Category Archives: SQLite

The Datasmith’s Hammer

Posted on June 10, 2011 | 2 Comments

Although my microETL add-in is very powerful, it can be a bit intimidating for those without a programming background. It was after all, designed for my needs primarily and being a professional programmer I tend to see the world from that perspective. Hence microETL’s ability to forge vast and complex datasets in parallel to those of Excel; to share those datasets not just with Excel and VBA but also with the powerful tool that is CPython. But microETLs genesis was not as an all-powerful ETL tool but as means to quickly and accurately handle tabular data in Excel. The original xLite (which begat microETL) started out with just two functions, Join two tables or Left Outer Join two tables. That was it, but it was still useful.

I’ve been intending for some time to build an offshoot of microETL that would be less powerful but perhaps more approachable and have fewer moving parts. This week I finished it, still needs some more testing, but the basic product is in place. It’s a single file add-in (a .xll) , called DATASMITH. At its heart is a single function called HAMMER, there will be other helper functions that in the main will wrap the HAMMER function, but in essence it is the datasmith’s HAMMER.

If microETL is a datasmith’s forge or indeed a mirco-foundry,HAMMER is a datasmith’s everyday portable tool (with perhaps Excel as the anvil, and your CPUs as the fire?). Talking of CPUs; multi-core CPUs are now the norm and since version 2007, Excel can utilise such multi-cores. MicroETL being VBA-based cannot however take advantage of this, to do so requires an .xll add-in; another reason to build the HAMMER.

So what will this new functionality look like:

Examples:

=HAMMER(Invoices[#All],InvoiceLine[#All],”JOIN”)

…will take the two ranges (you need the [#All] to pick-up the header and data sections of a 2007/2010 Excel Table) and join them using the columns with the same name as the join fields. The JOIN command expects the last two preceding arguments to be tables (aka arrays with a header line).

=HAMMER(“SALES”,DeptSales!A1:F2101,”SELECT * from table2 where dept=’:1′”,”SQL”)

… the 1st argument is loaded as Arg(1) (:1 in SQL), the 2nd argument is loaded (being an array) into a table named table2; if they were the other way round, it would be table1 and Arg(2). The 3rd argument is loaded as Arg(3) and the 4th is a command: SQL. SQL looks back at the preceding argument (Arg(3) in this case) and executes its contents as SQLite SQL. If the preceding argument was a table, it would expect to find a list of SQL statements for execution in the first column. The output of the last issued SELECT statement is then returned to Excel as an array.

=HAMMER(DeptTargets!A1:D20,DeptSales!A1:F2101,SalesTagetScript,”PYTHON”)

…this is similar to the previous SQL example but this time the command is PYTHON which will execute the Python Script passed in via the command’s preceding argument. The script will most likely return a table to Excel, but it could also, if not the last argument, create a table associated with its position, in this case table4, which could then be accessed by subsequent PYTHON or SQL scripts.

Up to 25 arguments can be passed, the last table produced is returned to Excel (either via a load, which wouldn’t be terribly useful, or more likely as a result of a command such as JOIN, SQL or PYTHON). HAMMER functions can of course be nested and can also issue “internal” HAMMER requests. The “flow” of commands is from left to right, with the preceding args usually setting the stage for subsequent commands. Alongside the all-powerful SQL and PYTHON commands, I’ll most likely add a set of “noSQL” offerings such as JOIN, LOJOIN (left-outer), DISTINCT, REDUCE (a SELECT .. GROUP BY… with PYTHON as the MAP?), UNION, INTERSECT. These will likely also be available through helper functions such as =JOIN(ThisTable,ThatTable).

Unlike microETL, there’s no persistence across function calls i.e. HAMMER will play by Excel’s Functions no-side-effects rule. Each call to a function will build up and tear down its in-memory SQLite environment (including calls to HAMMER from within PYTHON).(UPDATE: Specify the 1st COMMAND as “APPDB” to simulate microETL’s persistence across function calls)  Likewise each call to PYTHON will be a separate engine instance. Likewise each call to a HAMMER function will create its own PYTHON engine. Tables  and Python artefacts created by prior “steps” in a single HAMMER call are available to subsequent steps.

