Devart Python Connector for Microsoft Excel: Run SQL on Excel and Calc Files from Python
Devart Python Connector for Microsoft Excel is a paid Python module, $74.95 a year for one developer, that opens Excel workbook files (.xlsx, and .xls read-only) and OpenOffice or LibreOffice Calc files (.ods) directly from Python 3.10-3.15 on Windows, macOS and Linux, and lets you query and change them with SQL through the standard DB API 2.0 interface. Neither Excel nor Microsoft's Access Database Engine has to be installed.
It works on files, not on a running copy of Excel or a cloud service: you point it at a workbook path and its worksheets behave like tables. That makes it a fit for scripts that pick up spreadsheets dropped in a folder, on a Linux server or a Mac as easily as on Windows. For workbooks stored in OneDrive or SharePoint, Devart sells a separate Python Connector for Microsoft Excel Online.
Quick facts
- Version: 1.3.0, released October 22, 2026; first released July 29, 2024
- Developer: Devart
- Package:
pip install devart-excel-connector, or wheel files from Devart's download page (4.56 MB for Windows, 6.74 MB for macOS) - Python: 3.10 to 3.15 (1.3 dropped 3.7 to 3.9)
- File formats: Excel 2007-2021 .xlsx; Excel 97-2003 .xls (read-only); OpenOffice and LibreOffice Calc .ods
- Platforms: Windows and Windows Server 32-bit and 64-bit, macOS Intel and Apple silicon, Linux 64-bit
- Price: $74.95 a year or $149.95 perpetual for a Single (one developer) license
- Trial: 30 days, for evaluation only, not for production use
- Refund: the order page promises a 30-day money-back guarantee "if you are not satisfied", while Devart's ordering FAQ grants refunds within 30 days only when a technical problem support can't fix stops the product working
- What we checked: the order page charged $74.95 for a Single license with one year of updates on a yearly subscription, and PyPI and Devart's download page both showed version 1.3.0 with Windows, macOS and Linux wheels, on November 19, 2026. We didn't open a workbook with it
Download Free Trial Buy Now
Overview
Connecting takes one line: excel.connect(Database="C:/reports/sales.xlsx"). After that you create a cursor, run execute("SELECT * FROM employees") and read rows with fetchall(), passing values through ? placeholders as in any DB API module. Worksheets are the tables, and HeaderRow=True makes the first row's values the column names.
The connector reads the file format itself instead of going through Excel's COM automation or the Access Database Engine's OLE DB and ODBC drivers, which only exist on Windows and are a common source of 32-bit versus 64-bit trouble. Version 1.3 added a CREATE RANGE statement that names a block of cells as a database object, so a table that doesn't start in cell A1 can still be queried cleanly.

Devart's diagram of the direct connection: the connector inside a Python application reads the Excel workbook itself, with nothing in between.
Features
No Excel or Access Database Engine needed
Python Connector for Microsoft Excel reads and writes workbooks with its own code, so it runs on machines without Microsoft Office and on macOS and Linux, where the Access Database Engine doesn't exist. Deploying a script to many workstations means installing one Python wheel and nothing else.
SQL reads and writes on worksheets
Python Connector for Microsoft Excel supports SELECT, INSERT, UPDATE and DELETE on worksheets, so a script can filter rows, append records or correct values with SQL instead of walking cells. Several update statements can be sent to the workbook as one batch. Legacy .xls files are read-only.
Control over how sheets become tables
Python Connector for Microsoft Excel lets you use the first row as column names (HeaderRow), choose how many rows it samples to work out the columns (DescribeAmount, 25 by default), skip or keep empty rows (IncludeEmptyRows) and show or hide a ROWID pseudo column (HideRowId). Version 1.3 added CREATE RANGE for cell ranges and an ExactFieldLength option that sizes text fields from the data.
Single-user, shared read-only, or exclusive
Python Connector for Microsoft Excel opens a workbook for one user by default. ReadOnly mode lets several users or processes read the same file at once, and ExclusiveLock enforces single-user access and overrides ReadOnly, which matters when a file sits on a shared network drive.
