mirrors/postgres
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Respond to Jeremy Drake's original gripe that \copy needs to recognize

Browse Source 
E'...' syntax for strings in order to track the backend.
This commit is contained in:
Tom Lane 2006-06-01 01:28:00 +00:00
parent 6178762fcf
commit 87e8014d31
3 changed files with 75 additions and 78 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

124
src/bin/psql/copy.c
View File

@ -3,7 +3,7 @@
* *
* Copyright (c) 2000-2006, PostgreSQL Global Development Group * Copyright (c) 2000-2006, PostgreSQL Global Development Group
* *
* $PostgreSQL: pgsql/src/bin/psql/copy.c,v 1.63 2006/06/01 00:15:36 tgl Exp $ * $PostgreSQL: pgsql/src/bin/psql/copy.c,v 1.64 2006/06/01 01:28:00 tgl Exp $
*/ */
#include "postgres_fe.h" #include "postgres_fe.h"
#include "copy.h" #include "copy.h"
@ -127,7 +127,7 @@ parse_slash_copy(const char *args)
result = pg_calloc(1, sizeof(struct copy_options)); result = pg_calloc(1, sizeof(struct copy_options));
token = strtokx(line, whitespace, ".,()", "\"", token = strtokx(line, whitespace, ".,()", "\"",
0, false, pset.encoding); 0, false, false, pset.encoding);
if (!token) if (!token)
goto error; goto error;
@ -135,7 +135,7 @@ parse_slash_copy(const char *args)
{ {
result->binary = true; result->binary = true;
token = strtokx(NULL, whitespace, ".,()", "\"", token = strtokx(NULL, whitespace, ".,()", "\"",
0, false, pset.encoding); 0, false, false, pset.encoding);
if (!token) if (!token)
goto error; goto error;
} }
@ -143,7 +143,7 @@ parse_slash_copy(const char *args)
result->table = pg_strdup(token); result->table = pg_strdup(token);
token = strtokx(NULL, whitespace, ".,()", "\"", token = strtokx(NULL, whitespace, ".,()", "\"",
0, false, pset.encoding); 0, false, false, pset.encoding);
if (!token) if (!token)
goto error; goto error;
@ -156,12 +156,12 @@ parse_slash_copy(const char *args)
/* handle schema . table */ /* handle schema . table */
xstrcat(&result->table, token); xstrcat(&result->table, token);
token = strtokx(NULL, whitespace, ".,()", "\"", token = strtokx(NULL, whitespace, ".,()", "\"",
0, false, pset.encoding); 0, false, false, pset.encoding);
if (!token) if (!token)
goto error; goto error;
xstrcat(&result->table, token); xstrcat(&result->table, token);
token = strtokx(NULL, whitespace, ".,()", "\"", token = strtokx(NULL, whitespace, ".,()", "\"",
0, false, pset.encoding); 0, false, false, pset.encoding);
if (!token) if (!token)
goto error; goto error;
} }
@ -173,12 +173,12 @@ parse_slash_copy(const char *args)
for (;;) for (;;)
{ {
token = strtokx(NULL, whitespace, ".,()", "\"", token = strtokx(NULL, whitespace, ".,()", "\"",
0, false, pset.encoding); 0, false, false, pset.encoding);
if (!token || strchr(".,()", token[0])) if (!token || strchr(".,()", token[0]))
goto error; goto error;
xstrcat(&result->column_list, token); xstrcat(&result->column_list, token);
token = strtokx(NULL, whitespace, ".,()", "\"", token = strtokx(NULL, whitespace, ".,()", "\"",
0, false, pset.encoding); 0, false, false, pset.encoding);
if (!token) if (!token)
goto error; goto error;
xstrcat(&result->column_list, token); xstrcat(&result->column_list, token);
@ -188,7 +188,7 @@ parse_slash_copy(const char *args)
goto error; goto error;
} }
token = strtokx(NULL, whitespace, ".,()", "\"", token = strtokx(NULL, whitespace, ".,()", "\"",
0, false, pset.encoding); 0, false, false, pset.encoding);
if (!token) if (!token)
goto error; goto error;
} }
@ -199,13 +199,13 @@ parse_slash_copy(const char *args)
if (pg_strcasecmp(token, "with") == 0) if (pg_strcasecmp(token, "with") == 0)
{ {
token = strtokx(NULL, whitespace, NULL, NULL, token = strtokx(NULL, whitespace, NULL, NULL,
