From 78dc63669de188cb87d219d30709aa9c366ac366 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Motiejus=20Jak=C5=A1tys?= <motiejus@jakstys.lt>
Date: Tue, 22 Feb 2022 15:04:59 +0200
Subject: [PATCH] add tests for upper limits

---
 README.md    | 34 ++++++++++++++++++++++++----------
 src/user.zig | 20 ++++++++++----------
 2 files changed, 34 insertions(+), 20 deletions(-)

diff --git a/README.md b/README.md
index febe278..f29d11a 100644
--- a/README.md
+++ b/README.md
@@ -130,9 +130,10 @@ Now that we've sketched the implementation of `id(3)`, it's clearer to
 understand which operations need to be fast; in order of importance:
 
 1. lookup gid -> group info (this is on hot path in id) without members.
-2. lookup uid -> user.
-3. lookup groupname -> group.
-4. lookup username -> user.
+2. lookup username -> user's group memberships.
+3. lookup uid -> user.
+4. lookup groupname -> group.
+5. lookup username -> user.
 
 These indices can use perfect hashing like [bdz from cmph][cmph]: a perfect
 hash hashes a list of bytes to a sequential list of integers. Perfect hashing
@@ -292,8 +293,25 @@ Varint is an efficiently encoded integer (packed for small values). Same as
 [protocol buffer varints][varint], except the largest possible value is `u64`.
 They compress integers well.
 
-`groupmembers`, `additional_gids`
----------------------------------
+Group memberships
+-----------------
+
+There are two group memberships at play:
+
+1. given a username, resolve user's group gids (for `initgroups(3)`).
+2. given a group (gid/name), resolve the members' names (e.g. `getgrgid`).
+
+When user's groups are resolved in (1), the additional userdata is not
+requested (there is no way to return it). Therefore, it is reasonable to store
+the user's memberships completely out-of-bound, keyed by the hash of the
+username.
+
+When group's memberships are resolved in (2), the same call also requires other
+group information: gid and group name. Therefore it makes sense to store a
+pointer to the group members in the group information itself. However, the
+memberships are not *always* necessary (see remarks about `id(1)` in this
+document), therefore the memberships will be stored separately, outside of the
+groups section.
 
 `groupmembers` and `additional_gids` store group and user memberships
 respectively: for each group, a list of pointers (offsets) to User records, and
@@ -307,16 +325,12 @@ pseudo-code:
 ```
 const PackedList = struct {
     length: varint,
-    members: []varint
+    members: [length]varint
 }
 const Groupmembers = PackedList;
 const AdditionalGids = PackedList;
 ```
 
-An entry in `members` field points to the offset into a respective `User` or
-`Group` entry (number of bytes relative to the first entry of the type).
-`members` in `PackedList` are sorted the same way as in the input.
-
 A packed list is a list of varints.
 
 Complete file structure
diff --git a/src/user.zig b/src/user.zig
index 079adc4..ec455d4 100644
--- a/src/user.zig
+++ b/src/user.zig
@@ -32,15 +32,15 @@ pub const PackedUser = packed struct {
     }
 
     pub fn realHomeLen(self: *const PackedUser) usize {
-        return self.home_len + 1;
+        return @as(u32, self.home_len) + 1;
     }
 
     pub fn realNameLen(self: *const PackedUser) usize {
-        return self.name_len + 1;
+        return @as(u32, self.name_len) + 1;
     }
 
     pub fn realShellLen(self: *const PackedUser) usize {
-        return self.shell_len_or_idx + 1;
+        return @as(u32, self.shell_len_or_idx) + 1;
     }
 
     pub fn realGecosLen(self: *const PackedUser) usize {
@@ -180,7 +180,7 @@ pub const UserReader = struct {
             if (u.name_is_a_suffix) {
                 const name_start = u.realHomeLen() - u.realNameLen();
                 name = entry.blob[name_start..u.realHomeLen()];
-                pos = u.home_len + 1;
+                pos = u.realHomeLen();
             } else {
                 const name_start = u.realHomeLen();
                 name = entry.blob[name_start .. name_start + u.realNameLen()];
@@ -269,12 +269,12 @@ test "construct PackedUser blob" {
         .home = "/home/service1",
         .shell = "/usr/bin/nologin",
     }, User{
-        .uid = 1002,
-        .gid = 1002,
-        .name = "user-account",
-        .gecos = "",
-        .home = "/a",
-        .shell = "/bin/zsh",
+        .uid = 0,
+        .gid = 4294967295,
+        .name = "n" ** 32,
+        .gecos = "g" ** 255,
+        .home = "h" ** 64,
+        .shell = "s" ** 64,
     } };
     for (users) |user| {
         try writer.appendUser(user);