Using AWK to match the first expression, if it exists find next pattern, then insert - bash

Question

I have this following snippet of file, dbScripts/product.js:
db.ShoppingCart.update(
{ 'identifier': 'COFFEE' },
{
'name': 'Coffee',
'identifier': 'COFFEE',
'category': 'Beverages',
'type': [
{
'name': 'Rage',
'identifier': 'COFFEE_RAGE',
'desc':
'Coffee by Rage'
},
{
'name': 'Sleepy Owl',
'identifier': 'COFFEE_SLEEPY_OWL',
'desc':
'Coffee by Sleepy Owl'
}
]
},
{ upsert: true }
);
I wish to first match {'identifier': 'COFFEE'}, if it exists, then and only then look for next subsequent pattern 'type': [
If both conditions are satisfied, the code should insert similar object for other coffee brand. But if the identifier entered is not coffee, it would create the entire structure for it. [The latter part has been accomplished].
FILE_PATH="./dbScripts/products.js"
# example, COFFEE
echo "Enter product"
read -r PRODUCT_NAME
DESCRIPTION="Coffee by Nescafe"
IDENTIFIER="COFFEE_NESCAFE"
FOUND=$(awk -v ITEM="{ \'identifier\': \'${PRODUCT_NAME}\' }" 'BEGIN {FOUND=0}/ITEM/{++FOUND} END {print FOUND+0}' "${FILE_PATH}")
echo ${FOUND}
if [ ${FOUND} -eq 1 ]; then
SCRIPT_TEMPLATE="./AutomationUtil/object.tpl"
local SCRIPT=$(eval "echo \"$(cat "${SCRIPT_TEMPLATE}")\"")
awk -v text="${SCRIPT}" '1;/'type': | 'types': /{printf text}' "${FILE_PATH}" > "${FILE_PATH}_tmp" && mv "${FILE_PATH}_tmp" "${FILE_PATH}"
fi
I am trying to use awk-variable FOUND as flag to check if the pattern is found or not. Proceed with next pattern if and only if FOUND=1
Object.tpl
{'name': '${PRODUCT_NAME}','identifier': '${PRODUCT_IDENTIFIER}','desc':'${DESCRIPTION}'},
So, if both subsequent patterns are found, the above template will be evaluated, and inserted. If not, the Entire Object Structure from dbScripts/product.js will created at the end of the file, example, user enters Tea instead of Coffee.
In the example below, user enters COFFEE as identifier, which matches our first pattern {'identifier': 'COFFEE'}, then it finds the line, 'type': [, and inserts the evaluated Object.tpl below it.
Expected OUTPUT:
db.ShoppingCart.update(
{ 'identifier': 'COFFEE' },
{
'name': 'Coffee',
'identifier': 'COFFEE',
'category': 'Beverages',
'type': [
{
'name': 'Nescafe',
'identifier': 'COFFEE_NESCAFE',
'desc':
'Coffee by Nesafe'
},
{
'name': 'Rage',
'identifier': 'COFFEE_RAGE',
'desc':
'Coffee by Rage'
},
{
'name': 'Sleepy Owl',
'identifier': 'COFFEE_SLEEPY_OWL',
'desc':
'Coffee by Sleepy Owl'
}
]
},
{ upsert: true }
);
db.ShoppingCart.update(
{ 'identifier': 'NOODLES' },
{
'name': 'Noodles',
'identifier': 'NOODLES',
'category': 'Instant Food',
'type': [
{
'name': 'Ramen',
'identifier': 'NOODELS_RAMEN',
'desc':
'Noodles by Ramen'
}
]
}, {upsert: true}
);
I came up with this theoretically, but can't seem to make it work. I always get FOUND=0.
I am sure, this code is very stupid, but any help is appreciated. Again, both patterns should be found in order to insert the object.
NOTE: The formatting is being handled by prettier, so spacing should be ignored in case of new inserted text
I do not wish to use SED because MacOS doesn't come with GNU-SED.
awk version 20200816

Answer 1

There are a few issues with the current code but since the main question appears to be the wrong value assigned to the (bash) variable FOUND, I'll focus on just this issue ...
The main issue is the testing for the search pattern stored in the awk variable ITEM; one fix:
/ITEM/ {...} # wrong
$0 ~ ITEM {...} # fix
Incorporating this into OP's code, and plugging in some values for INPUT and FILE_PATH we get:
FOUND=$(awk -v ITEM="{ 'identifier' : '${INPUT}' }" '$0 ~ ITEM {++FOUND} END {print FOUND+0}' product.js)
$ echo "${FOUND}"
1
NOTES:
escaped single quotes (defining the awk variable ITEM) are not needed
awk variables have an initial value of 0 so no need for the BEGIN{} block
this solution assumes the input has the same exact white space as defined in the awk variable ITEM
As for the rest of the question ...
OP's 2nd awk script is looking for the strings item: or items: but no such string exists in the sample input, so I wouldn't expect anything to be 'inserted'; can't tell at this point if this is a typo/issue with the proposed awk code or faulty sample input
the sample Object.tpl file shows a one-line element definition but the expected output shows this inserted as multiple lines; this result certainly won't be generated by the proposed (2nd) awk script so not sure if OP is also looking for an awk solution that inserst the line feeds or ... ??
the entire FOUND=$(awk ...); if ... awk ...;fi block of code can probably be replaced with a single awk script
OP's awk version (awk version 20200816) looks a bit odd so not sure what flavor/version of awk this really is or what limitations there may be with said version of awk (eg, does OP's awk version support -i inplace for allowing awk to perform an 'inplace' update of the product.js file?)