Select from where in excel

Представьте себе ситуацию, Вы получили целевую выборку из одной базы данных, но для полноты картины, как всегда, нужны дополнительные данные. Проблема может быть в том, что нужная информация хранится в другой базе данных и возможности создать на ней свою таблицу нет, подключиться используя link тоже нельзя, да и количество элементов, по которым нужно получить данные, несколько больше, чем допустимое на данном источнике. Вот и получается, что возможность написать SQL запрос и получить нужные данные есть, но написать придется не один запрос, а потом потратить время на объединение полученных данных.

Выйти из подобной ситуации поможет Excel.

Уверен, что ни для кого не секрет, что MS Excel имеет встроенный модуль VBA и надстройки, позволяющие подключаться к внешним источникам данных, то есть по сути является мощным инструментом для аналитики, а значит идеально подходит для решения подобных задач.

Для того чтобы обойти проблему, нам потребуется таблица с целевой выборкой, в которой содержатся идентификаторы, по которым можно достаточно корректно получить недостающую информацию (это может быть уникальный идентификатор, назовем его ID, или набор из данных, находящихся в разных столбцах), ПК с установленным MS Excel, и доступом к БД с недостающей информацией и, конечно, желание получить ту самую информацию.

Создаем в MS Excel книгу, на листе которой размещаем таблицу с идентификаторами, по которым будем в дальнейшем формировать запрос (если у нас есть уникальный идентификатор, для обеспечения максимальной скорости обработки таблицу лучше представить в виде одного столбца), сохраняем книгу в формате *.xlsm, после чего приступаем к созданию макроса.

Через меню «Разработчик» открываем встроенный VBA редактор и начинаем творить.

Sub job_sql() — Пусть наш макрос называется job_sql.

Пропишем переменные для подключения к БД, записи данных и запроса:

Опишем параметры подключения:

Объявим процедуру свойства, для присвоения значения:

Вот теперь можно приступать непосредственно к делу.

Организуем цикл:

Как вы уже поняли конечное значение i=1000 здесь только для примера, а в реальности конечное значение соответствует количеству строк в Вашей таблице. В целях унификации можно использовать автоматический способ подсчета количества строк, например, вот такую конструкцию:

Тогда открытие цикла будет выглядеть так:

Как я уже говорил выше MS Excel является мощным инструментом для аналитики, и возможности Excel VBA не заканчиваются на простом переборе значений или комбинаций значений. При наличии известных Вам закономерностей можно ограничить объем выгружаемой из БД информации путем добавления в макрос простых условий, например:

Итак, мы определились с объемом и условиями выборки, организовали подключение к БД и готовы формировать запрос. Предположим, что нам нужно получить информацию о размере ежемесячного платежа [Ежемесячный платеж] из таблицы [payments].[refinans_credit], но только по тем случаям, когда размер ежемесячного платежа больше 0

Если значений для формирования запроса несколько, соответственно прописываем их в запросе:

В целях самоконтроля я обычно записываю сформированный макросом запрос, чтобы иметь возможность проверить его корректность и работоспособность, для этого добавим вот такую строчку:

в третьем столбце записываются запросы.

Выполняем SQL запрос:

А чтобы хоть как-то наблюдать за выполнением макроса выведем изменение i в статус-бар

Теперь нам нужно записать полученные результаты. Для этого будем использовать оператор Do While:

Указываем ячейки для вставки полученных данных (4 в примере это номер столбца с которого начинаем запись результатов)

— закрываем цикл If, если вводили дополнительные условия

— закрываем макрос.

В дополнение хочу отметить, что данный макрос позволяет обращаться как к БД на MS SQL так и к БД Oracle, разница будет только в параметрах подключения и собственно в синтаксисе SQL запроса.

В приведенном примере для авторизации при подключении к БД используется доменная аутентификация.

