# 
# patch "ChangeLog"
#  from [63c37b88656196a7dd549876c6f79062f756db93]
#    to [321bfa7b9d52da2e3b3dbe2eff53794d10630cb1]
# 
# patch "file_io.cc"
#  from [a887c87104212bab175b54ec1ceb1fa5153024e5]
#    to [da07ec92b669445496f03005c735d6b41911e422]
# 
# patch "file_io.hh"
#  from [f104a889390ecd8eb031a56a81862cfef3aec01a]
#    to [16f383f0e4eb6fa8454fce1071880f066fc69bdb]
# 
# patch "lua.cc"
#  from [60a450c0fcb35a460fdfe3a6fe4d74b5dd9612e0]
#    to [289aa9474f82e484b42a4db7f6280c4c6d33c088]
# 
# patch "std_hooks.lua"
#  from [d7bd0756c48ace573926197709e53eb24dae5f5f]
#    to [a5d4006c7e8837fcd3e816c309c4fb7bee69393a]
# 
========================================================================
--- ChangeLog	63c37b88656196a7dd549876c6f79062f756db93
+++ ChangeLog	321bfa7b9d52da2e3b3dbe2eff53794d10630cb1
@@ -1,9 +1,3 @@
-2005-08-09  Eric Anderson  <address@hidden>
-
-	* tests/perf-test.sh: A repeatable performance test harness
-	* tests/parse-accounting.pl: A script that parses the accounting
-	output into a nice tabular format
-
 2005-08-31  Matthew Gregan  <address@hidden>
 
 	* paths.cc (test_bookkeeping_path, test_system_path): Second
@@ -730,6 +724,12 @@
 	(external_path): Rename to system_path.
 	Misc. other updates.
 
+2005-08-21  Eric Anderson  <address@hidden>
+	* file_io.cc, file_io.hh, lua.cc, std_hooks.lua: determine if a
+	file is binary by looking at it incrementally, rather than reading
+	it in entirely.  Prepare for making it possible to control what
+	characters are considered "binary"
+
 2005-08-20  Nathaniel Smith  <address@hidden>
 
 	* paths.cc (is_absolute): New function.
@@ -1429,6 +1429,12 @@
 	requirements, zlib as a runtime requirement.
 
 2005-08-09  Eric Anderson  <address@hidden>
+
+	* tests/perf-test.sh: A repeatable performance test harness
+	* tests/parse-accounting.pl: A script that parses the accounting
+	output into a nice tabular format
+
+2005-08-09  Eric Anderson  <address@hidden>
  
 	* Changes to significantly improve network pull performance
 	* string_queue.hh: created to store pending data and allow for
========================================================================
--- file_io.cc	a887c87104212bab175b54ec1ceb1fa5153024e5
+++ file_io.cc	da07ec92b669445496f03005c735d6b41911e422
@@ -1,3 +1,4 @@
+// -*- mode: C++; c-file-style: "gnu"; indent-tabs-mode: nil -*-
 // copyright (C) 2002, 2003 graydon hoare <address@hidden>
 // all rights reserved.
 // licensed to the public under the terms of the GNU GPL (>= 2)
@@ -123,15 +124,44 @@
   return true;
 }
 
-bool guess_binary(string const & s)
+static bool did_char_is_binary_init;
+static bool char_is_binary[256];
+
+void
+set_char_is_binary(char c, bool is_binary)
 {
+    char_is_binary[c] = is_binary;
+}
+
+static void
+init_char_is_binary()
+{
   // these do not occur in ASCII text files
   // FIXME: this heuristic is (a) crap and (b) hardcoded. fix both these.
-  if (s.find_first_of('\x00') != string::npos ||
-      s.find_first_of("\x01\x02\x03\x04\x05\x06\x0e\x0f"
-                      "\x10\x11\x12\x13\x14\x15\x16\x17\x18"
-                      "\x19\x1a\x1c\x1d\x1e\x1f") != string::npos)
-    return true;
+  // Should be calling a lua hook here that can use set_char_is_binary()
+  // That will at least fix (b)
+  string nontext_chars("\x01\x02\x03\x04\x05\x06\x0e\x0f"
+		       "\x10\x11\x12\x13\x14\x15\x16\x17\x18"
+		       "\x19\x1a\x1c\x1d\x1e\x1f");
+  set_char_is_binary('\0',true);
+  for(int i=0;i<nontext_chars.size();++i) 
+    {
+      set_char_is_binary(nontext_chars[i],true);
+    }
+}
+
+bool guess_binary(string const & s)
+{
+  if (did_char_is_binary_init == false) 
+    {
+      init_char_is_binary();
+    }
+
+  for(int i = 0;i<s.size(); ++i) 
+    {
+      if (char_is_binary[s[i]])
+        return true;
+    }
   return false;
 }
 
