geo cel格式读取_用于将公式导出为ex​​cel和其他格式的新python库

One of the typical challenges when doing export of numerical computations is how to include the way how you achieve the results. Typically if results should be in Excel sheet at the end, you want to see the computation formula if you click on each cell. There is a native way using some Excel driver (like XlsxWriter) available in Python, but if you are dealing with a big Excel sheet, it becomes quite inconvenient. This article is about a new library called Portable Spreadsheet that can easily overcome these problems.

导出数值计算时的典型挑战之一是如何包括实现结果的方式。 通常，如果结果应在末尾的Excel工作表中，则单击每个单元格都希望看到计算公式。 使用Python中可用的某些Excel驱动程序(如XlsxWriter)有一种本机方式，但是如果您要处理一个很大的Excel工作表，它将变得非常不便。 本文是关于一个名为Portable Spreadsheet的新库，它可以轻松克服这些问题。

简单的问题 (Simple problem)

Say that you have an e-shop written in Python and want to export some simple Excel sheet that summarizes revenues per each category. There can be just two columns, one representing the category and another one representing revenues.

假设您有一个用Python编写的电子商店，并且要导出一些简单的Excel表格，其中汇总了每个类别的收入。 只能有两列，一列代表类别，另一列代表收入。

In the end, you want to see some aggregation functions, like the sum of all revenues, the article with minimal revenue, and say the sum of food and drinks as a new category (called grocery). What is crucial, the spreadsheet should be editable, it means, each computation should be exported as a formula (as illustrated in the figure above).

最后，您希望看到一些汇总功能，例如所有收入的总和，收入最小的商品，并将食品和饮料的总和作为一个新类别(称为食品杂货)说出来。 至关重要的是，电子表格应该是可编辑的，这意味着每个计算都应作为公式导出(如上图所示)。

使用可用的Excel驱动程序的方法 (Approach using available Excel drivers)

You can tackle this issue directly using some driver for Excel available in Python. Let’s say you choose XlsxWriter as your driver. Then you have to define each cell value and for the computations the way how you compute it plus the actual value after the computation. The code you have to write would follow the logic:

您可以使用Python中提供的一些Excel驱动程序直接解决此问题。 假设您选择XlsxWriter作为驱动程序。 然后，您必须定义每个像元值，并为计算定义计算方式以及计算后的实际值。 您必须编写的代码将遵循以下逻辑：

import xlsxwriter
workbook = xlsxwriter.Workbook('revenues.xlsx')
worksheet = workbook.add_worksheet()
food_rev = 1000
drinks_rev = 100
headphones_rev = 300
# Write categories
worksheet.write(0, 0, "Food")
worksheet.write(1, 0, "Drinks")
worksheet.write(2, 0, "Headphones")
# Write revenues
worksheet.write(0, 1, food_rev)
worksheet.write(1, 1, drinks_rev)
worksheet.write(2, 1, headphones_rev)
# Write results
worksheet.write_formula(3, 1, "=SUM(B1:B3)",
                        value=(food_rev + drinks_rev + headphones_rev))
worksheet.write_formula(4, 1, "=MIN(B1:B3)",
                        value=min(food_rev, drinks_rev, headphones_rev))
worksheet.write_formula(5, 1, "=B1+B2",
                        value=food_rev + drinks_rev)
# Close the sheet (in order to write it to the file)
workbook.close()

As you can see the code is not simple (nor complete) and you already had to hardwire the positions of each cell in it. You also had to use values for computations multiple times as you cannot directly use the values in the cell.

如您所见，代码并不简单(也不完整)，并且您已经不得不硬编码其中每个单元的位置。 您还必须多次使用值进行计算，因为您无法直接使用单元格中的值。

What more is that when you decide to export to the different format (say JSON), you have to write this code again.