HAMMER embeds Python under the guise of IronPython, so a lot of the power and speed of CPython will not be available, but on the other hand, the full power of the .NET CLR will be, not a bad swap.

And a huge advantage, HAMMER will be an asynchronous function (i.e. will run in its own thread). This will allow multiple long running transforms to be handled within the same Excel instance, a major shortcoming of microETL. This requires Excel 2007 or 2010, but will still work synchronously for Excel 97-2003. Having said that, there’s a requirement for Net4.o for IronPython so this add-in is more suited for modern versions of Excel and for OS >= XP SP3.

HAMMER is an array function, yeah I know, normal folk tend to steer clear of Excel arrays. Which is to be expected, they’re not the most intuitive end-user-facing construct that the industry has ever come up with. But, they are the “Excel way” for passing tables in and out of formulas and, once mastered, open up a new world of power to Excel users. I will be adding helper functions to make using arrays a bit easier (an autoSize wrapper function, that’ll resize the selection area if it’s too small or too big, at the cost of a 2nd pass at the enclosed functions, might also port microETL’s SQL “paste” functionality).

I had intended to have an example to download with this post, but have discovered a last-minute bug that needs fixing. So it’ll be most likely next week before I’ve a version to show.

UPDATE:

I’ve managed to sort out the bug, so here’s an example…

http://bit.ly/datasmith

Here’s a list of the HAMMER commands implemented so far …

Use setup.xls to install (or simply open in Excel), see IrelandFOIexample_hammer (2007/2010) for examples (there’s also a copy of the same functionality via microETL in IrelandFOIexample_microETL.xls – note the speed difference).  There’s Excel 2007/2010 32bit and  Excel 2010 64bit versions included and there’s also an un-tested 2003 version.

→ 2 Comments

Posted in ETL, excel, HAMMER, microETL, Python, SQLite, VBA

Tagged , , ,

Attach a SQLite database into Excel’s memory via microETL

Posted on May 20, 2011 | 1 Comment

In my previous post I described the various methods of accessing SQLite databases from within Excel using microETL. Via comments on the post, Michael Römer suggested a change to how microETL loads into memory an external SQLite database (not only suggested, but also provided the C code changes to enable the change; thanks Michael).

The existing xLiteLoadUnLoad(filename[,unload]) function loads a SQLite file into “main” i.e. the primary database (which is usually a :memory: db) overwriting any existing data. Michael’s suggestion was to allow loading into another in-memory database with a different alias; thus keeping the main database intact but allowing the benefits of in-memory access to the externally attached database. This feature has now been added.

I’ve kept the existing xLiteLoadUnLoad as is, but added a new optional argument to the xLiteAttachDB function so.

  •  xliteAttachDB(databaseName,alias)  becomes xliteAttachDB(databaseName,alias,[loadInMemory=False]). The optional loadInMemory argument defaults to FALSE, so acts like the old version (i.e. issues a standard SQLite Attach statement). But if set to TRUE; the function will first Attach a “:memory: database” named as the alias, then will load the external database file into that in-memory database. Once this happens the on-disk database is not referenced, so any changes will not be reflected back to disk. To enable changes to be persisted to disk, I’ve added another new function…
  • xLiteDBSaveAs(alias,outDatabaseFile) will save a copy of the database named alias (with could be “main” if you wished to backup the default in-memory database) to the file outDatabaseFile. I’ve also added a …
  • xLiteDetachDB(alias) to issue a SQLite DETACH statement. You might ask why not simply use the SQL() function to issue DETACH (or indeed ATTACH) statements? Statements such as ATTACH/DETACH cannot be issued by the SQL() functions as its pre-processor (for table() functionality) wraps SQL statements in a SAVEPOINT (nested SQL transaction). You can however use the fastSQL() or xliteRawSQL() functions to issue such commands.

There’s another (this time breaking) change to the SQLScript TIMER  (see here …) command. The existing function used an ActiveX control (the Internet Explorer control as a provider for JavaScript timer functionality); ActiveX controls do not work under 64bit Excel, so I’ve reverted back to using Application.OnTime as my timer mechanism. The breaking change is the 3rd argument, which previously expected a value indicating the number of thousands-of-a-second to wait, now it represents whole seconds.

