def __init__(self, filename=None, content=NEWWORKBOOK):
self.streams = {}
self.sheets = RecordList(BoundSheet)
self.fonts = RecordListRW(FontRecord)
self.numberformats = RecordList(NumberFormat)
self.extendedformats = RecordListRW(ExtendedFormat)
self.staticstrings = StaticStrings()
self.names = RecordList(NameRecord)
self.supbooks = RecordList(SupBookRecord)
if filename or content:
self.read(filename, content)
def __init__(self, filename=None,content=NEWWORKBOOK):
self.streams={}
self.sheets=RecordList(BoundSheet)
self.fonts=RecordListRW(FontRecord)
self.numberformats=RecordList(NumberFormat)
self.extendedformats=RecordListRW(ExtendedFormat)
self.staticstrings=StaticStrings()
self.names=RecordList(NameRecord)
self.supbooks=RecordList(SupBookRecord)
if filename or content:
self.read(filename,content)
class HSSFWorkbook(RecordContainer):
MAX_ROW = 0xFFFF
MAX_COLUMN = 0x00FF
# The maximum number of cell styles in a .xls workbook.
# The 'official' limit is 4,000, but POI allows a slightly larger number.
# This extra delta takes into account built-in styles that are automatically
# created for new workbooks
#
# See http://office.microsoft.com/en-us/excel-help/excel-specifications-and-limits-HP005199291.aspx
MAX_STYLES = 4030
def __init__(self, filename=None,content=NEWWORKBOOK):
self.streams={}
self.sheets=RecordList(BoundSheet)
self.fonts=RecordListRW(FontRecord)
self.numberformats=RecordList(NumberFormat)
self.extendedformats=RecordListRW(ExtendedFormat)
self.staticstrings=StaticStrings()
self.names=RecordList(NameRecord)
self.supbooks=RecordList(SupBookRecord)
if filename or content:
self.read(filename,content)
def write(self, filename):
cfb = CFBWriter()
for name in sorted(self.streams.iterkeys()):
cfb.put(name,self.streams[name])
cfb.put(('Workbook',),self.getdata())
cfb.write(filename)
def getdata(self):
self.staticstrings.newstrings=[]
self.staticstrings.newstring_map={}
for sheet in self.sheets:
if sheet.sheet:
sheet.sheetdata=sheet.sheet.getdata()
self.staticstrings.strings=self.staticstrings.newstrings
del self.staticstrings.newstring_map
if 0x0085 not in self.urecord:
self.last_record.next=self.sheets
result=[]
first=self.records.next
sheetpos=-1
while first:
if sheetpos==-1 and first is self.sheets:
sheetpos=len(result)
result.append(first.get_data(self))
first=first.next
reslen=sum(map(len,result))
result.append(self.staticstrings.getdata(self,reslen))
result.append(struct.pack('<HH',0x000A,0)) #EOF
reslen+=len(result[-2])+4
for sheet in self.sheets:
sheet.position_of_BOF=reslen
result.append(sheet.sheetdata)
reslen+=len(sheet.sheetdata)
result[sheetpos]=self.sheets.get_data(self)
return ''.join(result)
def read(self, filename, content):
if filename:
filehandle=open(filename)
cfb = CFBReader(filehandle)
self.streams = cfb.dirtree
del cfb
# Normally, the Workbook will be in a POIFS Stream
# called "Workbook". However, some XLS generators use "WORKBOOK"
workbook=None
for wb in ('Book','BOOK','WORKBOOK','Workbook'):
if (wb,) in self.streams:
workbook = self.streams.pop((wb,))
if not workbook:
raise IOError('The file does not contain a Workbook-entry')
content=workbook.data
loaders={
0x0018: self.names.read,
0x0031: self.fonts.read,
0x0059: self.read_xct,
0x005A: self.read_crn,
0x0085: self.sheets.read,
0x00e0: self.extendedformats.read,
0x00fc: self.staticstrings.read,
0x00ff: Record.ignore,
0x01ae: self.supbooks.read,
0x041E: self.numberformats.read,
}
urecord={}
self.records=Record(0,0)
last_record=self.records
ofs=0
for sid, data in record_stream(content):
if sid==0x000A: #EOF
break
new_record=loaders.get(sid,Record)(sid,data)
if new_record:
last_record.next=new_record
last_record=new_record
if sid not in urecord:
urecord[sid]=new_record
#if new_record.__class__==Record:
# print '%04x(%08x): %s'%(sid,ofs,poi.utils.DEBUG_RECORDS.get(sid))
ofs+=len(data)
self.urecord=urecord
self.last_record=last_record
pos=len(content)
for sheet in sorted(self.sheets,key=lambda s:s.position_of_BOF,reverse=True):
sheet.sheetdata=content[sheet.position_of_BOF:pos]
pos=sheet.position_of_BOF
self.numberformats_map=dict(std_format_strings)
self.numberformats_map.update(dict([(nf.index,nf.format) for nf in self.numberformats]))
def read_xct(self, sid, data):
cnt, itab = struct.unpack_from('<hH',data+'\0\0',4)
supbook=self.supbooks[-1]
sheet=SupBookSheet(supbook.sheets[itab],cnt)
supbook.sheets[itab]=sheet
self.supbooksheet=sheet
def read_crn(self, sid, data):
self.supbooksheet.append(data)
def add_extformat(self, xf):
if not hasattr(xf,'put_record'):
xf = self.extendedformats[xf]
return xf.put_record(self)
class HSSFWorkbook(RecordContainer):
