what should i fill in each blank of smart service in process model, or in interface smart service code?
Discussion posts and replies are publicly visible
This is part of the Excel Tools plugin (https://community.appian.com/b/appmarket/posts/excel-tools#) and a readme PDF is available in the download. As of version 2.2.16 the readme provides the following info on the inputs to the smart service.
what is delimiter?
Not sure if the sheet number starts with 0 or 1. Do you only have one sheet? If sheet number 1 isn't working just try 0.
seems it got some changes. But do you get an idea for this, seems i miss something
What's your table name and data source name?
the table name is the constant of datastore entity sub belonging table, data source i just like yours jdbc/appian,because i use appian database.
Table name is text not datastore entity. Literally use the text name of the table (eg. "APN_TableName")
ajhick said: "APN_TableName"
table name should be appian database table name or the datatype name?
The Appian database table name
I almost there, do you know how to transfer excel text type into datetime type?
As far as I know this isn't possible. What I did was to create a staging table where the datetime was actually of type Text so that the import will work. I then created a stored procedure to transform that text into the actual datetime in the real table.
what if it report some longtext impossible to import in?
Well, yes. You've said "This field can be a maximum of <THIS> long" and then you're trying to put more than <THIS> in. So it should fail.
What's your character limit on `groupname` and how much are you trying to put in?
I used longtext,there are some rows with alphabets more than 300 include semicolon & spaces, i just checked, it was all successfully import but why the process model report error?
You'll either need to change the table or somehow ensure the excel file doesn't exceed the limits on your table. No way around that one.
any way to let specific column to match specific database column though they have different column names?
I don't think so but it really sounds like you need that interim table (staging table) that exactly matches your Excel spreadsheet and then some way (for example, a stored procedure) that transforms and moves across the data to it's final destination.
Do you have a hint what mistake this for? Thank you
Only guessing but is there a 1 to 1 relationship between the column headers in your excel spreadsheet and the columns in your table? Do you have an auto increment primary key column? Is any column marked as required in the database?
I have heard that auto increment primary key columns can cause issues if they are created from Appian rather than from the database. Was the table you are importing into created from a CDT in Appian?
I create datatypes from existing table of appian back end database, precautiously not let it couldn't insert duplicated data, i create a rowid which as number to be auto increment. Is this because i don't have any column called rowid since it as primary key? But i let it auto increment show have a number input in by itself successfully since my previous case did like this.