Patch #923098: Share interned strings in marshal.

Doc/api/utilities.tex

@@ -283,20 +283,31 @@ data must be opened in binary mode.
 
 Numeric values are stored with the least significant byte first.
 
-\begin{cfuncdesc}{void}{PyMarshal_WriteLongToFile}{long value, FILE *file}
+The module supports two versions of the data format: version 0 is the
+historical version, version 1 (new in Python 2.4) shares interned
+strings in the file, and upon unmarshalling. \var{Py_MARSHAL_VERSION}
+indicates the current file format (currently 1).
+
+\begin{cfuncdesc}{void}{PyMarshal_WriteLongToFile}{long value, FILE *file, int version}
   Marshal a \ctype{long} integer, \var{value}, to \var{file}.  This
   will only write the least-significant 32 bits of \var{value};
   regardless of the size of the native \ctype{long} type.
+
+  \versionchanged[\var{version} indicates the file format]{2.4}
 \end{cfuncdesc}
 
 \begin{cfuncdesc}{void}{PyMarshal_WriteObjectToFile}{PyObject *value,
-                                                     FILE *file}
+                                                     FILE *file, int version}
   Marshal a Python object, \var{value}, to \var{file}.
+
+  \versionchanged[\var{version} indicates the file format]{2.4}
 \end{cfuncdesc}
 
-\begin{cfuncdesc}{PyObject*}{PyMarshal_WriteObjectToString}{PyObject *value}
+\begin{cfuncdesc}{PyObject*}{PyMarshal_WriteObjectToString}{PyObject *value, int version}
   Return a string object containing the marshalled representation of
   \var{value}.
+
+  \versionchanged[\var{version} indicates the file format]{2.4}
 \end{cfuncdesc}
 
 The following functions allow marshalled values to be read back in.

Doc/lib/libmarshal.tex

@@ -73,6 +73,9 @@ The module defines these functions:
   a \exception{ValueError} exception is raised --- but garbage data
   will also be written to the file.  The object will not be properly
   read back by \function{load()}.
+
+  \versionadded[The \var{version} argument indicates the data
+  format that \code{dumps} should use.]{2.4}
 \end{funcdesc}
 
 \begin{funcdesc}{load}{file}
@@ -86,11 +89,14 @@ The module defines these functions:
   \code{None} for the unmarshallable type.}
 \end{funcdesc}
 
-\begin{funcdesc}{dumps}{value}
+\begin{funcdesc}{dumps}{value\optional{, version}}
   Return the string that would be written to a file by
   \code{dump(\var{value}, \var{file})}.  The value must be a supported
   type.  Raise a \exception{ValueError} exception if value has (or
   contains an object that has) an unsupported type.
+
+  \versionadded[The \var{version} argument indicates the data
+  format that \code{dumps} should use.]{2.4}
 \end{funcdesc}
 
 \begin{funcdesc}{loads}{string}
@@ -98,3 +104,13 @@ The module defines these functions:
   \exception{EOFError}, \exception{ValueError} or
   \exception{TypeError}.  Extra characters in the string are ignored.
 \end{funcdesc}
+
+In addition, the following constants are defined:
+
+\begin{datadesc}{version}
+  Indicates the format that the module uses. Version 0 is the
+  historical format, version 1 (added in Python 2.4) shares
+  interned strings. The current version is 1.
+
+  \versionadded{2.4}
+\end{datadesc}
\ No newline at end of file

Include/marshal.h

@@ -7,9 +7,11 @@
 extern "C" {
 #endif
 
-PyAPI_FUNC(void) PyMarshal_WriteLongToFile(long, FILE *);
-PyAPI_FUNC(void) PyMarshal_WriteObjectToFile(PyObject *, FILE *);
-PyAPI_FUNC(PyObject *) PyMarshal_WriteObjectToString(PyObject *);
+#define Py_MARSHAL_VERSION 1
+
+PyAPI_FUNC(void) PyMarshal_WriteLongToFile(long, FILE *, int);
+PyAPI_FUNC(void) PyMarshal_WriteObjectToFile(PyObject *, FILE *, int);
+PyAPI_FUNC(PyObject *) PyMarshal_WriteObjectToString(PyObject *, int);
 
