diff --git a/doc/src/sgml/ref/copy.sgml b/doc/src/sgml/ref/copy.sgml
index 83a51362c5b20b47756670ab47fb3b70f1abadd1..c8e4debea715b2f7f8b83b96c5aacd7311a0eb67 100644
--- a/doc/src/sgml/ref/copy.sgml
+++ b/doc/src/sgml/ref/copy.sgml
@@ -1,5 +1,5 @@
<!--
-$Header: /cvsroot/pgsql/doc/src/sgml/ref/copy.sgml,v 1.52 2003/09/30 01:56:11 tgl Exp $
+$Header: /cvsroot/pgsql/doc/src/sgml/ref/copy.sgml,v 1.53 2003/10/06 02:38:53 tgl Exp $
PostgreSQL documentation
-->
@@ -257,7 +257,7 @@ COPY <replaceable class="parameter">tablename</replaceable> [ ( <replaceable cla
Columns in a row are separated by the delimiter character.
The column values themselves are strings generated by the
output function, or acceptable to the input function, of each
- attribute's data type. The specified null-value string is used in
+ attribute's data type. The specified null string is used in
place of columns that are null.
<command>COPY FROM</command> will raise an error if any line of the
input file contains more or fewer columns than are expected.
@@ -282,6 +282,15 @@ COPY <replaceable class="parameter">tablename</replaceable> [ ( <replaceable cla
newline, carriage return, and the current delimiter character.
</para>
+ <para>
+ The specified null string is sent by <command>COPY TO</command> without
+ adding any backslashes; conversely, <command>COPY FROM</command> matches
+ the input against the null string before removing backslashes. Therefore,
+ a null string such as <literal>\N</literal> cannot be confused with
+ the actual data value <literal>\N</literal> (which would be represented
+ as <literal>\\N</literal>).
+ </para>
+
<para>
The following special backslash sequences are recognized by
<command>COPY FROM</command>:
@@ -335,10 +344,12 @@ COPY <replaceable class="parameter">tablename</replaceable> [ ( <replaceable cla
</para>
<para>
- Never put a backslash before a data character <literal>N</> or period
- (<literal>.</>). Such pairs will be mistaken for the default null string
- or the end-of-data marker, respectively. Any other backslashed character
- that is not mentioned in the above table will be taken to represent itself.
+ Any other backslashed character that is not mentioned in the above table
+ will be taken to represent itself. However, beware of adding backslashes
+ unnecessarily, since that might accidentally produce a string matching the
+ end-of-data marker (<literal>\.</>) or the null string (<literal>\N</> by
+ default). These strings will be recognized before any other backslash
+ processing is done.
</para>
<para>
diff --git a/src/backend/commands/copy.c b/src/backend/commands/copy.c
index aac9f3c1d4fa850baf1a2a2b913dfec628583a1e..26de12ea9b7d5df6b515a1c7187f43d79ade15e5 100644
--- a/src/backend/commands/copy.c
+++ b/src/backend/commands/copy.c
@@ -8,7 +8,7 @@
*
*
* IDENTIFICATION
- * $Header: /cvsroot/pgsql/src/backend/commands/copy.c,v 1.212 2003/09/29 22:06:40 tgl Exp $
+ * $Header: /cvsroot/pgsql/src/backend/commands/copy.c,v 1.213 2003/10/06 02:38:53 tgl Exp $
*
*-------------------------------------------------------------------------
*/
@@ -122,7 +122,7 @@ static StringInfoData attribute_buf;
* to server encoding, and then extract individual attribute fields into
* attribute_buf. (We used to have CopyReadAttribute read the input source
* directly, but that caused a lot of encoding issues and unnecessary logic
- * complexity).
+ * complexity.)
*/
static StringInfoData line_buf;
static bool line_buf_converted;
@@ -133,7 +133,8 @@ static void CopyTo(Relation rel, List *attnumlist, bool binary, bool oids,
static void CopyFrom(Relation rel, List *attnumlist, bool binary, bool oids,
char *delim, char *null_print);
static bool CopyReadLine(void);
-static char *CopyReadAttribute(const char *delim, CopyReadResult *result);
+static char *CopyReadAttribute(const char *delim, const char *null_print,
+ CopyReadResult *result, bool *isnull);
static Datum CopyReadBinaryAttribute(int column_no, FmgrInfo *flinfo,
Oid typelem, bool *isnull);
static void CopyAttributeOut(char *string, char *delim);
@@ -1014,6 +1015,17 @@ CopyTo(Relation rel, List *attnumlist, bool binary, bool oids,
tmp = 0;
CopySendInt32(tmp);
}
+ else
+ {
+ /*
+ * For non-binary copy, we need to convert null_print to client
+ * encoding, because it will be sent directly with CopySendString.
