...
| Code Block |
|---|
{id}'s full name: <{ln}, {fn}??withheld>
|
This example includes the following rules:
...
| Code Block |
|---|
{A}
{A.B[2] + C('text')}
{A.B(C)}
|
The angle bracket indicates a format group, where constant text is only included if one or both adjoining references are non-empty. The expression to the right of the ?? (coalesce) operator is used if the format group evaluates to empty. The ?? section is optional.
...
The following examples will try to illustrate the various paths that an evaluation of a formatting group can take.
| Code Block |
|---|
<{A}, {B}>
|
A | B | Result |
|---|---|---|
| ||
1 | 2 | 1, 2 |
1 | 1 | |
2 | 2 |
| Code Block |
|---|
<{A};{B};{C}: Message>
|
A | B | C | Result |
|---|---|---|---|
| |||
1 | 2 | 3 | 1;2;3: Message |
1 | 2 | 1;2 | |
1 | 1 | ||
2 | 3 | 2;3: Message |
| Code Block |
|---|
<{A} items??empty>
|
A | Result |
|---|---|
1 | 1 items |
empty |
| Code Block |
|---|
<{A}, {B}.??<{C}??Nothing defined!>>
|
A | B | C | Result |
|---|---|---|---|
1 | 2 | 3 | 1, 2 |
2 | 3 | 2 | |
1 | 3 | 1 | |
3 | 3 | ||
Nothing defined! |
Advanced Examples
A real-world example with several data fields can take a few more iterations to get them right. Let's take a look below.
...
| Code Block |
|---|
My name is <{LastName}, {FirstName} {MiddleInitial}.> and my phone number is <{MobNumber}??<{TelNumber}??<{FaxNumber}>>>.
|
LastName | FirstName | MiddleInitial | MobNumber | TelNumber | FaxNumber | Result |
|---|---|---|---|---|---|---|
Doe | John | Q | 0765551111 | 0765552222 | 0765553333 | My name is Doe |
Doe | John | Q | 0765552222 | 0765553333 | My name is Doe | |
Doe | John | 0765552222 | 0765553333 | My name is Doe | ||
Doe | 0765553333 | My name is Doe and my phone number is 0765553333. | ||||
Doe | My name is Doe and my phone number is . | |||||
John | My name is John and my phone number is . | |||||
0765553333 | My name is and my phone number is 0765553333. | |||||
My name is and my phone number is . |
We can see from the sample input above that there are several instances where a phrase is included although all the data values are empty. The expression would work much better if the two formatting groups were merged.
...
| Code Block |
|---|
<My name is <{LastName}, {FirstName}, {MiddleInitial}.> and my phone number is <{MobNumber}??<{TelNumber}??<{FaxNumber}>>>.??No data>
|
LastName | FirstName | MiddleInitial | MobNumber | TelNumber | FaxNumber | Result |
|---|---|---|---|---|---|---|
Doe | John | Q | 0765551111 | 0765552222 | 0765553333 | My name is Doe |
Doe | John | Q | 0765552222 | 0765553333 | My name is Doe | |
Doe | John | 0765552222 | 0765553333 | My name is Doe | ||
Doe | 0765553333 | My name is Doe and my phone number is 0765553333. | ||||
Doe | My name is Doe | |||||
John | My name is John | |||||
0765553333 | 0765553333. | |||||
No data |
As you can see, the fallbacks for missing data are now better, printing "No data" when absolutely nothing has been specified.
Writing tests
The following code illustrates how to test an expression using C# and NUnit with various inputs.
The test is written so that you can copy/paste most of it to your own solution and change the number of and the names of the parameters.
...
.
...