You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.

43 lines
2.0 KiB

The tools in this directory are a scheme for autogenerating crypto doc support.
The central file is data/crypto_support.json, which is a JSON file that contains
the supported API levels for each crypto algorithm that has ever been supported
by Android. The categories are identified as described in the JCA, along with
a couple extra classes.
Each algorithm has up to three properties:
name - The name of the algorithm. These are currently normalized to
all-uppercase versions, but this produces substandard results in some cases
(eg, PBEWITHHMACSHA versus PBEWithHmacSHA), so it could be improved.
supported_api_levels: The set of Android API levels that this algorithm was/is
supported for. It should be a sequence of zero or more "N" values or "N-M"
ranges separated by commas followed by a "N+" value if the algorithm is
currently supported. For example: "17+", "1-8,22+", "1-10,15,22-24".
deprecated: Whether the algorithm is unsupported at the current API level. This
is equivalent to supported_api_levels.endswith('+'), but is included for
clarity's sake.
Updating the documentation is a three-step process: get the set of supported
algorithms, update the data file, then generate HTML based on the data file.
Getting the set of supported algorithms is done using
src/java/libcore/java/security/ListProviders.java. It's important that this is
run using vogar in mode=activity rather than the default of mode=device, as
the Android frameworks (in particular, Android Keystore) can add additional
support.
The data from ListProviders is fed into update_crypto_support.py, which rewrites
the data file. Any newly-added algorithms will be added to the file's data,
and any newly-removed algorithms will have their supported API levels updated.
The preceding two steps can be performed together by running
run_update_crypto_support.sh.
Finally, format_supported_algorithm_table.py reads the data file and outputs
a collection of HTML tables suitable for copying-and-pasting into the docs
to stdout. Add --javadoc to produce output appropriate for inclusion in
Javadoc.