Tcl Source Code

Check-in [3b76dae179]
Login
Bounty program for improvements to Tcl and certain Tcl packages.
Tcl 2019 Conference, Houston/TX, US, Nov 4-8
Send your abstracts to [email protected]
or submit via the online form by Sep 9.

Many hyperlinks are disabled.
Use anonymous login to enable hyperlinks.

Overview
Comment:In TclGetByteArrayFromObj(), no longer assume that Tcl_GetByteArrayFromObj() never returns NULL. And don't use \uXXXX construct in "testbytestring" command, since it's only supposed to work with pure bytearrays anyway.
Downloads: Tarball | ZIP archive | SQL archive
Timelines: family | ancestors | descendants | both | trunk
Files: files | file ages | folders
SHA3-256: 3b76dae1794432b5ed21823dce9230268aeeb6e20f95d7eecb88eeed17c90f96
User & Date: jan.nijtmans 2019-02-24 16:59:24
Context
2019-02-24
17:20
Merge 8.7 check-in: fb23ac1e71 user: jan.nijtmans tags: trunk
16:59
In TclGetByteArrayFromObj(), no longer assume that Tcl_GetByteArrayFromObj() never returns NULL. And... check-in: 3b76dae179 user: jan.nijtmans tags: trunk
2019-02-22
18:01
merge 8.7 check-in: adfe04dbbd user: dgp tags: trunk
Changes
Hide Diffs Unified Diffs Ignore Whitespace Patch

Changes to generic/tclInt.h.

4374
4375
4376
4377
4378
4379
4380

4381

4382
4383
4384
4385
4386
4387
4388
4389
4390
4391
4392
4393
4394
4395
4396
4397
4398
4399
4400
4401
4402
4403
   static inline Tcl_UniChar *TclGetUnicodeFromObj(Tcl_Obj *objPtr, size_t *lenPtr) {
      Tcl_UniChar *response = Tcl_GetUnicodeFromObj(objPtr, NULL);
      *(lenPtr) = *((size_t *) (objPtr)->internalRep.twoPtrValue.ptr1);
      return response;
   }
   static inline unsigned char *TclGetByteArrayFromObj(Tcl_Obj *objPtr, size_t *lenPtr) {
      unsigned char *response = Tcl_GetByteArrayFromObj(objPtr, NULL);

      *(lenPtr) = *((size_t *) (objPtr)->internalRep.twoPtrValue.ptr1);

      return response;
   }
#else
#define TclGetStringFromObj(objPtr, lenPtr) \
    (((objPtr)->bytes \
	    ? NULL : Tcl_GetString((objPtr)), \
	    *(lenPtr) = (objPtr)->length, (objPtr)->bytes))
#define TclGetUnicodeFromObj(objPtr, lenPtr) \
    (Tcl_GetUnicodeFromObj(objPtr, NULL), \
	    *(lenPtr) = *((size_t *) (objPtr)->internalRep.twoPtrValue.ptr1), \
	    Tcl_GetUnicodeFromObj(objPtr, NULL))
#define TclGetByteArrayFromObj(objPtr, lenPtr) \
    (Tcl_GetByteArrayFromObj(objPtr, NULL), \
	    *(lenPtr) = *((size_t *) (objPtr)->internalRep.twoPtrValue.ptr1), \
	    (unsigned char *)(((size_t *) (objPtr)->internalRep.twoPtrValue.ptr1) + 2))
#endif

/*
 *----------------------------------------------------------------
 * Macro used by the Tcl core to clean out an object's internal
 * representation. Does not actually reset the rep's bytes. The ANSI C
 * "prototype" for this macro is:






>
|
>












|
|
|







4374
4375
4376
4377
4378
4379
4380
4381
4382
4383
4384
4385
4386
4387
4388
4389
4390
4391
4392
4393
4394
4395
4396
4397
4398
4399
4400
4401
4402
4403
4404
4405
   static inline Tcl_UniChar *TclGetUnicodeFromObj(Tcl_Obj *objPtr, size_t *lenPtr) {
      Tcl_UniChar *response = Tcl_GetUnicodeFromObj(objPtr, NULL);
      *(lenPtr) = *((size_t *) (objPtr)->internalRep.twoPtrValue.ptr1);
      return response;
   }
   static inline unsigned char *TclGetByteArrayFromObj(Tcl_Obj *objPtr, size_t *lenPtr) {
      unsigned char *response = Tcl_GetByteArrayFromObj(objPtr, NULL);
      if (response) {
          *(lenPtr) = *((size_t *) (objPtr)->internalRep.twoPtrValue.ptr1);
      }
      return response;
   }
#else
#define TclGetStringFromObj(objPtr, lenPtr) \
    (((objPtr)->bytes \
	    ? NULL : Tcl_GetString((objPtr)), \
	    *(lenPtr) = (objPtr)->length, (objPtr)->bytes))
#define TclGetUnicodeFromObj(objPtr, lenPtr) \
    (Tcl_GetUnicodeFromObj(objPtr, NULL), \
	    *(lenPtr) = *((size_t *) (objPtr)->internalRep.twoPtrValue.ptr1), \
	    Tcl_GetUnicodeFromObj(objPtr, NULL))
#define TclGetByteArrayFromObj(objPtr, lenPtr) \
    (Tcl_GetByteArrayFromObj(objPtr, NULL) ? \
	(*(lenPtr) = *((size_t *) (objPtr)->internalRep.twoPtrValue.ptr1), \
	(unsigned char *)(((size_t *) (objPtr)->internalRep.twoPtrValue.ptr1) + 2)) : (*(lenPtr) = 0, NULL))
#endif

/*
 *----------------------------------------------------------------
 * Macro used by the Tcl core to clean out an object's internal
 * representation. Does not actually reset the rep's bytes. The ANSI C
 * "prototype" for this macro is:

Changes to tests/utf.test.

