def get_workbooks(self):
if self.temp_path is None:
return
filenames = []
for name in os.listdir(self.temp_path):
d = name.split('-',1)
d.append(name)
filenames.append(d)
filenames.sort()
for i,filename,pathname in filenames:
yield (
# We currently don't open with on_demand=True here
# as error filters should be lastish in the chain
# so there's not much win.
# However, if we did, getting rid of the temp dirs
# becomes a problem as, on Windows, they can't be
# deleted until the xlrd.Book object is done with
# and we don't know when that might be :-(
xlrd.open_workbook(
os.path.join(self.temp_path,pathname),
pickleable=0,
formatting_info=1,
on_demand=False,
ragged_rows=True
),
filename
)
def parse(cls, template_path):
"""Read and parse the template file specified by *template_path*."""
self = cls()
w = xlrd.open_workbook(template_path, formatting_info=True)
for idx, sht in enumerate(w.sheets()):
self.tmp_groups = {}
sheet_macro = Macro(sht.name)
self.meta[idx] = meta = SheetMetaInfo()
meta.sheet_macro = sheet_macro
for rx in xrange(sht.nrows):
cols = ((cx, sht.cell_value(rx, cx)) for cx in xrange(sht.ncols))
cols = [col for col in cols if len(unicode(col[-1])) > 0]
if len(cols) == 0:
continue
macros, groups = self.parse_column(cols)
meta.macros += [MacroDef(rx, cx, m, *es) for cx, m, es in macros]
group_starts = [(i, m, es) for (i, m, es) in groups if m._is_group_start]
if len(group_starts) > 0:
group_macros = GroupMacroDef.parse(group_starts)
head = group_starts[0][0]
for g in group_macros:
self.register_group(rx, head, g)
group_ends = ((i, m) for (i, m, es) in groups if not m._is_group_start)
for g in group_ends:
self.process_group(idx, rx, *g)
w = None
return self
def get_workbooks(self):
"""
If the data to be processed is not stored in files or if
special parameters need to be passed to xlrd.open_workbook
then this method must be overriden.
Any implementation must return an iterable sequence of tuples.
The first element of which must be an xlrd.Book object and the
second must be the filename of the file from which the book
object came.
"""
for path in self.get_filepaths():
yield (
xlrd.open_workbook(
path,
pickleable=0,
formatting_info=1,
on_demand=True,
ragged_rows=True),
os.path.split(path)[1]
)
def create_copy(fpath):
wt = open_workbook(fpath, formatting_info=True)
w = copy_book(wt)
return CneBook(w, fpath)
def __init__(self, book, old_path):
self.oldbook = open_workbook(old_path, formatting_info=True)
self.workbook = book
self.style_list = get_xlwt_style_list(self.oldbook)
def check_file(fname, verbose, do_punc=False, fmt_info=0, encoding='ascii', onesheet=''):
print
print fname
if do_punc:
checker = ispunc
else:
checker = None
try:
book = open_workbook(fname, formatting_info=fmt_info, on_demand=True)
except TypeError:
try:
book = open_workbook(fname, formatting_info=fmt_info)
except TypeError:
# this is becoming ridiculous
book = open_workbook(fname)
totold = totnew = totnotnull = 0
if onesheet is None or onesheet == "":
shxrange = range(book.nsheets)
else:
try:
shxrange = [int(onesheet)]
except ValueError:
shxrange = [book.sheet_names().index(onesheet)]
for shx in shxrange:
sheet = book.sheet_by_index(shx)
ngoodrows = number_of_good_rows(sheet, checker)
ngoodcols = number_of_good_cols(sheet, checker, nrows=ngoodrows)
oldncells = sheet.nrows * sheet.ncols
newncells = ngoodrows * ngoodcols
totold += oldncells
totnew += newncells
nnotnull = 0
sheet_density_pct_s = ''
if verbose >= 2:
colxrange = range(ngoodcols)
for rowx in xrange(ngoodrows):
rowtypes = sheet.row_types(rowx)
for colx in colxrange:
if rowtypes[colx] not in null_cell_types:
nnotnull += 1
totnotnull += nnotnull
sheet_density_pct = (nnotnull * 100.0) / max(1, newncells)
sheet_density_pct_s = "; den = %5.1f%%" % sheet_density_pct
if verbose >= 3:
# which rows have non_empty cells in the right-most column?
lastcolx = sheet.ncols - 1
for rowx in xrange(sheet.nrows):
cell = sheet.cell(rowx, lastcolx)
if cell.ctype != XL_CELL_EMPTY:
print "%s (%d, %d): type %d, value %r" % (
cellname(rowx, lastcolx), rowx, lastcolx, cell.ctype, cell.value)
if (verbose
or ngoodrows != sheet.nrows
or ngoodcols != sheet.ncols
or (verbose >= 2 and ngoodcells and sheet_density_pct < 90.0)
):
if oldncells:
pctwaste = (1.0 - float(newncells) / oldncells) * 100.0
else:
pctwaste = 0.0
shname_enc = safe_encode(sheet.name, encoding)
print "sheet #%2d: RxC %5d x %3d => %5d x %3d; %4.1f%% waste%s (%s)" \
% (shx, sheet.nrows, sheet.ncols,
ngoodrows, ngoodcols, pctwaste, sheet_density_pct_s, shname_enc)
if hasattr(book, 'unload_sheet'):
book.unload_sheet(shx)
if totold:
pctwaste = (1.0 - float(totnew) / totold) * 100.0
else:
pctwaste = 0.0
print "%d cells => %d cells; %4.1f%% waste" % (totold, totnew, pctwaste)