To download the latest version see the http://www.gobansaor.com/microetl page.

Update:

For another method of loading SQLite databases within Excel/VBA see my new .NET-centric micro ETL tool  http://blog.gobansaor.com/category/hammer/

→ 1 Comment

Posted in excel, microETL, SQLite, VBA

Tagged

Accessing SQLite databases from Excel via microETL

Posted on May 11, 2011 | 8 Comments

MicroETL makes two SQLite instances available to Excel. The main database is the App database, by default an in-memory instance; the second is a Helper instance, again by default in-memory, which is used to hold logs and to enable SQLite to perform certain activities that would not be possible without two separate database instances. The main commands such as SQL() deal exclusively with the App instance, and for the most part that’s the only database that a microETL user needs to be aware of.

Within the microETL project, VBA routines can gain access to the App instance via the gAppDB global variable (gHelperDB for the Helper instance) while Python routines use the global dictionary element connDict(“App”) or connDict(“xlite”) for the App database (and connDict(“Helper”) for the Helper).

SQLScript (and in-cell formulas) can gain access to the Helper instance via the xLiteSQL(inSQL,outRange,NoHeader,PasteOver,useHelperDB) command, by setting useHelperDB to True. (The SQL() function is a thin-wrapper on this function to handle positional substitution tokens (:1, :2 …), the main functionality,such as the table() preprocessor commands and output to a range, are provided by xLiteSQL.)

These default in-memory databases are blank databases that are  instantiated in memory at workbook start-up and any tables or data created during the workbook session are discarded at shutdown. The standard use-case assumes that any data to be saved or fetched will come from Excel or perhaps via external CSVs or corporate database connections as would be the case with a normal Excel workbook, i.e. the main use for SQLite is assumed to be a provider of a SQL layer on top of Excel not a primary data-store.

However, SQLite is a superb data-store, one which is used by 1000s of applications (quite likely many of apps on your smart-phone or iPad use SQLite as their database). Its ACID nature and small foot-print makes it an ideal means of storing client-side information that would otherwise need to be communicated to a “proper” server-side database such as MySQL or SQL Server.

MicroETL supports the use of disk-based databases for both the main App and companion Helper instances via:

  • xliteAttachDB(databaseName,alias); this command will attach a database to the App database and assign it an alias. This is a wrapper on the SQLite Attach command  http://www.sqlite.org/lang_attach.html. Note: the SQL Attach command can not be used via the SQL() command, but can be used by the “raw” fastSQL() function.
  • xLiteLoadUnLoad(filename[,unload]); this function is both useful & dangerous, it will replace the main (App) database’s contents with the contents of the SQLite file provided by the first parameter. Typically the target database is the default empty in-memory (:memory:) database so no harm done; but if the default database was a file-based database, whose contents you wished to preserve, this might not be what you wanted. The second parameter if set to TRUE (FALSE being the default) works in the opposite direction i.e. replaces the specified database’s contents with those of the main (App) database.
  • Setting a custom property of xLiteDB to a SQLite filename in a driver workbook, this database will then be used as the default App database. As well as file based databases, the xLiteDB property can be set to ”:temp:” or “:memory:”, :temp: is like :memory: but will use RAM & a temporary file while :memory: only uses RAM. The driver workbook is the first workbook to initiate microETL, so be careful of  multiple workbooks with different App “expectations” being accessed in the same Excel instance.
  • The similar custom property of xLiteHelperDB controls what Helper (i.e. Logging) database to use.
  • Another option is to use the custom properties xLiteAttachDB & xLiteAttachDBAlias to auto-attach a database at startup
  • Yet another method is provided by the custom properties xLiteDBSeed and xLiteHelperDBSeed. Does the the equivalent of xLiteLoadUnLoad(databaseName,FALSE) at startup. There’s an xLiteClose() function which release’s resources held by the App & Helper databases and if they were seeded via xLiteDBSeed or XLiteHelperDBSeed, will save back any changes to the seed databases.

UPDATE:

As per Michael’s suggestions below, I’ve add the ability to attach an external database into an “aliased” :memory: database, see http://blog.gobansaor.com/2011/05/20/attach-a-sqlite-database-into-excels-memory-via-microetl/

→ 8 Comments

Posted in excel, microETL, SQLite

microETL – the ultimate datasmithing tool?