Pooling and type mapping
Python Connector for Microsoft Excel pools connections (with min_size, max_size, lifetime and validate settings, or DisablePooling per connection) and maps Excel data types to Python types with options to change the mapping. Version 1.2 improved how column types are detected and fixed reading of date-formatted empty cells.
Benefits
- Runs where Office doesn't. Linux servers, Docker containers and Macs can process Excel files the same way Windows PCs do.
- SQL instead of cell loops. One query filters and sorts a sheet that would otherwise take a loop over openpyxl rows.
- Calc files too. The same code handles .ods spreadsheets from LibreOffice and OpenOffice users.
- Same API as your databases. Devart's Python connectors share connect, cursor and license calls, so moving data from a workbook into PostgreSQL or SQL Server looks the same on both sides.
Real-World Uses
- Loading supplier price lists. A nightly job on a Linux server reads each supplier's .xlsx price list with SELECT and inserts the rows into the product database.
- Cleaning up a shared tracker. A script runs UPDATE ... WHERE on a workbook to correct status values before it's sent on.
- Mixed Excel and LibreOffice teams. One import script accepts both .xlsx and .ods timesheets.
- Reading an offset table. CREATE RANGE names the block of cells under a report header so it can be queried like a table.
Who It's For, and Who Should Skip It
The connector suits Python developers who need SQL access to Excel or Calc files, especially on Linux or macOS where Microsoft's drivers aren't available, and teams that already use Devart's Python connectors for their databases.
Skip it if pandas' read_excel() or openpyxl already does what you need: both are free and widely documented, and most Excel jobs in Python are a single read into a DataFrame. Skip it as well if your workbooks live in OneDrive or SharePoint (that's the Excel Online connector's job), or if you must write to old .xls files, which this connector only reads.
Usage Guide
- Install with
pip install devart-excel-connector, or download the zip of wheels for your Python version and platform and run pip on the .whl file. - After buying, run
devart.excel.license.activate("<key>") once with the key from the order email or Devart's customer portal. - Connect:
conn = excel.connect(Database="sales.xlsx", HeaderRow=True); add ReadOnly=True when other processes read the same file. - Query:
cur = conn.cursor(), cur.execute("SELECT * FROM Sheet1 WHERE Region = ?", ["West"]), cur.fetchall(). - Write changes with INSERT or UPDATE, then commit before closing the connection.
Before you buy
By default a workbook can only be opened by one user at a time, so plan for ReadOnly mode when several scripts or people need the same file. Files in the old .xls format can be read but not changed; save them as .xlsx first if you need to write. The subscription license needs internet access to activate and check its status.
Where it falls short
- Free tools cover most jobs. pandas, openpyxl and xlrd read Excel files at no cost on every platform.
- Spreadsheets aren't tidy tables. Merged cells, several header rows and notes under the data need HeaderRow, CREATE RANGE or a cleaner layout before SQL gives clean results.
- Files, not live Excel. It works on saved workbooks, not on Excel's open window or on cloud workbooks.
- No asyncio API. It's a synchronous DB API 2.0 driver.
What Users Say
Reviews of this connector
Python Connector for Microsoft Excel hasn't collected reviews of its own on Capterra, G2, GetApp or TrustRadius. Feedback on Devart's Python connectors as a line is thin too: G2's Devart Python Connectors page shows a single 5-star review, too few for G2 to rate the product.
PeerSpot
A product engineer at a tech vendor with more than 10,000 employees reviewed Devart's Python Connectors on PeerSpot in November 2025, rating them 10 out of 10 after using them to feed an employee dashboard from an Oracle PL/SQL back end on AWS: "Since we started to use Python Connectors, we found it very reliable, and there are many measurable benefits of this." The same reviewer liked that they are "compliance-friendly, very secure, and has no plain text passwords or anything," but also wrote: "It is much costlier than other products."
What that adds up to
The little feedback there is praises reliability and security and names price as the drawback, and none of it is about Excel files. Run the 30-day trial on your real workbooks, including any with merged cells, extra header rows or .ods files, before you buy.
