21
0
xls/workbook.go

324 lines
7.6 KiB
Go
Raw Normal View History

2015-03-19 10:39:41 +01:00
package xls
import (
"bytes"
"encoding/binary"
2018-12-03 22:16:44 +01:00
"golang.org/x/text/encoding/charmap"
2015-03-19 10:39:41 +01:00
"io"
2016-05-28 04:39:46 +02:00
"os"
2015-03-19 10:39:41 +01:00
"unicode/utf16"
)
2015-09-30 04:40:01 +02:00
//xls workbook type
2015-03-19 10:39:41 +01:00
type WorkBook struct {
2015-09-30 05:17:25 +02:00
Is5ver bool
Type uint16
Codepage uint16
Xfs []st_xf_data
Fonts []Font
Formats map[uint16]*Format
//All the sheets from the workbook
sheets []*WorkSheet
Author string
rs io.ReadSeeker
sst []string
continue_utf16 uint16
continue_rich uint16
continue_apsb uint32
dateMode uint16
2015-03-19 10:39:41 +01:00
}
2015-09-30 04:40:01 +02:00
//read workbook from ole2 file
func newWorkBookFromOle2(rs io.ReadSeeker) *WorkBook {
2015-03-19 10:39:41 +01:00
wb := new(WorkBook)
wb.Formats = make(map[uint16]*Format)
// wb.bts = bts
wb.rs = rs
2015-09-30 05:17:25 +02:00
wb.sheets = make([]*WorkSheet, 0)
2015-03-19 10:39:41 +01:00
wb.Parse(rs)
return wb
}
func (w *WorkBook) Parse(buf io.ReadSeeker) {
2015-09-30 04:40:01 +02:00
b := new(bof)
bof_pre := new(bof)
2015-03-19 10:39:41 +01:00
// buf := bytes.NewReader(bts)
offset := 0
for {
2015-09-30 04:40:01 +02:00
if err := binary.Read(buf, binary.LittleEndian, b); err == nil {
bof_pre, b, offset = w.parseBof(buf, b, bof_pre, offset)
2015-03-19 10:39:41 +01:00
} else {
break
}
}
}
func (w *WorkBook) addXf(xf st_xf_data) {
w.Xfs = append(w.Xfs, xf)
}
func (w *WorkBook) addFont(font *FontInfo, buf io.ReadSeeker) {
2016-06-23 04:22:57 +02:00
name, _ := w.get_string(buf, uint16(font.NameB))
2015-03-19 10:39:41 +01:00
w.Fonts = append(w.Fonts, Font{Info: font, Name: name})
}
2015-03-25 07:47:26 +01:00
func (w *WorkBook) addFormat(format *Format) {
2016-05-28 04:39:46 +02:00
if w.Formats == nil {
os.Exit(1)
}
2015-03-25 07:47:26 +01:00
w.Formats[format.Head.Index] = format
2015-03-19 10:39:41 +01:00
}
2015-09-30 04:40:01 +02:00
func (wb *WorkBook) parseBof(buf io.ReadSeeker, b *bof, pre *bof, offset_pre int) (after *bof, after_using *bof, offset int) {
2015-03-19 10:39:41 +01:00
after = b
after_using = pre
2015-03-19 10:39:41 +01:00
var bts = make([]byte, b.Size)
binary.Read(buf, binary.LittleEndian, bts)
buf_item := bytes.NewReader(bts)
switch b.Id {
case 0x809:
2015-09-30 04:40:01 +02:00
bif := new(biffHeader)
2015-03-19 10:39:41 +01:00
binary.Read(buf_item, binary.LittleEndian, bif)
if bif.Ver != 0x600 {
wb.Is5ver = true
}
wb.Type = bif.Type
case 0x042: // CODEPAGE
binary.Read(buf_item, binary.LittleEndian, &wb.Codepage)
case 0x3c: // CONTINUE
if pre.Id == 0xfc {
var size uint16
var err error
2016-06-23 04:22:57 +02:00
if wb.continue_utf16 >= 1 {
size = wb.continue_utf16
wb.continue_utf16 = 0
} else {
err = binary.Read(buf_item, binary.LittleEndian, &size)
}
for err == nil && offset_pre < len(wb.sst) {
2016-06-23 04:22:57 +02:00
var str string
2018-12-03 22:16:44 +01:00
str, err = wb.get_string(buf_item, size)
wb.sst[offset_pre] = wb.sst[offset_pre] + str
2017-08-18 05:15:10 +02:00
if err == io.EOF {
break
}
2015-03-19 10:39:41 +01:00
offset_pre++
err = binary.Read(buf_item, binary.LittleEndian, &size)
2015-03-19 10:39:41 +01:00
}
}
offset = offset_pre
after = pre
after_using = b
2015-03-19 10:39:41 +01:00
case 0xfc: // SST
info := new(SstInfo)
binary.Read(buf_item, binary.LittleEndian, info)
wb.sst = make([]string, info.Count)
