eit: Add optional 2nd match subexpression for subtitle (#4791)

If the regex for the subtitle contains a second subexpression, and a
match is made, use the first subexpression for the subtitle and replace
the summary with the second subexpression.

For example, a UK Freeview subtitle regex might choose, when matching a
summary 'Subtitle: Text', to set the subtitle to 'Subtitle' and set the
summary to 'Text' to avoid repetition of the subtitle.

Update the scraper test script to support a test field 'new_summary'. As
the 'uk' scraper does not include any second subexpressions, do not
update the test data for now.

Issue: #4791

diff --git a/src/epggrab/module/eit.c b/src/epggrab/module/eit.c
index f261f56e2195e5e14902ada90b2fc171227ecda9..cdd3d4da61e09e35d058d8e8cc4a34ad261af2f0 100644 (file)
--- a/src/epggrab/module/eit.c
+++ b/src/epggrab/module/eit.c
@@ -703,13 +703,21 @@ static int _eit_process_event_one
        */
       const char *summary = lang_str_get(ev.summary, ev.default_charset);
       char buffer[2048];
-      if (eit_pattern_apply_list(buffer, sizeof(buffer), summary, &eit_mod->p_scrape_subtitle)) {
+      char buffer2[2048];
+      char *bufs[2] = { buffer, buffer2 };
+      size_t sizes[2] = { sizeof(buffer), sizeof(buffer2) };
+      if (eit_pattern_apply_list_2(bufs, sizes, summary, &eit_mod->p_scrape_subtitle)) {
         tvhtrace(LS_TBL_EIT, "  scrape subtitle '%s' from '%s' using %s on channel '%s'",
                  buffer, summary, mod->id,
                  ch ? channel_get_name(ch, channel_blank_name) : "(null)");
         lang_str_t *ls = lang_str_create2(buffer, ev.default_charset);
         *save |= epg_episode_set_subtitle(ee, ls, &changes4);
         lang_str_destroy(ls);
+        if (bufs[1]) {
+            ls = lang_str_create2(buffer2, ev.default_charset);
+            *save |= epg_broadcast_set_summary(ebc, ls, &changes2);
+            lang_str_destroy(ls);
+        }
       } else {
         /* No subtitle found in summary buffer. */
         *save |= epg_episode_set_subtitle(ee, ev.summary, &changes4);

diff --git a/src/epggrab/module/eitpatternlist.c b/src/epggrab/module/eitpatternlist.c
index 76d87b59f333b3a467955416a8e021b4e0c24e47..21d8baa2e9bf741fcde44d564b7731ea328c06ab 100644 (file)
--- a/src/epggrab/module/eitpatternlist.c
+++ b/src/epggrab/module/eitpatternlist.c
@@ -16,6 +16,7 @@
  *  along with this program.  If not, see <http://www.gnu.org/licenses/>.
  */
 
+#include <assert.h>
 #include <ctype.h>
 #include "tvheadend.h"
 #include "eitpatternlist.h"
@@ -47,23 +48,45 @@ void eit_pattern_compile_list ( eit_pattern_list_t *list, htsmsg_t *l )
 
 void *eit_pattern_apply_list(char *buf, size_t size_buf, const char *text, eit_pattern_list_t *l)
 {
-  regmatch_t match[2];
+    char *b[2] = { buf, NULL };
+    size_t s[2] = { size_buf, 0 };
+    return eit_pattern_apply_list_2(b, s, text, l);
+}
+
+void *eit_pattern_apply_list_2(char *buf[2], size_t size_buf[2], const char *text, eit_pattern_list_t *l)
+{
+  regmatch_t match[3];
   eit_pattern_t *p;
   ssize_t size;
 
