AlignTestCasesSection#
Align *** Test Cases ***
section to columns.
Enabling the transformer
AlignTestCasesSection is not included in default transformers, that’s why you need to call it with --transform
explicitly:
robotidy --transform AlignTestCasesSection src
Or configure enabled
parameter:
robotidy --configure AlignTestCasesSection:enabled=True
Align keyword calls and settings into columns with predefined width in non-templated test cases.
There are two possible alignment types (configurable via alignment_type
):
fixed
(default): pad the tokens to the fixed width of the columnauto
: pad the tokens to the width of the longest token in the column
The width of the column sets limit to the maximum width of the column. Default width is 24
(see Widths for information how to configure it).
With fixed
alignment each column have fixed width (and tokens that does not fit
go into overflow
state - see Overflow).
auto
alignment align tokens to the longest token in the column - the column width can be shorter than
configured width (but no longer).
See examples of the alignment types:
*** Test Cases ***
Test case
${short} Short Keyword short arg
${other_val} Short Keyword
... arg
... value
*** Test Cases ***
Test case
${short} Short Keyword short arg
${other_val} Short Keyword
... arg
... value
*** Test Cases ***
Test case
${short} Short Keyword short arg
${other_val} Short Keyword
... arg
... value
The auto
alignment often leads to more compact code. But fixed
setting offers more stability - adding new,
slightly longer variable or keyword call will not change alignment of the other lines.
Widths#
The column width is configurable via widths
parameter. The default value is 24
.
It’s possible to configure width of the several columns (using comma separated list of integers):
robotidy -c AlignKeywordsSection:widths=20
robotidy -c AlignKeywordsSection:widths=10,10,24,30
The last width will be used for the remaining columns. In previous example we configured widths for the 4 columns.
The last width (30
) will be used for 5th, 6th.. and following columns.
Use width 0
to disable column width limit. In auto
alignment type it will always align whole column to the
longest token (no matter how long the token is).
Overflow#
Tokens that do not fit in the column go into overflow
state. There are several ways to deal with them (configurable
via handle_too_long
parameter):
overflow
(default): align token to the next columncompact_overflow
: try to fit next token between current (overflowed) token and the next columnignore_rest
: ignore remaining tokens in the lineignore_line
: ignore whole line
See example (for fixed
alignment type and default width 24
):
*** Test Cases ***
Test case
# keyword call Looo.. does not fit default column width (24)
${assign} Looooooooonger Keyword Name ${argument} last
Short Short Short Short
Single
Multi ${arg}
... ${arg}
*** Test Cases ***
Test case
# we are "overflowing" to the next column - taking 24 * 2 = 48 width
${assign} Looooooooonger Keyword Name ${argument} Short
Short Short Short Short
Single
Multi ${arg}
... ${arg}
*** Test Cases ***
Test case
# ${argument} is fit between columns, and next argument ("last") is aligned correctly
${assign} Looooooooonger Keyword Name ${argument} last
Short Short Short Short
Single
Multi ${arg}
... ${arg}
*** Test Cases ***
Test case
# tokens after too long token are not aligned
${assign} Looooooooonger Keyword Name ${argument} Short
Short Short Short Short
Single
Multi ${arg}
... ${arg}
*** Test Cases ***
Test case
# the wole line containing too long token is ignored
${assign} Looooooooonger Keyword Name ${argument} Short
Short Short Short Short
Single
Multi ${arg}
... ${arg}
Compact overflow#
Compact overflow tries to fit too long tokens between the alignment columns.
This behaviour is controlled with compact_overflow_limit = 2
parameter. If more than configured limit columns are
misaligned in a row, the overflow
mode is used instead of compact_overflow
.
This behaviour helps in avoiding the situation, where compact_overflow
misalign whole line if most of the tokens
does not fit in the column.
Below example was run with config:
robotidy -c AlignKeywordsSection:enabled=True:handle_too_long=compact_overflow:widths=24,28,20 src.robot
*** Test Cases ***
Test case
# compact overflow will be used as we only "misalign" two columns in a row
LäVa_VastaanottotapahtumatTarkista ${VASTAANOTTO} ${TILAUS} ${OSTOTILAUS} ${LÄHETYS} ${TILAUSPVM}
# more than two columns are misaligned - using "overflow" instead
LäVa_VastaanottotapahtumatTarkista ${VASTAANOTTO_LONGER} ${TILAUS_OCC} ${OSTOTILAUS} ${LÄHETYS} ${TILAUSPVM}
*** Test Cases ***
Test case
# compact overflow will be used as we only "misalign" two columns in a row
LäVa_VastaanottotapahtumatTarkista ${VASTAANOTTO} ${TILAUS} ${OSTOTILAUS} ${LÄHETYS} ${TILAUSPVM}
# more than two columns are misaligned - using "overflow" instead
LäVa_VastaanottotapahtumatTarkista ${VASTAANOTTO_LONGER} ${TILAUS_OCC} ${OSTOTILAUS} ${LÄHETYS} ${TILAUSPVM}
Alignment of the indented blocks#
Indented blocks (FOR
, IF
, WHILE
, TRY..EXCEPT..
) are aligned independently.
*** Test Cases ***
Test case
${assign} Keyword
FOR ${var} IN 1 2 3
${variable} Keyword ${var}
Another Keyword
FOR ${var2} IN 1 2 3
Short 1 2
${assign} Longer Keyword
... ${multiline} ${arg}
END
END
Keyword Call ${value} # aligned together with keyword call before FOR loop
*** Test Cases ***
Test case
${assign} Keyword
FOR ${var} IN 1 2 3
${variable} Keyword ${var}
Another Keyword
FOR ${var2} IN 1 2 3
Short 1 2
${assign} Longer Keyword
... ${multiline} ${arg}
END
END
Keyword Call ${value} # aligned together with keyword call before FOR loop
Currently, inline IFs are ignored. Block headers (FOR ${var} IN @{LIST}
or IF $condition
) are not aligned.
Split too long lines#
AlignTestCasesSection
will split the lines if the lines after the alignment would exceed the limits set
in the SplitTooLongLine transformer.
Note
Currently, only --configure SplitTooLongLine:split_on_every_arg=True
mode is supported.
Using this configuration (SplitTooLongLine
is enabled by default):
robotidy -c AlignTestCasesSection:enabled=True:widths=14,24 --line-length 80 src
will result in the following transformation:
*** Test Cases ***
Test case
# fits now but it will not fit after the alignment
Keyword argument1 argument2 argument3 argument4
# does not fit before the alignment
Longer Keyword Name That Could Happen argument value with sentence that goes over
# fits, will not be split
Keyword argument
*** Test Cases ***
Test case
# fits now but it will not fit after the alignment
Keyword
... argument1
... argument2
... argument3
... argument4
# does not fit before the alignment
Longer Keyword Name That Could Happen
... argument value with sentence that goes over
# fits, will be aligned but not split
Keyword argument
Skip formatting#
It is possible to use the following arguments to skip formatting of the code:
It is highly recommended to use one of the skip
options if you wish to use the alignment but you have part of the code
that looks better with manual alignment. It is also possible to use disablers (Disablers) but skip
option
makes it easier to skip all instances of given type of the code.