map[string]Keyword {
+	keywords := map[string]Keyword{}
+	threshold := 8.0
+
+	for k, v := range ngmap {
+		w := v.Weight()
+
+		if w > threshold {
+			keywords[k] = Keyword{
+				Keyword: k,
+				Weight:  w,
+			}
+		}
+	}
+
+	return keywords
+}