From: Gerhard Sittig <redacted>
Date: Sun, 18 Jun 2017 17:12:11 +0000 (+0200)
Subject: decoder: Unbreak execution of Decoder.wait() with empty condition
X-Git-Url: https://sigrok.org/gitaction?a=commitdiff_plain;h=96e1cee7eb34f5ff5b6da6011ab89ad8bc1ac794;p=libsigrokdecode.git

decoder: Unbreak execution of Decoder.wait() with empty condition

The Decoder.wait() method works as expected when non-empty conditions
are specified by the caller. For empty conditions the implementation was
incomplete, and ended up in an infinite loop because the sample number
got stuck in the current location. Code review revealed more issues like
not getting more input data chunks when needed.

Detect when empty wait() conditions were specified, and re-use existing
code paths for match handling as much as possible. This is achieved by
the manual creation of a SKIP term with the appropriate count value.
It's assumed that naive decoder implementations will run this kind of
condition-less query for every individual sample, which means that
efficiency is important.

Make sample number 0 available to condition-less calls, too. Don't skip
the first sample in the input stream.
---

diff --git a/type_decoder.c b/type_decoder.c
index 9ed3339..d7b690f 100644
--- a/type_decoder.c
+++ b/type_decoder.c
@@ -585,9 +585,45 @@ static int set_new_condition_list(PyObject *self, PyObject *args)
 	return ret;
 }
 
+/**
+ * Create a SKIP condition list for condition-less .wait() calls.
+ *
+ * @param di Decoder instance.
+ * @param count Number of samples to skip.
+ *
+ * @retval SRD_OK The new condition list was set successfully.
+ * @retval SRD_ERR There was an error setting the new condition list.
+ *                 The contents of di->condition_list are undefined.
+ *
+ * This routine is a reduced and specialized version of the @ref
+ * set_new_condition_list() and @ref create_term_list() routines which
+ * gets invoked when .wait() was called without specifications for
+ * conditions. This minor duplication of the SKIP term list creation
+ * simplifies the logic and avoids the creation of expensive Python
+ * objects with "constant" values which the caller did not pass in the
+ * first place. It results in maximum sharing of match handling code
+ * paths.
+ */
+static int set_skip_condition(struct srd_decoder_inst *di, uint64_t count)
+{
+	struct srd_term *term;
+	GSList *term_list;
+
+	condition_list_free(di);
+	term = g_malloc0(sizeof(*term));
+	term->type = SRD_TERM_SKIP;
+	term->num_samples_to_skip = count;
+	term->num_samples_already_skipped = 0;
+	term_list = g_slist_append(NULL, term);
+	di->condition_list = g_slist_append(di->condition_list, term_list);
+
+	return SRD_OK;
+}
+
 static PyObject *Decoder_wait(PyObject *self, PyObject *args)
 {
 	int ret;
+	uint64_t skip_count;
 	unsigned int i;
 	gboolean found_match;
 	struct srd_decoder_inst *di;
@@ -607,10 +643,25 @@ static PyObject *Decoder_wait(PyObject *self, PyObject *args)
 		return NULL;
 	}
 	if (ret == 9999) {
-		/* Empty condition list, automatic match. */
-		PyObject_SetAttrString(di->py_inst, "matched", Py_None);
-		/* Leave self.samplenum unchanged (== di->abs_cur_samplenum). */
-		return get_current_pinvalues(di);
+		/*
+		 * Empty condition list, automatic match. Arrange for the
+		 * execution of regular match handling code paths such that
+		 * the next available sample is returned to the caller.
+		 * Make sure to skip one sample when "anywhere within the
+		 * stream", yet make sure to not skip sample number 0.
+		 */
+		if (di->abs_cur_samplenum)
+			skip_count = 1;
+		else if (!di->condition_list)
+			skip_count = 0;
+		else
+			skip_count = 1;
+		ret = set_skip_condition(di, skip_count);
+		if (ret < 0) {
+			srd_dbg("%s: %s: Cannot setup condition-less wait().",
+				di->inst_id, __func__);
+			return NULL;
+		}
 	}
 
 	while (1) {