g.gisenv: Fix Resource Leak issue in main.c

[ShubhamDesai]

This pull request fixes issue identified by Coverity Scan (CID : 1207681)
Used G_free() to fix this issue.

[ShubhamDesai]

@nilason won't the G_free directly work here?

[lbartoletti]

I think that you should also add a G_free at line 218 before the G_fatal_error

[nilason]

It wouldn't be bad, but introduce unnecessary code. G_fatal_error will print an error message and exit the process. Thus, in-memory will be freed safely and quickly by the system.

[nilason]

@nilason won't the G_free directly work here?

Not quite sure what you mean.

[nilason]

@nilason won't the G_free directly work here?

Not quite sure what you mean.

If you mean G_free(name) directly after u_name = G_store(name);, then yes.

[ShubhamDesai]

@nilason won't the G_free directly work here?

Not quite sure what you mean.

If you mean G_free(name) directly after u_name = G_store(name);, then yes.

I used G_free() directly but still it failed.
why this issue arised? can you please help.

[nilason]

@nilason won't the G_free directly work here?

Not quite sure what you mean.

If you mean G_free(name) directly after u_name = G_store(name);, then yes.

I used G_free() directly but still it failed. why this issue arised? can you please help.

I now see you mean the tests are failing. The reason wasn't obvious to me at first. This may be a good exercise :).
Hint: understand the purpose of the function, and study the function line-by-line, in particular the use of the variable value, therein lies the root of the problem.

[ShubhamDesai]

@nilason won't the G_free directly work here?

Not quite sure what you mean.

If you mean G_free(name) directly after u_name = G_store(name);, then yes.

I used G_free() directly but still it failed. why this issue arised? can you please help.

I now see you mean the tests are failing. The reason wasn't obvious to me at first. This may be a good exercise :). Hint: understand the purpose of the function, and study the function line-by-line, in particular the use of the variable value, therein lies the root of the problem.

The parse_variable function parses a string such as "VAR=value", separating the variable name (VAR) and value (value). Based on your hint I updated the function and tested locally with following testcases
Input with a variable and value:
Input: "VAR=value"
Result: "VAR"
Value: "value"

Input with only a variable (no =):
Input: "VAR"
Result: "VAR"
Value: (null)

Input with an empty value:
Input: "VAR="
Result: "VAR"
Value: (null)

All of them were as expected. It passed on my local windows machine
Could you please review once.