Re: Variable substitution in psql backtick expansion - Mailing list pgsql-hackers

Corey Huinker <corey.huinker@gmail.com> writes:
> On Thu, Mar 30, 2017 at 1:33 PM, Tom Lane <tgl@sss.pgh.pa.us> wrote:
>> single-quoted according to Unix shell conventions.  (So the
>> processing would be a bit different from what it is for the
>> same notation in SQL contexts.)

> +1

Here's a proposed patch for this.  I used the existing appendShellString()
logic, which already knows how to quote stuff safely on both Unix and
Windows.  A small problem is that appendShellString() rejects LF and CR
characters.  As written, it just printed an error and did exit(1), which
is more or less OK for existing callers but seemed like a pretty bad idea
for psql.  I refactored it to get the behavior proposed in the patch,
which is that we print an error and decline to substitute the variable's
value, leading to executing the backtick command with the :'variable'
text still in place.  This is more or less the same thing that happens
for encoding-check failures in the other variable-substitution cases,
so it didn't seem too unreasonable.

Perhaps it would be preferable to prevent execution of the backtick
command and/or fail the calling metacommand, but that seems to require
some very invasive refactoring (or else magic global variables), which
I didn't have the appetite for.

            regards, tom lane

diff --git a/doc/src/sgml/ref/psql-ref.sgml b/doc/src/sgml/ref/psql-ref.sgml
index b51b11b..ad463e7 100644
*** a/doc/src/sgml/ref/psql-ref.sgml
--- b/doc/src/sgml/ref/psql-ref.sgml
*************** testdb=>
*** 770,786 ****
      </para>

      <para>
-     Within an argument, text that is enclosed in backquotes
-     (<literal>`</literal>) is taken as a command line that is passed to the
-     shell. The output of the command (with any trailing newline removed)
-     replaces the backquoted text.
-     </para>
-
-     <para>
      If an unquoted colon (<literal>:</literal>) followed by a
      <application>psql</> variable name appears within an argument, it is
      replaced by the variable's value, as described in <xref
      linkend="APP-PSQL-interpolation" endterm="APP-PSQL-interpolation-title">.
      </para>

      <para>
--- 770,801 ----
      </para>

      <para>
      If an unquoted colon (<literal>:</literal>) followed by a
      <application>psql</> variable name appears within an argument, it is
      replaced by the variable's value, as described in <xref
      linkend="APP-PSQL-interpolation" endterm="APP-PSQL-interpolation-title">.
+     The forms <literal>:'<replaceable>variable_name</>'</literal> and
+     <literal>:"<replaceable>variable_name</>"</literal> described there
+     work as well.
+     </para>
+
+     <para>
+     Within an argument, text that is enclosed in backquotes
+     (<literal>`</literal>) is taken as a command line that is passed to the
+     shell.  The output of the command (with any trailing newline removed)
+     replaces the backquoted text.  Within the text enclosed in backquotes,
+     no special quoting or other processing occurs, except that appearances
+     of <literal>:<replaceable>variable_name</></literal> where
+     <replaceable>variable_name</> is a <application>psql</> variable name
+     are replaced by the variable's value.  Also, appearances of
+     <literal>:'<replaceable>variable_name</>'</literal> are replaced by the
+     variable's value suitably quoted to become a single shell command
+     argument.  (The latter form is almost always preferable, unless you are
+     very sure of what is in the variable.)  Because carriage return and line
+     feed characters cannot be safely quoted on all platforms, the
+     <literal>:'<replaceable>variable_name</>'</literal> form prints an
+     error message and does not substitute the variable value when such
+     characters appear in the value.
      </para>

      <para>
diff --git a/src/bin/psql/common.c b/src/bin/psql/common.c
index b06ae97..a2f1259 100644
*** a/src/bin/psql/common.c
--- b/src/bin/psql/common.c
*************** setQFout(const char *fname)
*** 116,134 ****
   * If the specified variable exists, return its value as a string (malloc'd
   * and expected to be freed by the caller); else return NULL.
   *
!  * If "escape" is true, return the value suitably quoted and escaped,
!  * as an identifier or string literal depending on "as_ident".
!  * (Failure in escaping should lead to returning NULL.)
   *
   * "passthrough" is the pointer previously given to psql_scan_set_passthrough.
   * In psql, passthrough points to a ConditionalStack, which we check to
   * determine whether variable expansion is allowed.
   */
  char *
! psql_get_variable(const char *varname, bool escape, bool as_ident,
                    void *passthrough)
  {
!     char       *result;
      const char *value;

