v2v: Move the shell_unquote function to src/utils.c.

This function is useful outside virt-v2v so move it into the core
utilities library.

The function has been rewritten from OCaml into C, but it should be
functionally identical.

src/guestfs-internal-frontend.h

@@ -90,6 +90,7 @@ extern void guestfs_int_fadvise_random (int fd);
 extern void guestfs_int_fadvise_noreuse (int fd);
 //extern void guestfs_int_fadvise_dontneed (int fd);
 //extern void guestfs_int_fadvise_willneed (int fd);
+extern char *guestfs_int_shell_unquote (const char *str);
 
 struct uefi_firmware {
   const char *code;		/* code file (NULL = end of list) */

src/utils.c

@@ -539,3 +539,63 @@ guestfs_int_fadvise_willneed (int fd)
 #endif
 }
 #endif
+
+/**
+ * Unquote a shell-quoted string.
+ *
+ * Augeas passes strings to us which may be quoted, eg. if they come
+ * from files in F</etc/sysconfig>.  This function can do simple
+ * unquoting of these strings.
+ *
+ * Note this function does not do variable substitution, since that is
+ * impossible without knowing the file context and indeed the
+ * environment under which the shell script is run.  Configuration
+ * files should not use complex quoting.
+ *
+ * C<str> is the input string from Augeas, a string that may be
+ * single- or double-quoted or may not be quoted.  The returned string
+ * is unquoted, and must be freed by the caller.  C<NULL> is returned
+ * on error and C<errno> is set accordingly.
+ *
+ * For information on double-quoting in bash, see
+ * L<https://www.gnu.org/software/bash/manual/html_node/Double-Quotes.html>
+ */
+char *
+guestfs_int_shell_unquote (const char *str)
+{
+  size_t len = strlen (str);
+  char *ret;
+
+  if (len >= 2) {
+    if (str[0] == '\'' && str[len-1] == '\'') {
+                                /* single quoting */
+      ret = strndup (&str[1], len-2);
+      if (ret == NULL)
+        return NULL;
+      return ret;
+    }
+    else if (str[0] == '"' && str[len-1] == '"') {
+                                /* double quoting */
+      size_t i, j;
+
+      ret = malloc (len + 1);   /* strings always get smaller */
+      if (ret == NULL)
+        return NULL;
+
+      for (i = 1, j = 0; i < len-1 /* ignore final quote */; ++i, ++j) {
+        if (i < len-2 /* ignore final char before final quote */ &&
+            str[i] == '\\' &&
+            (str[i+1] == '$' || str[i+1] == '`' || str[i+1] == '"' ||
+             str[i+1] == '\\' || str[i+1] == '\n'))
+          ++i;
+        ret[j] = str[i];
+      }
+
+      ret[j] = '\0';
+
+      return ret;
+    }
+  }
+
+  return strdup (str);
+}

v2v/convert_linux.ml

@@ -516,7 +516,7 @@ let rec convert ~keep_serial_console (g : G.guestfs) inspect source rcaps =
         fun line ->
           if Str.string_match rex line 0 then (
             let path = Str.matched_group 1 line in
-            let path = Linux.shell_unquote path in
+            let path = Utils.shell_unquote path in
             if String.length path >= 1 && path.[0] = '/' then (
               let vboxuninstall = path ^ "/uninstall.sh" in
               Some vboxuninstall

v2v/linux.ml

@@ -178,47 +178,3 @@ let rec file_owner g inspect path =
 and is_file_owned g inspect path =
   try file_owner g inspect path; true
   with Not_found -> false
-
-let rec shell_unquote str =
-  let len = String.length str in
-  if len >= 2 then (
-    if String.is_prefix str "'" && String.is_suffix str "'" then
-      String.sub str 1 (len-2)
-    else if String.is_prefix str "\"" && String.is_suffix str "\"" then
-      shell_unquote_double str len
-    else
-      str
-  )
-  else str
-
-(* https://www.gnu.org/software/bash/manual/html_node/Double-Quotes.html
- * but note we don't do any variable expansion etc so really we just
- * handle backslash here.
- *)
-and shell_unquote_double str len =
-  let i = ref 1 and j = ref 0 in
-  while !i < len-1 (* ignore final quote *) do
-    if is_backslash_sequence str !i len then
-      incr i;
-    incr i;
-    incr j
-  done;
-
-  let outlen = !j in
-  let outstr = String.create outlen in
-  let i = ref 1 and j = ref 0 in
-  while !i < len-1 do
-    if is_backslash_sequence str !i len then
-      incr i;
-    outstr.[!j] <- str.[!i];
-    incr i;
-    incr j
-  done;
-
-  outstr
-
-and is_backslash_sequence str i len =
-  i < len-2 (* ignore final character before the final quote *) &&
-  str.[i] = '\\' &&
-    (str.[i+1] = '$' || str.[i+1] = '`' || str.[i+1] = '"'
-     || str.[i+1] = '\\' || str.[i+1] = '\n')

v2v/linux.mli

@@ -39,11 +39,3 @@ val file_owner : Guestfs.guestfs -> Types.inspect -> string -> string
 
 val is_file_owned : Guestfs.guestfs -> Types.inspect -> string -> bool
 (** Returns true if the file is owned by an installed package. *)
-
-val shell_unquote : string -> string
-(** If the string looks like a shell quoted string, then attempt to
-    unquote it.
-
-    This is just intended to deal with quoting in configuration files
-    (like ones under /etc/sysconfig), and it doesn't deal with some
-    situations such as $variable interpolation. *)

v2v/utils-c.c

@@ -21,12 +21,20 @@
 #include <stdio.h>
 #include <stdlib.h>
 #include <unistd.h>
+#include <errno.h>
 
