Prev: Handling text lines from files with some (few) starnge chars
Next: [RELEASE] Python 2.7 release candidate 1 released
From: noydb on 5 Jun 2010 19:47
Is there a way to save a .xls file (the first worksheet) as a .dbf
or .csv without opening an instance of Excel with win32com.client
(been awhile, is this the best module these days for v2.5)? In case a
computer does not have Excel (2007) installed.
From: Tim Chase on 5 Jun 2010 21:31
On 06/05/2010 06:47 PM, noydb wrote:
> Is there a way to save a .xls file (the first worksheet) as a .dbf
> or .csv without opening an instance of Excel with win32com.client
> (been awhile, is this the best module these days for v2.5)? In case a
> computer does not have Excel (2007) installed.

Use the "xlrd" module[1]

############################
import csv
import xlrd

FILE_NAME = 'example.xls'
wb = xlrd.open_workbook(FILE_NAME)

for name in wb.sheet_names():
out = file('%s.csv' % name, 'wb')
writer = csv.writer(out)
sheet = wb.sheet_by_name(name)
for row in xrange(sheet.nrows):
writer.writerow([
sheet.cell_value(row, col)
for col in xrange(sheet.ncols)
])
out.close()
#############################

You say you only want the first sheet, so adjust accordingly.

-tkc

[1]
http://pypi.python.org/pypi/xlrd/




From: noydb on 6 Jun 2010 19:59
On Jun 5, 9:31 pm, Tim Chase <python.l...(a)tim.thechases.com> wrote:
> On 06/05/2010 06:47 PM, noydb wrote:
>
> > Is there a way to save a .xls file (the first worksheet) as a .dbf
> > or .csv without opening an instance of Excel with win32com.client
> > (been awhile, is this the best module these days for v2.5)?  In case a
> > computer does not have Excel (2007) installed.
>
> Use the "xlrd" module[1]
>
> ############################
> import csv
> import xlrd
>
> FILE_NAME = 'example.xls'
> wb = xlrd.open_workbook(FILE_NAME)
>
> for name in wb.sheet_names():
>    out = file('%s.csv' % name, 'wb')
>    writer = csv.writer(out)
>    sheet = wb.sheet_by_name(name)
>    for row in xrange(sheet.nrows):
>      writer.writerow([
>        sheet.cell_value(row, col)
>        for col in xrange(sheet.ncols)
>        ])
>    out.close()
> #############################
>
> You say you only want the first sheet, so adjust accordingly.
>
> -tkc
>
> [1]http://pypi.python.org/pypi/xlrd/

Many thanks Tim, this worked well!

In the interest of learning, anyone have a XLS to DBF solution?
From: Tim Chase on 6 Jun 2010 21:47
On 06/06/2010 06:59 PM, noydb wrote:
> On Jun 5, 9:31 pm, Tim Chase<python.l...(a)tim.thechases.com> wrote:
>> [1]http://pypi.python.org/pypi/xlrd/
>
> Many thanks Tim, this worked well!
>
> In the interest of learning, anyone have a XLS to DBF solution?

This becomes considerably trickier unless you're willing to have
all your DBF fields be CHAR-format rather than their actual
data-type. In an Excel file, one column can hold multiple
data-types (text, numbers, formulas, dates, currency, etc)
whereas (from my humble understanding) a single column in a DBF
must be of the same data-type.

That said, if you want to attempt it, I'd suggest using

http://dbfpy.sourceforge.net/

to create the DBF files using code similar to my CSV version
(which mostly iterates over the Excel workbook and dumps the
contents to CSV files -- just change it to dump the contents to
your DBF).

-tkc





 | 
Pages: 1
Prev: Handling text lines from files with some (few) starnge chars
Next: [RELEASE] Python 2.7 release candidate 1 released