0, false, pset.encoding); 0, false, false, pset.encoding);
if (!token || pg_strcasecmp(token, "oids") != 0) if (!token || pg_strcasecmp(token, "oids") != 0)
goto error; goto error;
result->oids = true; result->oids = true;
token = strtokx(NULL, whitespace, NULL, NULL, token = strtokx(NULL, whitespace, NULL, NULL,
0, false, pset.encoding); 0, false, false, pset.encoding);
if (!token) if (!token)
goto error; goto error;
} }
@ -218,7 +218,7 @@ parse_slash_copy(const char *args)
goto error; goto error;
token = strtokx(NULL, whitespace, NULL, "'", token = strtokx(NULL, whitespace, NULL, "'",
nonstd_backslash, true, pset.encoding); 0, false, true, pset.encoding);
if (!token) if (!token)
goto error; goto error;
@ -242,7 +242,7 @@ parse_slash_copy(const char *args)
} }
token = strtokx(NULL, whitespace, NULL, NULL, token = strtokx(NULL, whitespace, NULL, NULL,
0, false, pset.encoding); 0, false, false, pset.encoding);
/* /*
* Allows old COPY syntax for backward compatibility 2002-06-19 * Allows old COPY syntax for backward compatibility 2002-06-19
@ -250,19 +250,19 @@ parse_slash_copy(const char *args)
if (token && pg_strcasecmp(token, "using") == 0) if (token && pg_strcasecmp(token, "using") == 0)
{ {
token = strtokx(NULL, whitespace, NULL, NULL, token = strtokx(NULL, whitespace, NULL, NULL,
0, false, pset.encoding); 0, false, false, pset.encoding);
if (!(token && pg_strcasecmp(token, "delimiters") == 0)) if (!(token && pg_strcasecmp(token, "delimiters") == 0))
goto error; goto error;
} }
if (token && pg_strcasecmp(token, "delimiters") == 0) if (token && pg_strcasecmp(token, "delimiters") == 0)
{ {
token = strtokx(NULL, whitespace, NULL, "'", token = strtokx(NULL, whitespace, NULL, "'",
nonstd_backslash, false, pset.encoding); nonstd_backslash, true, false, pset.encoding);
if (!token) if (!token)
goto error; goto error;
result->delim = pg_strdup(token); result->delim = pg_strdup(token);
token = strtokx(NULL, whitespace, NULL, NULL, token = strtokx(NULL, whitespace, NULL, NULL,
0, false, pset.encoding); 0, false, false, pset.encoding);
} }
if (token) if (token)
@ -273,7 +273,7 @@ parse_slash_copy(const char *args)
*/ */
if (pg_strcasecmp(token, "with") == 0) if (pg_strcasecmp(token, "with") == 0)
token = strtokx(NULL, whitespace, NULL, NULL, token = strtokx(NULL, whitespace, NULL, NULL,
0, false, pset.encoding); 0, false, false, pset.encoding);
while (token) while (token)
{ {
@ -292,10 +292,10 @@ parse_slash_copy(const char *args)
else if (pg_strcasecmp(token, "delimiter") == 0) else if (pg_strcasecmp(token, "delimiter") == 0)
{ {
token = strtokx(NULL, whitespace, NULL, "'", token = strtokx(NULL, whitespace, NULL, "'",
nonstd_backslash, false, pset.encoding); nonstd_backslash, true, false, pset.encoding);
if (token && pg_strcasecmp(token, "as") == 0) if (token && pg_strcasecmp(token, "as") == 0)
token = strtokx(NULL, whitespace, NULL, "'", token = strtokx(NULL, whitespace, NULL, "'",
nonstd_backslash, false, pset.encoding); nonstd_backslash, true, false, pset.encoding);
if (token) if (token)
result->delim = pg_strdup(token); result->delim = pg_strdup(token);
else else
@ -304,10 +304,10 @@ parse_slash_copy(const char *args)
else if (pg_strcasecmp(token, "null") == 0) else if (pg_strcasecmp(token, "null") == 0)
{ {
token = strtokx(NULL, whitespace, NULL, "'", token = strtokx(NULL, whitespace, NULL, "'",
nonstd_backslash, false, pset.encoding); nonstd_backslash, true, false, pset.encoding);
if (token && pg_strcasecmp(token, "as") == 0) if (token && pg_strcasecmp(token, "as") == 0)
token = strtokx(NULL, whitespace, NULL, "'", token = strtokx(NULL, whitespace, NULL, "'",