      /* In an inactive \if branch, suppress all variable substitutions */
--- 116,134 ----
   * If the specified variable exists, return its value as a string (malloc'd
   * and expected to be freed by the caller); else return NULL.
   *
!  * If "quote" isn't PQUOTE_PLAIN, then return the value suitably quoted and
!  * escaped for the specified quoting requirement.  (Failure in escaping
!  * should lead to printing an error and returning NULL.)
   *
   * "passthrough" is the pointer previously given to psql_scan_set_passthrough.
   * In psql, passthrough points to a ConditionalStack, which we check to
   * determine whether variable expansion is allowed.
   */
  char *
! psql_get_variable(const char *varname, PsqlScanQuoteType quote,
                    void *passthrough)
  {
!     char       *result = NULL;
      const char *value;

      /* In an inactive \if branch, suppress all variable substitutions */
*************** psql_get_variable(const char *varname, b
*** 139,178 ****
      if (!value)
          return NULL;

!     if (escape)
      {
!         char       *escaped_value;

!         if (!pset.db)
!         {
!             psql_error("cannot escape without active connection\n");
!             return NULL;
!         }

!         if (as_ident)
!             escaped_value =
!                 PQescapeIdentifier(pset.db, value, strlen(value));
!         else
!             escaped_value =
!                 PQescapeLiteral(pset.db, value, strlen(value));

!         if (escaped_value == NULL)
!         {
!             const char *error = PQerrorMessage(pset.db);

!             psql_error("%s", error);
!             return NULL;
!         }

!         /*
!          * Rather than complicate the lexer's API with a notion of which
!          * free() routine to use, just pay the price of an extra strdup().
!          */
!         result = pg_strdup(escaped_value);
!         PQfreemem(escaped_value);
      }
-     else
-         result = pg_strdup(value);

      return result;
  }
--- 139,212 ----
      if (!value)
          return NULL;

!     switch (quote)
      {
!         case PQUOTE_PLAIN:
!             result = pg_strdup(value);
!             break;
!         case PQUOTE_SQL_LITERAL:
!         case PQUOTE_SQL_IDENT:
!             {
!                 /*
!                  * For these cases, we use libpq's quoting functions, which
!                  * assume the string is in the connection's client encoding.
!                  */
!                 char       *escaped_value;

!                 if (!pset.db)
!                 {
!                     psql_error("cannot escape without active connection\n");
!                     return NULL;
!                 }

!                 if (quote == PQUOTE_SQL_LITERAL)
!                     escaped_value =
!                         PQescapeLiteral(pset.db, value, strlen(value));
!                 else
!                     escaped_value =
!                         PQescapeIdentifier(pset.db, value, strlen(value));

!                 if (escaped_value == NULL)
!                 {
!                     const char *error = PQerrorMessage(pset.db);

!                     psql_error("%s", error);
!                     return NULL;
!                 }

!                 /*
!                  * Rather than complicate the lexer's API with a notion of
!                  * which free() routine to use, just pay the price of an extra
!                  * strdup().
!                  */
!                 result = pg_strdup(escaped_value);
!                 PQfreemem(escaped_value);
!                 break;
!             }
!         case PQUOTE_SHELL_ARG:
!             {
!                 /*
!                  * For this we use appendShellStringNoError, which is
!                  * encoding-agnostic, which is fine since the shell probably
!                  * is too.  In any case, the only special character is "'",
!                  * which is not known to appear in valid multibyte characters.
!                  */
!                 PQExpBufferData buf;
!
!                 initPQExpBuffer(&buf);
!                 if (!appendShellStringNoError(&buf, value))
!                 {
!                     psql_error("shell command argument contains a newline or carriage return: \"%s\"\n",
!                                value);
!                     free(buf.data);
!                     return NULL;
!                 }
!                 result = buf.data;
!                 break;
!             }
!
!             /* No default: we want a compiler warning for missing cases */
      }

      return result;
  }
diff --git a/src/bin/psql/common.h b/src/bin/psql/common.h
index 3d8b8da..1ceb8ae 100644
*** a/src/bin/psql/common.h
--- b/src/bin/psql/common.h
***************
*** 12,22 ****

  #include "libpq-fe.h"
  #include "fe_utils/print.h"

  extern bool openQueryOutputFile(const char *fname, FILE **fout, bool *is_pipe);
  extern bool setQFout(const char *fname);

