feat: support numbering for preview images

examples/cli/main.cpp

@@ -81,7 +81,7 @@ struct SDCliParams {
              &metadata_format},
             {"",
              "--preview-path",
-             "path to write preview image to (default: ./preview.png). Multi-frame previews support .avi, .webm, and animated .webp",
+             "path to write preview image to (default: ./preview.png). For image generation, the filename can have %03d placeholder for sequential numbering. Multi-frame previews support .avi, .webm, and animated .webp",
              0,
              &preview_path},
             {"",
@@ -377,21 +377,47 @@ bool load_images_from_dir(const std::string dir,
     return true;
 }
 
+std::string format_frame_idx(std::string pattern, int frame_idx) {
+    std::smatch match;
+    std::string result = pattern;
+    while (std::regex_search(result, match, format_specifier_regex)) {
+        std::string specifier = match.str(1);
+        char buffer[32];
+        snprintf(buffer, sizeof(buffer), specifier.c_str(), frame_idx);
+        result.replace(match.position(1), match.length(1), buffer);
+    }
+
+    // Then replace all '%%' with '%'
+    size_t pos = 0;
+    while ((pos = result.find("%%", pos)) != std::string::npos) {
+        result.replace(pos, 2, "%");
+        pos += 1;
+    }
+    return result;
+}
+
+int continuous_preview_counter = 0;
+
 void step_callback(int step, int frame_count, sd_image_t* image, bool is_noisy, void* data) {
     (void)step;
     (void)is_noisy;
     SDCliParams* cli_params = (SDCliParams*)data;
     // is_noisy is set to true if the preview corresponds to noisy latents, false if it's denoised latents
     // unused in this app, it will either be always noisy or always denoised here
     if (frame_count == 1) {
-        if (!write_image_to_file(cli_params->preview_path,
+        fs::path path = cli_params->preview_path;
+        if (encoded_image_format_from_path(path.string()) == EncodedImageFormat::UNKNOWN)
+            path += ".png";
+        if (std::regex_search(path.string(), format_specifier_regex))
+            path = fs::path(format_frame_idx(path.string(), continuous_preview_counter++));
+        if (!write_image_to_file(path.string(),
                                  image->data,
                                  image->width,
                                  image->height,
                                  image->channel,
                                  "",
                                  cli_params->compression_quality)) {
-            LOG_ERROR("save preview image to '%s' failed", cli_params->preview_path.c_str());
+            LOG_ERROR("save preview image to '%s' failed", path.string().c_str());
         }
     } else {
         if (create_video_from_sd_images(cli_params->preview_path.c_str(), image, frame_count, cli_params->preview_fps, cli_params->compression_quality) != 0) {
@@ -400,25 +426,6 @@ void step_callback(int step, int frame_count, sd_image_t* image, bool is_noisy,
     }
 }
 
-std::string format_frame_idx(std::string pattern, int frame_idx) {
-    std::smatch match;
-    std::string result = pattern;
-    while (std::regex_search(result, match, format_specifier_regex)) {
-        std::string specifier = match.str(1);
-        char buffer[32];
-        snprintf(buffer, sizeof(buffer), specifier.c_str(), frame_idx);
-        result.replace(match.position(1), match.length(1), buffer);
-    }
-
-    // Then replace all '%%' with '%'
-    size_t pos = 0;
-    while ((pos = result.find("%%", pos)) != std::string::npos) {
-        result.replace(pos, 2, "%");
-        pos += 1;
-    }
-    return result;
-}
-
 static fs::path get_video_audio_sidecar_path(const SDCliParams& cli_params) {
     fs::path out_path     = cli_params.output_path;
     fs::path base_path    = out_path;