 PyAPI_FUNC(long) PyMarshal_ReadLongFromFile(FILE *);
 PyAPI_FUNC(int) PyMarshal_ReadShortFromFile(FILE *);

Misc/NEWS

@@ -12,6 +12,9 @@ What's New in Python 2.4 alpha 1?
 Core and builtins
 -----------------
 
+- marshal now shares interned strings. This change introduces
+  a new .pyc magic.
+
 - Bug #966623. classes created with type() in an exec(, {}) don't
   have a __module__, but code in typeobject assumed it would always
   be there.

Python/import.c

@@ -26,9 +26,9 @@ extern time_t PyOS_GetLastModificationTime(char *, FILE *);
    a .pyc file in text mode the magic number will be wrong; also, the
    Apple MPW compiler swaps their values, botching string constants.
 
-   Apparently, there was a distinction made between even and odd
-   bytecodes that is related to Unicode.  The details aren't clear,
-   but the magic number has been odd for a long time.
+   The magic numbers must be spaced apart atleast 2 values, as the
+   -U interpeter flag will cause MAGIC+1 being used. They have been
+   odd numbers for some time now.
 
    There were a variety of old schemes for setting the magic number.
    The current working scheme is to increment the previous value by
@@ -47,9 +47,9 @@ extern time_t PyOS_GetLastModificationTime(char *, FILE *);
        Python 2.3a0: 62011
        Python 2.3a0: 62021
        Python 2.3a0: 62011 (!)
-       Python 2.4a0: 62031
+       Python 2.4a0: 62041
 */
-#define MAGIC (62031 | ((long)'\r'<<16) | ((long)'\n'<<24))
+#define MAGIC (62041 | ((long)'\r'<<16) | ((long)'\n'<<24))
 
 /* Magic word as global; note that _PyImport_Init() can change the
    value of this global to accommodate for alterations of how the
@@ -797,10 +797,10 @@ write_compiled_module(PyCodeObject *co, char *cpathname, long mtime)
 				"# can't create %s\n", cpathname);
 		return;
 	}
-	PyMarshal_WriteLongToFile(pyc_magic, fp);
+	PyMarshal_WriteLongToFile(pyc_magic, fp, Py_MARSHAL_VERSION);
 	/* First write a 0 for mtime */
-	PyMarshal_WriteLongToFile(0L, fp);
-	PyMarshal_WriteObjectToFile((PyObject *)co, fp);
+	PyMarshal_WriteLongToFile(0L, fp, Py_MARSHAL_VERSION);
+	PyMarshal_WriteObjectToFile((PyObject *)co, fp, Py_MARSHAL_VERSION);
 	if (fflush(fp) != 0 || ferror(fp)) {
 		if (Py_VerboseFlag)
 			PySys_WriteStderr("# can't write %s\n", cpathname);
@@ -811,7 +811,7 @@ write_compiled_module(PyCodeObject *co, char *cpathname, long mtime)
 	}
 	/* Now write the true mtime */
 	fseek(fp, 4L, 0);
-	PyMarshal_WriteLongToFile(mtime, fp);
+	PyMarshal_WriteLongToFile(mtime, fp, Py_MARSHAL_VERSION);
 	fflush(fp);
 	fclose(fp);
 	if (Py_VerboseFlag)

Python/marshal.c

@@ -27,6 +27,8 @@
 #define TYPE_COMPLEX	'x'
 #define TYPE_LONG	'l'
 #define TYPE_STRING	's'
+#define TYPE_INTERNED	't'
+#define TYPE_STRINGREF	'R'
 #define TYPE_TUPLE	'('
 #define TYPE_LIST	'['
 #define TYPE_DICT	'{'
@@ -42,6 +44,7 @@ typedef struct {
 	PyObject *str;
 	char *ptr;
 	char *end;
+	PyObject *strings; /* dict on marshal, list on unmarshal */
 } WFILE;
 