! extern char *psql_get_variable(const char *varname, bool escape, bool as_ident,
                    void *passthrough);

  extern void psql_error(const char *fmt,...) pg_attribute_printf(1, 2);
--- 12,23 ----

  #include "libpq-fe.h"
  #include "fe_utils/print.h"
+ #include "fe_utils/psqlscan.h"

  extern bool openQueryOutputFile(const char *fname, FILE **fout, bool *is_pipe);
  extern bool setQFout(const char *fname);

! extern char *psql_get_variable(const char *varname, PsqlScanQuoteType quote,
                    void *passthrough);

  extern void psql_error(const char *fmt,...) pg_attribute_printf(1, 2);
diff --git a/src/bin/psql/psqlscanslash.l b/src/bin/psql/psqlscanslash.l
index 319afdc..db7a1b9 100644
*** a/src/bin/psql/psqlscanslash.l
--- b/src/bin/psql/psqlscanslash.l
*************** other            .
*** 242,249 ****
                                                               yytext + 1,
                                                               yyleng - 1);
                          value = cur_state->callbacks->get_variable(varname,
!                                                                    false,
!                                                                    false,
                                                                     cur_state->cb_passthrough);
                          free(varname);

--- 242,248 ----
                                                               yytext + 1,
                                                               yyleng - 1);
                          value = cur_state->callbacks->get_variable(varname,
!                                                                    PQUOTE_PLAIN,
                                                                     cur_state->cb_passthrough);
                          free(varname);

*************** other            .
*** 268,281 ****
                  }

  :'{variable_char}+'    {
!                     psqlscan_escape_variable(cur_state, yytext, yyleng, false);
                      *option_quote = ':';
                      unquoted_option_chars = 0;
                  }


  :\"{variable_char}+\"    {
!                     psqlscan_escape_variable(cur_state, yytext, yyleng, true);
                      *option_quote = ':';
                      unquoted_option_chars = 0;
                  }
--- 267,282 ----
                  }

  :'{variable_char}+'    {
!                     psqlscan_escape_variable(cur_state, yytext, yyleng,
!                                              PQUOTE_SQL_LITERAL);
                      *option_quote = ':';
                      unquoted_option_chars = 0;
                  }


  :\"{variable_char}+\"    {
!                     psqlscan_escape_variable(cur_state, yytext, yyleng,
!                                              PQUOTE_SQL_IDENT);
                      *option_quote = ':';
                      unquoted_option_chars = 0;
                  }
*************** other            .
*** 337,345 ****

  <xslashbackquote>{
      /*
!      * backticked text: copy everything until next backquote, then evaluate.
!      *
!      * XXX Possible future behavioral change: substitute for :VARIABLE?
       */

  "`"                {
--- 338,345 ----

  <xslashbackquote>{
      /*
!      * backticked text: copy everything until next backquote (expanding
!      * variable references, but doing nought else), then evaluate.
       */

  "`"                {
*************** other            .
*** 350,355 ****
--- 350,393 ----
                      BEGIN(xslasharg);
                  }

+ :{variable_char}+    {
+                     /* Possible psql variable substitution */
+                     if (cur_state->callbacks->get_variable == NULL)
+                         ECHO;
+                     else
+                     {
+                         char       *varname;
+                         char       *value;
+
+                         varname = psqlscan_extract_substring(cur_state,
+                                                              yytext + 1,
+                                                              yyleng - 1);
+                         value = cur_state->callbacks->get_variable(varname,
+                                                                    PQUOTE_PLAIN,
+                                                                    cur_state->cb_passthrough);
+                         free(varname);
+
+                         if (value)
+                         {
+                             appendPQExpBufferStr(output_buf, value);
+                             free(value);
+                         }
+                         else
+                             ECHO;
+                     }
+                 }
+
+ :'{variable_char}+'    {
+                     psqlscan_escape_variable(cur_state, yytext, yyleng,
+                                              PQUOTE_SHELL_ARG);
+                 }
+
+ :'{variable_char}*    {
+                     /* Throw back everything but the colon */
+                     yyless(1);
+                     ECHO;
+                 }
+
  {other}|\n        { ECHO; }