而且，当您决定导出为其他格式(例如JSON)时，必须再次编写此代码。

便携式电子表格方法 (Portable Spreadsheet approach)

Another way how to deal with all these issues elegantly is to use a library called Portable Spreadsheet (installable easily via pip install portable-spreadsheet). Consider the following code as the opposite of the previous one:

优雅地处理所有这些问题的另一种方法是使用一个名为Portable Spreadsheet的库(可通过pip install Portable-spreadsheet轻松安装)。 考虑以下代码与上一个相反：

import portable_spreadsheet as ps
# Create the spreadsheet with the correct labels
sheet = ps.Spreadsheet.create_new_sheet(
    6, 1,
    rows_labels=['Food', 'Drink', 'Headphones', 'Total', 'Minimal', 'Grocery'],
    columns_labels=['Revenue']
)
# Set the values
sheet.loc['Food', 'Revenue'] = 15000
sheet.loc['Drink', 'Revenue'] = 16000
sheet.loc['Headphones', 'Revenue'] = 1000
# Set the computations
sheet.loc['Total', 'Revenue'] = sheet.loc['Food':'Total', 'Revenue'].sum()
sheet.loc['Minimal', 'Revenue'] = sheet.loc['Food':'Total', 'Revenue'].min()
sheet.loc['Grocery', 'Revenue'] = sheet.loc['Food', 'Revenue'] + \
                                  sheet.loc['Drink', 'Revenue']
# Export result to Excel
sheet.to_excel("rev.xlsx")
# Export result to JSON
json = sheet.to_json()

As you can see, you completely removed redundancies in the code, it is shorter, easily readable, and what’s best, you can export not only to Excel but to every reasonable format (JSON, CSV, HTML, etc.). If you are familiar with Pandas DataFrame concept, you can see the straight motivation.

如您所见，您完全删除了代码中的冗余，它更短，更易于阅读，并且最好的是，您不仅可以导出到Excel，还可以导出为每种合理的格式(JSON，CSV，HTML等)。 如果您熟悉Pandas DataFrame概念，则可以看到直接的动机。

它是如何工作的？ (How does it work under the hood?)

Internally, the principle of the Portable Spreadsheet library lies in the word constructing algorithm. Basically, each formula (in Excel, JSON, etc.) can be considered as a world in some language. This language is the Context-Free language defined by Context-Free grammar (see Chomsky hierarchy).

在内部，Portable Spreadsheet库的原理在于单词构造算法。 基本上，每个公式(在Excel，JSON等中)都可以被视为某种语言的世界。 此语言是无上下文语法定义的无上下文语言(请参阅Chomsky层次结构)。

This practically means that all that is needed is the definition of the prefixes and suffices for each operation. Operands here are the coordinates of the cells.

实际上，这意味着所需要做的就是定义每个操作的前缀和充分性。 这里的操作数是单元的坐标。

So, if you want to compute the sum of the items in the column, your operands are the coordinates of slice start and end, the operation is the sum (you have two inputs: operation and operands).

因此，如果要计算列中各项的总和，则您的操作数是切片起点和终点的坐标，操作是总和(您有两个输入：操作和操作数)。

In the picture above, you can see prefixes and suffixes. Some prefixes and suffixes are empty sets. This picture is just an illustration, but it shows how each word can be constructed. All these prefixes and suffixes (and other rules) are defined as the grammars for each language in a separate file.

在上图中，您可以看到前缀和后缀。 一些前缀和后缀是空集。 这张图片只是一个插图，但是它显示了每个单词的构造方式。 所有这些前缀和后缀(以及其他规则)在单独的文件中定义为每种语言的语法。

The rest of the Portable Spreadsheet library is simple storing of words for each cell defined by its coordinates. Each word (like that one with the sum mentioned above) can be the part of the other word (acting as an operand). This approach allows defining grammars for each language that should be part of the export (like native language description, Python language, Excel).

便携式电子表格库的其余部分是简单存储每个由其坐标定义的单元格的单词。 每个单词(就像上面提到的总和一样)可以是另一个单词的一部分(充当操作数)。 这种方法允许为应该作为导出一部分的每种语言定义语法(例如本地语言描述，Python语言，Excel)。

摘要 (Summary)

The purpose of this article is to present a new library for Python called Portable Spreadsheet (pip install portable-spreadsheet) that allows you to easily export computation formulas in a spreadsheet to various type of formats (mainly Excel and JSON). So you can see how each computation in a spreadsheet is done without manually coding each cell.

本文的目的是为Python提供一个新的库，称为可移植电子表格(pip install Portable-spreadsheet)，该库使您可以轻松地将电子表格中的计算公式导出为各种格式(主要是Excel和JSON)。 因此，您可以看到电子表格中的每个计算是如何完成的，而无需手动编码每个单元格。