Creating .csv files for import into Assembler Outputs Setup

Dear G203,

G203 said:

but Cadence doesn't read my .CSV file if it contains the expressions for gmax and gp, or virtually anything that contains a double quote ("). I eventually found out that even though my expressions for gmax and gp are generated correctly in my Excel and exported .CSV file, when I open them with Notepad, a lot of extra double quotes get added, which causes Virtuoso to generate an error (Images attached below).

I understand. The double-quotes are actually part of the standard for a comma-separated variable file. As you have discovered, the file format is not interpreted as per the standard by Assembler. I have experience this same issue. However, a simple bash script will remove the added quotations and preserve this required in your expressions I believe. I just tried it on some csv files generated by Excel using valid Cadence output expressions and it appears to do what I hoped. The script is attached as a txt file in order that I could upload it.

Remove the .txt extension, make it executable and it should run on your UNIX platform. The syntax is:

translate_csv.sh <input_csv_filename> <output_csv_filename>

Here is an example file and the resultant output:

$ cat test_nmos_outputs_test.csv
Name,Type,Output,Plot,Save
,terminal,/id/PLUS,,t
,terminal,/MN0/D,,
,net,/vd,,t
,net,/vdrain,,t
,net,/vgate,,t
,terminal,/M0/D,,t
gm_over_id,expr,"(OS(""/M0"" ""gm"" ""/scratch/noclean/dcd_serdes/sml/simulation/spectre/test_nmos/spectre/schematic"") / i(""/M0/D"" ?result ""dc""))",,
id_over_W,expr,"(i(""/M0/D"" ?result ""dc"") / VAR(""W""))",,t
gm_over_id/Id_over_W,expr,waveVsWave(?x id_over_W ?y gm_over_id),t,
$
$ translate_csv.sh test_nmos_outputs_test.csv assembler_test_nmos_outputs_test.csv
Done. Output filename is "assembler_test_nmos_outputs_test.csv".
$ cat assembler_test_nmos_outputs_test.csv
Name,Type,Output,Plot,Save
,terminal,/id/PLUS,,t
,terminal,/MN0/D,,
,net,/vd,,t
,net,/vdrain,,t
,net,/vgate,,t
,terminal,/M0/D,,t
gm_over_id,expr,(OS("/M0" "gm" "/scratch/noclean/dcd_serdes/sml/simulation/spectre/test_nmos/spectre/schematic") / i("/M0/D" ?result "dc")),,
id_over_W,expr,(i("/M0/D" ?result "dc") / VAR("W")),,t
gm_over_id/Id_over_W,expr,waveVsWave(?x id_over_W ?y gm_over_id),t,
$
$

I hope this helps and I understood your question G203!

Shawn

#!/bin/bash

if [ $# -ne 2 ]; then
   echo "Usage: translate_csv_file <input_csv_filename> <output_csv_filename>"
   exit 127
fi

cat $1 | tr -s '""' '"' | sed '1,$s/\,"/\,/g' | sed '1,$s/"\,/\,/g' | sed '1,$s/"$//g' > $2

echo "Done. Output filename is \"$2\"."

Thank you very much, this helped! ShawnLogan

Dear G203,

I am very happy to read that you have made some good progress - great! Thank you for letting us know!

Shawn