А как быть если для аутентификации необходимо ввести логин и пароль? Ничего невозможного нет. Изменим часть макроса, которая отвечает за подключение к БД следующим образом:

Но в этом случае при использовании макроса возникает риск компрометации Ваших учетных данных. Поэтому лучше программно удалять учетные данные после выполнения макроса. Разместим поля для ввода пароля и логина на листе и изменим макрос следующим образом:

Место для расположения текстовых полей не принципиально, можно расположить их на листе с таблицей в первых строках, но мне удобней размещать поля на отдельном листе. Чтобы введенные учетные данные не сохранялись вместе с результатом выполнения макроса в конце исполняемого кода дописываем:

Sheets(«Выгрузка»).TextBox1.Value = «« Sheets(»Выгрузка«).TextBox2.Value = »»

То есть просто присваиваем текстовым полям пустые значения, таким образом после выполнения макроса поля для ввода пароля и логина окажутся пустыми.

Вот такое вполне жизнеспособное решение, позволяющее сократить трудозатраты при получении и обработке данных, я использую. Надеюсь мой опыт применения SQL запросов в Excel будет полезен и вам в решении текущих задач.

I want to return rows from a table, where certain conditions are met, in excel.

ProdGroup   Width (mm)  Diameter (mm)   Date
Prod1           1120            1000   2016-01-11
Prod2           600             1000   2017-10-18
Prod1           930             800    2015-04-11
Prod3           1250            1200   2016-04-18
Prod2           840             1000   2019-06-27
Prod2           840              900   2018-03-21

I want to be make an equivalent of: «SELECT * FROM Table WHERE ProdGroup = «Prod2″ AND Diameter = 1000»
in excel. My idea is that I enter the values into two cells and that rows are returned based on what I write in the two cells.

I have tried using the =INDEX() function but I only managed to search for rows matching only 1 condition. Furthermore, I only succeeded return one row.

=INDEX(B2:D6,MATCH(A10,A2:A6,0),1)

This only gives me one IN parameter. With Prod2 this would only return one row.

With inputs «Prod2» and 1000:

ProdGroup         Diameter
Prod2             1000

I want an outcome like this:

Prod2           600             1000            2017-12-18
Prod2           840             1000            2019-06-27

I have no idea on how to go about doing this. Can somebody please help?

Kind regards
pbdude

26 Sep ’22 by Antonio Nakić-Alfirević

SQL Query function in Excel

If you’re reading this article you probably know that Google Sheets has a QUERY function that allows you to run SQL-like queries against data in the sheet. This function lets you do all sorts of gymnastics with the data in your sheet, be it filtering, aggregating, or pivoting data.

Being a fully-fledged desktop app, Excel tends to be more feature-rich than Google Sheets. This is especially true in the data analytics department where Excel shines with advanced Excel functions as well as Power Query functionality.

However, Excel doesn’t natively have a QUERY function that you can use in cells on the sheet.

In this blog post, I’m going to show you how to add a QUERY function to Excel and give a few examples of how to use it.

First look

Let’s start by taking a look at the function in action.

The function is pretty straightforward. It accepts the SQL query as the first parameter and returns a table with the results of the query.

The results automatically spill to the necessary amount of space. This spilling behavior relies on the dynamic array functionality that’s available in Excel 365 (but isn’t in earlier versions of Microsoft Excel).

Works with Excel tables

In Google Sheets, the QUERY function references data by address (e.g. “A1:B10”) while columns are referenced by letters (e.g. A, B, C…).

This works but has some drawbacks:

• It makes the query sensitive to the location of the data. If the data is moved or if columns are reordered, the query will break.
• It makes the query difficult to read since it uses range addresses and column letters instead of table and column names (e.g. Employees, DateOfBirth…)
• Adding or removing rows can break the query. For example, if the provided range is “A1:H10” the query will only take into account the first 10 rows. If additional rows are added, the query will not take them into account. You can get around this by omitting the end row number (e.g. “A1:H”), but this means that there must be no other content below the data range.

Excel, on the other hand, allows explicitly defining tables (aka ListObjects) that delineate the areas that hold data. Each Excel table has a name, as do its columns. This makes Excel tables very similar to database tables and makes them easier to work with from SQL.

