t0ha / ezodf

ezodf is a Python package to create new or open existing OpenDocument (ODF) files to extract, add, modify or delete document data, forked from dead project https://bitbucket.org/mozman/ezodf

EzODF.py

If you want to support us

Abstract

ezodf is a Python package to create new or open existing OpenDocument (ODF) files to extract, add, modify or delete document data.

a simple example:

from ezodf import newdoc, Paragraph, Heading, Sheet

odt = newdoc(doctype='odt', filename='text.odt')
odt.body += Heading("Chapter 1")
odt.body += Paragraph("This is a paragraph.")
odt.save()

ods = newdoc(doctype='ods', filename='spreadsheet.ods')
sheet = Sheet('SHEET', size=(10, 10))
ods.sheets += sheet
sheet['A1'].set_value("cell with text")
sheet['B2'].set_value(3.141592)
sheet['C3'].set_value(100, currency='USD')
sheet['D4'].formula = "of:=SUM([.B2];[.C3])"
pi = sheet[1, 1].value
ods.save()

for more examples see: /examples folder

Dependencies

• lxml <http://codespeak.net/lxml/> for painless serialisation with prefix declaration (xlmns:prefix="global:namespace:specifier") in the root element. Declarations for unused prefixes are also possible.
• nose <https://nose.readthedocs.org> for testing

For CPython 2.6 compatibility:

• weakrefset <https://pypi.python.org/pypi/weakrefset> for fixing incompatibility with weakref module before 2.7
• unittest2 <https://pypi.python.org/pypi/unittest2> for asserts like in python 2.7+

The target platform is CPython 2.7 and CPython 3.2+, work on compability with CPython 2.6 is in progress.

Installation

with pip:

pip install ezodf

or from source:

python setup.py install

Documentation

http://packages.python.org/ezodf

send feedback to [email protected]

ezodf can be found on GitHub at:

https://github.com/T0ha/ezodf

ezodf's Issues

ODS sheets: Wrong results with sparse tables

At least the following function return wrong results when I create a sheet that contains data only in the cells (for example) G1, Z1, CD1, CE1:
sheet.ncols()
sheet["CD1"].value
and many more.
It seems as if the table:number-columns-repeated attribute is simply ignored.

Support for hyperlinks in ODS

ODS sheets seem to be missing support for adding hyperlinks.

GPL-3.0 license text in __init__.py

ezodf/ezodf/__init__.py

Line 9 in f9675c1

# This program is free software: you can redistribute it and/or modify

still contains the GPL-3.0 license text despite this project being relicensed to MIT years ago. Should that text be removed?

newdoc without doctype

See this from your docu

# create a new text document
doc = ezodf.newdoc(doctype='odt', filename='text.odt')
# or open an existing text document
doc = ezodf.opendoc('text.odt')

opendoc doesn't need a doctype because it recoginize it from the filename or the filecontent.
newdoc shouldn't need doctype, too. Because it can recoginize the needed type from the filename suffix.

doctype should be optional (per default None)

pyexcel-ezodf, a fork of this reprository will take the maintenance task

Due to the long deplay of #18, I have decided to publish the my version of ezodf out to pypi. Namespace and other functionality are backward compatible.

Due to my limited knowledge of open document specs, anyone interested join the maintainer team will be welcome to join.

When @T0ha will be coming back to continue the leadership, I am happy to push the changes back.

pyexcel#2

Insert images to ods files

Hi T0ha!

We would like to add images to ods files using your library.

Table in text document

Is it possible to add a table in a text document ? I don't find any information in the documentation

ODS: Add rows and columns automatically

Hi, this is more a nice to have enhancement request than an issue.

Is it possible/easy for spreadsheets to create columns and rows automatically when adding a value instead of raising an index out of range error?

For example:

sheet = ezodf.Table('Sheet')
print(sheet.nrows(), sheet. ncols())
#0 0
sheet[10, 20].set_value('A value')
print(sheet.nrows(), sheet. ncols())
#10 20

Cheers

Improve README.md with project status

Dear folks,

could you please update the README with a section describing your project organisation and the current status.

I assume this is a one-person-freetime-project. But people need to know.

And the question is if that project is still maintained. The question is if new contributions (e.g. PRs) will be handled.

Or is the project still orphaned?

Release tag

Could someone please tag a release? There hasn't been one since December, and 0.3.0 has issues with python3.

Even just a 0.3.1 on github would be helpful https://github.com/T0ha/ezodf/releases

Thanks!

how to assign styles to columns in a table in a odt file ?

Hello,
I would like to add a table in a odt file with columns with precise dimensions using existing styles in this file as already done for an existing table.

<text:p text:style-name="P21"/><table:table table:name="Exigence" table:style-name="Exigence"><table:table-column table:style-name="Exigence.A"/><table:table-column table:style-name="Exigence.B"/><table:table-column table:style-name="Exigence.C"/><table:table-column table:style-name="Exigence.D

I can assign a style to a cell in a table creates :

table = ezodf.Table(size=(1, 4))
table.set_cell(pos=(0,0),cell=ezodf.Cell(style_name="Exigence.A1"))

So how can I assign a style in the columns?

