I create lots of(50+) *.xls file by C++ , if these files are read directly by pandas, like this
df = pd.read_excel(’./1.xls’)
it is not possible, if 1.xls is opened and than save as 2.xls manually, pandas can read 2.xls well.
so I want to use calc script to do this job , open an *.xls file, and save as another xls file ,how to write calc script ?
import shutil
shutil.copyfile('1.xls', '2.xls')
it doesn’t work
May be you should describe one or two bits more verbose what you are doing and what exactly doesnt work
I write the code as your post, the size of 2.xls is the same as 1.xls,
df = pd.read_excel(’./2.xls’) not work
when I open the 1.xls genereated by c++, the calc prompts information as attached picture
using default setting of calc, save opened xls file as 2.xls, 2.xls has larger size than 1.xls
the initila 1.xls lacks some information needed by calcs, I guess
Let me try to phrase the question as I understand it.
“I create XLS files using some C++ code (some library?); and the resulting XLS files can’t be opened in Pandas (likely, the generated XLS files are somehow defective). So I want LibreOffice to fix them for me; what can I do to have LibreOffice open and save all the generated XLS files?”
I’d use:
soffice --convert-to xls --outdir /path/to/outdir /path/to/indir/*.xls
And I’d definitely file a bug against the used C++ library.
using C++ ofsteam
std::ofstream Fs(“1.xls”);
if (!Fs.is_open()) {
std::cout << “error!” << std::endl;
return;
}
Thanks for your reminding, its csv file, it’s wrong that I think its an xls file
soffice --headless --convert-to xlsx --infilter=“csv:9,59,34,UTF8” 1.csv
This works for me, 9 is ASCII value of \t
again … pandas can read .csv and also it is able to write xlsx
yes, I change pd.read_excel() to pd.read_csv()