I have a power automate script that every hour refreshes a table in C:\SCADA\orders.csv with new orders. The csv file just have a simple table with the list of orders, with it’s date, number, description, qty, etc…
I want to show this table in a perspective view, I been trying with chatgpt to get a script that reads the csv file and put the contents into a dataset tag, but how is normal with chatgpt, the script doesn’t work, and it doesn’t seem to be able to find a solution.
The biggest problem seems to be that ignition can’t just access C:\SCADA at all. The user that is running the service in the computer is a local admin, and the folder is shared with the ignition service, but still reports that there’s no file.
I’m trying to find information about this but I can’t. Is there anything special that needs to be done for the gateway to be able to read local files?
Thank you.
PS: This is the script:
import csv
def getOrders():
path = r"C:\SCADA\orders.csv"
rows = []
headers = []
with open(path, 'r') as f:
reader = csv.reader(f)
for i, row in enumerate(reader):
if i == 0:
headers = row
else:
rows.append(row)
return system.dataset.toDataSet(headers, rows)
Also tried with path = r"C:\Program Files\Inductive Automation\Ignition\data\orders.csv" with no luck.
This function is called by a timer script:
`def handleTimerEvent():
try:
dataset = project.readorders.getOrders()
system.tag.writeBlocking(["[default]orders/OrdersDataset"], [dataset])
except Exception as e:
system.util.getLogger("getorderstimer").error("Timer error: " + str(e))`
Are you calling it from a gateway timer event?
If not, move it there.
Also double the backslashes in the path, and then write the results to a dataset tag in the call itself.
path = r"C:\\SCADA\\orders.csv"
blah blah
system.tag.writeBlocking("[default]orders/OrdersDataset", system.dataset.toDataSet(headers, rows))
This is how it looks right now, as you can see the tag is still null. The double \\ didn’t seem to help. I don’t know if this may be a file access problem more than a code problem.
As a sanity check, see if you can write a file.
It should at least let you know if it’s a permissions issue.
1 Like
tip: csv.reader is iterable, so you can use
headers = next(reader)
rows = list(reader)
using next will consume the first line, so it won't appear in rows.
I usually do something like
headers = next(reader)
data = [dict(zip(headers, row)) for row in reader]
This builds a list of dicts that you can pass directly to a perspective table.
As another sanity check, is the project enabled?
1 Like
The r indicates a raw string and you don’t need to double the slashes
After some debugging, following the tips you all gave me, found some indentation problems from copy-pasting code in the timer. Now the function is properly called, and after checking if I can write (I can) I found the problem was something as stupid as replacing csv.reader(f) for csv.reader(f, delimiter=';')
After some cleanup i’ll leave the script that ChatGPT ended making in case someone finds this post in the future:
# spaces-only indentation (no tabs)
import csv
import io
def getOrders():
logger = system.util.getLogger("readorders")
path = r"C:\SCADA\orders.csv"
encodings = ["utf-8-sig", "utf-8", "cp1252", "latin-1"]
used_encoding = None
headers = []
rows = []
# Try multiple encodings until one succeeds
for enc in encodings:
try:
with io.open(path, 'r', encoding=enc, newline='') as f:
reader = csv.reader(
f,
delimiter='\t', # TAB-delimited
quotechar='"',
escapechar='\\'
)
headers = []
rows = []
for i, row in enumerate(reader):
# Trim whitespace in cells
row = [c.strip() for c in row]
if i == 0:
headers = row
else:
rows.append(row)
used_encoding = enc
break
except UnicodeDecodeError:
# Try next encoding
continue
# Last-resort: decode with replacement to avoid crashes
if used_encoding is None:
with io.open(path, 'r', encoding='latin-1', errors='replace', newline='') as f:
reader = csv.reader(
f,
delimiter='\t',
quotechar='"',
escapechar='\\'
)
for i, row in enumerate(reader):
row = [c.strip() for c in row]
if i == 0:
headers = row
else:
rows.append(row)
used_encoding = "latin-1(replace)"
# Clean headers: keep content inside [...] when present; ensure uniqueness
cleaned = []
seen = set()
for h in headers:
h = h.strip()
if '[' in h and ']' in h:
inside = h[h.find('[') + 1:h.rfind(']')].strip()
if inside:
h = inside
base = h or "Column"
name = base
idx = 1
while name in seen:
idx += 1
name = "%s_%d" % (base, idx)
seen.add(name)
cleaned.append(name)
# Normalize rows to header length
colcount = len(cleaned)
normalized = []
for r in rows:
if len(r) < colcount:
r = r + [''] * (colcount - len(r))
elif len(r) > colcount:
r = r[:colcount]
normalized.append(r)
ds = system.dataset.toDataSet(cleaned, normalized)
logger.info(
"getOrders: path=%s, rows=%d, cols=%d, encoding=%s" %
(path, ds.getRowCount(), ds.getColumnCount(), used_encoding)
)
return ds`
I’m pretty sure this code is unnecessarily cluttered, but for now it just works for my .csv file
