The variable is only defined at initialization, not at run time. If you have a look at the code example from the documentation, you’ll notice that it is used in __init__, not in a callback. I think this is on purpose, although it could be more clear in the documentation.

I’m turning this ticket into a documentation enhancement request. If you would rather it be a request to have the environment variable also available at run time, from callback, please let me know.

Thanks for the quick answer.

I think it could be nice to have it available for the callback running in check mode.
At least it would make sense to me that the ENV variable is set all the time while check.

But from there I can definitely build the check awareness feature that I want from __init__ very easily.

EDIT: After giving a little bit more thought I think I will stay with doc enhancement as in my case it's more a trick in my spider that is forcing another trick!

While I agree that the current behaviour does match the docs, I wonder if this isn't a reasonable use case. From a very quick look, it seems like it could be solved by just including the next code block in the set_environ context manager:

diff --git scrapy/commands/check.py scrapy/commands/check.py
index 9d4437a4..09a76ca7 100644
--- scrapy/commands/check.py
+++ scrapy/commands/check.py
@@ -78,19 +78,19 @@ class Command(ScrapyCommand):
                 elif tested_methods:
                     self.crawler_process.crawl(spidercls)
 
-        # start checks
-        if opts.list:
-            for spider, methods in sorted(contract_reqs.items()):
-                if not methods and not opts.verbose:
-                    continue
-                print(spider)
-                for method in sorted(methods):
-                    print('  * %s' % method)
-        else:
-            start = time.time()
-            self.crawler_process.start()
-            stop = time.time()
-
-            result.printErrors()
-            result.printSummary(start, stop)
-            self.exitcode = int(not result.wasSuccessful())
+            # start checks
+            if opts.list:
+                for spider, methods in sorted(contract_reqs.items()):
+                    if not methods and not opts.verbose:
+                        continue
+                    print(spider)
+                    for method in sorted(methods):
+                        print('  * %s' % method)
+            else:
+                start = time.time()
+                self.crawler_process.start()
+                stop = time.time()
+
+                result.printErrors()
+                result.printSummary(start, stop)
+                self.exitcode = int(not result.wasSuccessful())

Any thoughts?

I have no strong opinion either way.

@Matthijsy Do you remember if you had a reason not to apply the environment variable at run time?

I am not sure why I did it that way. I think it would be fine to include the check run itself in the environment variable block as well.

Is someone working on this? If not I'd like to work on this.

@Gallaecio @Matthijsy
The referenced PR has no diff. Does the PR fix the issue? What am I missing?

The code has an added tab to include it within the context of set_environ(SCRAPY_CHECK='true'):, this makes sure that the environment variable is set during the execution of the code

Got it. My bad.
Thanks for clarifying.

#4298, from @faizan2700, is only missing tests. If @faizan2700 is busy, maybe someone else can work on tests, and send a pull request to the corresponding branch in his fork, or create a new pull request built on top of his changes.

when i run tests using tox command in scrapy directory instead of running tests I am getting exception (ValueError : No closing quotation). I tried to check this online but couldn't solve the problem. Can someone please help me on this (@Gallaecio ) . Thank You