  }
diff --git a/src/fe_utils/psqlscan.l b/src/fe_utils/psqlscan.l
index 19b3e57..27689d7 100644
*** a/src/fe_utils/psqlscan.l
--- b/src/fe_utils/psqlscan.l
*************** other            .
*** 699,706 ****
                                                           yyleng - 1);
                      if (cur_state->callbacks->get_variable)
                          value = cur_state->callbacks->get_variable(varname,
!                                                                    false,
!                                                                    false,
                                                                     cur_state->cb_passthrough);
                      else
                          value = NULL;
--- 699,705 ----
                                                           yyleng - 1);
                      if (cur_state->callbacks->get_variable)
                          value = cur_state->callbacks->get_variable(varname,
!                                                                    PQUOTE_PLAIN,
                                                                     cur_state->cb_passthrough);
                      else
                          value = NULL;
*************** other            .
*** 737,747 ****
                  }

  :'{variable_char}+'    {
!                     psqlscan_escape_variable(cur_state, yytext, yyleng, false);
                  }

  :\"{variable_char}+\"    {
!                     psqlscan_escape_variable(cur_state, yytext, yyleng, true);
                  }

      /*
--- 736,748 ----
                  }

  :'{variable_char}+'    {
!                     psqlscan_escape_variable(cur_state, yytext, yyleng,
!                                              PQUOTE_SQL_LITERAL);
                  }

  :\"{variable_char}+\"    {
!                     psqlscan_escape_variable(cur_state, yytext, yyleng,
!                                              PQUOTE_SQL_IDENT);
                  }

      /*
*************** psqlscan_extract_substring(PsqlScanState
*** 1415,1421 ****
   */
  void
  psqlscan_escape_variable(PsqlScanState state, const char *txt, int len,
!                          bool as_ident)
  {
      char       *varname;
      char       *value;
--- 1416,1422 ----
   */
  void
  psqlscan_escape_variable(PsqlScanState state, const char *txt, int len,
!                          PsqlScanQuoteType quote)
  {
      char       *varname;
      char       *value;
*************** psqlscan_escape_variable(PsqlScanState s
*** 1423,1429 ****
      /* Variable lookup. */
      varname = psqlscan_extract_substring(state, txt + 2, len - 3);
      if (state->callbacks->get_variable)
!         value = state->callbacks->get_variable(varname, true, as_ident,
                                                 state->cb_passthrough);
      else
          value = NULL;
--- 1424,1430 ----
      /* Variable lookup. */
      varname = psqlscan_extract_substring(state, txt + 2, len - 3);
      if (state->callbacks->get_variable)
!         value = state->callbacks->get_variable(varname, quote,
                                                 state->cb_passthrough);
      else
          value = NULL;
diff --git a/src/fe_utils/string_utils.c b/src/fe_utils/string_utils.c
index d1a9ddc..dc84d32 100644
*** a/src/fe_utils/string_utils.c
--- b/src/fe_utils/string_utils.c
*************** appendByteaLiteral(PQExpBuffer buf, cons
*** 425,437 ****
--- 425,454 ----
   * arguments containing LF or CR characters.  A future major release should
   * reject those characters in CREATE ROLE and CREATE DATABASE, because use
   * there eventually leads to errors here.
+  *
+  * appendShellString() simply prints an error and dies if LF or CR appears.
+  * appendShellStringNoError() omits those characters from the result, and
+  * returns false if there were any.
   */
  void
  appendShellString(PQExpBuffer buf, const char *str)
  {
+     if (!appendShellStringNoError(buf, str))
+     {
+         fprintf(stderr,
+                 _("shell command argument contains a newline or carriage return: \"%s\"\n"),
+                 str);
+         exit(EXIT_FAILURE);
+     }
+ }
+
+ bool
+ appendShellStringNoError(PQExpBuffer buf, const char *str)
+ {
  #ifdef WIN32
      int            backslash_run_length = 0;
  #endif
+     bool        ok = true;
      const char *p;

      /*
*************** appendShellString(PQExpBuffer buf, const
*** 442,448 ****
          strspn(str, "abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789-_./:") == strlen(str))
      {
          appendPQExpBufferStr(buf, str);
!         return;
      }

  #ifndef WIN32
--- 459,465 ----
          strspn(str, "abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789-_./:") == strlen(str))
      {
          appendPQExpBufferStr(buf, str);
!         return ok;
      }