Comparisons
Devart vs. pandas and openpyxl
pandas' read_excel() and openpyxl are free and open source, run everywhere and give you DataFrames or cell-level control, including formatting. Devart's connector costs $74.95 a year and gives you SQL queries, INSERT and UPDATE through a DB API cursor, .ods support in the same API and connection pooling. Choose pandas for analysis, Devart for SQL-style read and write code.
Devart vs. pyodbc with the Access Database Engine
On Windows, pyodbc plus Microsoft's free Access Database Engine also lets you query Excel files with SQL, but it needs that engine installed at the right bitness and doesn't exist on macOS or Linux. Devart's connector has no such dependency and runs on all three platforms.
Excel vs. Excel Online connector
This connector reads workbook files on a disk or network share. Devart's Python Connector for Microsoft Excel Online is a separate, Windows-only product that works with Excel Online workbooks through Microsoft's web API.
Editions and Pricing
| License | 1-year subscription | Perpetual | Who can use it |
|---|
| Single | $74.95 | $149.95 | One developer at one company site |
| Team | $189.95 | On the order page | Up to 4 developers at one company site |
| Site | $339.95 | On the order page | Unlimited developers at one company site |
Buy Now opens the order for Python Connector for Microsoft Excel with a Single license on a 1-year subscription; Team, Site, the perpetual license and 2- or 3-year terms are chosen on the order page. An OEM license for shipping the connector inside your own product is arranged with Devart separately.
FAQs
Does Python Connector for Microsoft Excel need Microsoft Excel installed?
No. Python Connector for Microsoft Excel reads and writes workbook files directly, without Microsoft Excel or the Microsoft Access Database Engine Redistributable.
Can Python Connector for Microsoft Excel write to .xls files?
No. Python Connector for Microsoft Excel opens Excel 97-2003 .xls workbooks in read-only mode; it can write to .xlsx and .ods files.
Does Python Connector for Microsoft Excel run on Linux and Apple silicon Macs?
Yes. Python Connector for Microsoft Excel ships wheels for 64-bit Linux, macOS on Intel and Apple silicon, and 32-bit and 64-bit Windows.
Can two scripts read the same workbook at once with Python Connector for Microsoft Excel?
Yes, in ReadOnly mode. Python Connector for Microsoft Excel opens a workbook for a single user by default; setting ReadOnly to True lets several users read it at the same time.
Can Python Connector for Microsoft Excel open LibreOffice spreadsheets?
Yes. Python Connector for Microsoft Excel supports OpenDocument Spreadsheet (.ods) files from Apache OpenOffice Calc and LibreOffice Calc.
Which Python versions does Python Connector for Microsoft Excel support?
Python Connector for Microsoft Excel 1.3, released October 2026, supports Python 3.10 to 3.15; earlier versions also supported 3.7 to 3.9.
Can I ship the connector inside an application I sell?
Only with an OEM license. Python Connector for Microsoft Excel's Single, Team and Site licenses cover developing and deploying your own scripts and applications within one company.
Can I get a refund on Python Connector for Microsoft Excel?
Possibly. Python Connector for Microsoft Excel's order page offers a 30-day money-back guarantee if you're not satisfied, but Devart's ordering FAQ describes refunds within 30 days only for technical problems its support can't fix. Use the 30-day trial to be sure before you buy.
System Requirements
- Python 3.10, 3.11, 3.12, 3.13, 3.14 or 3.15 with pip
- Windows or Windows Server 32-bit or 64-bit, macOS (Intel or Apple silicon) or 64-bit Linux
- Workbooks in .xlsx, .xls (read-only) or .ods format
- No Microsoft Excel or Access Database Engine required
- Internet access for activating and checking a subscription license
Conclusion
Devart Python Connector for Microsoft Excel is worth $74.95 a year when you want SQL reads and writes on Excel or Calc files from Python, above all on Linux or macOS servers where Microsoft's own drivers can't go, or in a shop that already runs Devart's database connectors. If you only load spreadsheets into pandas, the free tools do that well and you don't need it. Try it on your messiest real workbook with the 30-day trial before deciding.
Download Free Trial Buy Now