nonstd_backslash, false, pset.encoding); nonstd_backslash, true, false, pset.encoding);
if (token) if (token)
result->null = pg_strdup(token); result->null = pg_strdup(token);
else else
@ -316,10 +316,10 @@ parse_slash_copy(const char *args)
else if (pg_strcasecmp(token, "quote") == 0) else if (pg_strcasecmp(token, "quote") == 0)
{ {
token = strtokx(NULL, whitespace, NULL, "'", token = strtokx(NULL, whitespace, NULL, "'",
nonstd_backslash, false, pset.encoding); nonstd_backslash, true, false, pset.encoding);
if (token && pg_strcasecmp(token, "as") == 0) if (token && pg_strcasecmp(token, "as") == 0)
token = strtokx(NULL, whitespace, NULL, "'", token = strtokx(NULL, whitespace, NULL, "'",
nonstd_backslash, false, pset.encoding); nonstd_backslash, true, false, pset.encoding);
if (token) if (token)
result->quote = pg_strdup(token); result->quote = pg_strdup(token);
else else
@ -328,10 +328,10 @@ parse_slash_copy(const char *args)
else if (pg_strcasecmp(token, "escape") == 0) else if (pg_strcasecmp(token, "escape") == 0)
{ {
token = strtokx(NULL, whitespace, NULL, "'", token = strtokx(NULL, whitespace, NULL, "'",
nonstd_backslash, false, pset.encoding); nonstd_backslash, true, false, pset.encoding);
if (token && pg_strcasecmp(token, "as") == 0) if (token && pg_strcasecmp(token, "as") == 0)
token = strtokx(NULL, whitespace, NULL, "'", token = strtokx(NULL, whitespace, NULL, "'",
nonstd_backslash, false, pset.encoding); nonstd_backslash, true, false, pset.encoding);
if (token) if (token)
result->escape = pg_strdup(token); result->escape = pg_strdup(token);
else else
@ -340,7 +340,7 @@ parse_slash_copy(const char *args)
else if (pg_strcasecmp(token, "force") == 0) else if (pg_strcasecmp(token, "force") == 0)
{ {
token = strtokx(NULL, whitespace, ",", "\"", token = strtokx(NULL, whitespace, ",", "\"",
0, false, pset.encoding); 0, false, false, pset.encoding);
if (pg_strcasecmp(token, "quote") == 0) if (pg_strcasecmp(token, "quote") == 0)
{ {
/* handle column list */ /* handle column list */
@ -348,7 +348,7 @@ parse_slash_copy(const char *args)
for (;;) for (;;)
{ {
token = strtokx(NULL, whitespace, ",", "\"", token = strtokx(NULL, whitespace, ",", "\"",
0, false, pset.encoding); 0, false, false, pset.encoding);
if (!token || strchr(",", token[0])) if (!token || strchr(",", token[0]))
goto error; goto error;
if (!result->force_quote_list) if (!result->force_quote_list)
@ -356,7 +356,7 @@ parse_slash_copy(const char *args)
else else
xstrcat(&result->force_quote_list, token); xstrcat(&result->force_quote_list, token);
token = strtokx(NULL, whitespace, ",", "\"", token = strtokx(NULL, whitespace, ",", "\"",
0, false, pset.encoding); 0, false, false, pset.encoding);
if (!token || token[0] != ',') if (!token || token[0] != ',')
break; break;
xstrcat(&result->force_quote_list, token); xstrcat(&result->force_quote_list, token);
@ -365,7 +365,7 @@ parse_slash_copy(const char *args)
else if (pg_strcasecmp(token, "not") == 0) else if (pg_strcasecmp(token, "not") == 0)
{ {
token = strtokx(NULL, whitespace, ",", "\"", token = strtokx(NULL, whitespace, ",", "\"",
0, false, pset.encoding); 0, false, false, pset.encoding);
if (pg_strcasecmp(token, "null") != 0) if (pg_strcasecmp(token, "null") != 0)
goto error; goto error;
/* handle column list */ /* handle column list */
@ -373,7 +373,7 @@ parse_slash_copy(const char *args)
for (;;) for (;;)
{ {
token = strtokx(NULL, whitespace, ",", "\"", token = strtokx(NULL, whitespace, ",", "\"",
0, false, pset.encoding); 0, false, false, pset.encoding);
if (!token || strchr(",", token[0])) if (!token || strchr(",", token[0]))
goto error; goto error;