Full SQL syntax support (SQLite)

Under the hood, the Windy.Query function is powered by SQLite – a small but powerful embedded database engine.

When called, the function passes the query to the built-in SQLite engine which has an adapter that lets it use Excel tables as its data source.

This means that the entire SQLite syntax is available for use in queries. In comparison, in Google Sheets, the query syntax is rather limited. It only supports a single table (no joins) and a very small set of built-in functions.

Examples of use

Since the engine under the hood is SQLite, queries can use all operations available in SQLite, including table joins, temp tables, column table expressions, window functions etc… Let’s go over some examples of how to use these in Excel.

Joining tables

Here’s an example of a simple one-to-many join:

The usual way of doing a simple operation such as this one in Excel would be to use xlookup or PowerQuery, but SQL is now another option. And if we needed anything more complex than a simple join, SQL would quickly shine as the most powerful and convenient option of the three.

Merging table rows (union)

Another way we might want to combine two (or more) tables is to combine their rows. We can do this with a SQL UNION operator.

The tables might have some rows in common. If we want to keep only one instance of such rows we would use the regular UNION operator. If we want to keep both versions of rows that are in common, we would use the UNION ALL operator.

Finding differences between two tables

In the previous example, we had two tables that had some rows in common and some rows not. Let’s assume, for example, that the first table contains last year’s list of employees and the second table is the new list of employees.

If we wanted to find out the differences between the two tables, we could easily do that with a bit of SQL.

All of the rows that are in the first table but not in the second one we will mark as “deleted”. All of the rows that are in the second table but not in the first one we will mark as “added”. Here’s what that SQL query looks like:

select
	id, name, 'deleted'
from
	employees e where not exists (select * from Employees_New en where e.id == en.Id)
union
select
	id, name, 'added'
from
	employees_new en where not exists (select * from Employees e where e.id == en.Id)

And here’s what the result looks like:

Ranking rows

Another useful thing we might want to do is rank rows based on some criteria. For example, suppose we have a table with a list of cities. For each city we have its population and the country it belongs to.

Our task is to find the top 3 cities in each country based on population. Here’s how we might do that in SQL.

-- we use this CTE so we can reference the calculated 'rank_pop' column in the where clause
with cte as (
	select
		city,
		country,
		population,
		-- using the RANK() window function
		RANK() OVER (PARTITION BY country ORDER BY population) as rank_pop
	from
		cities c)
select
	*
from
	cte
where
	-- filtering by the 'rank_pop' column from the CTE
	rank_pop <= 3
order by country, rank_pop

This query is a bit more complex than the previous ones. It uses a common table expression and a window function (the rank function), and showcases the ability to write complex SQL in queries.

Queries can also make use of dozens of built-in SQLite functions. Various specialized extended functions such as RegexReplace, GPSDist (GPS distance between two points) and LevDist (fuzzy text matching) are also available.

Updating tables

OK, this next example is a bit of a hack, but a useful one… The query you supply doesn’t need to be a SELECT query. You can do UPDATE/INSERT/DELETE statements as well, and these will modify the data in the target Excel tables.

This can be a handy way to clean and transform data in your tables in place, without having to export/import the data to an external database (e.g. SQL Server, MySql, Postgres…).

This works because the SQLite engine isn’t copying the data. Rather, it’s using an adapter that lets it access live data in the Excel table.

How does the function see Excel tables?

At first glance, it might seem strange that the query can access your workbook tables. After all, we did not pass them in as parameters, and functions normally only work with parameters that are passed to them.

However, the Windy.Query function is aware of the workbook it’s being called from and it can read data from the workbook’s tables without the need for passing them in as parameters. This makes the function much easier to call especially when working with multiple tables.

Column Headers

Results returned by the Windy.Query function can optionally include headers. This is controlled by the second parameter of the function.

The texts in the column headers are determined by the SQL query itself. You can easily rename result columns by aliasing them in the select list.

Automatically refresh results

By default, the SQL query runs as a one-off operation when you enter the formula but does not refresh if the source tables change. However, if you want the query to refresh whenever one of the source tables changes, you can easily do so by setting the autoRefresh argument to true.

