Fix minor incompatibilities in inflate behavior

When presented with multiple valid but concatenated streams, the
old driver returned an empty result once the end of the first
stream was reached, and kept doing so even if fed new data. The
new driver/NIF returned a data_error instead.

zlib:inflateInit/3 has been added to control this behavior, but is
not yet ready for public use.

4 files changed, 84 insertions, 91 deletions

diff --git a/erts/emulator/nifs/common/zlib_nif.c b/erts/emulator/nifs/common/zlib_nif.c
index a1a65e1946..a9c5b05e47 100644
--- a/erts/emulator/nifs/common/zlib_nif.c
+++ b/erts/emulator/nifs/common/zlib_nif.c
@@ -69,11 +69,27 @@ typedef enum {
     ST_DEFLATE = 1,
     ST_INFLATE = 2,
     ST_CLOSED = 3
-} zlib_state;
+} zlib_state_t;
+
+/* Controls what to do when the user attempts to decompress more data after
+ * Z_STREAM_END has been returned:
+ *
+ * - 'cut' wipes all further input and returns empty results until reset by
+ * the user. This is the default behavior, matching that of the old driver.
+ * - 'reset' resets the state without discarding any input, making it possible
+ * to decompress blindly concatenated streams.
+ * - 'error' crashes with a data error. */
+typedef enum {
+    EOS_BEHAVIOR_ERROR = 0,
+    EOS_BEHAVIOR_RESET = 1,
+    EOS_BEHAVIOR_CUT = 2
+} zlib_eos_behavior_t;
 
 typedef struct {
     z_stream s;
-    zlib_state state;
+    zlib_state_t state;
+
+    zlib_eos_behavior_t eos_behavior;
 
     /* These refer to the plaintext CRC, and are only needed for zlib:crc32/1
      * which is deprecated. */
@@ -102,7 +118,6 @@ typedef struct {
 static ERL_NIF_TERM zlib_open(ErlNifEnv *env, int argc, const ERL_NIF_TERM argv[]);
 static ERL_NIF_TERM zlib_close(ErlNifEnv *env, int argc, const ERL_NIF_TERM argv[]);
 static ERL_NIF_TERM zlib_deflateInit(ErlNifEnv *env, int argc, const ERL_NIF_TERM argv[]);
-static ERL_NIF_TERM zlib_deflateInit2(ErlNifEnv *env, int argc, const ERL_NIF_TERM argv[]);
 static ERL_NIF_TERM zlib_deflateSetDictionary(ErlNifEnv *env, int argc, const ERL_NIF_TERM argv[]);
 static ERL_NIF_TERM zlib_deflateReset(ErlNifEnv *env, int argc, const ERL_NIF_TERM argv[]);
 static ERL_NIF_TERM zlib_deflateEnd(ErlNifEnv *env, int argc, const ERL_NIF_TERM argv[]);
@@ -110,7 +125,6 @@ static ERL_NIF_TERM zlib_deflateParams(ErlNifEnv *env, int argc, const ERL_NIF_T
 static ERL_NIF_TERM zlib_deflate(ErlNifEnv *env, int argc, const ERL_NIF_TERM argv[]);
 
 static ERL_NIF_TERM zlib_inflateInit(ErlNifEnv *env, int argc, const ERL_NIF_TERM argv[]);
-static ERL_NIF_TERM zlib_inflateInit2(ErlNifEnv *env, int argc, const ERL_NIF_TERM argv[]);
 static ERL_NIF_TERM zlib_inflateSetDictionary(ErlNifEnv *env, int argc, const ERL_NIF_TERM argv[]);
 static ERL_NIF_TERM zlib_inflateGetDictionary(ErlNifEnv *env, int argc, const ERL_NIF_TERM argv[]);
 static ERL_NIF_TERM zlib_inflateReset(ErlNifEnv *env, int argc, const ERL_NIF_TERM argv[]);
@@ -130,8 +144,7 @@ static ERL_NIF_TERM zlib_enqueue_input(ErlNifEnv *env, int argc, const ERL_NIF_T
 
 static ErlNifFunc nif_funcs[] = {
     /* deflate */
-    {"deflateInit_nif", 2, zlib_deflateInit},
-    {"deflateInit_nif", 6, zlib_deflateInit2},
+    {"deflateInit_nif", 6, zlib_deflateInit},
     {"deflateSetDictionary_nif", 2, zlib_deflateSetDictionary},
     {"deflateReset_nif", 1, zlib_deflateReset},
     {"deflateEnd_nif", 1, zlib_deflateEnd},
@@ -139,8 +152,7 @@ static ErlNifFunc nif_funcs[] = {
     {"deflate_nif", 4, zlib_deflate},
 