  #ifndef WIN32
*************** appendShellString(PQExpBuffer buf, const
*** 451,460 ****
      {
          if (*p == '\n' || *p == '\r')
          {
!             fprintf(stderr,
!                     _("shell command argument contains a newline or carriage return: \"%s\"\n"),
!                     str);
!             exit(EXIT_FAILURE);
          }

          if (*p == '\'')
--- 468,475 ----
      {
          if (*p == '\n' || *p == '\r')
          {
!             ok = false;
!             continue;
          }

          if (*p == '\'')
*************** appendShellString(PQExpBuffer buf, const
*** 481,490 ****
      {
          if (*p == '\n' || *p == '\r')
          {
!             fprintf(stderr,
!                     _("shell command argument contains a newline or carriage return: \"%s\"\n"),
!                     str);
!             exit(EXIT_FAILURE);
          }

          /* Change N backslashes before a double quote to 2N+1 backslashes. */
--- 496,503 ----
      {
          if (*p == '\n' || *p == '\r')
          {
!             ok = false;
!             continue;
          }

          /* Change N backslashes before a double quote to 2N+1 backslashes. */
*************** appendShellString(PQExpBuffer buf, const
*** 524,529 ****
--- 537,544 ----
      }
      appendPQExpBufferStr(buf, "^\"");
  #endif   /* WIN32 */
+
+     return ok;
  }


diff --git a/src/include/fe_utils/psqlscan.h b/src/include/fe_utils/psqlscan.h
index 0cc632b..e9c8143 100644
*** a/src/include/fe_utils/psqlscan.h
--- b/src/include/fe_utils/psqlscan.h
*************** typedef enum _promptStatus
*** 48,60 ****
      PROMPT_COPY
  } promptStatus_t;

  /* Callback functions to be used by the lexer */
  typedef struct PsqlScanCallbacks
  {
!     /* Fetch value of a variable, as a pfree'able string; NULL if unknown */
      /* This pointer can be NULL if no variable substitution is wanted */
!     char       *(*get_variable) (const char *varname, bool escape,
!                                            bool as_ident, void *passthrough);
      /* Print an error message someplace appropriate */
      /* (very old gcc versions don't support attributes on function pointers) */
  #if defined(__GNUC__) && __GNUC__ < 4
--- 48,69 ----
      PROMPT_COPY
  } promptStatus_t;

+ /* Quoting request types for get_variable() callback */
+ typedef enum
+ {
+     PQUOTE_PLAIN,                /* just return the actual value */
+     PQUOTE_SQL_LITERAL,            /* add quotes to make a valid SQL literal */
+     PQUOTE_SQL_IDENT,            /* quote if needed to make a SQL identifier */
+     PQUOTE_SHELL_ARG            /* quote if needed to be safe in a shell cmd */
+ } PsqlScanQuoteType;
+
  /* Callback functions to be used by the lexer */
  typedef struct PsqlScanCallbacks
  {
!     /* Fetch value of a variable, as a free'able string; NULL if unknown */
      /* This pointer can be NULL if no variable substitution is wanted */
!     char       *(*get_variable) (const char *varname, PsqlScanQuoteType quote,
!                                              void *passthrough);
      /* Print an error message someplace appropriate */
      /* (very old gcc versions don't support attributes on function pointers) */
  #if defined(__GNUC__) && __GNUC__ < 4
diff --git a/src/include/fe_utils/psqlscan_int.h b/src/include/fe_utils/psqlscan_int.h
index b4044e8..af62f5e 100644
*** a/src/include/fe_utils/psqlscan_int.h
--- b/src/include/fe_utils/psqlscan_int.h
*************** extern char *psqlscan_extract_substring(
*** 141,146 ****
                             const char *txt, int len);
  extern void psqlscan_escape_variable(PsqlScanState state,
                           const char *txt, int len,
!                          bool as_ident);

  #endif   /* PSQLSCAN_INT_H */
--- 141,146 ----
                             const char *txt, int len);
  extern void psqlscan_escape_variable(PsqlScanState state,
                           const char *txt, int len,
!                          PsqlScanQuoteType quote);

  #endif   /* PSQLSCAN_INT_H */
diff --git a/src/include/fe_utils/string_utils.h b/src/include/fe_utils/string_utils.h
index 6fb7f5e..c682343 100644
*** a/src/include/fe_utils/string_utils.h
--- b/src/include/fe_utils/string_utils.h
*************** extern void appendByteaLiteral(PQExpBuff
*** 42,47 ****
--- 42,48 ----
                     bool std_strings);

  extern void appendShellString(PQExpBuffer buf, const char *str);
+ extern bool appendShellStringNoError(PQExpBuffer buf, const char *str);
  extern void appendConnStrVal(PQExpBuffer buf, const char *str);
  extern void appendPsqlMetaConnect(PQExpBuffer buf, const char *dbname);