vinci1it2000 / xlref

Excel table reader.

xlref: Excel table reader.

release:	1.2.4
date:	2024-02-27 15:30:00
repository:	https://github.com/vinci1it2000/xlref
pypi-repo:	https://pypi.org/project/xlref/
docs:	http://xlref.readthedocs.io/
wiki:	https://github.com/vinci1it2000/xlref/wiki/
download:	http://github.com/vinci1it2000/xlref/releases/
donate:	https://donorbox.org/xlref
keywords:	data, excel, tables, parser, reference, ranges
developers:
license:	EUPL 1.1+

About xlref

xlref is an useful library to capture by a simple reference (e.g., A1(RD):..:RD) a table with non-empty cells from Excel-sheets when its exact position is not known beforehand.

This code was inspired by the xleash module of the pandalone library. The reason of developing a similar tool was to have a smaller library to install and improve the performances of reading .xlsx files.

Installation

To install it use (with root privileges):

$ pip install xlref

Or download the last git version and use (with root privileges):

$ python setup.py install

Reference Syntax

The capturing is preformed according to an excel like reference syntax and the non-empty cells of the targeted excel-sheet. The syntax is defined as follows:

[<excel>]#[<sheet>!]<st-cel>[(<moves>)][:<nd-cel>[(<moves>)]][:<expansion>][<filters>]

Note

The fields between square parenthesis are optionals.

Follows the description of the parameters:

• excel: excel file path relative to the parent reference file directory. If not defined, the parent reference excel is inherited.

• sheet: excel sheet name if not defined, the parent reference excel sheet name is inherited.

• st-cel: first cell coordinate of excel range. The cell coordinate (i.e., <column><row>) is defined by a column (letter) and row (number), like in excel. xlref allows two special characters ^ and _, that represents the leftmost/topmost and rightmost/bottommost non-empty cell column/row.

• moves: the sequence of primitive directions (i.e., L:left, U: up, R: right, D: down) that xlref uses iteratively for finding the first non-empty cell. The allowed primitive direction combinations are L, U, R, D, LD, LU, UL, UR, RU, RD, DL, and DR. The following diagram shows the graphically the moves from the starting cell X:

          U
   UL◄───┐▲┌───►UR
  LU     │││     RU
   ▲     │││     ▲
   │     │││     │
   └─────┼│┼─────┘
  L◄──────X──────►R
   ┌─────┼│┼─────┐
   │     │││     │
   ▼     │││     ▼
  LD     │││     RD
   DL◄───┘▼└───►DR
          D
  

• nd-cel: second cell coordinate of excel range. It has the same syntax of the st-cel, but it has and extra special character .. This represents the column or row of the st-cel after the application of the moves.

• expansion: the sequence of primitive directions to expand the captured range.

• filters: list of string and or dictionaries that defines the filters to apply iteratively on the captured range.

Reference Reading Steps

The library performs the following steps to read a reference:

1. Open the excel file or inherits the parent's one,
2. Open the sheet by its name or inherits the parent's one,
3. Set the first range cell,
4. Move the first cell according to the specified moves until it finds the first non-empty cell,
5. Set the second range cell or inherits the moved first range cell,
6. Move the second cell like in point 4,
7. Expand the range according to the defined expansions,
8. Apply the iteratively the filters on the captured range.

Tutorial

.. testsetup::

    >>> import os.path as osp
    >>> from setup import mydir
    >>> _ref = osp.join(mydir, 'tests/files/excel.xlsx#ref!A1(RD):RD[%s]')

A typical example is capturing a table with a "header" row and convert into a dictionary. The code below shows how to do it:

>>> import xlref as xl
>>> _ref = 'excel.xlsx#ref!A1(RD):RD[%s]'  # doctest: +SKIP
>>> ref = xl.Ref(_ref % '"dict"')
>>> ref.range  # Captured range.
B2:C28
>>> values = ref.values; values  # Captured values.
{...}
>>> values['st-cell-move']
'#D5(RU):H1(DL)'

You can notice from the code above that all the values of the dictionary are references. To parse it recursively, there are two options:

1. add the "recursive" filter before the "dict":

   >>> values = xl.Ref(_ref % '"recursive", "dict"').values
   >>> values['st-cell-move'].tolist()
   [[1.0, 2.0, 3.0],
    [4.0, 5.0, 6.0],
    [7.0, 8.0, 9.0]]

2. apply a filter onto dictionary' values using the extra functionality of the "dict" filter:

   >>> values = xl.Ref(_ref % '{"fun": "dict", "value":"ref"}').values
   >>> values['st-cell-move'].tolist()
   [[1.0, 2.0, 3.0],
    [4.0, 5.0, 6.0],
    [7.0, 8.0, 9.0]]

You have also the possibility to define and use your custom filters as follows:

>>> import numpy as np
>>> xl.FILTERS['my-filter'] = lambda parent, x: np.sum(x)
>>> xl.Ref('#D5(RU):H1(DL)["my-filter"]', ref).values
45.0

An alternative way is to use directly the methods of the filtered results as follows:

>>> xl.Ref('#D5(RU):H1(DL)["sum"]', ref).values
45.0