========================================================================
--- file_io.hh	f104a889390ecd8eb031a56a81862cfef3aec01a
+++ file_io.hh	16f383f0e4eb6fa8454fce1071880f066fc69bdb
@@ -46,6 +46,7 @@
 
 // returns true if the string content is binary according to monotone heuristic
 bool guess_binary(std::string const & s);
+void set_char_is_binary(char c, bool is_binary);
 
 void mkdir_p(any_path const & path);
 void make_dir_for(any_path const & p);
========================================================================
--- lua.cc	60a450c0fcb35a460fdfe3a6fe4d74b5dd9612e0
+++ lua.cc	289aa9474f82e484b42a4db7f6280c4c6d33c088
@@ -1,3 +1,4 @@
+// -*- mode: C++; c-file-style: "gnu"; indent-tabs-mode: nil -*-
 // copyright (C) 2002, 2003 graydon hoare <address@hidden>
 // all rights reserved.
 // licensed to the public under the terms of the GNU GPL (>= 2)
@@ -22,6 +23,7 @@
 
 #include <set>
 #include <map>
+#include <fstream>
 
 #include "app_state.hh"
 #include "file_io.hh"
@@ -518,11 +520,31 @@
   }
 
   static int
-  monotone_guess_binary_for_lua(lua_State *L)
+  monotone_guess_binary_filename_for_lua(lua_State *L)
   {
     const char *path = lua_tostring(L, -1);
     N(path, F("guess_binary called with an invalid parameter"));
-    lua_pushboolean(L, guess_binary(std::string(path, lua_strlen(L, -1))));
+
+    std::ifstream file(path, ios_base::binary);
+    if (!file) 
+      {
+        lua_pushnil(L);
+        return 1;
+      }
+    const int bufsize = 8192;
+    string buf;
+    while(file.good()) 
+      {
+        buf.resize(bufsize);
+        file.read(&buf[0],bufsize);
+        buf.resize(file.gcount());
+        if (guess_binary(buf)) 
+          {
+            lua_pushboolean(L, true);
+            return 1;
+          }
+      }
+    lua_pushboolean(L, false);
     return 1;
   }
   
@@ -600,7 +622,7 @@
   lua_register(st, "wait", monotone_wait_for_lua);
   lua_register(st, "kill", monotone_kill_for_lua);
   lua_register(st, "sleep", monotone_sleep_for_lua);
-  lua_register(st, "guess_binary", monotone_guess_binary_for_lua);
+  lua_register(st, "guess_binary_filename", monotone_guess_binary_filename_for_lua);
   lua_register(st, "include", monotone_include_for_lua);
   lua_register(st, "includedir", monotone_includedir_for_lua);
 }
========================================================================
--- std_hooks.lua	d7bd0756c48ace573926197709e53eb24dae5f5f
+++ std_hooks.lua	a5d4006c7e8837fcd3e816c309c4fb7bee69393a
@@ -131,10 +131,7 @@
    if (string.find(lowname, "%.sql$")) then return false end
    -- unknown - read file and use the guess-binary 
    -- monotone built-in function
-   filedata=read_contents_of_file(name, "rb")
-   if (filedata ~= nil) then return guess_binary(filedata) end
-   -- still unknown (file empty or unreadable) - report it as nil
-   return nil
+   return guess_binary_filename(name)
 end
 
 function edit_comment(basetext, user_log_message)