From 183a16d953328a8d83f9bdfc241b27c8c619d36f Mon Sep 17 00:00:00 2001
From: Marcus Holland-Moritz <github@mhxnet.de>
Date: Wed, 3 Aug 2022 18:59:47 +0200
Subject: [PATCH] fsst: deterministic symbol tables (needed to fix #91)

This fixes what I believe is a bug in the fsst library that causes
symbol tables to be non-deterministic. There's an open issue/PR for
the library, so it's not yet clear if this fix is correct/optimal.
---
 fsst/libfsst.cpp | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/fsst/libfsst.cpp b/fsst/libfsst.cpp
index d35dad3e..069d9cca 100644
--- a/fsst/libfsst.cpp
+++ b/fsst/libfsst.cpp
@@ -111,6 +111,9 @@ SymbolTable *buildSymbolTable(Counters& counters, vector<u8*> line, size_t len[]
             while (true) {
 	       u8* old = cur;
                counters.count1Inc(pos1);
+               if (cur==end) {
+                  break;
+               }
                // count single symbol (i.e. an option is not extending it)
                if (st->symbols[pos1].length() != 1)
                   counters.count1Inc(*cur);
@@ -130,8 +133,6 @@ SymbolTable *buildSymbolTable(Counters& counters, vector<u8*> line, size_t len[]
                      pos2 = st->byteCodes[word & 0xFF] & FSST_CODE_MASK;
                      cur += 1;
                   }
-               } else if (cur==end) { 
-                  break;
                } else {
                   assert(cur<end);
                   pos2 = st->findLongestSymbol(cur, end);