     /* inflate */
-    {"inflateInit_nif", 1, zlib_inflateInit},
-    {"inflateInit_nif", 2, zlib_inflateInit2},
+    {"inflateInit_nif", 3, zlib_inflateInit},
     {"inflateSetDictionary_nif", 2, zlib_inflateSetDictionary},
     {"inflateGetDictionary_nif", 1, zlib_inflateGetDictionary},
     {"inflateReset_nif", 1, zlib_inflateReset},
@@ -509,9 +521,11 @@ static ERL_NIF_TERM zlib_open(ErlNifEnv *env, int argc, const ERL_NIF_TERM argv[
     d->s.opaque = d;
     d->s.data_type = Z_BINARY;
 
-    d->state = ST_NONE;
+    d->eos_behavior = EOS_BEHAVIOR_CUT;
     d->eos_seen = 0;
 
+    d->state = ST_NONE;
+
     d->want_output_crc = 0;
     d->want_input_crc = 0;
     d->is_raw_stream = 0;
@@ -551,42 +565,6 @@ static ERL_NIF_TERM zlib_close(ErlNifEnv *env, int argc, const ERL_NIF_TERM argv
 
 static ERL_NIF_TERM zlib_deflateInit(ErlNifEnv *env, int argc, const ERL_NIF_TERM argv[]) {
     zlib_data_t *d;
-    int level, res;
-
-    if(argc != 2 || !get_zlib_data(env, argv[0], &d) ||
-                    !enif_get_int(env, argv[1], &level)) {
-        return enif_make_badarg(env);
-    } else if(!zlib_process_check(env, d)) {
-        return enif_raise_exception(env, am_not_on_controlling_process);
-    } else if(d->state != ST_NONE) {
-        return enif_raise_exception(env, am_already_initialized);
-    }
-
-    res = deflateInit(&d->s, level);
-
-    if(res == Z_OK) {
-        d->state = ST_DEFLATE;
-        d->eos_seen = 0;
-
-        /* FIXME: crc32/1 is documented as returning "the current calculated
-         * checksum," but failed to mention that the old implementation only
-         * calculated it when WindowBits < 0 (See zlib_deflateInit2).
-         *
-         * We could fix this behavior by setting d->want_input_crc to 1 here,
-         * but we've decided to retain this quirk since the performance hit is
-         * quite significant. */
-        d->want_output_crc = 0;
-        d->want_input_crc = 0;
-
-        d->output_crc = crc32(0L, Z_NULL, 0);
-        d->input_crc = crc32(0L, Z_NULL, 0);
-    }
-
-    return zlib_return(env, res);
-}
-
-static ERL_NIF_TERM zlib_deflateInit2(ErlNifEnv *env, int argc, const ERL_NIF_TERM argv[]) {
-    zlib_data_t *d;
     int level, method, windowBits, memLevel, strategy, res;
 
     if(argc != 6 || !get_zlib_data(env, argv[0], &d) 
@@ -741,39 +719,12 @@ static ERL_NIF_TERM zlib_deflate(ErlNifEnv *env, int argc, const ERL_NIF_TERM ar
 
 static ERL_NIF_TERM zlib_inflateInit(ErlNifEnv *env, int argc, const ERL_NIF_TERM argv[]) {
     zlib_data_t *d;
-    int res;
 
-    if(argc != 1 || !get_zlib_data(env, argv[0], &d)) {
-        return enif_make_badarg(env);
-    } else if(!zlib_process_check(env, d)) {
-        return enif_raise_exception(env, am_not_on_controlling_process);
-    } else if(d->state != ST_NONE) {
-        return enif_raise_exception(env, am_already_initialized);
-    }
-
-    res = inflateInit(&d->s);
-
-    if(res == Z_OK) {
-        d->state = ST_INFLATE;
-        d->eos_seen = 0;
-
-        d->want_output_crc = 0;
-        d->want_input_crc = 0;
-        d->is_raw_stream = 0;
-
-        d->output_crc = crc32(0L, Z_NULL, 0);
-        d->input_crc = crc32(0L, Z_NULL, 0);
-    }
-
-    return zlib_return(env, res);
-}
+    int windowBits, eosBehavior, res;
 