Note that the auto-refresh functionality relies on Excel’s RTD (Real-Time Data) server. The RTD server usually throttles updates so functions don’t overwhelm Excel with frequent updates. The default throttle interval is 2s meaning that the function will not update more than once every 2s. To improve responsiveness, you can lower this value to something like 20ms. The simplest way to do this is through the “Configure” dialog in the QueryStorm runtime’s ribbon.

Passing parameters

When needed, SQL queries can use values from cells as parameters. To use a cell as a parameter in a query, start by giving the cell a name (named range).

Once the cell has a name, you can reference it in the query using the @paramName or $paramName syntax.

If automatic refresh is turned on, results will automatically refresh whenever one of the parameter cells changes its value.

Performance

This is all well and good for small tables, you might think, but how does it handle large data sets? Well, it handles them quite well. The function can read source tables of 100k rows and 10 columns within a few milliseconds and can return this amount of data in a second or two. In addition to this, all columns are automatically indexed so searches and joins are extremely performant as well.

This makes the function perform very well, both from the data throughput standpoint as well as from the computational one.

OK, so is this better than the Google Sheets version of the QUERY function?

Yes, dah. Did you read the previous chapters? 😛

Installing the Windy.Query function

So how do you install this function into your Excel? It’s a simple 2-step process.

Step 1 is to install the QueryStorm Runtime add-in (if you don’t already have it). This is a free, 4MB add-in for Excel that lets you install and use various extensions for Excel. It’s basically an app store for Excel.

Step 2 is to click the “Extensions” button in the “QueryStorm” tab in the Excel ribbon, find the Windy.Query package in the “Online” tab, and install it.

What happens if I share the workbook with a user who doesn’t have the function?

Nothing bad. If the other user doesn’t have the function installed, they will see the last results of the query that were returned on your machine. They just won’t be able to refresh the results.

Advanced SQL query editor

Writing SQL queries in the formula bar can get a bit unwieldy. To make queries easier to write, it’s better to use a proper editor, preferably one that offers syntax highlighting and code completion for SQL and knows about the tables in your workbook.

For this purpose, I recommend using the QueryStorm IDE. This is an advanced IDE that lets you use SQL in Excel. You can write the query in the QueryStorm code editor and then paste the query into the Windy.Query function when you’re happy with it (if needed).

The IDE does more than just allow using SQL in Excel. You can use it to create and share functions and addins for Excel. In fact the QueryStorm IDE was used to create the Windy.Query function itself.

The IDE has a free community version for individuals and small companies, while users in larger companies can make use of the free trial license. For paid licenses, check out the pricing page.

You can read more in this blog post that’s dedicated to the QueryStorm SQL IDE.

Video demonstration

For a video demonstration of the Windy.Query function, take a look the following video:

07 Aug 3 Ways to Perform an Excel SQL Query

Howdee! Excel is a great tool for performing data analysis. However, sometimes getting the data we need into Excel can be cumbersome and take a lot of time when going through other systems. You’re also at the mercy of how a disparate system exports data, and may need an additional step between exporting and getting the data into the format you need. If you have access to the database where the data is housed, you can circumvent these steps and create your own custom Excel SQL query.

To follow along with my below demos, you’ll need to have an instance of SQL server installed on your desktop. If you don’t, you can download the trial version, developer version, or free express version here. I’ll be working with the free developer version in this article. I’m also using a sample database that you can download here. The easiest way to install this is using SQL Server Management Studio (SSMS). That download is available here. Once you open SSMS, it should automatically detect your local server instance. You must ensure your SQL Server User is running as the “Local Client” and then you can create a blank database, and restore that database from the backup file. If you have issues accomplishing this, let me know in the comments and I’ll elaborate on how this is done.

