r.info: Output comments as one string in JSON (#4216)

The part of history which appears under the comments key can be multiple lines and is stored in a format which limits line length. This is basically how r.info shows it by default in the plain output for humans. Long lines can be split and continuation is marked using a backslash.

This adds a function which puts the continued lines back together and adds newlines between the other lines. There is no newline at the end. This string is then serialized in JSON instead of the original list of strings.

This also avoids serializing the history twice when the h flag is not used, but it does change the behavior for what keys are included.

raster/r.info/main.c

@@ -40,6 +40,7 @@ enum OutputFormat { PLAIN, JSON };
 /* local prototypes */
 static void format_double(const double, char *);
 static void compose_line(FILE *, const char *, ...);
+static char *history_as_string(struct History *hist);
 
 int main(int argc, char **argv)
 {
@@ -773,16 +774,14 @@ int main(int argc, char **argv)
                     json_object_set_string(
                         root_object, "description",
                         Rast_get_history(&hist, HIST_KEYWRD));
-                    JSON_Value *comments_value = json_value_init_array();
-                    JSON_Array *comments = json_array(comments_value);
-                    if (Rast_history_length(&hist)) {
-                        for (i = 0; i < Rast_history_length(&hist); i++) {
-                            json_array_append_string(
-                                comments, Rast_history_line(&hist, i));
-                        }
+                    char *buffer = history_as_string(&hist);
+                    if (buffer) {
+                        json_object_set_string(root_object, "comments", buffer);
+                        G_free(buffer);
+                    }
+                    else {
+                        json_object_set_null(root_object, "comments");
                     }
-                    json_object_set_value(root_object, "comments",
-                                          comments_value);
                 }
                 else {
                     json_object_set_null(root_object, "source1");
@@ -794,7 +793,7 @@ int main(int argc, char **argv)
             }
         }
 
-        if (hflag->answer || format == JSON) {
+        if (hflag->answer) {
             if (hist_ok) {
                 switch (format) {
                 case PLAIN:
@@ -823,16 +822,15 @@ int main(int argc, char **argv)
                     json_object_set_string(
                         root_object, "description",
                         Rast_get_history(&hist, HIST_KEYWRD));
-                    JSON_Value *comments_value = json_value_init_array();
-                    JSON_Array *comments = json_array(comments_value);
-                    if (Rast_history_length(&hist)) {
-                        for (i = 0; i < Rast_history_length(&hist); i++) {
-                            json_array_append_string(
-                                comments, Rast_history_line(&hist, i));
-                        }
+                    char *buffer = history_as_string(&hist);
+                    if (buffer) {
+                        json_object_set_string(root_object, "comments", buffer);
+                        G_free(buffer);
+                    }
+                    else {
+                        json_object_set_null(root_object, "comments");
                     }
-                    json_object_set_value(root_object, "comments",
-                                          comments_value);
+
                     break;
                 }
             }
@@ -874,3 +872,46 @@ static void compose_line(FILE *out, const char *fmt, ...)
     printline(line);
     G_free(line);
 }
+
+static char *history_as_string(struct History *hist)
+{
+    int history_length = Rast_history_length(hist);
+    char *buffer = NULL;
+    if (history_length) {
+        size_t buffer_size = 0;
+        size_t total_length = 0;
+        for (int i = 0; i < history_length; i++) {
+            const char *line = Rast_history_line(hist, i);
+            size_t line_length = strlen(line);
+
+            // +1 for the null character
+            size_t required_size = total_length + line_length + 1;
+            if (required_size > buffer_size) {
+                // This is heuristic for reallocation based on remaining
+                // iterations and current size which is a good estimate for the
+                // first iteration and possible overshoot later on reducing the
+                // number of reallocations.
+                buffer_size = required_size * (history_length - i);
+                buffer = (char *)G_realloc(buffer, buffer_size);
+                if (total_length == 0)
+                    buffer[0] = '\0';
+            }
+            if (line_length >= 1 && line[line_length - 1] == '\\') {
+                // Ending backslash is line continuation.
+                strncat(buffer, line, line_length - 1);
+                total_length += line_length - 1;
+            }
+            else {
+                strncat(buffer, line, line_length);
+                total_length += line_length;
+                if (i < history_length - 1) {
+                    // Add newline to separate lines, but don't and newline at
+                    // the end of last (or only) line.
+                    strcat(buffer, "\n");
+                    ++total_length;
+                }
+            }
+        }
+    }
+    return buffer;
+}

raster/r.info/testsuite/test_r_info.py

@@ -47,7 +47,7 @@ def setUpClass(cls):
             "source1": "",
             "source2": "",
             "description": "generated by r.mapcalc",
-            "comments": ["1 * lakes_large"],
+            "comments": "1 * lakes_large",
         }
 
     @classmethod
@@ -136,6 +136,51 @@ def test_sflag_format_json(self):
         self._test_format_json_helper(module, expected_json_with_stats)
 
 
+class TestComments(TestCase):
+    """Check printing of comments"""
+
+    def test_comments_one_line(self):
+        """Check that one line is text without any newlines"""
+        module = SimpleModule("r.info", map="lakes", format="json")
+        self.runModule(module)
+        result = json.loads(module.outputs.stdout)
+        self.assertFalse(result["comments"].endswith("\n"))
+        self.assertEqual(result["comments"], "1 * lakes_large")
+
+    def test_comments_continued_line(self):
+        """Check that continued lines are merged"""
+        module = SimpleModule("r.info", map="elevation", format="json")
+        self.runModule(module)
+        result = json.loads(module.outputs.stdout)
+        self.assertFalse(result["comments"].endswith("\n"))
+        self.assertEqual(
+            result["comments"],
+            'r.proj input="ned03arcsec" location="northcarolina_latlong" '
+            'mapset="helena" output="elev_ned10m" method="cubic" resolution=10',
+        )
+
+    def test_comments_multiple_lines(self):
+        """Check multiple lines are preserved"""
+        module = SimpleModule("r.info", map="lsat7_2002_30", format="json")
+        self.runModule(module)
+        result = json.loads(module.outputs.stdout)
+        self.assertFalse(result["comments"].endswith("\n"))
+
+        lines = result["comments"].splitlines()
+        self.assertEqual(
+            len(lines),
+            31,
+        )
+        self.assertEqual(
+            lines[0],
+            'r.in.gdal input="p016r035_7t20020524_z17_nn30_nc_spm_wake.tif" output="lsat7_2002_30"',
+        )
+        self.assertEqual(
+            lines[-1],
+            'i.landsat.rgb "b=lsat7_2002_10" "g=lsat7_2002_20" "r=lsat7_2002_30"',
+        )
+
+
 if __name__ == "__main__":
     from grass.gunittest.main import test