content about app_xid in the alteryx connect h2 entry table

[jerrylau]

Hi all,

I am using 2022.3 Connect now and I am looking into some previous workflow (which worked on older Connect version) developed by member left already

The workflow tries to extract data from the STD_XFORMS.ENTRY to get app_xid = file, file_table and file_table_column respectively to rebuild a hierarchy structure where file / file_table are parent of file_table_column

However, when i tested this in a separate environment (which I have already pushed some excel / yxdb files into Connect), I cant find any data matching app_xid = "file_table_column"

Does anyone have any idea that how are entries supposed to load into Connect with the "file_table_column"?

Thanks!

Jerry

[PetrH]

Hi @jerrylau , 
I would bet that Data profiling while executing "Files loader" was not checked.

That's why the data about columns(csv, yxdb) or sheets(excel) was not analyzed.

see print screen

[jerrylau]

Hi Petrh,

Thanks for your prompt reply, actually I have checked that option, and in the Connect page, I can see the column info for excel->sheet and yxdb too

I can also find some column info in another table called files_table_meta but that one seem not include info like current_version

[PetrH]

Hi @jerrylau , 

I checked the sync job settings for files.
Actually its related to data profiling, but in case that the connect is switched to mode, where it treats columns (database, and files) as individual assets.
In older versions of connect we treated the columns as individual assets having all possible settings, like owners, descriptions or tags. (it was pretty hi demanding to database and hardware - with combination of large size content)
(the enabling such feature might have some performance impact)
In files loader sync job there is section of creating "file_table_columns" this section is executed only (based on where clause) the Connects treats the individual asset details. (this setting is for system level, there is no option to enable it only for files, database columns or specific servers)

as a summary I can say that this section is set only if you enable the individual columns detail in admin/performance. than the individual page for each column will be created, and data profiling on columns would be visible. In current presentation where we treat columns as a part of table(simple table presentation) there is no space to list such details in one row. 
again one print screen with mentioned detail 

[jerrylau]

Thanks PetrH!

So is there any documentation to explain how Connect did the data injection and the mechanism explanation?

I am comparing the sync job xml files between 2019.3 (old version we used before) and 2022.3 (current version), but i barely find big difference, so I am not sure what changes were made to introduce such a difference now. (and not really familiar with what the script is doing actually)

Jerry

[jerrylau]

and one more question is, i have many folders under a root folder, as there are too many files

i try to scan them batch by batch e.g. <root>\A, <root>\B... each time (by slightly modifying the file_loader at the beginning, add a filtering)

but it turns out that i scanned A, the connect console appears <root>\A, next i scan B, the connect consoles show up B but disappeared A

i can still find out folder A in the h2 database but the entry_current flag becomes False

so what's wrong with it?