Hi Ankur & Friends,
Iam very new to this blog...Iam happy to find out such a blog where people interacts and share their views...I hope my query will also be answered here... By writing this blog I just want to find out a way to automate a kind of monotonous work that we are doing for a long time..I want to automate this using QTp.... the actual process is as follows..
1) There are two excel sheets one is the "Base file" and the other is the "New file"...for each release a new "New file" will be created and we need to compare (regression testing) it with the "Base file"
2) Both of the files contain minimum of 500 rows and 50 columns
3) Following are the possible outputs
a) New file matches exactly with the Base file - no issues
b) New file has some extra rows/columns when compared to the Base file - to be noted
c) New file has some missing rows/columns when compared to base file - to be noted
d) There is some mismatching values between the New file and Base file - to be noted
------ The rows in both the excels may not be in the same order...cannot even be arranged in the same order...due to extra/missing rows possibility
In all teh cases except 'a', the values under issue are validated against the application (purely manual)
For comparing excels, we are using an XLcomparator tool....this compares both the files and generated the difference report..... the limitation with this tool is that it can have maximum of four indices only ( Indices are the column names in a row...which put together uniquely identifies the row).....
The output of the tool are not satisfactory...most of the times it reports many duplications (same row repeating in the file many times..) as it uses only 4 indices ...just using four indices we cannot uniquely identify each row .....so we use to do all permutations and combinations with the available cloumn names and change the indices and run the tool again and agin till it gives the least duplications...this is very tedious and as the number of releases increase, it is troubling us like anything.... for a single release we have to compare 1014 such Base files and New files....
I just want to replace this comparator tool with a QTP script ...to get the best results...is that possible...please suggest me ur views and thoughts...
A sample "New file" is attached for you to have the feel how they look like..... the colourings are all done by the comparator tool to mark the duplication/missing/extra/mismatch data....
Thanks,
Karthik
Iam very new to this blog...Iam happy to find out such a blog where people interacts and share their views...I hope my query will also be answered here... By writing this blog I just want to find out a way to automate a kind of monotonous work that we are doing for a long time..I want to automate this using QTp.... the actual process is as follows..
1) There are two excel sheets one is the "Base file" and the other is the "New file"...for each release a new "New file" will be created and we need to compare (regression testing) it with the "Base file"
2) Both of the files contain minimum of 500 rows and 50 columns
3) Following are the possible outputs
a) New file matches exactly with the Base file - no issues
b) New file has some extra rows/columns when compared to the Base file - to be noted
c) New file has some missing rows/columns when compared to base file - to be noted
d) There is some mismatching values between the New file and Base file - to be noted
------ The rows in both the excels may not be in the same order...cannot even be arranged in the same order...due to extra/missing rows possibility
In all teh cases except 'a', the values under issue are validated against the application (purely manual)
For comparing excels, we are using an XLcomparator tool....this compares both the files and generated the difference report..... the limitation with this tool is that it can have maximum of four indices only ( Indices are the column names in a row...which put together uniquely identifies the row).....
The output of the tool are not satisfactory...most of the times it reports many duplications (same row repeating in the file many times..) as it uses only 4 indices ...just using four indices we cannot uniquely identify each row .....so we use to do all permutations and combinations with the available cloumn names and change the indices and run the tool again and agin till it gives the least duplications...this is very tedious and as the number of releases increase, it is troubling us like anything.... for a single release we have to compare 1014 such Base files and New files....
I just want to replace this comparator tool with a QTP script ...to get the best results...is that possible...please suggest me ur views and thoughts...
A sample "New file" is attached for you to have the feel how they look like..... the colourings are all done by the comparator tool to mark the duplication/missing/extra/mismatch data....
Thanks,
Karthik
