Expand description

Computers, for any content or directory, the first revision or release that contains it.

The algorithm is:

  1. Initialize an array of timestamps (as AtomicI64), one for each node, to the maximum timestamp
  2. For each revision/release (in parallel): a. Get its author date (if none, skip the revision/release) b. traverse all contents and directories contained by that revision/release. For each content/directory, atomically set the timestamp to the current rev/rel’s timestamp if it is lower than the existing one
  3. Write the array

Structs

  • Args 🔒
    Returns a directory of CSV files with header ‘author_date,revrel_SWHID,cntdir_SWHID’ and a row for each of the contents and directories with the earliest revision/release that contains them.

Functions

  • Mark any content reachable from the root revrel as having a first occurrence older or equal to this revision