Password from stdin

[EvilBit]

fixes #86

At the moment, this forbids \0 in input stream, but allows \n, etc. Not sure if we want to restrict --password to single-line, printable ASCII only, or arbitrary binary secrets (and what about \0 in that case).

[codecov]

Codecov Report

Merging #89 (c9cbceb) into master (d4b8937) will increase coverage by 0.08%.
The diff coverage is 61.90%.

❗ Current head c9cbceb differs from pull request most recent head 2d29a8a. Consider uploading reports for the commit 2d29a8a to get more accurate results

@@            Coverage Diff             @@
##           master      #89      +/-   ##
==========================================
+ Coverage   81.67%   81.75%   +0.08%     
==========================================
  Files           4        4              
  Lines         802      822      +20     
==========================================
+ Hits          655      672      +17     
- Misses        147      150       +3     

Impacted Files	Coverage Δ
src/tpm2-totp.c	76.57% <61.90%> (-0.53%)	⬇️
src/libtpm2-totp.c	85.74% <0.00%> (+0.71%)	⬆️

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update d4b8937...2d29a8a. Read the comment docs.

[diabonas]

LGTM! Could you rebase to include 8cf5b58 ("test: should not recover secret with incorrect password") into b2500a6 ("feat: read password from stdin"), please? This will make future bisects easier because you don't have to deal with transient CI failures (in this case from scan-build).

[EvilBit]

LGTM! Could you rebase to include 8cf5b58 ("test: should not recover secret with incorrect password") into b2500a6 ("feat: read password from stdin"), please? This will make future bisects easier because you don't have to deal with transient CI failures (in this case from scan-build).

Sure. But I believe you meant c9cbceb fix: close potential memory leak in reading from stdin, not 8cf5b58 test: should not recover secret with incorrect password?

[diabonas]

Sure. But I believe you meant c9cbceb fix: close potential memory leak in reading from stdin, not 8cf5b58 test: should not recover secret with incorrect password?

Oops, you are right, I meant to squash c9cbceb into b2500a6. Also please rebase onto the current master to resolve the conflict after merging #88 :)

[EvilBit]

Also please rebase onto the current master to resolve the conflict after merging #88 :)

Sure ;)
I knew that would eventually be necessary when submitting multiple PRs simultaneously, but I wanted to develop all features independently based on current master, in case one wouldn't be accepted upstream.

If you know of a nicer way to manage this kind of situation, please let me know. :)

[diabonas]

If you know of a nicer way to manage this kind of situation, please let me know. :)

Not really, unfortunately - I usually base on the current master and rebase if necessary as well