WEBVTT

1
00:00:00.800 --> 00:00:02.180
Understand the model.

2
00:00:02.360 --> 00:00:03.400
Review the change.

3
00:00:03.740 --> 00:00:04.540
Keep the evidence.

4
00:00:07.800 --> 00:00:12.316
A proposed sales change should begin with
the definition you actually have.

5
00:00:12.916 --> 00:00:18.876
Here, Northstar Sales is a fictional
calculation view, opened from an XML

6
00:00:18.886 --> 00:00:19.441
file.

7
00:00:20.041 --> 00:00:25.041
The workbench parses that source and
builds an interactive scenario, with

8
00:00:25.051 --> 00:00:27.425
details and output columns alongside it.

9
00:00:28.025 --> 00:00:30.585
This demonstration uses offline mode.

10
00:00:30.765 --> 00:00:33.845
No database connection or customer data is
involved.

11
00:00:36.067 --> 00:00:40.087
Select a node to see its inputs and the
mappings that feed its output.

12
00:00:40.788 --> 00:00:46.829
For this example, net amount is calculated
from amount using a factor of zero

13
00:00:46.839 --> 00:00:47.510
point nine.

14
00:00:48.110 --> 00:00:51.050
The properties panel exposes the parsed
formula.

15
00:00:51.230 --> 00:00:55.210
These are source definitions, not verified
database results.

16
00:00:55.920 --> 00:01:00.300
Keep that distinction in mind before
accepting any proposed change.

17
00:01:02.333 --> 00:01:05.677
Before changing net amount,
open its column lineage.

18
00:01:06.277 --> 00:01:10.837
The upstream path shows the
recognized dependencies inside

19
00:01:10.847 --> 00:01:11.896
this loaded view.

20
00:01:12.496 --> 00:01:15.516
Show the path in the graph to
connect those relationships

21
00:01:15.526 --> 00:01:16.816
with the model itself.

22
00:01:17.438 --> 00:01:20.698
Highlighted fields and
connections provide context.

23
00:01:21.258 --> 00:01:24.358
External definitions and
unsupported expressions remain

24
00:01:24.368 --> 00:01:25.878
explicit coverage limits.

25
00:01:28.333 --> 00:01:33.293
Generate SQL for the published output to
inspect how the model is translated.

26
00:01:33.926 --> 00:01:38.826
The preview contains a candidate query,
including the fictional source object

27
00:01:38.836 --> 00:01:40.586
and calculated expression.

28
00:01:41.248 --> 00:01:45.048
Copy it, or download the SQL for a
separate review.

29
00:01:45.737 --> 00:01:48.197
The workbench does not execute that query.

30
00:01:48.377 --> 00:01:51.637
You still need independent validation on
your target system.

31
00:01:53.700 --> 00:01:57.960
Analyze View checks the model,
then performs an internal

32
00:01:57.970 --> 00:02:00.320
static scan of its generated
SQL.

33
00:02:01.138 --> 00:02:03.958
This controlled example
contains an identity

34
00:02:03.968 --> 00:02:05.998
projection with an eligible
cleanup.

35
00:02:06.298 --> 00:02:07.849
Read the evidence and its
scope.

36
00:02:08.449 --> 00:02:12.189
Agree to that application fix,
then apply the selected

37
00:02:12.199 --> 00:02:12.889
change.

38
00:02:13.542 --> 00:02:17.522
A finding without an eligible
fix remains a review task.

39
00:02:17.782 --> 00:02:20.242
Severity alone never
authorizes a rewrite.

40
00:02:22.467 --> 00:02:25.667
The result is a separate XML
candidate.

41
00:02:25.677 --> 00:02:28.387
The original source and graph
are preserved.

42
00:02:29.266 --> 00:02:33.946
Inspect the exact changes and
the optimized XML before

43
00:02:33.956 --> 00:02:34.700
downloading it.

44
00:02:35.300 --> 00:02:38.060
You can also generate SQL from
the candidate.

45
00:02:38.320 --> 00:02:42.100
If a selected node was
removed, the app falls back to

46
00:02:42.110 --> 00:02:42.498
Output.

47
00:02:43.098 --> 00:02:46.278
Structural checks do not
establish runtime equivalence

48
00:02:46.288 --> 00:02:47.178
or better performance.

49
00:02:49.433 --> 00:02:55.029
Now compare the original sales definition
with a proposed formula change.

50
00:02:55.629 --> 00:02:59.349
Each side can use an open view or an XML
file.

51
00:02:59.749 --> 00:03:01.709
Here, choose the two fictional files.

52
00:03:02.452 --> 00:03:03.592
Run the comparison.

53
00:03:03.832 --> 00:03:06.592
Yellow highlights mark observed structural
differences.

54
00:03:07.418 --> 00:03:11.678
Switch to XML to inspect the changed
expression directly.

55
00:03:12.018 --> 00:03:15.718
A difference identifies a review point,
not a deployment verdict.

56
00:03:18.433 --> 00:03:21.453
Structured reports keep the
relationships and coverage

57
00:03:21.553 --> 00:03:23.853
readable in worksheet-style
tabs.

58
00:03:24.536 --> 00:03:28.496
Choose a sheet, inspect full
cell values, and filter the

59
00:03:28.506 --> 00:03:29.256
rows you need.

60
00:03:29.862 --> 00:03:33.602
Export Excel includes all
report tabs and rows.

61
00:03:33.612 --> 00:03:36.542
CSV exports the filtered
current sheet.

62
00:03:37.207 --> 00:03:40.847
These are real downloads from
the application, ready for

63
00:03:40.857 --> 00:03:41.767
your review process.

64
00:03:43.967 --> 00:03:48.867
When a workflow needs explanation, Help
Center opens separately from your

65
00:03:48.877 --> 00:03:49.447
working view.

66
00:03:50.047 --> 00:03:52.187
Search by a feature or a question.

67
00:03:52.567 --> 00:03:55.944
These guides match the active offline
application mode.

68
00:03:56.544 --> 00:04:00.444
Connected installations have their own
login and catalog guidance.

69
00:04:00.804 --> 00:04:04.324
This recording makes no claim to
demonstrate those integrations.

70
00:04:06.467 --> 00:04:10.107
Inspect the source, trace the field, and
review the candidate.

71
00:04:10.467 --> 00:04:12.007
Calculation View Workbench.
