add --eof-in-length option

src/greedy_packer.rs

@@ -21,7 +21,7 @@ pub fn pack(
         if let Some(m) = match_finder.matches(pos).next() {
             let max_offset = 1 << (m.length * 3 - 1).min(31);
             let offset = pos - m.pos;
-            if offset < max_offset {
+            if offset < max_offset && m.length >= config.min_length() {
                 lz::Op::Match {
                     offset: offset as u32,
                     len: m.length as u32,
@@ -40,7 +40,7 @@ pub fn pack(
                     .zip(data[(pos - offset)..].iter())
                     .take_while(|(a, b)| a == b)
                     .count();
-                if length > 0 {
+                if length >= config.min_length() {
                     lz::Op::Match {
                         offset: offset as u32,
                         len: length as u32,

src/lib.rs

@@ -22,6 +22,7 @@ pub struct Config {
     pub simplified_prob_update: bool,
 
     pub no_repeated_offsets: bool,
+    pub eof_in_length: bool,
 }
 
 impl Default for Config {
@@ -39,6 +40,17 @@ impl Default for Config {
             simplified_prob_update: false,
 
             no_repeated_offsets: false,
+            eof_in_length: false,
+        }
+    }
+}
+
+impl Config {
+    pub fn min_length(&self) -> usize {
+        if self.eof_in_length {
+            2
+        } else {
+            1
         }
     }
 }

src/lz.rs

@@ -40,11 +40,12 @@ impl Op {
                         coder,
                         state,
                         256 * state.parity_contexts + 1,
-                        offset + 1,
+                        offset + if config.eof_in_length { 0 } else { 1 },
                         config,
                     );
                     state.last_offset = offset;
                 }
+                assert!(!config.eof_in_length || len > 1);
                 encode_length(coder, state, 256 * state.parity_contexts + 65, len, config);
                 state.prev_was_match = true;
                 state.pos += len as usize;
@@ -60,15 +61,20 @@ pub fn encode_eof(coder: &mut dyn EntropyCoder, state: &mut CoderState, config:
         state.pos % state.parity_contexts * 256,
         config.is_match_bit,
     );
-    if !state.prev_was_match {
+    if !state.prev_was_match && !config.no_repeated_offsets {
         encode_bit(
             coder,
             state,
             256 * state.parity_contexts,
-            config.new_offset_bit,
+            config.new_offset_bit ^ config.eof_in_length,
         );
     }
-    encode_length(coder, state, 256 * state.parity_contexts + 1, 1, config);
+    if !config.eof_in_length || config.no_repeated_offsets {
+        encode_length(coder, state, 256 * state.parity_contexts + 1, 1, config);
+    }
+    if config.eof_in_length {
+        encode_length(coder, state, 256 * state.parity_contexts + 65, 1, config);
+    }
 }
 
 fn encode_bit(
@@ -183,7 +189,7 @@ pub fn unpack_internal(
                     &mut contexts,
                     256 * config.parity_contexts + 1,
                     &config,
-                ) - 1;
+                ) - if config.eof_in_length { 0 } else { 1 };
                 if offset == 0 {
                     break;
                 }
@@ -194,6 +200,9 @@ pub fn unpack_internal(
                 256 * config.parity_contexts + 65,
                 &config,
             );
+            if config.eof_in_length && length == 1 {
+                break;
+            }
             if let Some(ref mut result) = result {
                 for _ in 0..length {
                     result.push(result[result.len() - offset]);

src/main.rs

@@ -30,12 +30,14 @@ fn main() -> Result<()> {
                 config.bitstream_is_big_endian = true;
             }
             Long("no-repeated-offsets") => config.no_repeated_offsets = true,
+            Long("eof-in-length") => config.eof_in_length = true,
 
             Long("z80") => {
                 config.use_bitstream = true;
                 config.bitstream_is_big_endian = true;
                 config.invert_bit_encoding = true;
                 config.simplified_prob_update = true;
+                level = 9;
             }
             Long("x86") => {
                 config.use_bitstream = true;

src/parsing_packer.rs

@@ -110,6 +110,9 @@ fn parse(
         max_arrivals: usize,
         config: &crate::Config,
     ) {
+        if length < config.min_length() {
+            return;
+        }
         cost_counter.reset();
         let mut state = arrival.state.clone();
         let op = lz::Op::Match {