Hi Yashwanth,
You took quite a considerable time to write your question. 
Please feel free to revert back. Also I would recommend to test each scenario as I have tried to explain as much as I can.
I would like to explain you the concepts of Upper Limit and Lower limits first and then you will have your answers by yourself (well, most of it).
First, Upper Limit:
This is related to upper limit of selection criteria that is used in generic extraction. Let's take an example:
Say I have extracted a data at 12:00 and then again I am extracting a data at 12:30. Now there can be 2 cases:
1. Upper Limit Given (say 300 secs = 5 Min): In this case When the data is extracted at 12:30 then the selection criteria sent for generic extraction will be low = 12:00 and high = 12:25 (this is because 5 minutes are subtracted from upper limit 12:30). Here then the delta time in RSA7 would be set as 12:25.
And when the extraction happen at 1:00 again then the selection criteria sent would be low = 12:25 and high = 12:55.
2. Upper Limit not given: In this case When the data is extracted at 12:30 then the selection criteria sent for generic extraction will be low = 12:00 and high = 12:30 (here nothing will be subtracted from upper limit 12:30). Here then the delta time in RSA7 would be set as 12:30.
And when the extraction happen at 1:00 again then the selection criteria sent would be low = 12:30 and high = 1:00.
Now why we do set upper limit? - If we do not set then,
Let's say for example, a document- is created at 12:25 but not saved until 12:35. Now this record will not be contained in data extraction happening at 12:30 as it is not saved. Also it will not be contained in the extracted data happening at 1:00 because the timestamp of that record is 12:25 and the selection critera for extraction at 1:00 is 12:30 to 1:00.
So if we set the upper limit it will ensure the extraction of unsaved records which might take time for saving.
Coming to Lower Limit:
This is related to lower limit of selection criteria that is used in generic extraction. Let's take same above example:
Say I have extracted a data at 12:00 and then again I am extracting a data at 12:30. Now there can be 2 cases:
1. Lower Limit Given (say 300 secs = 5 Min): In this case When the data is extracted at 12:30 then the selection criteria sent for generic extraction will be low = 11:55 and high = 12:30 (this is because 5 minutes are subtracted from lower limit 12:00). Here then the delta time in RSA7 would be set as 12:30.
And when the extraction happen at 1:00 again then the selection criteria sent would be low = 12:25 and high = 1:00.
2. Lower Limit not given: In this case When the data is extracted at 12:30 then the selection criteria sent for generic extraction will be low = 12:00 and high = 12:30 (here nothing will be subtracted from Lower limit 12:30). Here then the delta time in RSA7 would be set as 12:30.
And when the extraction happen at 1:00 again then the selection criteria sent would be low = 12:30 and high = 1:00.
Now why we do set lower limit? - If we do not set then, same example applies to lower limit too.
So if we set the lower limit it will ensure the extraction of unsaved records which might take time for saving.
----------------------------------------------------------------------------------------------------------
Now coming to your questions:
CALDAY case:
1.consider 2014-08-01 i have initialised the delta, while initialising i mentioned 1 as upper limit, now i have triggered the infopack on same day(2014-08-01) what will happen? after delta load what will be the pointer in RSA7. then again 2014-08-02 i ran the delta at same time frame, what will be the pointer. when to use upper limit.
---> After initialization if you try to extract delta again on same day - it will not extract anything. The reason is becuase the selection criteria sent to generic extractor would be low = '2014-08-01' and high = '2014-07-31' (2014-08-01 minus 1 day). So this is a invlaid condition and nothing will be selected. This will not effect the upper limit of delta limit in RSA7 and it will remain 2014-08-01.
Now if you extract on 2014-08-02, then data of only 2014-08-01 will be extracted as the limit for generic extraction would be low = '2014-08-01' and high = '2014-08-01' (2014-08-02 minus 1 day). So this will repeated data of previous day.
Upper limit set to 1.
2014-08-01 - initialization - All previous data till 2014-08-01 will be extracted - RSA7 limit set to 2014-08-01
2014-08-02 - first Delta - all data from '2014-08-01' and high = '2014-08-01' - RSA7 limit set to 2014-08-01
2014-08-03 - second Delta - all data from '2014-08-01' and high = '2014-08-02' - RSA7 limit set to 2014-08-02
2014-08-04 - third Delta - all data from '2014-08-02' and high = '2014-08-03' - RSA7 limit set to 2014-08-03
I agree that this is kind of messy situation. So the upper limit is mostly used in case of timestamps where the time taken saving of records is considered into account. However this is a strategic decision taken by business.
2.consider 2014-08-01 i have initialised the delta, while initialising i mentioned 1 as lower limit, now i have triggered the infopack on same day(2014-08-01) what will happen? after delta load what will be the pointer in RSA7. then again 2014-08-02 i ran the delta at same time frame, what will be the pointer. when to use lower limit.
---> After initialization if you try to extract delta again on same day - it will extract repeated data again. The reason is becuase the selection criteria sent to generic extractor would be low = '2014-07-31' (2014-08-01 minus 1 day) and high = '2014-08-01' . So this is almost same as condition during initialization and repeated data of one day will be extracted. This will not effect the upper limit of delta limit in RSA7 and it will remain 2014-08-01.
Now if you extract on 2014-08-02, then data of 2014-08-02, 2014-08-01 and 2014-07-31 will be extracted as the limit for generic extraction would be low = '2014-07-31' (2014-08-01 minus 1 day) and high = '2014-08-02' (2014-08-01 minus 1 day). So this will repeated data of previous 2 day.
Now lets consider below situation:
Lower limit set to 1.
