lib/parse-sexp: support phrase queries.

[notmuch] / lib / parse-sexp.cc

diff --git a/lib/parse-sexp.cc b/lib/parse-sexp.cc
index 0d2c0ba85e2cabee764ff0fba04151c3fd5efd22..0917f50566ac5104b5a9839825e05e4e27b319ea 100644 (file)
--- a/lib/parse-sexp.cc
+++ b/lib/parse-sexp.cc
@@ -2,13 +2,14 @@
 
 #if HAVE_SFSEXP
 #include "sexp.h"
-
+#include "unicode-util.h"
 
 /* _sexp is used for file scope symbols to avoid clashing with
  * definitions from sexp.h */
 
 typedef enum {
-    SEXP_FLAG_NONE = 0,
+    SEXP_FLAG_NONE     = 0,
+    SEXP_FLAG_FIELD    = 1 << 0,
 } _sexp_flag_t;
 
 typedef struct  {
@@ -26,6 +27,8 @@ static _sexp_prefix_t prefixes[] =
       SEXP_FLAG_NONE },
     { "or",             Xapian::Query::OP_OR,           Xapian::Query::MatchNothing,
       SEXP_FLAG_NONE },
+    { "subject",        Xapian::Query::OP_AND,          Xapian::Query::MatchAll,
+      SEXP_FLAG_FIELD },
     { }
 };
 
@@ -64,6 +67,36 @@ _sexp_combine_query (notmuch_database_t *notmuch,
                                sx->next, output);
 }
 
+static notmuch_status_t
+_sexp_parse_phrase (std::string term_prefix, const char *phrase, Xapian::Query &output)
+{
+    Xapian::Utf8Iterator p (phrase);
+    Xapian::Utf8Iterator end;
+    std::vector<std::string> terms;
+
+    while (p != end) {
+       Xapian::Utf8Iterator start;
+       while (p != end && ! Xapian::Unicode::is_wordchar (*p))
+           p++;
+
+       if (p == end)
+           break;
+
+       start = p;
+
+       while (p != end && Xapian::Unicode::is_wordchar (*p))
+           p++;
+
+       if (p != start) {
+           std::string word (start, p);
+           word = Xapian::Unicode::tolower (word);
+           terms.push_back (term_prefix + word);
+       }
+    }
+    output = Xapian::Query (Xapian::Query::OP_PHRASE, terms.begin (), terms.end ());
+    return NOTMUCH_STATUS_SUCCESS;
+}
+
 /* Here we expect the s-expression to be a proper list, with first
  * element defining and operation, or as a special case the empty
  * list */
@@ -76,11 +109,13 @@ _sexp_to_xapian_query (notmuch_database_t *notmuch, const _sexp_prefix_t *parent
     if (sx->ty == SEXP_VALUE) {
        std::string term = Xapian::Unicode::tolower (sx->val);
        Xapian::Stem stem = *(notmuch->stemmer);
-       if (sx->aty == SEXP_BASIC)
-           term = "Z" + stem (term);
-
-       output = Xapian::Query (term);
-       return NOTMUCH_STATUS_SUCCESS;
+       std::string term_prefix = parent ? _find_prefix (parent->name) : "";
+       if (sx->aty == SEXP_BASIC && unicode_word_utf8 (sx->val)) {
+           output = Xapian::Query ("Z" + term_prefix + stem (term));
+           return NOTMUCH_STATUS_SUCCESS;
+       } else {
+           return _sexp_parse_phrase (term_prefix, sx->val, output);
+       }
     }
 
     /* Empty list */
@@ -97,6 +132,15 @@ _sexp_to_xapian_query (notmuch_database_t *notmuch, const _sexp_prefix_t *parent
 
     for (_sexp_prefix_t *prefix = prefixes; prefix && prefix->name; prefix++) {
        if (strcmp (prefix->name, sx->list->val) == 0) {
+           if (prefix->flags & SEXP_FLAG_FIELD) {
+               if (parent) {
+                   _notmuch_database_log (notmuch, "nested field: '%s' inside '%s'\n",
+                                          prefix->name, parent->name);
+                   return NOTMUCH_STATUS_BAD_QUERY_SYNTAX;
+               }
+               parent = prefix;
+           }
+
            return _sexp_combine_query (notmuch, parent, prefix->xapian_op, prefix->initial,
                                        sx->list->next, output);
        }