folder<\/strong>, not a file. Point it at the directory that holds the<\/li>\n<\/ul>\n.dbf<\/code>\/.fpt<\/code>\/.cdx<\/code> files; every table in it is offered on the next step.<\/p>\n\nDBF Format<\/code> = Auto<\/code><\/strong> \u2014 let it read the version byte instead of telling it the dialect. This<\/li>\n<\/ul>\nmatters: dBase III and VFP write different header sizes, and a tool that assumes one of them fails on the other with Unknown type n<\/code> or Type \"0\" of the field \"\"<\/code>.<\/p>\n\nCode Page<\/code> = Default<\/code><\/strong> \u2014 this is the field that decides whether your text arrives readable.<\/li>\n<\/ul>\nA Chinese, Cyrillic or Greek .dbf<\/code> written under a DOS code page needs the matching code page here; leave it on Default<\/code> and you get the right shape with the wrong characters.<\/p>\nPress Test<\/code><\/strong>; for a folder of .dbf<\/code> files it answers with a single Direct access<\/code> confirmation, and the status line under the buttons goes from Not Connected.<\/code> to Connected.<\/code><\/strong><\/p>\n![\"DBCopier](\"https:\/\/www.withdata.com\/wp\/wp-content\/uploads\/2026\/11\/dbf-01-logon-source.png\")
<\/p>\n
\n- Switch to
DB (Target)<\/code><\/strong> and set Target Database Type<\/code> to SQL Server<\/code><\/strong>: Server<\/code><\/li>\n<\/ol>\n(.\\SQLEXPRESS<\/code> for a named instance), Port<\/code>, Database<\/code>. With Authentication<\/code> left on Windows Authentication<\/code><\/strong> the Username<\/code>\/Password<\/code> fields stay empty. Test<\/code><\/strong> answers with the server version, Connect<\/code><\/strong> turns the status line into Connected.<\/code>, and only then does Ok<\/code><\/strong> close the dialog \u2014 press it earlier and it says Please connect to target database<\/code>.<\/p>\n![\"DBCopier](\"https:\/\/www.withdata.com\/wp\/wp-content\/uploads\/2026\/11\/dbf-02-logon-target.png\")
<\/p>\n
\n- The main window now names both sides \u2014 `DBCopier — Source DBF: Target SQL Server:<\/li>\n<\/ol>\n
:: \u2014 and lists the tables it found in the folder. Start the copy from <\/code>Wizard > Table To Table Copy<\/strong>; the <\/code>Start a task – DBCopier dialog offers the same thing as <\/code>Convert From Table`<\/strong>.<\/p>\n\nOpen Source<\/code><\/strong> \u2014 the Table<\/code> dropdown lists every .dbf<\/code> in the folder. Picking one fills the grid<\/li>\n<\/ol>\nwith the first 20 rows<\/strong>, which is where the D<\/code>-type dates, the numeric types from \u00a74 and the memo columns can be inspected before<\/em> anything moves.<\/p>\n![\"Open](\"https:\/\/www.withdata.com\/wp\/wp-content\/uploads\/2026\/11\/dbf-03-open-source.png\")
<\/p>\n
\nSet Target<\/code><\/strong> \u2014 Schema<\/code> and Table<\/code> on the SQL Server side, Load Type<\/code>, Save To<\/code> (Database Table<\/code><\/li>\n<\/ol>\nor SQL File<\/code>), Encoding<\/code>, and the Columns Map<\/code><\/strong>: one row per field with Include<\/code>, Source Column<\/code>, Table Field<\/code>, Date Format<\/code> (for the date fields) and NULLIF<\/code>. Fields with the same name are matched for you; Select All<\/code> ticks the ones you keep. If the target table does not exist yet, the button next to Table<\/code> opens Create Table<\/code><\/strong> (Field Type<\/code> = Auto Detect<\/code>), generates the DDL and runs it with Execute<\/code><\/strong>.<\/p>\n![\"Set](\"https:\/\/www.withdata.com\/wp\/wp-content\/uploads\/2026\/11\/dbf-04-set-target.png\")
