Fix buffer allocations in encoding conversion routines so that they won't

fail on zero-length inputs.  This isn't an issue in normal use because the
conversion infrastructure skips calling the converters for empty strings.
However a problem was created by yesterday's patch to check whether the
right conversion function is supplied in CREATE CONVERSION.  The most
future-proof fix seems to be to make the converters safe for this corner case.

diff --git a/src/backend/utils/mb/conversion_procs/cyrillic_and_mic/cyrillic_and_mic.c b/src/backend/utils/mb/conversion_procs/cyrillic_and_mic/cyrillic_and_mic.c
index 94203bd8ad84179f06cc6251a5605759c8e439d0..ac1ef86a9465ce7070afaa9d21e1582be9c45ef8 100644 (file)
--- a/src/backend/utils/mb/conversion_procs/cyrillic_and_mic/cyrillic_and_mic.c
+++ b/src/backend/utils/mb/conversion_procs/cyrillic_and_mic/cyrillic_and_mic.c
@@ -201,7 +201,7 @@ koi8r_to_win1251(PG_FUNCTION_ARGS)
 
        CHECK_ENCODING_CONVERSION_ARGS(PG_KOI8R, PG_WIN1251);
 
-       buf = palloc(len * ENCODING_GROWTH_RATE);
+       buf = palloc(len * ENCODING_GROWTH_RATE + 1);
        koi8r2mic(src, buf, len);
        mic2win1251(buf, dest, strlen((char *) buf));
        pfree(buf);
@@ -219,7 +219,7 @@ win1251_to_koi8r(PG_FUNCTION_ARGS)
 
        CHECK_ENCODING_CONVERSION_ARGS(PG_WIN1251, PG_KOI8R);
 
-       buf = palloc(len * ENCODING_GROWTH_RATE);
+       buf = palloc(len * ENCODING_GROWTH_RATE + 1);
        win12512mic(src, buf, len);
        mic2koi8r(buf, dest, strlen((char *) buf));
        pfree(buf);
@@ -237,7 +237,7 @@ koi8r_to_win866(PG_FUNCTION_ARGS)
 
        CHECK_ENCODING_CONVERSION_ARGS(PG_KOI8R, PG_WIN866);
 
-       buf = palloc(len * ENCODING_GROWTH_RATE);
+       buf = palloc(len * ENCODING_GROWTH_RATE + 1);
        koi8r2mic(src, buf, len);
        mic2win866(buf, dest, strlen((char *) buf));
        pfree(buf);
@@ -255,7 +255,7 @@ win866_to_koi8r(PG_FUNCTION_ARGS)
 
        CHECK_ENCODING_CONVERSION_ARGS(PG_WIN866, PG_KOI8R);
 
-       buf = palloc(len * ENCODING_GROWTH_RATE);
+       buf = palloc(len * ENCODING_GROWTH_RATE + 1);
        win8662mic(src, buf, len);
        mic2koi8r(buf, dest, strlen((char *) buf));
        pfree(buf);
@@ -279,7 +279,7 @@ win866_to_win1251(PG_FUNCTION_ARGS)
         * not in KOI8R. As we use MULE_INTERNAL/KOI8R as an intermediary, we
         * will fail to convert those characters.
         */
-       buf = palloc(len * ENCODING_GROWTH_RATE);
+       buf = palloc(len * ENCODING_GROWTH_RATE + 1);
        win8662mic(src, buf, len);
        mic2win1251(buf, dest, strlen((char *) buf));
        pfree(buf);
@@ -298,7 +298,7 @@ win1251_to_win866(PG_FUNCTION_ARGS)
        CHECK_ENCODING_CONVERSION_ARGS(PG_WIN1251, PG_WIN866);
 