2016-06-23 04:22:57 +02:00
var size uint16
var i = 0
2018-09-13 21:10:15 +02:00
// dont forget to initialize offset
offset = 0
2016-06-23 04:22:57 +02:00
for ; i < int(info.Count); i++ {
2017-02-21 09:59:50 +01:00
var err error
2018-12-03 22:16:44 +01:00
err = binary.Read(buf_item, binary.LittleEndian, &size)
if err == nil {
2016-06-23 04:22:57 +02:00
var str string
str, err = wb.get_string(buf_item, size)
wb.sst[i] = wb.sst[i] + str
2017-02-21 09:59:50 +01:00
}
2016-09-30 05:46:51 +02:00
2017-02-21 09:59:50 +01:00
if err == io.EOF {
break
2015-03-19 10:39:41 +01:00
}
}
2016-06-23 04:22:57 +02:00
offset = i
2018-12-03 22:16:44 +01:00
case 0x85: // boundsheet
var bs = new(boundsheet)
2015-03-19 10:39:41 +01:00
binary.Read(buf_item, binary.LittleEndian, bs)
// different for BIFF5 and BIFF8
wb.addSheet(bs, buf_item)
case 0x0e0: // XF
if wb.Is5ver {
xf := new(Xf5)
binary.Read(buf_item, binary.LittleEndian, xf)
wb.addXf(xf)
} else {
xf := new(Xf8)
binary.Read(buf_item, binary.LittleEndian, xf)
wb.addXf(xf)
}
case 0x031: // FONT
f := new(FontInfo)
binary.Read(buf_item, binary.LittleEndian, f)
wb.addFont(f, buf_item)
case 0x41E: //FORMAT
2016-05-28 04:39:46 +02:00
font := new(Format)
binary.Read(buf_item, binary.LittleEndian, &font.Head)
2016-06-23 04:22:57 +02:00
font.str, _ = wb.get_string(buf_item, font.Head.Size)
2016-05-28 04:39:46 +02:00
wb.addFormat(font)
case 0x22: //DATEMODE
binary.Read(buf_item, binary.LittleEndian, &wb.dateMode)
2015-03-19 10:39:41 +01:00
}
return
}
func decodeWindows1251(enc []byte) string {
2018-12-03 22:16:44 +01:00
dec := charmap.Windows1251.NewDecoder()
out, _ := dec.Bytes(enc)
return string(out)
}
2016-06-23 04:22:57 +02:00
func (w *WorkBook) get_string(buf io.ReadSeeker, size uint16) (res string, err error) {
2015-03-19 10:39:41 +01:00
if w.Is5ver {
var bts = make([]byte, size)
2016-06-23 04:22:57 +02:00
_, err = buf.Read(bts)
res = decodeWindows1251(bts)
//res = string(bts)
2015-03-19 10:39:41 +01:00
} else {
var richtext_num = uint16(0)
var phonetic_size = uint32(0)
2015-03-19 10:39:41 +01:00
var flag byte
2016-06-23 04:22:57 +02:00
err = binary.Read(buf, binary.LittleEndian, &flag)
2015-03-19 10:39:41 +01:00
if flag&0x8 != 0 {
2016-06-23 04:22:57 +02:00
err = binary.Read(buf, binary.LittleEndian, &richtext_num)
} else if w.continue_rich > 0 {
richtext_num = w.continue_rich
w.continue_rich = 0
2015-03-19 10:39:41 +01:00
}
if flag&0x4 != 0 {
2016-06-23 04:22:57 +02:00
err = binary.Read(buf, binary.LittleEndian, &phonetic_size)
} else if w.continue_apsb > 0 {
phonetic_size = w.continue_apsb
w.continue_apsb = 0
2015-03-19 10:39:41 +01:00
}
if flag&0x1 != 0 {
var bts = make([]uint16, size)
var i = uint16(0)
for ; i < size && err == nil; i++ {
err = binary.Read(buf, binary.LittleEndian, &bts[i])
}
2018-08-13 23:05:20 +02:00
2018-09-13 21:10:15 +02:00
// when eof found, we dont want to append last element
2018-08-13 23:05:20 +02:00
var runes []rune
if err == io.EOF {
2018-12-03 22:16:44 +01:00
i = i - 1
2018-08-13 23:05:20 +02:00
}
2018-12-03 22:16:44 +01:00
runes = utf16.Decode(bts[:i])
2018-08-13 23:05:20 +02:00
2015-03-19 10:39:41 +01:00
res = string(runes)
if i < size {
2018-12-03 22:16:44 +01:00
w.continue_utf16 = size - i
}
2018-12-03 22:16:44 +01:00
2015-03-19 10:39:41 +01:00
} else {
var bts = make([]byte, size)
2016-06-23 04:22:57 +02:00
var n int
n, err = buf.Read(bts)
if uint16(n) < size {
w.continue_utf16 = size - uint16(n)
err = io.EOF
}