-static ERL_NIF_TERM zlib_inflateInit2(ErlNifEnv *env, int argc, const ERL_NIF_TERM argv[]) {
-    zlib_data_t *d;
-    int windowBits, res;
-
-    if(argc != 2 || !get_zlib_data(env, argv[0], &d)
-                 || !enif_get_int(env, argv[1], &windowBits)) {
+    if(argc != 3 || !get_zlib_data(env, argv[0], &d)
+                 || !enif_get_int(env, argv[1], &windowBits)
+                 || !enif_get_int(env, argv[2], &eosBehavior)) {
         return enif_make_badarg(env);
     } else if(!zlib_process_check(env, d)) {
         return enif_raise_exception(env, am_not_on_controlling_process);
@@ -785,6 +736,8 @@ static ERL_NIF_TERM zlib_inflateInit2(ErlNifEnv *env, int argc, const ERL_NIF_TE
 
     if(res == Z_OK) {
         d->state = ST_INFLATE;
+
+        d->eos_behavior = eosBehavior;
         d->eos_seen = 0;
 
         d->is_raw_stream = (windowBits < 0);
@@ -934,6 +887,28 @@ static ERL_NIF_TERM zlib_inflate(ErlNifEnv *env, int argc, const ERL_NIF_TERM ar
         return enif_raise_exception(env, am_not_initialized);
     }
 
+    if(d->eos_seen) {
+        int res;
+
+        switch(d->eos_behavior) {
+        case EOS_BEHAVIOR_ERROR:
+            return zlib_return(env, Z_DATA_ERROR);
+        case EOS_BEHAVIOR_RESET:
+            res = inflateReset(&d->s);
+
+            if(res != Z_OK) {
+                return zlib_return(env, res);
+            }
+
+            d->eos_seen = 0;
+            break;
+        case EOS_BEHAVIOR_CUT:
+            zlib_reset_input(d);
+
+            return enif_make_tuple2(env, am_finished, enif_make_list(env, 0));
+        }
+    }
+
     return zlib_codec(&inflate, env, d, input_chunk_size, output_chunk_size, flush);
 }
 
diff --git a/erts/preloaded/ebin/zlib.beam b/erts/preloaded/ebin/zlib.beam
index 267b5cb0a8..5048bdb846 100644
--- a/erts/preloaded/ebin/zlib.beam
+++ b/erts/preloaded/ebin/zlib.beam
diff --git a/erts/preloaded/src/zlib.erl b/erts/preloaded/src/zlib.erl
index dca5a42779..7bd9234a90 100644
--- a/erts/preloaded/src/zlib.erl
+++ b/erts/preloaded/src/zlib.erl
@@ -23,7 +23,7 @@
 -export([open/0,close/1,deflateInit/1,deflateInit/2,deflateInit/6,
          deflateSetDictionary/2,deflateReset/1,deflateParams/3,
          deflate/2,deflate/3,deflateEnd/1,
-         inflateInit/1,inflateInit/2,
+         inflateInit/1,inflateInit/2,inflateInit/3,
          inflateSetDictionary/2,inflateGetDictionary/1, inflateReset/1,
          inflate/2,inflate/3,inflateEnd/1,
          inflateChunk/2,inflateChunk/1,
@@ -68,6 +68,13 @@
 
 -define(MAX_WBITS, 15).
 
+-define(DEFAULT_MEMLEVEL, 8).
+-define(DEFAULT_WBITS, 15).
+
+-define(EOS_BEHAVIOR_ERROR, 0).
+-define(EOS_BEHAVIOR_RESET, 1).
+-define(EOS_BEHAVIOR_CUT, 2).
+
 %% Chunk sizes are hardcoded on account of them screwing with the
 %% predictability of the system. zlib is incapable of trapping so we need to
 %% ensure that it never operates on any significant amount of data.
@@ -130,10 +137,7 @@ deflateInit(Z) ->
       Z :: zstream(),
       Level :: zlevel().
 deflateInit(Z, Level) ->
-    deflateInit_nif(Z, arg_level(Level)).
-
-deflateInit_nif(_Z, _Level) ->
-    erlang:nif_error(undef).
+    deflateInit(Z, Level, deflated, ?DEFAULT_WBITS, ?DEFAULT_MEMLEVEL, default).
 
 -spec deflateInit(Z, Level, Method, WindowBits, MemLevel, Strategy) -> 'ok' when
       Z :: zstream(),
@@ -225,16 +229,21 @@ deflateEnd_nif(_Z) ->
 -spec inflateInit(Z) -> 'ok' when
       Z :: zstream().
 inflateInit(Z) ->
-    inflateInit_nif(Z).
-inflateInit_nif(_Z) ->
-    erlang:nif_error(undef).
+    inflateInit(Z, ?DEFAULT_WBITS).
 