Add support for Flat XML OpenDocumentFormat files

When I try to open a .odt file, I get:

>>> textproc = ezodf.opendoc("foo.fodt")
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "/usr/local/lib/python2.7/dist-packages/ezodf/document.py", line 54, in opendoc
    mime_type = __detect_mime_type(fm)
  File "/usr/local/lib/python2.7/dist-packages/ezodf/document.py", line 69, in __detect_mime_type
    mime_type = MIMETYPES[ext]
KeyError: '.fodt'

Same for .ods files

Support for .fods, .fodt, etc. files

The .fxxx variants are the same as the standard variants, but not compressed. Thus, they play nicely with GitHub. See LibreOffice for reference... It should be easy to add support for these files, since they are the "same" just remove the decompression step.

support = operator

This is a wish.

Currently you have to do
sheet['B2'].set_value(3.141592)

IMO it would be more "natural" if it could be done like this, too:
sheet['B2'] = 3.141592

Conversion of midnight datetime string to date instead of datetime

I have to read a column of date/time, if I read the date "15/07/2014 23:00:00" there is no problem I get 2014-07-15T23:00:00. But if I read 15/07/2014 00:00:00, I get 2014-07-15 instead of 2014-07-15T00:00:00. I lose the time.

How to use nested list?

Hello,I have a trouble to make a nested list.
The code is :
temp = ezodf.newdoc("odt","")
a = ezodf.List()
b = ezodf.List()
a.append(ezodf.ListItem('1'))
b.append(ezodf.ListItem('2'))
a += b
temp.saveas("temp.odt")

And i load the file(temp.odt) with Openoffice4.1.2 , (python3.4 ,win10 64bit environment)
Openoffice only shows the "a" list in the document, i feel frustrated about the situation.
Did i misunderstand something about the module or the function?
Please let me know.

At the end, thanks for the efforts what you devoted.It saves my time.

[feature request] insert formula in odt, the writer's file

Just as the title says. is it possilbe to insert math formula in the writer document?

Useless error message if wrong file path is given

If one calls opendoc() with a non-existing path:

opened = ezodf.opendoc("/home/user/path/that/does/not/exist.ods")

This is what one gets:

Traceback (most recent call last):
  File "/home/yyyyyyy/loop_list_test.py", line 9, in <module>
    adapted_ods = ezodf.opendoc(adapted)
  File "/usr/local/lib/python2.7/dist-packages/ezodf/document.py", line 54, in opendoc
    mime_type = __detect_mime_type(fm)
  File "/usr/local/lib/python2.7/dist-packages/ezodf/document.py", line 69, in __detect_mime_type
    mime_type = MIMETYPES[ext]
KeyError: '.ods'

It's just a suggestion but a simple check if that file actually exists and a more meaningful error message would be very helpful here.

__detect_mime_type is broken

__detect_mime_type has a bug in line 66/67:

ezodf/ezodf/document.py

Lines 56 to 68 in f9675c1

	def __detect_mime_type(file_manager):
	mime_type = file_manager.get_text('mimetype')
	if mime_type is not None:
	return mime_type
	# Fall-through to next mechanism
	entry = file_manager.manifest.find('/')
	if entry is not None:
	mime_type = entry.get(CN('manifest:media-type'))
	else:
	# use file ext name
	ext = os.path.splitext(file_manager.zipname)[1]
	mime_type = MIMETYPES[ext]
	return mime_type

os.path.splitext returns the file extension with a leading dot (e. g. .ods). But MIMETYPES only contains the extensions without the dot:

ezodf/ezodf/const.py

Line 23 in f9675c1

'ots': "application/vnd.oasis.opendocument.spreadsheet-template",

Project status and relation to ezdof2

Hi. Do you know about this?
https://github.com/iwschris/ezodf2

The last commit there was 2013. But the README says that it is derived from "ezdof".
But your last commit is much more fresh. ;)

So can you give some more infos about it? I would assume that your "ezdof" is newer and more stable then "ezdof2"?

Importing requirements in setup

This library requires the lxml package, but there are imports in setup.py file causing the requirement to import before setup can be run and lxml can be installed.

    Traceback (most recent call last):
      File "<string>", line 14, in <module>
      File "/home/dave/venvs/individualcar/build/ezodf/setup.py", line 14, in <module>
        from ezodf import VERSION
      File "ezodf/__init__.py", line 27, in <module>
        from .document import opendoc, newdoc
      File "ezodf/document.py", line 16, in <module>
        from .xmlns import subelement, CN, etree, wrap, ALL_NSMAP, fake_element
      File "ezodf/xmlns.py", line 10, in <module>
        from lxml import etree
    ImportError: No module named lxml

Line break in ods

I tried to add line breaks in spreadsheet like CTRL+ENTER. Is there a way to perform this?

Best regards