Sed / Awk : replace the first occurrence of a pattern with the content of another file - awk

Question

With sed, I'm trying to replace the first occurrence of a comment in a script, like :
#ENTRYPOINT_CONTENT
by the content of a second file ($file_content) into another third file (/base.sh).
So, according to many docs, the string should be quite simple, something like :
sed "s|\#ENTRYPOINT_CONTENT|$file_content|" /base.sh
But I always end up with errors like :
sed: -e expression #1, char 23: unterminated s' command
or similar messages, also tried different delimiters, even with Awk instead, but without success, it seems to be fine after escaping the # in the search pattern, but I still can't get the file content as a variable in Sed.
Any ideas, either with Sed or Awk ?
Edit : --------------------
#Sundeep #James Brown:
Don't want to mix the subjects nor to be long :) the response to you clarification request is in bold at the end of this edit, but just to elaborate the context, my case is a Docker entrypoint script in bash (for a base Docker image) called /root/test/base :
#!/usr/bin/env bash
#ENTRYPOINT_CONTENT
if [[ -e "/root/test/custom" ]]; then
printf "\n\n#ENTRYPOINT_CONTENT\n" >> /root/test/custom
# Code from whjm's reply below (actually works but appends shebangs from custom files)
sed -e '0,/^#ENTRYPOINT_CONTENT/!b; /^#ENTRYPOINT_CONTENT/{ r /root/test/custom' -e 'd; }' /root/test/base.sh >> /root/test/base2.sh
mv /root/test/base2.sh /root/test/base.sh
rm -f /root/test/custom
fi
I just want to let users drop another bash script of their own on a specific path (say /root/test/custom), for example :
#!/usr/bin/env bash
echo 'My 2nd bash code'
The first script above (base) should insert the content of the custom file at #ENTRYPOINT_CONTENT position (in the base script itself, without removing this search string), like this :
#!/usr/bin/env bash
echo 'My 2nd bash code'
#ENTRYPOINT_CONTENT
if [[ -e "/root/test/custom" ]]; then
printf "\n\n#ENTRYPOINT_CONTENT\n" >> /root/test/custom
# Code from whjm's reply below
sed -e '0,/^#ENTRYPOINT_CONTENT/!b; /^#ENTRYPOINT_CONTENT/{ r /root/test/custom' -e 'd; }' /root/test/base.sh >> /root/test/base2.sh
mv /root/test/base2.sh /root/test/base.sh
rm -f /root/test/custom
fi
If another user later drops another custom script at the same path, we should have the code of this third custom script appended like this :
#!/usr/bin/env bash
echo 'My 2nd bash code'
echo 'My 3rd bash code'
#ENTRYPOINT_CONTENT
if [[ -e "/root/test/custom" ]]; then
# ... and so on
Regarding the shebangs from custom files, it's not a real issue if they are appended to the base file, the sed code from #whjm works as expected but appends them, while (surprisingly) both awk codes from #James Brown already (and gracefully :) ignore all additional shebangs from custom files (probably because they also start with # as the #ENTRYPOINT_CONTENT search string) but currently partly preprend the code like :
#!/usr/bin/env bash
echo 'My 3rd bash code'
echo 'My 2nd bash code'
#ENTRYPOINT_CONTENT
while I'm trying to get it appended like :
#!/usr/bin/env bash
echo 'My 2nd bash code'
echo 'My 3rd bash code'
#ENTRYPOINT_CONTENT
So, in short, #Sundeep, if you could just give me an updated version of your awk code for this, it would be perfect ! :D (couldn't find a way to invert this...) Thanks a lot.
Code from your previous post :
NR==FNR { b=b (FNR==1?"":ORS) $0; next }
{ r=r (FNR==1?"":ORS) $0 } #ENTRYPOINT_CONTENT
END { sub(/\#[^\n]+/,r,b); print b}

Answer 1

If you're using GNU sed:
[STEP 101] # cat file1
11
xx // replace me
44
xx // don't replace me
55
[STEP 102] # cat file2
22
33
[STEP 103] # sed -e '0,/^xx/!b; /^xx/{ r file2' -e 'd }' file1
11
22
33
44
xx // don't replace me
55
[STEP 104] #