+  assert(buf[0]);
+  assert(text);
+
   if (!l) return NULL;
   /* search and report the first match */
   TAILQ_FOREACH(p, l, p_links)
-    if (!regexec(&p->compiled, text, 2, match, 0) && match[1].rm_so != -1) {
-      size = MIN(match[1].rm_eo - match[1].rm_so, size_buf - 1);
+    if (!regexec(&p->compiled, text, 3, match, 0) && match[1].rm_so != -1) {
+      size = MIN(match[1].rm_eo - match[1].rm_so, size_buf[0] - 1);
       if (size > 0) {
         while (isspace(text[match[1].rm_so + size - 1]))
           size--;
-        memcpy(buf, text + match[1].rm_so, size);
+        memcpy(buf[0], text + match[1].rm_so, size);
+      }
+      buf[0][size] = '\0';
+      if (match[2].rm_so != -1 && buf[1]) {
+          size = MIN(match[2].rm_eo - match[2].rm_so, size_buf[1] - 1);
+          if (size > 0) {
+              while (isspace(text[match[2].rm_so + size - 1]))
+                  size--;
+              memcpy(buf[1], text + match[2].rm_so, size);
+          }
+          buf[1][size] = '\0';
+          tvhtrace(LS_EPGGRAB,"  pattern \"%s\" matches with '%s' & '%s'", p->text, buf[0], buf[1]);
+      } else {
+          buf[1] = NULL;
+          tvhtrace(LS_EPGGRAB,"  pattern \"%s\" matches with '%s'", p->text, buf[0]);
       }
-      buf[size] = '\0';
-      tvhtrace(LS_EPGGRAB,"  pattern \"%s\" matches with '%s'", p->text, buf);
-      return buf;
+      return buf[0];
     }
   return NULL;
 }

diff --git a/src/epggrab/module/eitpatternlist.h b/src/epggrab/module/eitpatternlist.h
index a2e863d44489c4756f6b3923589d2ce15c033dc3..2f43ecadf5fa6683b4d1005a5fa643d5964dd0b4 100644 (file)
--- a/src/epggrab/module/eitpatternlist.h
+++ b/src/epggrab/module/eitpatternlist.h
@@ -38,5 +38,11 @@ void eit_pattern_compile_list ( eit_pattern_list_t *list, htsmsg_t *l );
  * Return the buf or NULL if no match.
  */
 void *eit_pattern_apply_list(char *buf, size_t size_buf, const char *text, eit_pattern_list_t *l);
+/* As eit_pattern_apply_list(), but return up to 2 matches.
+ * buf[0] & size_buf[0] are the first match, buf[1] & size_buf[1] the second.
+ * If no second match is found, set buf[1] to NULL.
+ * Return the first buf or NULL if no match.
+ */
+void *eit_pattern_apply_list_2(char *buf[2], size_t size_buf[2], const char *text, eit_pattern_list_t *l);
 void eit_pattern_free_list ( eit_pattern_list_t *l );
 #endif

diff --git a/support/eitscrape_test.py b/support/eitscrape_test.py
index af60da7053733fbb97f2cde08894ccc5faa7358f..7c68e251141bfcf5d47838f40796c5e3135f1a95 100755 (executable)
--- a/support/eitscrape_test.py
+++ b/support/eitscrape_test.py
@@ -59,14 +59,14 @@ class EITScrapeTest(object):
     self.num_failed = 0;
     self.num_ok = 0;
 
-  def run_test_case_i(self, text, reg, expect, testing):
+  def run_test_case_i(self, text, reg, expect, testing, match=1):
     """Run a test case for text using the regular expression lists in reg,
     expecting the result of a match to be expect while running a test
     case for the string testing."""
     for iter in reg:
       m = iter.search(text)
       if (m is not None):
-        result = m.group(1)
+        result = m.group(match)
         if result == expect:
           print 'OK: Got correct result of "%s" testing "%s" for "%s" using "%s"' % (result, testing, text, iter.pattern)
           self.num_ok = self.num_ok + 1
@@ -99,7 +99,7 @@ class EITScrapeTest(object):
       if key in ('age', 'genre'):
         print 'Test case contains key "%s" which is not currently tested for "%s"' % (key, test)
 
-      if key not in ('age', 'airdate', 'comment', 'episode', 'genre', 'new_subtitle', 'season', 'summary'):
+      if key not in ('age', 'airdate', 'comment', 'episode', 'genre', 'new_subtitle', 'new_summary', 'season', 'summary'):
         print 'Test case contains invalid key "%s" (possible typo) for "%s"' % (key, test)
         raise SyntaxWarning('Test case contains invalid/unknown key "%s" (possible typo) for "%s"' % (key, test))
 
@@ -116,6 +116,8 @@ class EITScrapeTest(object):
       self.run_test_case_i(text, airdate_reg, test['airdate'], "airdate")
     if test.has_key('new_subtitle'):
       self.run_test_case_i(text, subtitle_reg, test['new_subtitle'], "new_subtitle")
+    if test.has_key('new_summary'):
+      self.run_test_case_i(text, subtitle_reg, test['new_summary'], "new_summary", match=2)