 -spec inflateInit(Z, WindowBits) -> 'ok' when
       Z :: zstream(),
       WindowBits :: zwindowbits().
 inflateInit(Z, WindowBits) ->
-    inflateInit_nif(Z, WindowBits).
-inflateInit_nif(_Z, _WindowBits) ->
+    inflateInit(Z, WindowBits, cut).
+
+-spec inflateInit(Z, WindowBits, EoSBehavior) -> 'ok' when
+      Z :: zstream(),
+      WindowBits :: zwindowbits(),
+      EoSBehavior :: error | reset | cut.
+inflateInit(Z, WindowBits, EoSBehavior) ->
+    inflateInit_nif(Z, arg_bitsz(WindowBits), arg_eos_behavior(EoSBehavior)).
+inflateInit_nif(_Z, _WindowBits, _EoSBehavior) ->
     erlang:nif_error(undef).
 
 -spec inflateSetDictionary(Z, Dictionary) -> 'ok' when
@@ -652,6 +661,11 @@ arg_strategy(_) -> erlang:error(bad_compression_strategy).
 arg_method(deflated) -> ?Z_DEFLATED;
 arg_method(_) -> erlang:error(bad_compression_method).
 
+arg_eos_behavior(error) -> ?EOS_BEHAVIOR_ERROR;
+arg_eos_behavior(reset) -> ?EOS_BEHAVIOR_RESET;
+arg_eos_behavior(cut) -> ?EOS_BEHAVIOR_CUT;
+arg_eos_behavior(_) -> erlang:error(bad_eos_behavior).
+
 -spec arg_bitsz(zwindowbits()) -> zwindowbits().
 arg_bitsz(Bits) when is_integer(Bits) andalso
                      ((8 =< Bits andalso Bits < 48) orelse
diff --git a/lib/kernel/test/zlib_SUITE.erl b/lib/kernel/test/zlib_SUITE.erl
index e246276262..b7e053b137 100644
--- a/lib/kernel/test/zlib_SUITE.erl
+++ b/lib/kernel/test/zlib_SUITE.erl
@@ -276,10 +276,10 @@ api_inflateInit(Config) when is_list(Config) ->
 			  ?m(ok,zlib:close(Z12))
 		  end, lists:seq(8,15)),
     ?m(?EXIT(badarg), zlib:inflateInit(gurka, -15)),
-    ?m(?EXIT(already_initialized), zlib:inflateInit(Z1, 7)),
-    ?m(?EXIT(already_initialized), zlib:inflateInit(Z1, -7)),
-    ?m(?EXIT(already_initialized), zlib:inflateInit(Z1, 48)),
-    ?m(?EXIT(already_initialized), zlib:inflateInit(Z1, -16)),
+    ?m(?EXIT(bad_windowbits), zlib:inflateInit(Z1, 7)),
+    ?m(?EXIT(bad_windowbits), zlib:inflateInit(Z1, -7)),
+    ?m(?EXIT(bad_windowbits), zlib:inflateInit(Z1, 48)),
+    ?m(?EXIT(bad_windowbits), zlib:inflateInit(Z1, -16)),
     ?m(ok, zlib:close(Z1)).
 
 %% Test inflateSetDictionary.
@@ -416,6 +416,9 @@ api_inflateChunk(Config) when is_list(Config) ->
     {more, Part1AsIOList} = zlib:inflateChunk(Z1, Compressed),
     {more, Part2AsIOList} = zlib:inflateChunk(Z1),
     {more, Part3AsIOList} = zlib:inflateChunk(Z1),
+
+    [] = zlib:inflateChunk(Z1),
+    [] = zlib:inflateChunk(Z1),
     [] = zlib:inflateChunk(Z1),
 
     ?m(Part1, iolist_to_binary(Part1AsIOList)),
@@ -483,7 +486,8 @@ api_safeInflate(Config) when is_list(Config) ->
 
     SafeInflateLoop(zlib:safeInflate(Z1, Compressed), []),
 
-    ?m(?EXIT(data_error), zlib:safeInflate(Z1, Compressed)),
+    ?m({finished, []}, zlib:safeInflate(Z1, Compressed)),
+    ?m({finished, []}, zlib:safeInflate(Z1, Compressed)),
 
     ?m(ok, zlib:inflateReset(Z1)),
     ?m(?EXIT(badarg), zlib:safeInflate(gurka, Compressed)),