        /* Use mic/KOI8R as intermediary, see comment in win866_to_win1251() */
-       buf = palloc(len * ENCODING_GROWTH_RATE);
+       buf = palloc(len * ENCODING_GROWTH_RATE + 1);
        win12512mic(src, buf, len);
        mic2win866(buf, dest, strlen((char *) buf));
        pfree(buf);
@@ -316,7 +316,7 @@ iso_to_koi8r(PG_FUNCTION_ARGS)
 
        CHECK_ENCODING_CONVERSION_ARGS(PG_ISO_8859_5, PG_KOI8R);
 
-       buf = palloc(len * ENCODING_GROWTH_RATE);
+       buf = palloc(len * ENCODING_GROWTH_RATE + 1);
        iso2mic(src, buf, len);
        mic2koi8r(buf, dest, strlen((char *) buf));
        pfree(buf);
@@ -334,7 +334,7 @@ koi8r_to_iso(PG_FUNCTION_ARGS)
 
        CHECK_ENCODING_CONVERSION_ARGS(PG_KOI8R, PG_ISO_8859_5);
 
-       buf = palloc(len * ENCODING_GROWTH_RATE);
+       buf = palloc(len * ENCODING_GROWTH_RATE + 1);
        koi8r2mic(src, buf, len);
        mic2iso(buf, dest, strlen((char *) buf));
        pfree(buf);
@@ -353,7 +353,7 @@ iso_to_win1251(PG_FUNCTION_ARGS)
        CHECK_ENCODING_CONVERSION_ARGS(PG_ISO_8859_5, PG_WIN1251);
 
        /* Use mic/KOI8R as intermediary, see comment in win866_to_win1251() */
-       buf = palloc(len * ENCODING_GROWTH_RATE);
+       buf = palloc(len * ENCODING_GROWTH_RATE + 1);
        iso2mic(src, buf, len);
        mic2win1251(buf, dest, strlen((char *) buf));
        pfree(buf);
@@ -372,7 +372,7 @@ win1251_to_iso(PG_FUNCTION_ARGS)
        CHECK_ENCODING_CONVERSION_ARGS(PG_WIN1251, PG_ISO_8859_5);
 
        /* Use mic/KOI8R as intermediary, see comment in win866_to_win1251() */
-       buf = palloc(len * ENCODING_GROWTH_RATE);
+       buf = palloc(len * ENCODING_GROWTH_RATE + 1);
        win12512mic(src, buf, len);
        mic2iso(buf, dest, strlen((char *) buf));
        pfree(buf);
@@ -391,7 +391,7 @@ iso_to_win866(PG_FUNCTION_ARGS)
        CHECK_ENCODING_CONVERSION_ARGS(PG_ISO_8859_5, PG_WIN866);
 
        /* Use mic/KOI8R as intermediary, see comment in win866_to_win1251() */
-       buf = palloc(len * ENCODING_GROWTH_RATE);
+       buf = palloc(len * ENCODING_GROWTH_RATE + 1);
        iso2mic(src, buf, len);
        mic2win866(buf, dest, strlen((char *) buf));
        pfree(buf);
@@ -410,7 +410,7 @@ win866_to_iso(PG_FUNCTION_ARGS)
        CHECK_ENCODING_CONVERSION_ARGS(PG_WIN866, PG_ISO_8859_5);
 
        /* Use mic/KOI8R as intermediary, see comment in win866_to_win1251() */
-       buf = palloc(len * ENCODING_GROWTH_RATE);
+       buf = palloc(len * ENCODING_GROWTH_RATE + 1);
        win8662mic(src, buf, len);
        mic2iso(buf, dest, strlen((char *) buf));
        pfree(buf);

diff --git a/src/backend/utils/mb/conversion_procs/euc_jp_and_sjis/euc_jp_and_sjis.c b/src/backend/utils/mb/conversion_procs/euc_jp_and_sjis/euc_jp_and_sjis.c
index 2dde019e5f625cecf56fb39f5ea830753c33768c..654b8185b6aa09ddd3ad35bab01a0309485354fd 100644 (file)
--- a/src/backend/utils/mb/conversion_procs/euc_jp_and_sjis/euc_jp_and_sjis.c
+++ b/src/backend/utils/mb/conversion_procs/euc_jp_and_sjis/euc_jp_and_sjis.c
@@ -27,8 +27,6 @@
  */
 #include "sjis.map"
 