if (!result->force_notnull_list) if (!result->force_notnull_list)
@ -381,7 +381,7 @@ parse_slash_copy(const char *args)
else else
xstrcat(&result->force_notnull_list, token); xstrcat(&result->force_notnull_list, token);
token = strtokx(NULL, whitespace, ",", "\"", token = strtokx(NULL, whitespace, ",", "\"",
0, false, pset.encoding); 0, false, false, pset.encoding);
if (!token || token[0] != ',') if (!token || token[0] != ',')
break; break;
xstrcat(&result->force_notnull_list, token); xstrcat(&result->force_notnull_list, token);
@ -395,7 +395,7 @@ parse_slash_copy(const char *args)
if (fetch_next) if (fetch_next)
token = strtokx(NULL, whitespace, NULL, NULL, token = strtokx(NULL, whitespace, NULL, NULL,
0, false, pset.encoding); 0, false, false, pset.encoding);
} }
} }
@ -415,6 +415,22 @@ error:
} }
/*
* Handle one of the "string" options of COPY. If the user gave a quoted
* string, pass it to the backend as-is; if it wasn't quoted then quote
* and escape it.
*/
static void
emit_copy_option(PQExpBuffer query, const char *keyword, const char *option)
{
appendPQExpBufferStr(query, keyword);
if (option[0] == '\'' ||
((option[0] == 'E' || option[0] == 'e') && option[1] == '\''))
appendPQExpBufferStr(query, option);
else
appendStringLiteralConn(query, option, pset.db);
}
/* /*
* Execute a \copy command (frontend copy). We have to open a file, then * Execute a \copy command (frontend copy). We have to open a file, then
@ -462,29 +478,11 @@ do_copy(const char *args)
/* Uses old COPY syntax for backward compatibility 2002-06-19 */ /* Uses old COPY syntax for backward compatibility 2002-06-19 */
if (options->delim) if (options->delim)
{ emit_copy_option(&query, " USING DELIMITERS ", options->delim);
/* if user gave a quoted string, use it as-is */
if (options->delim[0] == '\'')
appendPQExpBuffer(&query, " USING DELIMITERS %s", options->delim);
else
{
appendPQExpBuffer(&query, " USING DELIMITERS ");
appendStringLiteralConn(&query, options->delim, pset.db);
}
}
/* There is no backward-compatible CSV syntax */ /* There is no backward-compatible CSV syntax */
if (options->null) if (options->null)
{ emit_copy_option(&query, " WITH NULL AS ", options->null);
/* if user gave a quoted string, use it as-is */
if (options->null[0] == '\'')
appendPQExpBuffer(&query, " WITH NULL AS %s", options->null);
else
{
appendPQExpBuffer(&query, " WITH NULL AS ");
appendStringLiteralConn(&query, options->null, pset.db);
}
}
if (options->csv_mode) if (options->csv_mode)
appendPQExpBuffer(&query, " CSV"); appendPQExpBuffer(&query, " CSV");
@ -493,28 +491,10 @@ do_copy(const char *args)
appendPQExpBuffer(&query, " HEADER"); appendPQExpBuffer(&query, " HEADER");
if (options->quote) if (options->quote)
{ emit_copy_option(&query, " QUOTE AS ", options->quote);
/* if user gave a quoted string, use it as-is */
if (options->quote[0] == '\'')
appendPQExpBuffer(&query, " QUOTE AS %s", options->quote);
else
{
appendPQExpBuffer(&query, " QUOTE AS ");
appendStringLiteralConn(&query, options->quote, pset.db);
}
}
if (options->escape) if (options->escape)
{ emit_copy_option(&query, " ESCAPE AS ", options->escape);
/* if user gave a quoted string, use it as-is */
if (options->escape[0] == '\'')
appendPQExpBuffer(&query, " ESCAPE AS %s", options->escape);
else
{
appendPQExpBuffer(&query, " ESCAPE AS ");
appendStringLiteralConn(&query, options->escape, pset.db);
}
}
if (options->force_quote_list) if (options->force_quote_list)
appendPQExpBuffer(&query, " FORCE QUOTE %s", options->force_quote_list); appendPQExpBuffer(&query, " FORCE QUOTE %s", options->force_quote_list);

26
src/bin/psql/stringutils.c
View File

@ -3,7 +3,7 @@
* *