 #define w_byte(c, p) if (((p)->fp)) putc((c), (p)->fp); \
@@ -189,7 +192,24 @@ w_object(PyObject *v, WFILE *p)
 	}
 #endif
 	else if (PyString_Check(v)) {
-		w_byte(TYPE_STRING, p);
+		if (p->strings && PyString_CHECK_INTERNED(v)) {
+			PyObject *o = PyDict_GetItem(p->strings, v);
+			if (o) {
+				long w = PyInt_AsLong(o);
+				w_byte(TYPE_STRINGREF, p);
+				w_long(w, p);
+				goto exit;
+			}
+			else {
+				o = PyInt_FromLong(PyDict_Size(p->strings));
+				PyDict_SetItem(p->strings, v, o);
+				Py_DECREF(o);
+				w_byte(TYPE_INTERNED, p);
+			}
+		}
+		else {
+			w_byte(TYPE_STRING, p);
+		}
 		n = PyString_GET_SIZE(v);
 		w_long((long)n, p);
 		w_string(PyString_AS_STRING(v), n, p);
@@ -269,28 +289,32 @@ w_object(PyObject *v, WFILE *p)
 		w_byte(TYPE_UNKNOWN, p);
 		p->error = 1;
 	}
-
+   exit:
 	p->depth--;
 }
 
+/* version currently has no effect for writing longs. */
 void
-PyMarshal_WriteLongToFile(long x, FILE *fp)
+PyMarshal_WriteLongToFile(long x, FILE *fp, int version)
 {
 	WFILE wf;
 	wf.fp = fp;
 	wf.error = 0;
 	wf.depth = 0;
+	wf.strings = NULL;
 	w_long(x, &wf);
 }
 
 void
-PyMarshal_WriteObjectToFile(PyObject *x, FILE *fp)
+PyMarshal_WriteObjectToFile(PyObject *x, FILE *fp, int version)
 {
 	WFILE wf;
 	wf.fp = fp;
 	wf.error = 0;
 	wf.depth = 0;
+	wf.strings = (version > 0) ? PyDict_New() : NULL;
 	w_object(x, &wf);
+	Py_XDECREF(wf.strings);
 }
 
 typedef WFILE RFILE; /* Same struct with different invariants */
@@ -491,6 +515,7 @@ r_object(RFILE *p)
 		}
 #endif
 
+	case TYPE_INTERNED:
 	case TYPE_STRING:
 		n = r_long(p);
 		if (n < 0) {
@@ -506,6 +531,16 @@ r_object(RFILE *p)
 					"EOF read where object expected");
 			}
 		}
+		if (type == TYPE_INTERNED) {
+			PyString_InternInPlace(&v);
+			PyList_Append(p->strings, v);
+		}
+		return v;
+
+	case TYPE_STRINGREF:
+		n = r_long(p);
+		v = PyList_GET_ITEM(p->strings, n);
+		Py_INCREF(v);
 		return v;
 
 #ifdef Py_USING_UNICODE
@@ -673,6 +708,7 @@ PyMarshal_ReadShortFromFile(FILE *fp)
 {
 	RFILE rf;
 	rf.fp = fp;
+	rf.strings = NULL;
 	return r_short(&rf);
 }
 
@@ -681,6 +717,7 @@ PyMarshal_ReadLongFromFile(FILE *fp)
 {
 	RFILE rf;
 	rf.fp = fp;
+	rf.strings = NULL;
 	return r_long(&rf);
 }
 
@@ -747,22 +784,30 @@ PyObject *
 PyMarshal_ReadObjectFromFile(FILE *fp)
 {
 	RFILE rf;
+	PyObject *result;
 	rf.fp = fp;
-	return read_object(&rf);
+	rf.strings = PyList_New(0);
+	result = r_object(&rf);
+	Py_DECREF(rf.strings);
+	return result;
 }
 
 PyObject *
 PyMarshal_ReadObjectFromString(char *str, int len)
 {
 	RFILE rf;
+	PyObject *result;
 	rf.fp = NULL;
 	rf.ptr = str;
 	rf.end = str + len;
-	return read_object(&rf);
+	rf.strings = PyList_New(0);
+	result = r_object(&rf);
+	Py_DECREF(rf.strings);
+	return result;
 }
 
