I have a data file that I export from an internal web based program. The data are long text and when I tried exporting as xlsx and perhaps did not download in a usable format due to html code and length? I notice there was an xml option and chose that. I was able to import ok on Friday using External Data Import XML File. Today, I created a new download (only difference should be new data between Friday and Today) exported as xml and went through the wizard steps. I chose append to existing table, it did not let me choose the table, it created a new table. After some trial and error, got the data in - had to modify structure to long text and remove the @ symbol. In future, I'll copy the existing table to the table name xml import wizard expects and that will allow it to append to existing.
Since still in the process of setting this up as on-going and it refused to import/append to the existing table, I decided to compare the data to Friday, there were a couple of differences that I'm not sure how to fix in order to be able to rely on the data going forward. Three of the records (each record's issue was in a separate column) had what appears to be a difference.
I opened the two xml files in winmerge and located the id. When I scrolled across, I noticed html entities were different. Showing the highlighted part below...
Friday's download:
If I am to automate/rely on data not being truncated, any ideas how I could test, without having to keep old data, run a query to do a literal compare of each individual field in order to find if there is a problem or is this a problem?
Since still in the process of setting this up as on-going and it refused to import/append to the existing table, I decided to compare the data to Friday, there were a couple of differences that I'm not sure how to fix in order to be able to rely on the data going forward. Three of the records (each record's issue was in a separate column) had what appears to be a difference.
I opened the two xml files in winmerge and located the id. When I scrolled across, I noticed html entities were different. Showing the highlighted part below...
Friday's download:
Today's download:
Could that account for the difference, and I wonder why the entity would be different between Friday and Today when the rest of the text is identical.
If I am to automate/rely on data not being truncated, any ideas how I could test, without having to keep old data, run a query to do a literal compare of each individual field in order to find if there is a problem or is this a problem?
