Use the result of the fuzzy search in the current list as a lower bound for

the fuzzy search in the compendiums.

4 files changed, 36 insertions, 10 deletions

diff --git a/gettext-tools/src/ChangeLog b/gettext-tools/src/ChangeLog
index 68e3bd5..bfed490 100644
--- a/gettext-tools/src/ChangeLog
+++ b/gettext-tools/src/ChangeLog
@@ -1,5 +1,14 @@
 2008-09-14  Bruno Haible  <bruno@clisp.org>
 
+	* msgl-fsearch.h (message_fuzzy_index_search): Add 'lower_bound'
+	argument.
+	* msgl-fsearch.c (message_fuzzy_index_search): Likewise.
+	* msgmerge.c (definitions_search_fuzzy): Use the result of the fuzzy
+	search in the current list as a lower bound for the fuzzy search in the
+	compendiums.
+
+2008-09-14  Bruno Haible  <bruno@clisp.org>
+
 	* message.h (message_list_search_fuzzy, FUZZY_THRESHOLD): Clarify
 	documentation.
 
diff --git a/gettext-tools/src/msgl-fsearch.c b/gettext-tools/src/msgl-fsearch.c
index 8d7d541..e612f01 100644
--- a/gettext-tools/src/msgl-fsearch.c
+++ b/gettext-tools/src/msgl-fsearch.c
@@ -484,10 +484,13 @@ mult_index_list_free (struct mult_index_list *accu)
 }
 
 /* Find a good match for the given msgctxt and msgid in the given fuzzy index.
-   The match does not need to be optimal.  */
+   The match does not need to be optimal.
+   Ignore matches for which the fuzzy_search_goal_function is < LOWER_BOUND.
+   LOWER_BOUND must be >= FUZZY_THRESHOLD.  */
 message_ty *
 message_fuzzy_index_search (message_fuzzy_index_ty *findex,
-			    const char *msgctxt, const char *msgid)
+			    const char *msgctxt, const char *msgid,
+			    double lower_bound)
 {
   const char *str = msgid;
 
@@ -547,7 +550,7 @@ message_fuzzy_index_search (message_fuzzy_index_ty *findex,
 		    if (count > accu.nitems)
 		      count = accu.nitems;
 
-		    best_weight = FUZZY_THRESHOLD;
+		    best_weight = lower_bound;
 		    best_mp = NULL;
 		    for (ptr = accu.item; count > 0; ptr++, count--)
 		      {
@@ -589,7 +592,7 @@ message_fuzzy_index_search (message_fuzzy_index_ty *findex,
     if (!(lmax <= SHORT_MSG_MAX))
       abort ();
 
-    best_weight = FUZZY_THRESHOLD;
+    best_weight = lower_bound;
     best_mp = NULL;
     for (l = lmin; l <= lmax; l++)
       {
diff --git a/gettext-tools/src/msgl-fsearch.h b/gettext-tools/src/msgl-fsearch.h
index 6793cef..d5f7962 100644
--- a/gettext-tools/src/msgl-fsearch.h
+++ b/gettext-tools/src/msgl-fsearch.h
@@ -1,5 +1,5 @@
 /* Fast fuzzy searching among messages.
-   Copyright (C) 2006 Free Software Foundation, Inc.
+   Copyright (C) 2006, 2008 Free Software Foundation, Inc.
    Written by Bruno Haible <bruno@clisp.org>, 2006.
 
    This program is free software: you can redistribute it and/or modify
@@ -39,10 +39,13 @@ extern message_fuzzy_index_ty *
 				  const char *canon_charset);
 
 /* Find a good match for the given msgctxt and msgid in the given fuzzy index.
-   The match does not need to be optimal.  */
+   The match does not need to be optimal.
+   Ignore matches for which the fuzzy_search_goal_function is < LOWER_BOUND.
+   LOWER_BOUND must be >= FUZZY_THRESHOLD.  */
 extern message_ty *
        message_fuzzy_index_search (message_fuzzy_index_ty *findex,
-				   const char *msgctxt, const char *msgid);
+				   const char *msgctxt, const char *msgid,
+				   double lower_bound);
 
 /* Free a fuzzy index.  */
 extern void
diff --git a/gettext-tools/src/msgmerge.c b/gettext-tools/src/msgmerge.c
index b8df6a9..be765e9 100644
--- a/gettext-tools/src/msgmerge.c
+++ b/gettext-tools/src/msgmerge.c
@@ -767,20 +767,31 @@ definitions_search_fuzzy (definitions_ty *definitions,
 			       msgctxt, msgid);
   if (compendiums != NULL)
     {
+      double lower_bound_for_mp2;
       message_ty *mp2;
 
+      lower_bound_for_mp2 =
+	(mp1 != NULL
+	 ? fuzzy_search_goal_function (mp1, msgctxt, msgid, 0.0)
+	 : FUZZY_THRESHOLD);
+      /* This lower bound must be >= FUZZY_THRESHOLD.  */
+      if (!(lower_bound_for_mp2 >= FUZZY_THRESHOLD))
+	abort ();
+
       /* Create the fuzzy index lazily.  */
       if (definitions->comp_findex == NULL)
 	definitions_init_comp_findex (definitions);
 
       mp2 = message_fuzzy_index_search (definitions->comp_findex,
-					msgctxt, msgid);
+					msgctxt, msgid,
+					lower_bound_for_mp2);
 
       /* Choose the best among mp1, mp2.  */
       if (mp1 == NULL
 	  || (mp2 != NULL
-	      && (fuzzy_search_goal_function (mp2, msgctxt, msgid, 0.0)
-		  > fuzzy_search_goal_function (mp1, msgctxt, msgid, 0.0))))
+	      && (fuzzy_search_goal_function (mp2, msgctxt, msgid,
+					      lower_bound_for_mp2)
+		  > lower_bound_for_mp2)))
 	mp1 = mp2;
     }