Skip to content
Commits on Mar 9, 2015
  1. Merge pull request #16 from ManiacDC/master

    committed
    Variable naming and scope corrections
  2. @ManiacDC

    Fixed spelling of "AvailableTypes" in DBA.DatabaseFactory *BACKWARDS …

    ManiacDC committed
    …INCOMPATIBLE*
    
    Renamed some internally used variables and classes to more properly reflect on their global scope (potential for backwards incompatibility)
    Fixed scope of __MySQL_Query_Dump function
Commits on Aug 28, 2012
Commits on Aug 19, 2012
Commits on Aug 15, 2012
  1. general refactoring

    committed
Commits on Aug 13, 2012
  1. added image for blob tests

    committed
  2. added image for blob tests

    committed
Commits on Aug 11, 2012
Commits on Aug 5, 2012
Commits on Jul 7, 2012
  1. fixed non Select Querys: Check for rs-Object and if its Closed. Some …

    committed
    …provider return a closed RS in case of a non selection query.
Commits on Jun 29, 2012
  1. Merge pull request #14 from infogulch/cdecl

    committed
    missed a cdecl
Commits on Jun 28, 2012
  1. @infogulch

    missed a cdecl

    infogulch committed
Commits on Jun 27, 2012
  1. Merge pull request #13 from infogulch/DllCallCDecl

    committed
    Fix missing CDecl on sqlite dllcalls
  2. Merge pull request #12 from infogulch/isfilepathvalid

    committed
    Fix IsFilePathValid().
Commits on Jun 26, 2012
  1. @infogulch
  2. @infogulch
Commits on Jun 7, 2012
  1. Merge pull request #10 from infogulch/sqlite-prepared-insert

    committed
    Use prepared queries for DataBaseSQLite.InsertMany
Commits on Jun 6, 2012
  1. @infogulch

    Use prepared queries for DataBaseSQLite.InsertMany

    infogulch committed
    Using prepared queries is faster and more type safe. I tested with
    500,000+ row inserts and ran into "Maximum Length Of An SQL Statement"
    problems (see: http://www.sqlite.org/limits.html #3). Also using
    BeginTransaction, EndTransaction (and Rollback if necessary) along with
    prepared queries makes a disk-based db just as fast as :memory: for such
    large queries.
    
    The `prepare, loop { bind, step, reset }` method is highly reccomended
    vs a giant sql statement by both the docs and developers on irc.
  2. ignore *.sqlite files.

    committed
  3. removed test db

    committed
  4. several improvements, better exception handling. replaced sqlite.dll …

    committed
    …(32bit) with newest version which supports multi insert statements.
Commits on Jun 5, 2012
  1. fixed typo, thx @ Sheik

    committed
Commits on Jun 3, 2012
  1. Merge pull request #9 from infogulch/collections-count

    committed
    Fix performance issue in Collections.Count()
  2. Merge pull request #8 from infogulch/enumerators

    committed
    Implement special enumerators in Row and Table
Commits on Jun 1, 2012
  1. @infogulch

    Fix performance issue in Collections.Count()

    infogulch committed
    This originally looped through the entire object to get the number of
    items. For large objects, this could be a huge performance hit. Instead,
    use `this.SetCapacity(0)`, which returns the object's capacity after
    shrinking the object to fit. Any performance loss due to shrinking the
    object prematurely is gained back exponentially by removing a loop of ahk
    code.
    
    Consider:
    
        while this.Count() < 1000000
                this.insert(A_Index)
    
    This code would loop 500,000,500,000 extra times before this commit.
    Insane.
  2. @infogulch

    Implement special enumerators in Row and Table

    infogulch committed
    When someone tries to enumerate a table directly (not Table.rows), they
    probably mean to enumerate rows. This adds a simple wrapper to do that.
    
    When enumerating a Row, enumerate both the columns and fields Collections
    simultaneously, where the first param (key) is the column name, and the
    second param (value) is the field's value: `for col_name, value in Row`
Commits on May 24, 2012
  1. removed dbg line

    committed
Something went wrong with that request. Please try again.