2017-07-26 03:39:26 +02:00
var bts1 = make([]uint16, n)
2016-06-23 04:22:57 +02:00
for k, v := range bts[:n] {
2016-06-20 03:39:08 +02:00
bts1[k] = uint16(v)
}
runes := utf16.Decode(bts1)
res = string(runes)
2015-03-19 10:39:41 +01:00
}
if richtext_num > 0 {
2015-03-19 10:39:41 +01:00
var bts []byte
var seek_size int64
2015-03-19 10:39:41 +01:00
if w.Is5ver {
seek_size = int64(2 * richtext_num)
2015-03-19 10:39:41 +01:00
} else {
seek_size = int64(4 * richtext_num)
2015-03-19 10:39:41 +01:00
}
bts = make([]byte, seek_size)
2016-06-23 04:22:57 +02:00
err = binary.Read(buf, binary.LittleEndian, bts)
if err == io.EOF {
w.continue_rich = richtext_num
}
// err = binary.Read(buf, binary.LittleEndian, bts)
2015-03-19 10:39:41 +01:00
}
if phonetic_size > 0 {
2015-03-19 10:39:41 +01:00
var bts []byte
bts = make([]byte, phonetic_size)
2016-06-23 04:22:57 +02:00
err = binary.Read(buf, binary.LittleEndian, bts)
if err == io.EOF {
w.continue_apsb = phonetic_size
}
2015-03-19 10:39:41 +01:00
}
}
2016-06-23 04:22:57 +02:00
return
2015-03-19 10:39:41 +01:00
}
func (w *WorkBook) addSheet(sheet *boundsheet, buf io.ReadSeeker) {
2016-06-23 04:22:57 +02:00
name, _ := w.get_string(buf, uint16(sheet.Name))
2015-09-30 05:17:25 +02:00
w.sheets = append(w.sheets, &WorkSheet{bs: sheet, Name: name, wb: w})
2015-03-19 10:39:41 +01:00
}
2015-06-16 04:37:11 +02:00
//reading a sheet from the compress file to memory, you should call this before you try to get anything from sheet
2015-09-30 05:17:25 +02:00
func (w *WorkBook) prepareSheet(sheet *WorkSheet) {
2015-03-19 10:39:41 +01:00
w.rs.Seek(int64(sheet.bs.Filepos), 0)
2015-06-16 04:37:11 +02:00
sheet.parse(w.rs)
2015-03-19 10:39:41 +01:00
}
//Get one sheet by its number
2015-09-30 05:17:25 +02:00
func (w *WorkBook) GetSheet(num int) *WorkSheet {
if num < len(w.sheets) {
s := w.sheets[num]
if !s.parsed {
w.prepareSheet(s)
}
2015-09-30 05:17:25 +02:00
return s
} else {
return nil
}
}
//Get the number of all sheets, look into example
func (w *WorkBook) NumSheets() int {
return len(w.sheets)
}
2015-06-16 04:37:11 +02:00
//helper function to read all cells from file
2016-05-28 04:39:46 +02:00
//Notice: the max value is the limit of the max capacity of lines.
//Warning: the helper function will need big memeory if file is large.
2015-03-24 06:06:52 +01:00
func (w *WorkBook) ReadAllCells(max int) (res [][]string) {
res = make([][]string, 0)
2015-09-30 05:17:25 +02:00
for _, sheet := range w.sheets {
2015-03-24 06:06:52 +01:00
if len(res) < max {
max = max - len(res)
2015-09-30 05:17:25 +02:00
w.prepareSheet(sheet)
2015-03-24 06:06:52 +01:00
if sheet.MaxRow != 0 {
leng := int(sheet.MaxRow) + 1
if max < leng {
leng = max
}
temp := make([][]string, leng)
2017-02-21 09:27:01 +01:00
for k, row := range sheet.rows {
2015-03-24 06:06:52 +01:00
data := make([]string, 0)
2017-02-21 09:27:01 +01:00
if len(row.cols) > 0 {
for _, col := range row.cols {
2015-03-24 06:06:52 +01:00
if uint16(len(data)) <= col.LastCol() {
data = append(data, make([]string, col.LastCol()-uint16(len(data))+1)...)
}
str := col.String(w)
2016-09-30 05:46:51 +02:00
2015-03-24 06:06:52 +01:00
for i := uint16(0); i < col.LastCol()-col.FirstCol()+1; i++ {
data[col.FirstCol()+i] = str[i]
}
2015-03-19 10:39:41 +01:00
}
2015-03-24 06:06:52 +01:00
if leng > int(k) {
temp[k] = data
2015-03-19 10:39:41 +01:00
}
}
}
2015-03-24 06:06:52 +01:00
res = append(res, temp...)
2015-03-19 10:39:41 +01:00
}
}
}
return
}