rrm workspace: TRM/HRM/SRM code, Maze dataset, dynamical-analysis pipelineHEADmain

Curated export for clone-and-run Maze training (2x A6000) + diagnostics.
trm/hrm pretrain.py carry trajectory-augmentation code (backward-compatible).
Heavy artifacts (checkpoints/wandb/npz) gitignored; see PROVENANCE.md.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

1 files changed, 75 insertions, 0 deletions

diff --git a/research/flossing/external/julia-1.6.7/share/julia/stdlib/v1.6/SHA/docs/src/index.md b/research/flossing/external/julia-1.6.7/share/julia/stdlib/v1.6/SHA/docs/src/index.md
new file mode 100644
index 0000000..30c88e5
--- /dev/null
+++ b/research/flossing/external/julia-1.6.7/share/julia/stdlib/v1.6/SHA/docs/src/index.md
@@ -0,0 +1,75 @@
+# SHA
+
+
+Usage is very straightforward:
+```julia
+julia> using SHA
+
+julia> bytes2hex(sha256("test"))
+"9f86d081884c7d659a2feaa0c55ad015a3bf4f1b2b0b822cd15d6c15b0f00a08"
+```
+
+Each exported function (at the time of this writing, SHA-1, SHA-2 224, 256, 384 and 512, and SHA-3 224, 256, 384 and 512 functions are implemented) takes in either an `AbstractVector{UInt8}`, an `AbstractString` or an `IO` object.  This makes it trivial to checksum a file:
+
+```julia
+shell> cat /tmp/test.txt
+test
+julia> using SHA
+
+julia> open("/tmp/test.txt") do f
+           sha2_256(f)
+       end
+32-element Array{UInt8,1}:
+ 0x9f
+ 0x86
+ 0xd0
+ 0x81
+ 0x88
+ 0x4c
+ 0x7d
+ 0x65
+    ⋮
+ 0x5d
+ 0x6c
+ 0x15
+ 0xb0
+ 0xf0
+ 0x0a
+ 0x08
+```
+
+Due to the colloquial usage of `sha256` to refer to `sha2_256`, convenience functions are provided, mapping `shaxxx()` function calls to `sha2_xxx()`.  For SHA-3, no such colloquialisms exist and the user must use the full `sha3_xxx()` names.
+
+`shaxxx()` takes `AbstractString` and array-like objects (`NTuple` and `Array`) with elements of type `UInt8`.
+
+To create a hash from multiple items the `SHAX_XXX_CTX()` types can be used to create a stateful hash object that
+is updated with `update!` and finalized with `digest!`
+
+```julia
+julia> ctx = SHA2_256_CTX()
+SHA2 256-bit hash state
+
+julia> update!(ctx, b"some data")
+0x0000000000000009
+
+julia> update!(ctx, b"some more data")
+0x0000000000000017
+
+julia> digest!(ctx)
+32-element Vector{UInt8}:
+ 0xbe
+ 0xcf
+ 0x23
+ 0xda
+ 0xaf
+ 0x02
+    ⋮
+ 0x25
+ 0x52
+ 0x19
+ 0xa0
+ 0x8b
+ 0xc5
+```
+
+Note that, at the time of this writing, the SHA3 code is not optimized, and as such is roughly an order of magnitude slower than SHA2.