Example of of Crypto Deterministic Config using https://cloud.google.…

[hilliao]

…com/dlp/docs/pseudonymization#supported-methods to resolve #108

Thank you for opening a Pull Request! Before submitting your PR, there are a few things you can do to make sure it goes smoothly:

• Make sure to open an issue as a bug/issue before writing your code! That way we can discuss the change, evaluate designs, and agree on the general idea
• Ensure the tests and linter pass
• Code coverage does not decrease (if any source code was changed)
• Appropriate docs were updated (if necessary)

Fixes #<108> 🦕

[snippet-bot]

Here is the summary of changes.

You are about to add 2 region tags.

• samples/snippets/deid.py:291, tag dlp_deidentify_deterministic
• samples/snippets/deid.py:469, tag dlp_reidentify_deterministic

---

This comment is generated by snippet-bot.
If you find problems with this result, please file an issue at:
https://github.com/googleapis/repo-automation-bots/issues.
To update this comment, add snippet-bot:force-run label or use the checkbox below:

• Refresh this comment

[tmatsuo]

LGTM, with few small suggestions

[tmatsuo]

In case you don't have access to the build log, here is the lint output:

./deid.py:471:1: E303 too many blank lines (3)
# [START dlp_reidentify_deterministic]
^
./deid.py:472:1: E302 expected 2 blank lines, found 3
def reidentify_with_deterministic(
^

[hilliao]

I fixed the 5 places and fixed parameters in the unit tests. I had to push an amended commit as new commits failed https://www.conventionalcommits.org/en/v1.0.0/ checks. Should new commit's comment be identical to prior commit's comment? What's the rule @tmatsuo ?

[tmatsuo]

@hilliao
The conventional commit check seems good now.

[tmatsuo]

@hilliao Do you have access to the build log?
https://source.cloud.google.com/results/invocations/8de21aa8-ef52-4544-8e1b-93d75a7bac21/log

[hilliao]

I can see the build error but I don't have access to test_deidentify_with_deterministic 's output to see the output of the tokenized harmful string. I believe the error from the build is caused by using my own KMS key instead of projects/python-docs-samples-tests/locations/global/keyRings/dlp-test/cryptoKeys/dlp-test which I don't have access to.
build error:
debug_error_string = "{"created":"@1617263155.197209081","description":"Error received from peer ipv4:74.125.197.95:443","file":"src/core/lib/surface/call.cc","file_line":1067,"grpc_message":"Could not transform all content due to transformation errors. To force this request to succeed anyway (by skipping any transformations that cause errors), settransformation_error_handling.modetoleave_untransformedin the DeidentifyConfig; this will return the partially transformed content along with an error summary.\n
Lack of access:

echo -ne $RAW_KEY | gcloud kms encrypt --key $KEY_NAME --keyring $KEY_RING_NAME --location global --plaintext-file=- --ciphertext-file wrapped.key --project python-docs-samples-tests && cat wrapped.key | base64
ERROR: (gcloud.kms.encrypt) PERMISSION_DENIED: Permission 'cloudkms.cryptoKeyVersions.useToEncrypt' denied on resource 'projects/python-docs-samples-tests/locations/global/keyRings/dlp-test/cryptoKeys/dlp-test' (or it may not exist).

I believe the tokenized string is in the output of test_deidentify_with_deterministic but I can't find its output from the build error. Can you tell me the tokenized string or the whole build output so I can update the code line?

labeled_fpe_string = "My SSN is SSN_TOKEN(36):AUGFopzf7RiKAHWD2Sm5GJWS0Yfezsbc1OU="

[tmatsuo]

@hilliao I got it. I'll try to add a commit.

[tmatsuo]

Re: conventional commits
I also got an error. My guess is our bot can not parse your commit message well. I squashed the commit.

[tmatsuo]

Now the tests are passing. It might be better to get the string dynamically with a pytest fixture.

[tmatsuo]

@hilliao WDYT?

[hilliao]

I would like to merge this pull request first; then I will create a new issue to change the re-identification test method to de-identification then re-identification in the test method. Can @tmatsuo merge?

[tmatsuo]

@hilliao Sounds good

[tmatsuo]

@hilliao Thanks for your contribution!