If you are familiar enough with SQL and have access to your own data, you can skip these steps and use your data. Otherwise, I recommend downloading these tools before getting started. If you’re new to SQL, I highly recommend the SQL Essential Training courses on Lynda.com. Now, on to why you’re all here…

Excel SQL Query Using Get Data

This option is the most straight forward approach to creating an Excel SQL query. However, it is important to note that this approach is only available in Excel 2013 and later and will not currently work on Mac OSX. To get started, select “Get Data” à “From Database” à “From SQL Server Database” as shown in the screen grab. At this point it will pop-up a prompt to enter your server name and the target database you’re wanting to query (you can get this information from SSMS). You can enter this information and then select “OK”. This will allow you to browse available tables from that database to import. You can remove columns and filter tables before importing. If you do not know how to write SQL queries yet, this is one approach you can take.

However, if you select the “Advanced” dropdown arrow, you can create your own custom Excel SQL query. I usually create my query in SSMS or Visual Studio and then just paste the final query in this window. That is because there is no intellisense in this window and it can be difficult to spot errors in your query. Once you select OK, it will ask you to confirm credentials and you may get an error about encryption. This is common when connecting to databases in this manner and nothing to worry about. The next screen will provide an example of your data and you can select “Load” to import it.

This will create a table on a new tab and you’ll also notice a new pane on the right titled “Connections & Queries”. It will display the name of your query (defaults to “Query1”, “Query2”, etc.) and you can rename the query by right-clicking and selecting “Rename”. You can also edit the query from this location as well. It will open up an interface with a sample of your data and you can add/remove columns, filter your data, or edit your source query from here.

Now that you’ve set up this Excel SQL query, you can simply refresh the data set with fresh data anytime by clicking “Refresh All” on the “Data” ribbon. A quick side note here. If you pivot this data, “Refresh All” will refresh pivot tables first and then the query. To update your pivot table, you’ll need to refresh all twice or update your pivot table manually. To me, one of the downsides of this approach is the results are always returned in a table. I personally do not like working with tables in Excel. That’s where using VBA for your SQL query can come in handy.

Excel SQL Query Using VBA

Using VBA to create your Excel SQL query is not as straight forward as the previous approach, but can still be an extremely useful method depending on your situation. I particularly like that the data is not returned to a table unless you designate it to be so. This technique will work on older versions of Microsoft Excel but will not work on Mac OSX versions of Excel since it uses and ADO connection.

To get started, open up the VBA editor by pressing alt+F11. Before beginning to write your code, you’ll need to ensure that the “Microsoft ActiveX Data Objects 2.0 Library” is referenced from the VBA Project. To do this, click on “Tools” in the ribbon menu at the top of the VBA editor. In the popup, ensure the library is checked as shown below. This allows the project to use the ADO connectors to create the connection to your database. Next, let’s dimension a few variables.

Dim Conn As New ADODB.Connection
Dim recset As New ADODB.Recordset
Dim sqlQry As String, sConnect As String

The Conn variable is will be used to represent the connection between our VBA project and the SQL database. The receset variable will represent a new record set through which we will give the command to perform our Excel SQL query using the connection we’ve established. Finally, the sqlQry variable will represent a string variable that is our SQL query command, and the sConnect variable will be a string representing the connection string the database requires. Let’s look at how to use these variables to perform a SQL query.

sqlQry = "select top 1000 si.InvoiceID, si.InvoiceDate, sc.CustomerName from Sales.Invoices si" & _
             " left join sales.Customers sc on sc.CustomerID = si.CustomerID"

sConnect = "Driver={SQL Server};Server=[Your Server Name Here]; Database=[Your Database Here];Trusted_Connection=yes;"

Conn.Open sConnect

Set recset = New ADODB.Recordset

    recset.Open sqlQry, Conn
    Sheet2.Cells(2, 1).CopyFromRecordset recset
    recset.Close

Conn.Close

Set recset = Nothing

While this may look complex, each step is relatively simple. Firstly, we set our sqlQry variable equal to a string that represents the syntax of our SQL query. We then create a connection string we can use in our next command to connect to the database. So, “Conn.Open” is the command to open the connection and “sConnect” is the string it uses to do so. “Trusted_Connection=yes” means that the connection will attempt to be established using your Microsoft credentials for the account you’re logged in as.

