temp_code/go/merge_file_go/main.go

package main

import (
	"bytes"
	"fmt"
	"github.com/ledongthuc/pdf"
	"io/ioutil"
	"regexp"
	"strings"
)

func GetAllFile(pathname string, pdf_list []string, all_file_list []string) ([]string, []string, error) {
	rd, err := ioutil.ReadDir(pathname)
	if err != nil {
		fmt.Println("read dir fail:", err)
		return pdf_list, all_file_list, err
	}

	for _, fi := range rd {
		if !fi.IsDir() {
			fullName := fi.Name()
			if strings.Index(fullName, ".pdf") != -1 {
				pdf_list = append(pdf_list, fullName)
			}
			if strings.Index(fullName, ".xlsx") != -1 && strings.Index(fullName, ".pdf") != -1 {
				all_file_list = append(all_file_list, fullName)
			}
		}
	}
	return pdf_list, all_file_list, nil
}

func ReadPdf(path string) (string, error) {
	f, r, err := pdf.Open(path)
	// remember close file
	defer f.Close()
	if err != nil {
		return "", err
	}
	var buf bytes.Buffer
	b, err := r.GetPlainText()
	if err != nil {
		return "", err
	}
	buf.ReadFrom(b)
	return buf.String(), nil
}

func re(str string) (re_str string) {
	reg := regexp.MustCompile(`款号：(.+?)颜色`)
	re_str = reg.FindString(str)
	//re_str = reg.FindAllString(str, -1)
	if len(re_str) > 0 {
		return re_str
	} else {
		return
	}
}

func main() {
	var pdf_list []string
	var all_file_list []string

	// 读取文件夹内的所有 pdf && pdf&excel
	pdf_list, all_file_list, _ = GetAllFile(".", pdf_list, all_file_list)

	// 循环读取 pdf, 获取所有款号
	// var code []string
	for _, p := range pdf_list {
		str, _ := ReadPdf(p)
		re_str := re(str)
		fmt.Println(re_str)
	}
}