I tried to make it customizable, understandable and self documenting
rather than quick and dirty.
I am sure that you will be able to wrap it to nice script (in bash):
text="A,SDMMC0_CK,I/O,1B,QSPI0_SCK,O,1F,D0,I/O,2A,SDMMC0_DAT3,I/O,1B,QS
PI0_IO3,I/O,1F,D5,I/O,2A,SDMMC0_DAT4,I/O,1B,QSPI1_SCK,O,1D,TIOA5,I/O,1E
,FLEXCOM2_IO0,I/O,1F,D6,I/O,2"function row2col () {cat - | \ awk -v
FS=, ' function initRow ( col, colName, i) { # initialize
rowArr for (col=1;col<=colCount;++col)
{ colName=colArr[col]; for (i=0;i<fieldsPerCol;++i)
{ if (i==0) { rowArr[colName]=""; } else
{ rowArr[colName]=","
rowArr[colName]; } } } } function
printHeader ( col, i) { # prints table header for
(col=1;col<=colCount;++col) { for (i=1;i<=fieldsPerCol;++i)
{ if (col==1 && i==1) { printf("%s%d", colArr[col],
i); } else { printf(",%s%d", colArr[col],
i); } } } print ""; } function printRow
( col, colName) { # prints rowArr in order stored in
colArr for (col=1;col<=colCount;++col)
{ colName=colArr[col]; if (col==1)
{ printf("%s",rowArr[colName]); } else { if
(col==colCount)
{ printf(",%s\n",rowArr[colName]); } else
{ printf(",%s",rowArr[colName]); } } }
} BEGIN { # list columns and their
order colLst="A,B,C,D,E,F"; # How many fields per column A B
...) fieldsPerCol=3; # this is useful for easy, in order
iterations over
colLst colCount=split(colLst,colArr,","); initRow(); }
$1 == "A" { # A always starts new row if (rowCount != 0)
{ # print output row printRow(); initRow(); }
else { printHeader(); } ++rowCount; } $0 !~ /^[
\t]*$/ { # collect row data from i+1 cols (first col is a b
...) colName=$1 rowArr[colName]=$2; for
(i=2;i<=fieldsPerCol;++i) { rowArr[colName]=rowArr[colName] ","
$(i+1); } } END { # print last
row printRow(); }'}
echo "$text" | row2col
A1,A2,A3,B1,B2,B3,C1,C2,C3,D1,D2,D3,E1,E2,E3,F1,F2,F3SDMMC0_CK,I/O,1,QS
PI0_SCK,O,1,,,,,,,,,,D0,I/O,2SDMMC0_DAT3,I/O,1,QSPI0_IO3,I/O,1,,,,,,,,,
,D5,I/O,2SDMMC0_DAT4,I/O,1,QSPI1_SCK,O,1,,,,TIOA5,I/O,1,FLEXCOM2_IO0,I/
O,1,D6,I/O,2
you could also:echo "$text" | row2col | column -s , -t
A1 A2 A3 B1 B2 B3 C1 C2 C3 D1 D2 D3
E1 E2 E3 F1 F2 F3
SDMMC0_CK I/O 1 QSPI0_SCK O 1
D0 I/O 2
SDMMC0_DAT3 I/O 1 QSPI0_IO3 I/O 1
D5 I/O 2
SDMMC0_DAT4 I/O 1 QSPI1_SCK O 1 TIOA5 I/O 1
FLEXCOM2_IO0 I/O 1 D6 I/O 2
Have fun,Tomas
> import it into a spreadsheet. The data looks like this:
> A,SDMMC0_CK,I/O,1
> B,QSPI0_SCK,O,1
> F,D0,I/O,2
> A,SDMMC0_DAT3,I/O,1
> B,QSPI0_IO3,I/O,1
> F,D5,I/O,2
> A,SDMMC0_DAT4,I/O,1
> B,QSPI1_SCK,O,1
> D,TIOA5,I/O,1
> E,FLEXCOM2_IO0,I/O,1
> F,D6,I/O,2
>
> This data represents the available functionality of three GPIO pins.
> Think pin mux'ing. Each leading letter represents a possible
> function
> for a pin. In the above example, the first pin has functions A, B,
> and
> F. Each pin could have 6 possible functions (A through F), but
> typically has fewer. Thus the missing letters in the
> sequence. Note
> that not every pin has an 'A' function. The goal is to put the
> function
> info in the appropriate columns based on the leading letter. There
> should be one line per pin. For the above data, the end result
> should
> look something like this:
>
> SDMMC0_CK,I/O,1,QSPI0_SCK,O,1,,,,,,,,,,D0,I/O,2
> SDMMC0_DAT3,I/O,1,QSPI0_IO3,I/O,1,,,,,,,,,,D5,I/O,2
> SDMMC0_DAT4,I/O,1,QSPI1_SCK,O,1,TIOA5,I/O,1,FLEXCOM2_IO0,I/O,1,D6,I/O
> ,2
>
> I did this by hand, so the comma count may be off, but hopefully it
> gets
> the point across. Note that any of the functions A-F may be
> missing,
> but there will always be at least one function per pin.
>
> Suggestions on how to best accomplish this?
>
> If you're really bored, here's the datasheet that I'm scraping. Look
> at
> Table 6-2 beginning on page 13. I've already got most of the data
> from
> this table through outside channels, but I'm missing the data from
> the
> PIO Peripheral Signal column.
>
> <http://ww1.microchip.com/downloads/en/DeviceDoc/SAMA5D2-Plus-DDR2-LP
> DDR2-System-in-Package-(SIP)-60001484b.pdf>
>
>
> thanks,
> galen
_______________________________________________
PLUG mailing list
[email protected]
http://lists.pdxlinux.org/mailman/listinfo/plug