MAX_ROW = 0xFFFF
MAX_COLUMN = 0x00FF
# The maximum number of cell styles in a .xls workbook.
# The 'official' limit is 4,000, but POI allows a slightly larger number.
# This extra delta takes into account built-in styles that are automatically
# created for new workbooks
#
# See http://office.microsoft.com/en-us/excel-help/excel-specifications-and-limits-HP005199291.aspx
MAX_STYLES = 4030
def __init__(self, filename=None, content=NEWWORKBOOK):
self.streams = {}
self.sheets = RecordList(BoundSheet)
self.fonts = RecordListRW(FontRecord)
self.numberformats = RecordList(NumberFormat)
self.extendedformats = RecordListRW(ExtendedFormat)
self.staticstrings = StaticStrings()
self.names = RecordList(NameRecord)
self.supbooks = RecordList(SupBookRecord)
if filename or content:
self.read(filename, content)
def write(self, filename):
cfb = CFBWriter()
for name in sorted(self.streams.iterkeys()):
cfb.put(name, self.streams[name])
cfb.put(('Workbook', ), self.getdata())
cfb.write(filename)
def getdata(self):
self.staticstrings.newstrings = []
self.staticstrings.newstring_map = {}
for sheet in self.sheets:
if sheet.sheet:
sheet.sheetdata = sheet.sheet.getdata()
self.staticstrings.strings = self.staticstrings.newstrings
del self.staticstrings.newstring_map
if 0x0085 not in self.urecord:
self.last_record.next = self.sheets
result = []
first = self.records.next
sheetpos = -1
while first:
if sheetpos == -1 and first is self.sheets:
sheetpos = len(result)
result.append(first.get_data(self))
first = first.next
reslen = sum(map(len, result))
result.append(self.staticstrings.getdata(self, reslen))
result.append(struct.pack('<HH', 0x000A, 0)) #EOF
reslen += len(result[-2]) + 4
for sheet in self.sheets:
sheet.position_of_BOF = reslen
result.append(sheet.sheetdata)
reslen += len(sheet.sheetdata)
result[sheetpos] = self.sheets.get_data(self)
return ''.join(result)
def read(self, filename, content):
if filename:
filehandle = open(filename)
cfb = CFBReader(filehandle)
self.streams = cfb.dirtree
del cfb
# Normally, the Workbook will be in a POIFS Stream
# called "Workbook". However, some XLS generators use "WORKBOOK"
workbook = None
for wb in ('Book', 'BOOK', 'WORKBOOK', 'Workbook'):
if (wb, ) in self.streams:
workbook = self.streams.pop((wb, ))
if not workbook:
raise IOError('The file does not contain a Workbook-entry')
content = workbook.data
loaders = {
0x0018: self.names.read,
0x0031: self.fonts.read,
0x0059: self.read_xct,
0x005A: self.read_crn,
0x0085: self.sheets.read,
0x00e0: self.extendedformats.read,
0x00fc: self.staticstrings.read,
0x00ff: Record.ignore,
0x01ae: self.supbooks.read,
0x041E: self.numberformats.read,
}
urecord = {}
self.records = Record(0, 0)
last_record = self.records
ofs = 0
for sid, data in record_stream(content):
if sid == 0x000A: #EOF
break
new_record = loaders.get(sid, Record)(sid, data)
if new_record:
last_record.next = new_record
last_record = new_record
if sid not in urecord:
urecord[sid] = new_record
#if new_record.__class__==Record:
# print '%04x(%08x): %s'%(sid,ofs,poi.utils.DEBUG_RECORDS.get(sid))
ofs += len(data)
self.urecord = urecord
self.last_record = last_record
pos = len(content)
for sheet in sorted(self.sheets,
key=lambda s: s.position_of_BOF,
reverse=True):
sheet.sheetdata = content[sheet.position_of_BOF:pos]
pos = sheet.position_of_BOF
self.numberformats_map = dict(std_format_strings)
self.numberformats_map.update(
dict([(nf.index, nf.format) for nf in self.numberformats]))
def read_xct(self, sid, data):
cnt, itab = struct.unpack_from('<hH', data + '\0\0', 4)
supbook = self.supbooks[-1]
sheet = SupBookSheet(supbook.sheets[itab], cnt)
supbook.sheets[itab] = sheet
self.supbooksheet = sheet
def read_crn(self, sid, data):
self.supbooksheet.append(data)
def add_extformat(self, xf):
if not hasattr(xf, 'put_record'):
xf = self.extendedformats[xf]
return xf.put_record(self)