-#define ENCODING_GROWTH_RATE 4
-
 PG_FUNCTION_INFO_V1(euc_jp_to_sjis);
 PG_FUNCTION_INFO_V1(sjis_to_euc_jp);
 PG_FUNCTION_INFO_V1(euc_jp_to_mic);

diff --git a/src/backend/utils/mb/conversion_procs/euc_tw_and_big5/euc_tw_and_big5.c b/src/backend/utils/mb/conversion_procs/euc_tw_and_big5/euc_tw_and_big5.c
index be5f18e9a02fbbfa6c2df82741132e2b54be0e51..ead78c43af9f70034bbb8dcc9c19afc2b6b56ce8 100644 (file)
--- a/src/backend/utils/mb/conversion_procs/euc_tw_and_big5/euc_tw_and_big5.c
+++ b/src/backend/utils/mb/conversion_procs/euc_tw_and_big5/euc_tw_and_big5.c
@@ -57,7 +57,7 @@ euc_tw_to_big5(PG_FUNCTION_ARGS)
 
        CHECK_ENCODING_CONVERSION_ARGS(PG_EUC_TW, PG_BIG5);
 
-       buf = palloc(len * ENCODING_GROWTH_RATE);
+       buf = palloc(len * ENCODING_GROWTH_RATE + 1);
        euc_tw2mic(src, buf, len);
        mic2big5(buf, dest, strlen((char *) buf));
        pfree(buf);
@@ -75,7 +75,7 @@ big5_to_euc_tw(PG_FUNCTION_ARGS)
 
        CHECK_ENCODING_CONVERSION_ARGS(PG_BIG5, PG_EUC_TW);
 
-       buf = palloc(len * ENCODING_GROWTH_RATE);
+       buf = palloc(len * ENCODING_GROWTH_RATE + 1);
        big52mic(src, buf, len);
        mic2euc_tw(buf, dest, strlen((char *) buf));
        pfree(buf);

diff --git a/src/backend/utils/mb/conversion_procs/latin2_and_win1250/latin2_and_win1250.c b/src/backend/utils/mb/conversion_procs/latin2_and_win1250/latin2_and_win1250.c
index 57e6c4c0c99719ae7f978e3c1fc592b246e5b199..527a8fd69445a7d90e74ec433cd98bc674c0c6d2 100644 (file)
--- a/src/backend/utils/mb/conversion_procs/latin2_and_win1250/latin2_and_win1250.c
+++ b/src/backend/utils/mb/conversion_procs/latin2_and_win1250/latin2_and_win1250.c
@@ -113,7 +113,7 @@ latin2_to_win1250(PG_FUNCTION_ARGS)
 
        CHECK_ENCODING_CONVERSION_ARGS(PG_LATIN2, PG_WIN1250);
 
-       buf = palloc(len * ENCODING_GROWTH_RATE);
+       buf = palloc(len * ENCODING_GROWTH_RATE + 1);
        latin22mic(src, buf, len);
        mic2win1250(buf, dest, strlen((char *) buf));
        pfree(buf);
@@ -131,7 +131,7 @@ win1250_to_latin2(PG_FUNCTION_ARGS)
 
        CHECK_ENCODING_CONVERSION_ARGS(PG_WIN1250, PG_LATIN2);
 
-       buf = palloc(len * ENCODING_GROWTH_RATE);
+       buf = palloc(len * ENCODING_GROWTH_RATE + 1);
        win12502mic(src, buf, len);
        mic2latin2(buf, dest, strlen((char *) buf));
        pfree(buf);