Posted on April 29, 2011 | 1 Comment

Over the years I’ve accumulated and written thousands of lines of VBA code, I’ve also integrated Excel with countless other technologies. Most of this work was associated with the movement and managing of datasets; data for reporting on, for analysing and for collection. MicroETL is, in effect, a collection of the best (or at least the most useful) of these activities, packaged primarily so that I don’t have to continuously re-invent the wheel, so that when I’m working on a client’s problem, I can concentrate on the problem and not the technology. So although I share the code with all and sundry with the hope that others may find it useful (little choice as it’s VBA ;) ), its primary purpose is to the serve my needs, and indirectly, those of my clients.

VBA is of course at the heart of microETL. I’m often asked why not port the code to one of Microsoft’s more “modern” languages? There are three  main reasons for not doing so now (and one major one for not doing so in the past).

First the past. When MS introduced .NET the needs of those who program mainly around the Office platform were ignored, .NET add-ins for Excel were a pain to write, even more painful to deploy and even then, were slow compared to VBA add-ins. Things have improved, it’s now much easier to target Office from within Visual Studio (although still just as expensive). But even now, the best method of using .NET with Excel is a  superb 3rd-party open source tool, ExcelDNA. This is the tool I’ve used (and will continue to use) whenever I needed to merge Excel with the .NET platform.

I’m currently developing a .NET offshoot of microETL called HAMMER; using ExcelDNA, C#-SQLite and IronPython; not so much a replacement for microETL but a new type of tool to take advantage of the new capabilities offered by Excel 2007/2010 (multi-threading and 64bit memory addressing mainly) and .NET4 (IronPython and ease of multi-threading programming). It’s also, thanks to ExceDNAPack, a single file deployment. To keep an eye on its development follow the HAMMER tag on my blog.

So if ExcelDNA provides a pain-free (and cost-free) method of using C# or VB.NET with Excel, why stick with VBA? As I said, three reasons:

  • VBA is to Excel what JavaScript is to the web browser. It’s only remaining purpose as a language (with the death of VB6) is as an Office automation language and ,in effect, when you’re programming in Excel you’re largely working with the Excel Object model (in much the same way and the DOM is the focus of attention on JavaScript progamming). VBA is perfectly suited to this task and although it has its problems as a language (again like JavaScript) once you know them they’re not a problem. VBA is here for the long haul, Office 2010′s addition of the 64bit support to the language (Vb7 !) confirmed that. As the code is already in VBA, works very well as VBA and has a long future ahead of it, why change?
  • VBA is easy for a large subset of end-users to learn due to Excel’s macro-recording facility. The resulting code may not be pretty but it works both as a automation tool and as a learning tool. Millions of citizen programmers owe their skills to the VBA macro recorder. One of microETL’s goal is to make adding functionality as easy as possible and one such method of doing that is via simple VBA functions that can be called from SQLScript (microETL’s tabular scripting extension). C# or VB.NET are professional languages designed for professionals to use, VBA comes from a long line of “pro-am” scripting languages which open the world of programming to a much larger population.
  • Multiple deployment options. MicroETL can be installed like a normal add-in and consists of three files, the VBA add-in, the SQLite3 dll and the xLiteSQLite dll to wrap SQLite to make it VBA accessible. Python functionality and other options will add further files. But by embedding a small call-on-open macro in a workbook the add-in can be called without installation. This is how I use it for once-off tasks, as I can package the code and the associated data and workbooks in a single zip file for archiving purposes or for deployment (often running off a USB stick).  Because I can deploy my code as code-at-a-moment-in-time I can modify the codebase to suit the current requirements without affecting other microETL projects that are using, or might in the future use, a different version fo the code. In fact, microETL has a noSQL compile option that removes SQLIte and other DLL dependencies (with a reduction in power obviously) enabling the code to be embedded in a workbook allowing for a single-file deployment (which often, is the only option). This could not be done with a .NET add-in.