 PyObject *
-PyMarshal_WriteObjectToString(PyObject *x) /* wrs_object() */
+PyMarshal_WriteObjectToString(PyObject *x, int version)
 {
 	WFILE wf;
 	wf.fp = NULL;
@@ -773,7 +818,9 @@ PyMarshal_WriteObjectToString(PyObject *x) /* wrs_object() */
 	wf.end = wf.ptr + PyString_Size(wf.str);
 	wf.error = 0;
 	wf.depth = 0;
+	wf.strings = (version > 0) ? PyDict_New() : NULL;
 	w_object(x, &wf);
+	Py_XDECREF(wf.strings);
 	if (wf.str != NULL)
 		_PyString_Resize(&wf.str,
 		    (int) (wf.ptr -
@@ -796,7 +843,8 @@ marshal_dump(PyObject *self, PyObject *args)
 	WFILE wf;
 	PyObject *x;
 	PyObject *f;
-	if (!PyArg_ParseTuple(args, "OO:dump", &x, &f))
+	int version = Py_MARSHAL_VERSION;
+	if (!PyArg_ParseTuple(args, "OO|i:dump", &x, &f, &version))
 		return NULL;
 	if (!PyFile_Check(f)) {
 		PyErr_SetString(PyExc_TypeError,
@@ -808,7 +856,9 @@ marshal_dump(PyObject *self, PyObject *args)
 	wf.ptr = wf.end = NULL;
 	wf.error = 0;
 	wf.depth = 0;
+	wf.strings = (version > 0) ? PyDict_New() : 0;
 	w_object(x, &wf);
+	Py_XDECREF(wf.strings);
 	if (wf.error) {
 		PyErr_SetString(PyExc_ValueError,
 				(wf.error==1)?"unmarshallable object"
@@ -823,7 +873,7 @@ static PyObject *
 marshal_load(PyObject *self, PyObject *args)
 {
 	RFILE rf;
-	PyObject *f;
+	PyObject *f, *result;
 	if (!PyArg_ParseTuple(args, "O:load", &f))
 		return NULL;
 	if (!PyFile_Check(f)) {
@@ -832,16 +882,20 @@ marshal_load(PyObject *self, PyObject *args)
 		return NULL;
 	}
 	rf.fp = PyFile_AsFile(f);
-	return read_object(&rf);
+	rf.strings = PyList_New(0);
+	result = read_object(&rf);
+	Py_DECREF(rf.strings);
+	return result;
 }
 
 static PyObject *
 marshal_dumps(PyObject *self, PyObject *args)
 {
 	PyObject *x;
-	if (!PyArg_ParseTuple(args, "O:dumps", &x))
+	int version = Py_MARSHAL_VERSION;
+	if (!PyArg_ParseTuple(args, "O|i:dumps", &x, version))
 		return NULL;
-	return PyMarshal_WriteObjectToString(x);
+	return PyMarshal_WriteObjectToString(x, version);
 }
 
 static PyObject *
@@ -850,12 +904,16 @@ marshal_loads(PyObject *self, PyObject *args)
 	RFILE rf;
 	char *s;
 	int n;
-	if (!PyArg_ParseTuple(args, "s#:loads", &s, &n))
+	PyObject* result;
+	if (!PyArg_ParseTuple(args, "s#|i:loads", &s, &n))
 		return NULL;
 	rf.fp = NULL;
 	rf.ptr = s;
 	rf.end = s + n;
-	return read_object(&rf);
+	rf.strings = PyList_New(0);
+	result = read_object(&rf);
+	Py_DECREF(rf.strings);
+	return result;
 }
 
 static PyMethodDef marshal_methods[] = {
@@ -869,5 +927,6 @@ static PyMethodDef marshal_methods[] = {
 PyMODINIT_FUNC
 PyMarshal_Init(void)
 {
-	(void) Py_InitModule("marshal", marshal_methods);
+	PyObject *mod = Py_InitModule("marshal", marshal_methods);
+	PyModule_AddIntConstant(mod, "version", Py_MARSHAL_VERSION);
 }