* Copyright (c) 2000-2006, PostgreSQL Global Development Group * Copyright (c) 2000-2006, PostgreSQL Global Development Group
* *
* $PostgreSQL: pgsql/src/bin/psql/stringutils.c,v 1.42 2006/03/05 15:58:52 momjian Exp $ * $PostgreSQL: pgsql/src/bin/psql/stringutils.c,v 1.43 2006/06/01 01:28:00 tgl Exp $
*/ */
#include "postgres_fe.h" #include "postgres_fe.h"
@ -32,7 +32,8 @@ static void strip_quotes(char *source, char quote, char escape, int encoding);
* delim - set of non-whitespace separator characters (or NULL) * delim - set of non-whitespace separator characters (or NULL)
* quote - set of characters that can quote a token (NULL if none) * quote - set of characters that can quote a token (NULL if none)
* escape - character that can quote quotes (0 if none) * escape - character that can quote quotes (0 if none)
* del_quotes - if TRUE, strip quotes from the returned token, else return * e_strings - if TRUE, treat E'...' syntax as a valid token
* del_quotes - if TRUE, strip quotes from the returned token, else return
* it exactly as found in the string * it exactly as found in the string
* encoding - the active character-set encoding * encoding - the active character-set encoding
* *
@ -43,6 +44,9 @@ static void strip_quotes(char *source, char quote, char escape, int encoding);
* a single quote character in the data. If escape isn't 0, then escape * a single quote character in the data. If escape isn't 0, then escape
* followed by anything (except \0) is a data character too. * followed by anything (except \0) is a data character too.
* *
* The combination of e_strings and del_quotes both TRUE is not currently
* handled. This could be fixed but it's not needed anywhere at the moment.
*
* Note that the string s is _not_ overwritten in this implementation. * Note that the string s is _not_ overwritten in this implementation.
* *
* NB: it's okay to vary delim, quote, and escape from one call to the * NB: it's okay to vary delim, quote, and escape from one call to the
@ -55,6 +59,7 @@ strtokx(const char *s,
const char *delim, const char *delim,
const char *quote, const char *quote,
char escape, char escape,
bool e_strings,
bool del_quotes, bool del_quotes,
int encoding) int encoding)
{ {
@ -126,13 +131,24 @@ strtokx(const char *s,
return start; return start;
} }
/* check for E string */
p = start;
if (e_strings &&
(*p == 'E' || *p == 'e') &&
p[1] == '\'')
{
quote = "'";
escape = '\\'; /* if std strings before, not any more */
p++;
}
/* test if quoting character */ /* test if quoting character */
if (quote && strchr(quote, *start)) if (quote && strchr(quote, *p))
{ {
/* okay, we have a quoted token, now scan for the closer */ /* okay, we have a quoted token, now scan for the closer */
char thisquote = *start; char thisquote = *p++;
for (p = start + 1; *p; p += PQmblen(p, encoding)) for (; *p; p += PQmblen(p, encoding))
{ {
if (*p == escape && p[1] != '\0') if (*p == escape && p[1] != '\0')
p++; /* process escaped anything */ p++; /* process escaped anything */

3
src/bin/psql/stringutils.h
View File

@ -3,7 +3,7 @@
* *
* Copyright (c) 2000-2006, PostgreSQL Global Development Group * Copyright (c) 2000-2006, PostgreSQL Global Development Group
* *
* $PostgreSQL: pgsql/src/bin/psql/stringutils.h,v 1.23 2006/03/05 15:58:52 momjian Exp $ * $PostgreSQL: pgsql/src/bin/psql/stringutils.h,v 1.24 2006/06/01 01:28:00 tgl Exp $
*/ */
#ifndef STRINGUTILS_H #ifndef STRINGUTILS_H
#define STRINGUTILS_H #define STRINGUTILS_H
@ -15,6 +15,7 @@ extern char *strtokx(const char *s,
const char *delim, const char *delim,
const char *quote, const char *quote,
char escape, char escape,
bool e_strings,
bool del_quotes, bool del_quotes,
int encoding); int encoding);