Okay, so VBA is useful, but why embed SQLite and Python? As I’ve said above, microETL exists to make my life easier. I’m a database programmer, I’ve been using SQL (or its predecessors ) for  three decades. I like set-based logic (such as SQL  or PowerPivot’s DAX) because:

  • I’m good at it (practice does make perfect) enabling me to solve many data related problems efficiently and more accurately (fast is no good if the result is wrong) that would be the case with functional (Excel fromulas) or procedural (code) approaches.
  • I believe anybody who spends a significant proportion of their working lives managing datasets should learn basic SQL. If they don’t it’s like driving a car but never getting beyond 2nd gear.

SQLite allows me to bring the beautiful world of set logic to user-centric Excel world of functional programming (yes, all you Excel jocks, you’re functional programmers). It also brings some nice side-effects:

  • It is, like Excel itself, an outstanding single-file document-oriented datastore.
  • By default, microETL uses SQLite in-memory database functionality (you can also attach disk-based ones) so it offers a means of exploiting the vast amount of cheap RAM that modern PCs now offer. Excel has always stored its working datasets in-memory and it too now offers much greater capacity (1,000,000 rows per sheet in Excel 2007/2010). SQLite has no practical limit other than the available memory and the addressable-limit of the OS (2GB for 32bit Windows). With 64bit Windows (microETL is Excel 64bit enabled) that addressable limit is now effectively gone. This makes microETL a superb platform for all sorts of ETL tasks.
  • As SQLite offers Excel a secondary in-memory datastore it can be used to share models across programming platform boundaries. This allows me to embed Python in Excel and use the in-memory SQLite database as a shared data conduit. I’ve done the same with JavaScript (but only on 32bit) and with .NET. The result is the non-native languages can be used as if they were not operating with Excel, no need to mangle in Excel interface code, simple!

Which brings me to Python. Why embed a Python interpreter into microETL? Can I not just use VBA as I quite obviously know it inside out? Well of course for all things Excel-focused that is what I do, and for most other transformation or calculations that require some programming I would also opt for VBA. No reason not to, it works and works quickly. However, Python (and in particular C-Python) offers access to world of superb code. If something can be automated, or interfaced to, or turned into a computing algorithm, somebody (usually very bright) will have done it in Python. Python also works well with C, and again if a C library is useful somebody will have wrapped it in Python. Python means that if I’m asked “Can you integrate Excel with ……?”, I can say, “Yeah, no problem”.

Python as a language has the other advantage of being a superb tool to work through a problem with; it doesn’t get in the way; very little of my mind has to be diverted to the language itself, the problem at hand gets the attention (C would be an example of a language on the other extreme, you better have the solution to your domain problem figured out before you start to program in C as the programming task itself will command most of your attention).

Python is an ideal language for business logic. It is easy to learn, eminently readable and generally immune to programming gotchas. Because of this, it’s very popular as a citizen programmers’ language, especially amongst engineers, scientists and quants, so it blends in naturally with microETL the ultimate datasmithing tool!

So if microETL is built on a foundation of “civilian friendly” technologies such as Excel formulas, VBA, SQL and Python is it a tool that anybody could use? In theory yes, in practice it depends.

Only a small proportion of the population has the ability to master procedural programming and an even smaller precentage has the interest in so doing. (A much large proportion of the population shows the ability and the interest to tackle Excel’s native programming method: functional no-side-effects programming using Excel formulas, which accounts for the appeal of spreadsheets to the general populace!).

So although I’ve abstracted away a large level of the complexity of automating Excel by means of the SQLScript and SQL functions, it’s still programming.  But even if the full power of the tool is beyond most, the tabular sequential nature of  SQLScript and the relative approachability of SQL should mean that many can at least follow the logic and data flows and might even be able to modify existing scripts. Those users with a good knowledge of VBA and or SQL should find the tool enables them to be more productive and cost effective. While those with Excel formula skills can usually work in tandem with a framework of microETL functionality with relative ease.

For the latest versions and articles on using microETL follow the microETL tag on this blog …

→ 1 Comment

Posted in ETL, excel, microETL, SQLite, VBA

microETL’s SQL function

Posted on March 9, 2011 | 5 Comments

At the heart of microETL is the SQL() command (or more correctly microETL.SQL()). The primary purpose of microETL is to allow the power of SQL to be used within Excel and SQL() is how that is delivered. The command can either be issued from a cell, like a standard excel formula, or by using the microETL SQL Menu option or as a command in a SQLScript. It can be called from a VBA macro (using Application.Run(“microETL.SQL” … ) but not if that macro is to be used as a User Defined Function (aka user defined formula). It can also be called from within another SQL statement using the equivalent SQLite user defined function also called SQL().

