go-sh-manymanuals/exp/bleve.go

package main

import (
	"fmt"
	"log"
	"os"
	"path/filepath"
	"strings"

	"github.com/blevesearch/bleve/v2"
	pdf "github.com/johbar/go-poppler"
)

func readPDF(name string) (string, error) {
	doc, err := pdf.Open(name)
	if err != nil {
		return "", err
	}
	defer doc.Close()

	var txt string
	for i := 0; i < doc.GetNPages(); i++ {
		txt += doc.GetPage(i).Text()
	}

	return txt, nil
}

type datasheet struct {
	filename string
	filepath string
	contents string
}

func main() {
	// create or open a bleve index
	var index bleve.Index
	var err error
	mapping := bleve.NewIndexMapping()
	index, err = bleve.New("test.bleve", mapping)
	if err != nil {
		index, err = bleve.Open("test.bleve")
		if err != nil {
			log.Fatal(err)
		}
	}

	// gather all datasheets - filename, filepath, contents
	var datasheets []datasheet
	if err := filepath.Walk("../datasheets", func(path string, info os.FileInfo, err error) error {
		if err != nil {
			return err
		}

		name := info.Name()
		if strings.HasSuffix(name, "pdf") {
			contents, err := readPDF(path)
			if err != nil {
				return err
			}

			datasheet := datasheet{
				filename: name,
				filepath: path,
				contents: contents,
			}

			datasheets = append(datasheets, datasheet)
		}

		return nil
	}); err != nil {
		log.Fatal(err)
	}

	// index by filename and by content. no idea if this is The Way To Go
	for _, datasheet := range datasheets {
		contents, err := readPDF(datasheet.filepath)
		if err != nil {
			log.Fatal(err)
		}

		if err := index.Index(datasheet.filename, contents); err != nil {
			log.Fatal(err)
		}
	}

	// query for something! change the string to test other possibilities
	query := bleve.NewMatchQuery("Enhanced-Page-Mode")
	search := bleve.NewSearchRequest(query)
	searchResults, err := index.Search(search)
	if err != nil {
		log.Fatal(err)
	}

	// print out the results
	fmt.Println(searchResults)
}