Now that the connection is open, we can open a new record set and pass it the sql command using the sqlQry variable, and tell it which connection to use by passing it the Conn variable. We can then use the VBA command “CopyFromRecordset” to paste the recordset anywhere in our workbook. It’s important to close both the record set and connection at this point. You also want to set your recset variable equal to nothing so it does not eat up valuable resources.

One of the downsides to using this method is that you must explicitly tell Excel some things that the previous approach did automatically. For example, this SQL query will not return any column headers. Therefore, you must explicitly tell Excel what to label your columns. Secondly, the data is not automatically cleared and the new query imported. You must also explicitly tell Excel to do this as well. Here is the final code with those commands added.

Sub SQL_Example()
Dim Conn As New ADODB.Connection
Dim recset As New ADODB.Recordset
Dim sqlQry As String, sConnect As String

Sheet2.Cells.ClearContents

sqlQry = "select top 1000 si.InvoiceID, si.InvoiceDate, sc.CustomerName from Sales.Invoices si" & _
            " left join sales.Customers sc on sc.CustomerID = si.CustomerID"

sConnect = "Driver={SQL Server};Server=[Your Server Name Here]; Database=[Your Database Name Here];Trusted_Connection=yes;"

Conn.Open sConnect
Set recset = New ADODB.Recordset

    recset.Open sqlQry, Conn
    Sheet2.Cells(2, 1).CopyFromRecordset recset
    recset.Close

Conn.Close
Set recset = Nothing

Sheet2.Cells(1, 1) = "Invoice ID"
Sheet2.Cells(1, 2) = "Invoice Date"
Sheet2.Cells(1, 3) = "Customer Name"

End Sub

My preference for using this approach is when I want the user to be able to pass parameters to my Excel SQL query. For example, I might have a dropdown of customer names the user could select. By using this tactic, I can easily add a dropdown of customer names the user can select, and pass that value to my SQL query in a where clause.

As you can see, both the built in Excel SQL query and the VBA method have pros and cons. I employ both in my everyday work depending on what situation I find myself in.

Excel SQL Query Using Microsoft Query

This option is likely the most complex option, but it has the added advantage of being compatible with some versions of Mac OSX. I won’t pretend to be an expert at creating Mac OSX compatible tools for Excel, but I have successfully used this implementation to create an embedded Excel SQL query for Macs in the past.

I also like this method because you can create popup style parameters. For example, you can prompt the user to input date range parameters at the time the SQL query is ran. Like the first example, running this query is as easy as clicking “Refresh All” on the Data ribbon. Let’s dive in to the details.

To get started here, click “Get Data” on the Data ribbon. In the menu that dropdowns select “From Other Sources” and, finally “From Microsoft Query”.

This will open a wizard for you to choose your data source. Double click “<New Data Source>” and you’ll be prompted to enter some information about your data source. Option 1 can be anything you wish that describes your data source. Option 2 should be “SQL Server”. Click “Connect” and it will pop up a third window where you can enter information about the server and login information. Be sure you select the “Options>>” dropdown so you can select the database you’re wanting to connect to.

You’ll now have a new data source in the original window. Double click the data source to bring up a table import wizard. If you want to import an entire table, you can do so here and even filter and sort the data using the import wizard. However, if you want to use your own custom query as we have been, just select any field and go through the wizard and import the data. When you come to screen that asks you if you want to return the data to Excel or edit in a query, return the data to Excel. It will then prompt you to select where you want the data returned in your workbook.

The query will return the data in a table format. To change it to your own custom SQL query, let’s follow these steps:

• Click anywhere in the data table.
• On the Excel Data Ribbon, in the “Queries & Connections” group, properties will no longer be grayed out like it normally is. Click this.
• In the popup – you’ll see another properties icon. Click this.
• In this popup, select the “Definition” tab and paste your SQL query in the “Command Text” input box.