Example: click for a larger image (animated GIF)

=SQL() in action

SQL(sqlStatement,selectDestination(OR arg1), arg2 …. argn)

The 1st argument is the SQL statement(s) to execute. These can be any valid SQLite statements (see SQL as understood by SQLite); exceptions are the ATTACH statement and in some circumstances the DROP TABLE statement. The statement will be issued against the default microETL in-memory database i.e. an empty database that is instantiated in memory at workbook start-up and any tables or data created during the workbook session are discarded at shutdown.

So, if the database is ephemeral where will my data be stored? In the spreadsheet most likely. The other alternatives  are to use the SQLite UDFs load_CSV() and write_CSV() to load/save CSV files, or see http://blog.gobansaor.com/2011/05/11/accessing-sqlite-databases-from-excel-via-microetl/ for other options.

In order to make transferring data between Excel and the SQL environment easier and simpler a pre-processor has been added to the function. There are three types of pre-processor commands:

  • The TABLE(…) command tells SQL() to go fetch the relevant Excel range, loads the “table” into SQLite; when the SQL command finishes it writes back the data (if changed) to the originating Excel address, and cleans up the temporary SQLite-side table. Nearly all SQL commands can be issued against the resulting tables, including UPDATE, DELETE, DROP and CREATE TRIGGERS etc.
  • Two other pre-processor commands are CELL(…) and CELLS(…). Similar to TABLE(…), data is loaded from the Excel addresses specified and placed in the SQL  statement. CELL takes a single cell reference while CELLS will iterate through a multi-cell range and insert a comma-separated list into the SQL statement (in a format suitable for IN (…) expressions).
  • Positional substitution tokens in the form :1 to :n

Note these are pre-processor commands not SQLite commands and can only be accessed using the SQL() function. The pre-processor “functions” may appear to work like a normal SQLite function call but are less flexible in some ways e.g. no spaces allowed, parser expects “TABLE(” not “TABLE  ( ” but more flexible in others so a function requiring a table name can use the pre-processor to provide it e.g. =SQL(“Select load_CSV(‘TABLE(Sheet!A1)’,'c:\data\test.csv’);)”).

The sequence of events when a TABLE() pre-processor command is detected in a SQL statement is as follows:

  • SQLite SAVEPOINT is set if the SQL()  statement is not already running in the context of a previous SQL() function call. This wrapping in a transaction is the reason for ATTACH commands not been allowed. The purpose of the transaction is two-fold; 1st it will clean up any temporary tables and other updates in the case of a function failure, 2nd, SQLite inserts/updates are much faster when executed within a transaction, especially if writing to a disk-based database.
  • If the text between “TABLE(” and the next “)” is a valid range representation (or a named range pointing at such) a check is carried out to see whether that same range has been referenced before in this context (same context being within the same set of SQL statements or those of a prior “wrapping” SQL() function call). If already “handled” replace the TABLE(…) with the name of the temporary table associated with this range. If not, load the table data (as long as the range is an Excel table, or points to the top left-hand corner of a potential table) into a new temporary SQLite table and replace the TABLE(…) with its name.
  • When the last SQL statement within the current context is executed and if the temporary table has been modified (INSERT’ed, UPDATEd,DELETEd or DROPped ) the table is written back to the source range (or range cleared if a DROP statement).
  • The SAVEPOINT is released.

The remaining arguments to the SQL() function are also processed by the pre-processor, the arguments passed replace positional substitution tokens in the form :1 to :n where :1 is replaced by 2nd argument and :n is replaced by the n+1th argument.

The 2nd argument (that is the arg corresponding to the :1 token) is special. If no “:1″ token is found within the SQL Statement, the argument is assumed to be a destination for a SQL SELECT statement (think of it as a SELECT INTO). It can be either a string representation of a workbook range or a named range pointing at one. The range can be either a full range, top left-hand corner range or an Excel Table. (If none of the above, the SELECT will populate an ADO detached RecordSet and use the argument as the name of that recordset).