 #include <caml/alloc.h>
 #include <caml/fail.h>
 #include <caml/memory.h>
 #include <caml/mlvalues.h>
 
+#ifdef HAVE_CAML_UNIXSUPPORT_H
+#include <caml/unixsupport.h>
+#else
+#define Nothing ((value) 0)
+extern void unix_error (int errcode, char * cmdname, value arg) Noreturn;
+#endif
+
 #include "guestfs.h"
 #include "guestfs-internal-frontend.h"
 
@@ -104,3 +112,19 @@ v2v_utils_aavmf_firmware (value unitv)
 {
   return get_firmware (guestfs_int_aavmf_firmware);
 }
+
+value
+v2v_utils_shell_unquote (value strv)
+{
+  CAMLparam1 (strv);
+  CAMLlocal1 (retv);
+  char *ret;
+
+  ret = guestfs_int_shell_unquote (String_val (strv));
+  if (ret == NULL)
+    unix_error (errno, (char *) "guestfs_int_shell_unquote", Nothing);
+
+  retv = caml_copy_string (ret);
+  free (ret);
+  CAMLreturn (retv);
+}

v2v/utils.ml

@@ -142,6 +142,8 @@ let du filename =
   | line::_ -> Int64.of_string line
   | [] -> invalid_arg filename
 
+external shell_unquote : string -> string = "v2v_utils_shell_unquote"
+
 (* The following functions are only exported for unit tests. *)
 module UNIT_TESTS = struct
   let ovmf_i386_firmware = ovmf_i386_firmware

v2v/utils.mli

@@ -66,6 +66,14 @@ val du : string -> int64
     This can raise either [Failure] or [Invalid_argument] in case
     of errors. *)
 
+val shell_unquote : string -> string
+(** If the string looks like a shell quoted string, then attempt to
+    unquote it.
+
+    This is just intended to deal with quoting in configuration files
+    (like ones under /etc/sysconfig), and it doesn't deal with some
+    situations such as $variable interpolation. *)
+
 (**/**)
 
 (* The following functions are only exported for unit tests. *)

v2v/v2v_unit_tests.ml

@@ -767,16 +767,16 @@ let test_virtio_iso_path_matches_guest_os ctx =
 
 let test_shell_unquote ctx =
   let printer = identity in
-  assert_equal ~printer "a" (Linux.shell_unquote "a");
-  assert_equal ~printer "b" (Linux.shell_unquote "'b'");
-  assert_equal ~printer "c" (Linux.shell_unquote "\"c\"");
-  assert_equal ~printer "dd" (Linux.shell_unquote "\"dd\"");
-  assert_equal ~printer "e\\e" (Linux.shell_unquote "\"e\\\\e\"");
-  assert_equal ~printer "f\\" (Linux.shell_unquote "\"f\\\\\"");
-  assert_equal ~printer "\\g" (Linux.shell_unquote "\"\\\\g\"");
-  assert_equal ~printer "h\\-h" (Linux.shell_unquote "\"h\\-h\"");
-  assert_equal ~printer "i`" (Linux.shell_unquote "\"i\\`\"");
-  assert_equal ~printer "j\"" (Linux.shell_unquote "\"j\\\"\"")
+  assert_equal ~printer "a" (Utils.shell_unquote "a");
+  assert_equal ~printer "b" (Utils.shell_unquote "'b'");
+  assert_equal ~printer "c" (Utils.shell_unquote "\"c\"");
+  assert_equal ~printer "dd" (Utils.shell_unquote "\"dd\"");
+  assert_equal ~printer "e\\e" (Utils.shell_unquote "\"e\\\\e\"");
+  assert_equal ~printer "f\\" (Utils.shell_unquote "\"f\\\\\"");
+  assert_equal ~printer "\\g" (Utils.shell_unquote "\"\\\\g\"");
+  assert_equal ~printer "h\\-h" (Utils.shell_unquote "\"h\\-h\"");
+  assert_equal ~printer "i`" (Utils.shell_unquote "\"i\\`\"");
+  assert_equal ~printer "j\"" (Utils.shell_unquote "\"j\\\"\"")
 
 let test_find_uefi_firmware ctx =
   let rec printer = function
@@ -815,7 +815,7 @@ let suite =
       "Utils.drive_index" >:: test_drive_index;
       "Windows_virtio.virtio_iso_path_matches_guest_os" >::
         test_virtio_iso_path_matches_guest_os;
-      "Linux.shell_unquote" >:: test_shell_unquote;
+      "Utils.shell_unquote" >:: test_shell_unquote;
       "Utils.find_uefi_firmware" >:: test_find_uefi_firmware;
     ]