How to add a character to the beginning of a line and end of a line in python

I have a data set I get from my IT group. They have an automated extract they are unwilling to change.

The files looks like this (add more lines as requested)

col1#|#col2#|#col3#|#col4

data1#|#data2#|#data3#|#data4

data1#|#data2#|#data3#|#data4

cdata1#|#da#ta2#|#data3#|#data4

(line 4, this is an example where a # in my data screws up the quote character of only #)

The # look like they are attempting to be quote characters. I'd like to use them like that, but for whatever reason they don't include one at the start of each line or the end of each line. The files are of various column counts, so I'm trying to process them to basically add a # to the start and end of each line.

Also since # is often found in my data I'd like to convert the # into ### to make the import to my tool cleaner.

So I'd like

###col1###|###col2###|###col3###|###col4###|###col4###

How could I accomplish this?

current code being used to process csv:

csv_pointer = open(file, encoding=CSV_Encoding, errors=Error_Detection)
csv_reader = csv.reader(
    csv_pointer,
    delimiter=CSV_Seperator,
    quoting=csv.QUOTE_NONE

)
batch = list()
# for each row in csv reader
for row in csv_reader:
    # append the processed row to the batch list
    # processed row meaning we strip down the fields to remove redundant 
data
    # and add Nones if the length of the row is not up to the FIELDS_COUNT
    batch.append([k.strip() for k in row] + [None] * (FIELDS_COUNT - len(row)))
# check if the batch length is greater than ROWS_AT_ONCE
if len(batch) >= ROWS_AT_ONCE:
    # if it is use the executemany over the cursor to insert the data in the batch list to the database
    curr.executemany(insert_func(Table_Name), batch)
    # commit
    conn.commit()
    # set the batch to empty list again
    batch = list()
# if the batch list is not empty
if batch:
# if it is use the executemany over the cursor to insert the data in the batch list to the database
curr.executemany(insert_func(Table_Name), batch)
# commit
conn.commit()
# delete batch (just incase the program message up and it need to delete the batch)
del batch

I attempted to change my delimiter to #|#, which seems like it would fix my problem, but it's returning the error:
TypeError: "delimiter" must be a 1-character string

If I may, why not use the delimiting strategy of the csv format from your IT team?
You can split on "#|#" in parsing tool (if it's in python):

text="col1#|#col2#|#col3#|#col4"
values = text.split("#|#")
# values is ['col1', 'col2', 'col3', 'col4']

Using the csv module, you'll have to specify the quoting parameter. Specifically on line 2:

csv_reader = csv.reader(
    csv_pointer,
    delimiter=CSV_Seperator,
    quoting='#'
)

If that interferes with '#'s in your fields, then you may want to take a literal approach to this problem (without the csv library):

batch = []
with open(file, r) as f:
     for l in f.readlines()[1:]: # if there's a header, if not, then remove the [1:]
         batch.append(l.split("#|#"))

Would something like this work for you?

#Initial text
text="col1#|#col2#|#col3#|#col4"
#adds a ### to start and end
text='###{}###'.format(text)
#Replaces #|# with ###|### 
text=text.replace("#|#","###|###")

This returns:

###col1###|###col2###|###col3###|###col4###

Obviously, this would need to go in a loop of some kind to go through all the data you have, and could also be consolidated into one line but I split it up to try and make it clearer.