Now you’ve built an Excel SQL Query that can be refreshed anytime the workbook is refreshed. In my screengrab, the “Parameters” button is greyed out. If you want to add parameters to your query, you do so by adding “?” in your command text. That looks like this.

This creates a parameter the end user can interact with. You can have the user be prompted to enter an input when the workbook is refreshed, select a default value, or have it linked to a cell in the workbook. Even though this option is cumbersome to set up, I really enjoy using it. It allows me a lot of flexibility to have the user interact with the data. As I touched on in the beginning, I’ve had success using this option on Microsoft Office for Mac OSX. I don’t want to say this will work 100% of the time on a Mac because I’ve also had it fail. If anyone has any input on this, I’d love to hear from you.

Let me know your thoughts on these approaches in the comments! What other ways do you creatively get data into Excel from SQL data sources?

Cheers!

R

SelectCompare allows for data comparison of Excel spreadsheets, as well as other data sources with a simple, unified interface.

You may be initially anxious about writing an SQL statement to pull data from your Excel, but it is pretty easy!

This post will show you a few examples of writing queries against your spreadsheet that you will be able to use right away for your Excel data comparison.

If you want to find out how to connect to an Excel workbook, please read another post on this blog — Compare Excel Spreadsheets. You may also want to have a look at the blog post DSN-less ODBC Data Sources which shows a few cool connection strings to other data technologies, allowing you to compare data between Excel and CSV for example.

You will see below how to query Excel spreadsheets, specify ranges of columns and rows, and write simple filters that make SQL so powerful.

In a general sense, the data that is stored in the Excel spreadsheet can be represented in two ways — it either has a header with column names, or (more rarely) it doesn’t.

Excel spreadsheet with headers

Let’s assume we have a spreadsheet with headers in the first row as in the following picture.

The simplest query you can issue is 

Select * from [Customers$]

The query just selects all columns and all rows from the spreadsheet. Excel automatically returns only populated rows and columns.

If you need to limit the number of rows you select from the spreadsheet, use the TOP clause in the select statement:

Select top 3 * from [Customers$]

If you are interested only in specific columns, you can select only those that you need in the following way:

select CustomerID, CompanyName, ContactName, ContactTitle, Country
from [Customers$]
order by ContactName

What if my data doesn’t begin in the first row?

It often happens that your spreadsheet has a title and other information at the top. The actual data begin below the title:

In such case you can specify the range of data you want to select from:

select * from [Customers$A5:K92] order by CompanyName desc

Or just simply

select * from [Customers$A5:K] order by CompanyName desc

Filtering data in the spreadsheet

You can select only data that you need for comparison by adding a WHERE clause to the query:

select CustomerID, CompanyName, ContactName from [Customers$]
where ContactTitle = 'Owner'

order by ContactName

Excel spreadsheet without headers

Excel, by default, assumes that the column names of your table are stored in the first row of the table you select from. In some cases you may come across spreadsheets that do not have header information, like in the following illustration:

In this case, if you select data from the spreadsheet, the first row will be treated as column names, which is not what you want.

You have to modify the connection string for your Excel Workbook. Just add the parameter «HDR=NO» to the connection string as displayed below:

Now, when you select from the spreadsheet, you will notice that Excel provided automatic names for the columns:

Usually, you want to have meaningful names for your columns. You can easily apply aliases to your column names, like in the following illustration.

select F1 as CustomerID, F2 as CompanyName, F3 as CustomerName, F4 as CustomerTitle
from [CustomersNoHDR$A1:K]
where F4 ='Owner'

These are only a few simple examples of queries you can use to select data for your comparison of spreadsheets. You can write more complicated queries, like joining multiple spreadsheets or calling Excel functions in the query to perform certain operations on the attributes, but this is a topic for another occasion.

Other reading

Check also other articles describing various aspects of the usage of SelectCompare.

I described two interesting use cases that use Excel data templates to facilitate data comparison of data produced by two distinct business processes.

Other articles help to understand how to use SelectCompare.Some of them are listed below:

• Create an Excel baseline for your data
• How to compare data with a baseline

Помогла ли Вам статья?