« SQL Server Integration Services » : différence entre les versions
Aucun résumé des modifications |
|||
(12 versions intermédiaires par le même utilisateur non affichées) | |||
Ligne 76 : | Ligne 76 : | ||
== Data Conversion == | == Data Conversion == | ||
Create a new column with a data converted in a new type. | Create a new column with a data converted in a new type. | ||
{| class="wikitable wtp" | |||
|+ Tips to convert from Excel to MariaDB | |||
! Excel type | |||
! MariaDB type | |||
|- | |||
| DT_DATE || DT_DBDATE | |||
|} | |||
== Derived Column == | == Derived Column == | ||
Ligne 86 : | Ligne 93 : | ||
== Aggregate == | == Aggregate == | ||
Do an aggregate operation like {{boxx|Group by}} {{boxx|Count}} {{boxx|Sum}} {{boxx|Average}} {{boxx|Minimum}} {{boxx|Maximum}} | Do an aggregate operation like {{boxx|Group by}} {{boxx|Count}} {{boxx|Sum}} {{boxx|Average}} {{boxx|Minimum}} {{boxx|Maximum}} | ||
== Lookup == | |||
Creates 2 outputs, one for the matched rows and another one for the non-matched rows. | |||
== Conditional Split == | |||
Creates multiples outputs, one for each filter. | |||
== Union All == | |||
Combines the rows of 2 sources. | |||
== Data Warehousing == | |||
A way to handle separately new rows and rows to update.<br> | |||
To do so we can use the Slowly Changing Dimension Wizard (only for small datasets).<br> | |||
Or doing it manually: | |||
# Use a lookup to isolate new rows | |||
## Add new rows to destination | |||
## Use a conditional split to isolate rows to update from non-new rows | |||
### Update the rows to update | |||
= Expressions = | |||
== Concatenating values == | |||
<kode lang='tsql'> | |||
[Column1] + " - " + [Column2] | |||
-- if [Column1] or [Column2] is null, the result will be null | |||
-- handle null values | |||
(ISNULL([Column1]) ? "" : [Column1]) + " - " + (ISNULL([Column2]) ? "" : [Column2]) | |||
</kode> | |||
== Cast to string == | |||
<kode lang='tsql'> | |||
"Date: " + (DT_STR, 20, 1252)@[Column3] | |||
</kode> | |||
= Variables = | |||
Right-click on the designer → Variables | |||
* Filter: show system variables or variables of all scopes | |||
* Columns: add Namespace, Description, Raise event when var value changes | |||
= Erreurs = | = Erreurs = |
Dernière version du 9 mars 2023 à 16:33
Présentation
Permet de:
- extraire des données depuis une source: fichier, base de données
- transformer ces données
- charger ces données vers une destination: fichier, base de données
SSIS est un outil ETL (Extraction Transformation and Load)
Liens
Nouveau Projet
Visual Studio → File → New → Project → Integration Services Project
Control Flow
Récupération des données.
Exécuter du code SQL
Execute SQL Task
Télécharger un fichier depuis internet
- Spécifier l'url du fichier à télécharger: clique-droit dans Connection Managers → New Connection → Type: HTTP
- Spécifier l'emplacement où télécharger le fichier: clique-droit dans Connection Managers → New File Connection
- Ajouter une Script Task
ScriptMain.cs |
try { // Logging start of download bool fireAgain = true; Dts.Events.FireInformation(0, "Download File", "Start downloading " + Dts.Connections["HTTP Connection Manager"].ConnectionString, string.Empty, 0, ref fireAgain); // Get your newly added HTTP Connection Manager var mySSISConnection = Dts.Connections["HTTP Connection Manager"].AcquireConnection(null); // Create a new connection var myConnection = new HttpClientConnection(mySSISConnection); // Download file and use the File Connectionstring to save the file myConnection.DownloadFile(Dts.Connections["DownloadedFile"].ConnectionString, true); // Logging end of download Dts.Events.FireInformation(0, "Download File", "Finished downloading " + Dts.Connections["DownloadedFile"].ConnectionString, string.Empty, 0, ref fireAgain); // Quit Script Task succesful Dts.TaskResult = (int)ScriptResults.Success; } catch (Exception ex) { // Logging why download failed Dts.Events.FireError(0, "Download File", "Download failed: " + ex.Message, string.Empty, 0); // Quit Script Task unsuccesful Dts.TaskResult = (int)ScriptResults.Failure; } |
Data Flow
Mise en forme des données.
Fichier CSV
SSIS Toolbox Box → Other Destinations/Sources → Flat File Destination/Source
Multiple sources
Connection → MULTIFLATFILE | pour plusieurs fichiers CSV ayant exactement les mêmes colonnes |
Union All | pour réunir des données non triées |
Merge | pour réunir des données triées |
Merge Join | pour réunir des données découpées et liées |
Data Conversion
Create a new column with a data converted in a new type.
Excel type | MariaDB type |
---|---|
DT_DATE | DT_DBDATE |
Derived Column
Create a new column or replace an existing one with computed result as value.
-- convert a DT_NUMERIC to DT_DBDATE (2020[int] → 2020-01-01[date]) COLUMN1 == 0 ? NULL(DT_DBDATE) : (DT_DBDATE)((DT_STR,4,1252)COLUMN1 + "-01-01") |
Aggregate
Do an aggregate operation like Group by Count Sum Average Minimum Maximum
Lookup
Creates 2 outputs, one for the matched rows and another one for the non-matched rows.
Conditional Split
Creates multiples outputs, one for each filter.
Union All
Combines the rows of 2 sources.
Data Warehousing
A way to handle separately new rows and rows to update.
To do so we can use the Slowly Changing Dimension Wizard (only for small datasets).
Or doing it manually:
- Use a lookup to isolate new rows
- Add new rows to destination
- Use a conditional split to isolate rows to update from non-new rows
- Update the rows to update
Expressions
Concatenating values
[Column1] + " - " + [Column2] -- if [Column1] or [Column2] is null, the result will be null -- handle null values (ISNULL([Column1]) ? "" : [Column1]) + " - " + (ISNULL([Column2]) ? "" : [Column2]) |
Cast to string
"Date: " + (DT_STR, 20, 1252)@[Column3] |
Variables
Right-click on the designer → Variables
- Filter: show system variables or variables of all scopes
- Columns: add Namespace, Description, Raise event when var value changes
Erreurs
A required privilege is not held by the client
Getting this error while deploying.
This error happens after the installation of Visual Studio SSDT extension.
To fix it, run the repai process of SQL Server.
Unable to bulk copy data. You may need to run this package as an administrator
Utiliser OLE DB Destination au lieu de SQL Server Destination.
Installation
- New SQL stand-alone installation or add features to an existing installation
- Instance Features → Database Engine Services
- Shared Features → Integration Services
Enable named pipes
- C:\Windows\SysWOW64\SQLServerManager16.msc
- SQL Server Network Configuration → Protocols for MSSQLSERVER
- Named Pipes → Enabled
- TCP/IP → Enabled
SSIS Catalog
- SSMS → right-click on Integration Services Catalogs → Create Catalog
- Enable CLR integration
- Enable automatic execution of IS stored procedure at SQL Server startup
SQL Server Agent
- Start from SSMS or Services
Visual Studio SQL Server Integration Services Extension
- Visual Studio → Extensions → Manage Extensions
- SQL Server Integration Services Projects