+ */
+ if (server_encoding != client_encoding)
+ null_print = (char *)
+ pg_server_to_client((unsigned char *) null_print,
+ strlen(null_print));
+ }
mySnapshot = CopyQuerySnapshot();
@@ -1441,9 +1453,10 @@ CopyFrom(Relation rel, List *attnumlist, bool binary, bool oids,
if (file_has_oids)
{
- string = CopyReadAttribute(delim, &result);
+ string = CopyReadAttribute(delim, null_print,
+ &result, &isnull);
- if (strcmp(string, null_print) == 0)
+ if (isnull)
ereport(ERROR,
(errcode(ERRCODE_BAD_COPY_FILE_FORMAT),
errmsg("null OID in COPY data")));
@@ -1478,9 +1491,10 @@ CopyFrom(Relation rel, List *attnumlist, bool binary, bool oids,
errmsg("missing data for column \"%s\"",
NameStr(attr[m]->attname))));
- string = CopyReadAttribute(delim, &result);
+ string = CopyReadAttribute(delim, null_print,
+ &result, &isnull);
- if (strcmp(string, null_print) == 0)
+ if (isnull)
{
/* we read an SQL NULL, no need to do anything */
}
@@ -1880,25 +1894,33 @@ CopyReadLine(void)
return result;
}
-/*
+/*----------
* Read the value of a single attribute, performing de-escaping as needed.
*
+ * delim is the column delimiter string (must be just one byte for now).
+ * null_print is the null marker string. Note that this is compared to
+ * the pre-de-escaped input string.
+ *
* *result is set to indicate what terminated the read:
* NORMAL_ATTR: column delimiter
* END_OF_LINE: end of line
* In either case, the string read up to the terminator is returned.
*
- * Note: This function does not care about SQL NULL values -- it
- * is the caller's responsibility to check if the returned string
- * matches what the user specified for the SQL NULL value.
- *
- * delim is the column delimiter string.
+ * *isnull is set true or false depending on whether the input matched
+ * the null marker. Note that the caller cannot check this since the
+ * returned string will be the post-de-escaping equivalent, which may
+ * look the same as some valid data string.
+ *----------
*/
static char *
-CopyReadAttribute(const char *delim, CopyReadResult *result)
+CopyReadAttribute(const char *delim, const char *null_print,
+ CopyReadResult *result, bool *isnull)
{
char c;
char delimc = delim[0];
+ int start_cursor = line_buf.cursor;
+ int end_cursor;
+ int input_len;
/* reset attribute_buf to empty */
attribute_buf.len = 0;
@@ -1909,6 +1931,7 @@ CopyReadAttribute(const char *delim, CopyReadResult *result)
for (;;)
{
+ end_cursor = line_buf.cursor;
if (line_buf.cursor >= line_buf.len)
break;
c = line_buf.data[line_buf.cursor++];
@@ -1957,16 +1980,6 @@ CopyReadAttribute(const char *delim, CopyReadResult *result)
c = val & 0377;
}
break;
-
- /*
- * This is a special hack to parse `\N' as
- * <backslash-N> rather then just 'N' to provide
- * compatibility with the default NULL output. -- pe
- */
- case 'N':
- appendStringInfoCharMacro(&attribute_buf, '\\');
- c = 'N';
- break;
case 'b':
c = '\b';
break;
@@ -1993,6 +2006,14 @@ CopyReadAttribute(const char *delim, CopyReadResult *result)
appendStringInfoCharMacro(&attribute_buf, c);
}
+ /* check whether raw input matched null marker */
+ input_len = end_cursor - start_cursor;
+ if (input_len == strlen(null_print) &&
+ strncmp(&line_buf.data[start_cursor], null_print, input_len) == 0)
+ *isnull = true;
+ else
+ *isnull = false;
+
return attribute_buf.data;
}