A number of helper functions have been added to SQLite within microETL. Note: these are not preprocessor commands and so can be used by any SQL issued within microETL. (There are a number of SQL issuing functions other than SQL() and there’s also the SQL calls issued within Python functions.)

Using the Menu option SQL

The main SQL helper functions are …

x(formula,p1,p2,p3…)

Where formula is a spreadsheet formula or built-in function, followed by a variable list of parameters. The formula can take advantage of microETL “placeholders” e.g. x(“20+ :1 + :2″,20,40) will return 80, x(“upper(:1:)”,”abc”) returns ABC.

The formula’s text, with any placeholders replaced, is passed to Excel’s Application object’s Evaluate method, so not withstanding some of the limitations associated with this function (see here) it’s possible to use most Excel formulas within a SQL statement.

Only built-in functions (i.e not User Defined Functions) can be called in this manner.  To call a UDF use the udf() (or u() /mdf())  function.

When specifying the formula do not include the leading “=”, this will ensure formula is evaluated inline along with the rest of the SQL statement. If preceded by “=”, the formula is passed back as text. This can sometimes be useful either for testing purposes or to get around Application.Evaluate limitations.

Example: click for larger view:

udf(functionName,p1,p2,p3…)

This allows any VBA public function (i.e. a UDF) in the active workbook  to be called from within a SQLite SQL statement. The u() and mdf() functions are similar but the mdf() function will only run functions in the microETL project while u() will default to the microETL project but will also take references to UDFs in other workbooks.

Py(address,p1,p2,p3…)

The address can be either a string representation of an Excel range e.g. “Sheet2!A3:Sheet2!D6″ or a named range. If the range is a single cell the function will expect to find the full Python script within, if a multi-cell range then the script will utilise cells as Python indents.

The scripts are in fact anonymous parameter-less Python functions. Parameters are passed via xLite “placeholders” which are replaced before the script is passed to the Python interpreter e.g.

If in  this case 2nd parameter is less than 40, return 1st parameter increased by 10% otherwise return unchanged.

SQLScript(address,p1,p2,p3…)

See previous post http://blog.gobansaor.com/2011/03/04/sqlscript-microetls-sql-sequencer-utility/

SQL(sqlStatement,p1,p2,p3…)

This allows the microETL.SQL() function (along with its pre-processor goodness) to be called from an environment where the pre-processor is not supported e.g. within Python. It’s also useful to process a series of SQL statements held in a table.

Various Hierarchy functions

See this Handling Flat, Parent-Child and Nested Set Hierarchies and this.

write_CSV(table,filename)

This will write out the contents of the table in classic CSV format (UTF-8 encoded). Useful for external SQLite tables that may be too big to fit in memory but need transformation before loading as a CSV into the likes of PowerPivot.

load_CSV(table,filename)

Will load UFT-8 encoded CSV files into a table. Again, if a CSV source is too big to load into memory (thinking fact tables here) but needs some SQL love’n'care can be loaded into an external table (i.e. in a disk-based database) bypassing the need to fit in Excel’s working memory.

sqliteDate(dateTimeString)

Will attempt to convert a string to an ISO DateTime string ( the format used internally by SQLite i.e. “YYYY-MM-DD hh:mm:ss”). Usually microETL will handle date conversions between Excel and SQlite and back again (but when a date is pasted back it’ll be in Excel numeric representation, the column may need to be formatted to the required date/time representation). Sometimes however the date format of imported “string dates” may not be recognised (for example, Excel loaded CSVs often fail to recognise dates correctly). This function will usually solve the problem.

ISODate(Year,Month,Day)

Will take a date passed as 3 numbers (e.g. ISODate(2010,12,31)) and convert to SQLite date format.

REGEXP

I’ve also added REGEXP (regular expression) support so you can do stuff like so…

SELECT * FROM Foo WHERE account_id REGEXP ‘[0][0-9]*[x]‘

or

=SQL(“SELECT regexp(‘[0][0-9]*[x]‘,’0987f85x’);”) -> returns 0 for false (should be all digits between leading zero and final ‘x’).

→ 5 Comments

Posted in ETL, microETL, SQLite, VBA

Tagged