2014-08-01 - initialization - All previous data till 2014-08-01 will be extracted - RSA7 limit set to 2014-08-01
2014-08-02 - first Delta - all data from '2014-07-31' and high = '2014-08-02' - RSA7 limit set to 2014-08-02
2014-08-03 - second Delta - all data from '2014-08-01' and high = '2014-08-03' - RSA7 limit set to 2014-08-03
2014-08-04 - third Delta - all data from '2014-08-02' and high = '2014-08-04' - RSA7 limit set to 2014-08-04
The Lower limit can be used in case of timestamps and day based extraction. However this is a strategic decision taken by business.
3.consider 2014-08-01 i have initialised the delta, while initialising i mentioned 1 as upper limit and 1 as lower limit now i have triggered the infopack on same day(2014-08-01) what will happen? (can i do like this-----if yes then what cases,------if no ,why cant )
---> Having this case I believe you can yourself now figure out the values in generic extraction. I will just give you the timelimits set for this condition.
Lower limit set to 1. Upper Limit set to 1.
2014-08-01 - initialization - All previous data till 2014-08-01 will be extracted - RSA7 limit set to 2014-08-01
2014-08-02 - first Delta - all data from '2014-07-31' and high = '2014-08-01' - RSA7 limit set to 2014-08-02
2014-08-03 - second Delta - all data from '2014-08-01' and high = '2014-08-02' - RSA7 limit set to 2014-08-03
2014-08-04 - third Delta - all data from '2014-08-02' and high = '2014-08-03' - RSA7 limit set to 2014-08-04
I beleive its totally dependent on which case you want to use for extraction. I personally rely on Lower limits in all cases.
----------------------------------------------------------------------------------------------------------
Having explained so much about the limits in generic extraction I will just give you the time values in case of timestamp extraction and related upper and lower limits.
related to ' timestamp ' pointer as delta (yyyy-mm-dd ,09:00:00) every 1hour delta has to run
1.consider 2014-08-01,09:00:00 i have initialised the delta, while initialising i mentioned 1800 seconds as upper limit, now i have triggered the infopack. what will happen? after delta load what will be the pointer in RSA7. take 3 successive time intervals and explain, when we go for this setting
--->Upper limit set to 1800.
2014-08-01, 09:00:00 - initialization - All previous data till 2014-08-01 will be extracted - RSA7 limit set to 2014-08-01, 09:00:00
2014-08-01, 10:00:00 - first Delta - all data from '2014-08-01, 09:00:00' and high = '2014-08-01, 09:30:00' (1800 secs minus from 10:00) - RSA7 limit set to 2014-08-01, 09:30:00
2014-08-01, 11:00:00 - second Delta - all data from '2014-08-01, 09:30:00' and high = '2014-08-01, 10:30:00' (1800 secs minus from 11:00) - RSA7 limit set to 2014-08-01, 10:30:00
2014-08-01, 12:00:00 - third Delta - all data from '2014-08-01, 10:30:00' and high = '2014-08-01, 11:30:00' (1800 secs minus from 12:00) - RSA7 limit set to 2014-08-01, 11:30:00
2.consider 2014-08-01,09:00:00 i have initialised the delta, while initialising i mentioned 1800 seconds as lower limit, now i have triggered the infopack. what will happen? after delta load what will be the pointer in RSA7. take 3 successive time intervals and explain, when we go for this setting
--->Lower limit set to 1800.
2014-08-01, 09:00:00 - initialization - All previous data till 2014-08-01 will be extracted - RSA7 limit set to 2014-08-01, 09:00:00
2014-08-01, 10:00:00 - first Delta - all data from '2014-08-01, 08:30:00' (1800 secs minus from 09:00) and high = '2014-08-01, 10:00:00' - RSA7 limit set to 2014-08-01, 10:00:00
2014-08-01, 11:00:00 - second Delta - all data from '2014-08-01, 9:30:00' (1800 secs minus from 10:00) and high = '2014-08-01, 11:00:00' - RSA7 limit set to 2014-08-01, 11:00:00
2014-08-01, 12:00:00 - third Delta - all data from '2014-08-01, 10:30:00' (1800 secs minus from 11:00) and high = '2014-08-01, 12:00:00' - RSA7 limit set to 2014-08-01, 12:00:00
3.consider 2014-08-01,09:00:00 i have initialised the delta, while initialising i mentioned 1800 seconds as upper limit, 120 seconds as lower limit now i have triggered the infopack. what will happen? after delta load what will be the pointer in RSA7. take 3 successive time intervals and explain, when we go for this setting.(can i do like this-----if yes then what cases,------if no ,why cant ).
--->Lower limit set to 120. Upper Limit set to 1800
2014-08-01, 09:00:00 - initialization - All previous data till 2014-08-01 will be extracted - RSA7 limit set to 2014-08-01, 09:00:00
2014-08-01, 10:00:00 - first Delta - all data from '2014-08-01, 08:57:00' (120 secs minus from 09:00) and high = '2014-08-01, 10:30:00' (1800 secs minus from 10:00) - RSA7 limit set to 2014-08-01, 10:30:00
2014-08-01, 11:00:00 - second Delta - all data from '2014-08-01, 9:57:00' (120 secs minus from 10:00) and high = '2014-08-01, 11:30:00' (1800 secs minus from 11:00) - RSA7 limit set to 2014-08-01, 11:30:00
2014-08-01, 12:00:00 - third Delta - all data from '2014-08-01, 10:57:00' (120 secs minus from 11:00) and high = '2014-08-01, 12:30:00' (1800 secs minus from 11:00) - RSA7 limit set to 2014-08-01, 12:30:00
----------------------------------------------------------------------------------------------------------
For numeric pointer exactly same as above applies.
Please test and find out more and let me know if there are any doubts
.
Thanks
Amit