104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
test utf-4.1 {Tcl_NumUtfChars: zero length} testnumutfchars {
    testnumutfchars ""
} {0}
test utf-4.2 {Tcl_NumUtfChars: length 1} {testnumutfchars testbytestring} {
    testnumutfchars [testbytestring "\xC2\xA2"]
} {1}
test utf-4.3 {Tcl_NumUtfChars: long string} {testnumutfchars testbytestring} {
    testnumutfchars [testbytestring "abc\xC2\xA2\xe4\xb9\x8e\uA2\u4e4e"]
} {7}
test utf-4.4 {Tcl_NumUtfChars: #u0000} {testnumutfchars testbytestring} {
    testnumutfchars [testbytestring "\xC0\x80"]
} {1}
test utf-4.5 {Tcl_NumUtfChars: zero length, calc len} testnumutfchars {
    testnumutfchars "" 0
} {0}
test utf-4.6 {Tcl_NumUtfChars: length 1, calc len} {testnumutfchars testbytestring} {
    testnumutfchars [testbytestring "\xC2\xA2"] 2
} {1}
test utf-4.7 {Tcl_NumUtfChars: long string, calc len} {testnumutfchars testbytestring} {
    testnumutfchars [testbytestring "abc\xC2\xA2\xe4\xb9\x8e\uA2\u4e4e"] 10
} {7}
test utf-4.8 {Tcl_NumUtfChars: #u0000, calc len} {testnumutfchars testbytestring} {
    testnumutfchars [testbytestring "\xC0\x80"] 2
} {1}
# Bug [2738427]: Tcl_NumUtfChars(...) no overflow check
test utf-4.9 {Tcl_NumUtfChars: #u20AC, calc len, incomplete} {testnumutfchars testbytestring} {
    testnumutfchars [testbytestring "\xE2\x82\xAC"] 2






|











|







104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
test utf-4.1 {Tcl_NumUtfChars: zero length} testnumutfchars {
    testnumutfchars ""
} {0}
test utf-4.2 {Tcl_NumUtfChars: length 1} {testnumutfchars testbytestring} {
    testnumutfchars [testbytestring "\xC2\xA2"]
} {1}
test utf-4.3 {Tcl_NumUtfChars: long string} {testnumutfchars testbytestring} {
    testnumutfchars [testbytestring "abc\xC2\xA2\xE4\xB9\x8E\xA2\x4E"]
} {7}
test utf-4.4 {Tcl_NumUtfChars: #u0000} {testnumutfchars testbytestring} {
    testnumutfchars [testbytestring "\xC0\x80"]
} {1}
test utf-4.5 {Tcl_NumUtfChars: zero length, calc len} testnumutfchars {
    testnumutfchars "" 0
} {0}
test utf-4.6 {Tcl_NumUtfChars: length 1, calc len} {testnumutfchars testbytestring} {
    testnumutfchars [testbytestring "\xC2\xA2"] 2
} {1}
test utf-4.7 {Tcl_NumUtfChars: long string, calc len} {testnumutfchars testbytestring} {
    testnumutfchars [testbytestring "abc\xC2\xA2\xE4\xB9\x8E\xA2\x4E"] 10
} {7}
test utf-4.8 {Tcl_NumUtfChars: #u0000, calc len} {testnumutfchars testbytestring} {
    testnumutfchars [testbytestring "\xC0\x80"] 2
} {1}
# Bug [2738427]: Tcl_NumUtfChars(...) no overflow check
test utf-4.9 {Tcl_NumUtfChars: #u20AC, calc len, incomplete} {testnumutfchars testbytestring} {
    testnumutfchars [testbytestring "\xE2\x82\xAC"] 2

Changes to tests/util.test.

275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
    # pattern += Tcl_UtfToUniChar(pattern, &endChar);
    # get 1 UTF-8 character
    Wrapper_Tcl_StringMatch "a\[a\u4e4fc]c" "a\u4e4fc"
} 1
test util-5.18 {Tcl_StringMatch: UTF-8} testbytestring {
    # pattern += Tcl_UtfToUniChar(pattern, &endChar);
    # proper advance: wrong answer would match on UTF trail byte of \u4e4f
    Wrapper_Tcl_StringMatch {a[a\u4e4fc]c} [testbytestring a\u008fc]
} 0
test util-5.19 {Tcl_StringMatch: UTF-8} {
    # pattern += Tcl_UtfToUniChar(pattern, &endChar);
    # proper advance.
    Wrapper_Tcl_StringMatch {a[a\u4e4fc]c} "acc"
} 1
test util-5.20 {Tcl_StringMatch} {






|







275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
    # pattern += Tcl_UtfToUniChar(pattern, &endChar);
    # get 1 UTF-8 character
    Wrapper_Tcl_StringMatch "a\[a\u4e4fc]c" "a\u4e4fc"
} 1
test util-5.18 {Tcl_StringMatch: UTF-8} testbytestring {
    # pattern += Tcl_UtfToUniChar(pattern, &endChar);
    # proper advance: wrong answer would match on UTF trail byte of \u4e4f
    Wrapper_Tcl_StringMatch {a[a\u4e4fc]c} [testbytestring a\x8Fc]
} 0
test util-5.19 {Tcl_StringMatch: UTF-8} {
    # pattern += Tcl_UtfToUniChar(pattern, &endChar);
    # proper advance.
    Wrapper_Tcl_StringMatch {a[a\u4e4fc]c} "acc"
} 1
test util-5.20 {Tcl_StringMatch} {