<\/p>\n
\nSummary<\/code><\/strong> \u2014 From :<\/code> and To :<\/code> on one line each, the first ten rows as they will arrive,<\/li>\n<\/ol>\nConvert Rows<\/code>, Skip Rows<\/code>, Errors to allow<\/code><\/strong> and the Log File<\/code> path.<\/p>\n\nConvert Data<\/code><\/strong> \u2014 it asks Start Converting?<\/code><\/strong>, then reports<\/li>\n<\/ol>\n27 records, Elapsed: 00:00:00.109. From Table: DEPARTMENTS, To Table: [dbo].[DEPARTMENTS].<\/code> and closes the log with Total Convert Records: 27<\/code> \/ End Convert<\/code>.<\/p>\n![\"Convert](\"https:\/\/www.withdata.com\/wp\/wp-content\/uploads\/2026\/11\/dbf-05-convert-data.png\")
<\/p>\n
The reason this route wins for FoxPro data is not speed, it is that the memo file is read for you: the .fpt<\/code> blocks are followed by offset and length, so the 4 KB \/ 64 KB \/ 1 MB values in \u00a76 arrive whole, and the target column is created wide enough (varchar(max)<\/code> for a W<\/code>\/M<\/code> field) instead of being truncated at 8,000 bytes.<\/p>\nRoute B \u2014 scripting the same load<\/h3>\n
When the copy has to run inside your own pipeline, export from FoxPro to a delimited file and load it with bcp<\/code> or BULK INSERT<\/code>: set MAXERRORS<\/code> or the whole batch dies on the first bad row, pin the date format and the codepage on both ends, and load the memo fields in a separate pass<\/strong> joined on the primary key \u2014 a delimited export cannot carry them.<\/p>\nFor a cutover without downtime, do not read the live table while users are writing to it: copy the files and sync from the copy. Two patterns hold up \u2014 a high-watermark on an auto-increment id or a last-modified column (only if the FoxPro application actually maintains one), or a nightly full copy followed by a row count and a per-table hash, alerting on drift.<\/p>\n
Both are scripts you will own and re-test at every schema change; the GUI route above is the same load with the memo offsets and the type mapping already written.<\/p>\n
4. Type mapping: the four fields that need a decision<\/h2>\n
\n\n\n| VFP type<\/th>\n | Stored as<\/th>\n | SQL Server<\/th>\n | Watch for<\/th>\n<\/tr>\n<\/thead>\n |
\n\nCurrency<\/code> (Y<\/code>)<\/td>\n| 8-byte scaled integer, 4 decimal places<\/td>\n | money<\/code> or decimal(19,4)<\/code><\/td>\nmoney<\/code> rounds to 4 decimals; decimal<\/code> does not. Pick one and keep it<\/td>\n<\/tr>\n\nDateTime<\/code> (T<\/code>)<\/td>\n| 8 bytes: Julian day + milliseconds<\/td>\n | datetime2(3)<\/code><\/td>\ndatetime<\/code> truncates to 3.33 ms; datetime2<\/code> does not<\/td>\n<\/tr>\n\nMemo<\/code> (M<\/code>)<\/td>\nBlock in the .fpt<\/code><\/td>\nnvarchar(max)<\/code><\/td>\nThe content<\/em> is in another file; the column alone is not enough<\/td>\n<\/tr>\n\nGeneral<\/code> (G<\/code>)<\/td>\nOLE object block in the .fpt<\/code><\/td>\nvarbinary(max)<\/code><\/td>\n| It carries 4\u20138 bytes of OLE header in front of the actual document<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n The rest map cleanly: Integer<\/code> (I<\/code>) \u2192 int<\/code>, Double<\/code> (B<\/code>) \u2192 float<\/code>, Logical<\/code> (L<\/code>) \u2192 | | | | | | | | | | | | | |