screenkit 0.0.0 → 0.0.2

data/lib/screenkit/exporter/segment.rb

@@ -40,7 +40,7 @@ module ScreenKit
       end
 
       def voiceover_path
-        return episode.mute_sound_path unless episode.tts?
+        return episode.mute_sound_path unless episode.tts_available?
 
         dir = episode.root_dir.join("voiceovers")
         dir.glob("#{prefix}.{#{ContentType.audio.join(',')}}").first ||
@@ -75,15 +75,20 @@ module ScreenKit
             .export(video_path)
         when *ContentType.demotape
           Exporter::Demotape
-            .new(demotape_path: content_path, log_path:)
-            .export(video_path)
+            .new(
+              demotape_path: content_path,
+              log_path:,
+              options: episode.demotape_options
+            ).export(video_path)
         else
           raise "Unsupported content type: #{content_path.extname}"
         end
       end
 
       def crossfade_duration
-        episode.scenes.fetch(:segment).fetch(:crossfade_duration, 0.5)
+        Duration.parse(
+          episode.scenes.fetch(:segment).fetch(:crossfade_duration, 0.5)
+        )
       end
 
       def merge_audio_and_video(log_path:)
@@ -112,79 +117,19 @@ module ScreenKit
 
         audio_mix_inputs = ["[1:a]"]
         animation_duration = 0.2
+        current_input_index = 2 # Start after video (0) and voiceover (1)
 
         callouts.each_with_index do |callout, index|
-          type = callout[:type].to_sym
-          callout_config = episode.project_config.callouts[type]
-          in_sound = Sound.new(input: callout_config[:in_transition][:sound],
-                               source: episode.source)
-          out_sound = Sound.new(input: callout_config[:out_transition][:sound],
-                                source: episode.source)
-
-          starts_at = callout[:starts_at]
-          max_duration = [content_duration - 0.2, 0].max
-          duration = callout[:duration]
-                     .clamp(0, max_duration)
-                     .round(half: :down)
-          ends_at = starts_at + duration
-          callout_image_path = episode.output_dir.join("callouts",
-                                                       "#{prefix}-#{index}.png")
-          image_width, image_height = image_size(callout_image_path)
-
-          x, y = calculate_position(
-            anchor: Anchor.new(callout_config[:anchor]),
-            margin: Spacing.new(callout_config[:margin] || 0),
-            width: image_width,
-            height: image_height
-          )
-
-          inputs += [
-            "-loop", "1",
-            "-t", duration,
-            "-i", callout_image_path,
-
-            # Add sound for transition in
-            "-i", in_sound.path,
-
-            # Add sound for transition out
-            "-i", out_sound.path
-          ]
-
-          input_stream = "v#{index}"
-          output_stream = "v#{index + 1}"
-          callout_index = 2 + (index * 3)
-
-          animation_filters = AnimationFilters.new(
-            content_duration:,
-            callout_index:,
-            input_stream:,
-            output_stream:,
+          current_input_index = process_callout(
+            callout:,
             index:,
-            starts_at:,
-            ends_at:,
-            x:,
-            y:,
+            content_duration:,
             animation_duration:,
-            image_width:,
-            image_height:
-          ).send(callout_config[:animation])
-
-          filters.concat(animation_filters[:video])
-
-          # Delay and mix callout sounds
-          in_index = callout_index + 1
-          out_index = callout_index + 2
-
-          filters << "[#{in_index}:a]volume=#{in_sound.volume}," \
-                     "adelay=#{(starts_at * 1000).to_i}|" \
-                     "#{(starts_at * 1000).to_i}[in_#{index}]"
-          filters << "[#{out_index}:a]volume=#{out_sound.volume}," \
-                     "adelay=#{(animation_filters[:out_start] * 1000).to_i}|" \
-                     "#{(animation_filters[:out_start] * 1000).to_i}" \
-                     "[out_#{index}]"
-
-          audio_mix_inputs << "[in_#{index}]"
-          audio_mix_inputs << "[out_#{index}]"
+            inputs:,
+            filters:,
+            audio_mix_inputs:,
+            current_input_index:
+          )
         end
 
         # Mix all audio streams (voiceover + all sounds)
@@ -235,7 +180,7 @@ module ScreenKit
       def create_voiceover(log_path:)
         return if voiceover_path&.file? && !episode.options.overwrite
         return unless script_path.file?
-        return unless episode.tts?
+        return unless episode.tts_available?
 
         FileUtils.mkdir_p(voiceover_path.dirname)
 
@@ -253,6 +198,257 @@ module ScreenKit
 
         format(path.to_s, prefix:) if path
       end
+
+      def process_callout(
+        callout:,
+        index:,
+        content_duration:,
+        animation_duration:,
+        inputs:,
+        filters:,
+        audio_mix_inputs:,
+        current_input_index:
+      )
+        type = callout[:type].to_sym
+        callout_config = episode.callout_styles[type]
+        in_sound = Sound.new(input: callout_config[:in_transition][:sound],
+                             source: episode.source)
+        out_sound = Sound.new(input: callout_config[:out_transition][:sound],
+                              source: episode.source)
+
+        starts_at = TimeFormatter.parse(callout[:starts_at])
+        max_duration = [content_duration - 0.2, 0].max
+        duration = Duration.parse(callout[:duration])
+                           .clamp(0, max_duration)
+                           .round(half: :down)
+        ends_at = starts_at + duration
+        callout_duration = ends_at - starts_at
+
+        callout_path = find_callout_path(index)
+        video_callout = video_callout?(callout_path)
+        has_video_audio = has_audio?(callout_path)
+
+        callout_width, callout_height = image_size(callout_path)
+        x, y = calculate_callout_position(
+          video_callout:,
+          callout_config:,
+          callout_width:,
+          callout_height:
+        )
+
+        callout_index, current_input_index = add_callout_inputs(
+          inputs:,
+          current_input_index:,
+          callout_path:,
+          video_callout:,
+          has_video_audio:,
+          duration:
+        )
+
+        in_sound_index, out_sound_index, current_input_index =
+          add_transition_sound_inputs(
+            inputs:,
+            current_input_index:,
+            video_callout:,
+            in_sound:,
+            out_sound:
+          )
+
+        animation_filters = add_video_filters(
+          filters:,
+          index:,
+          callout_index:,
+          callout_config:,
+          video_callout:,
+          content_duration:,
+          starts_at:,
+          ends_at:,
+          x:,
+          y:,
+          animation_duration:,
+          callout_width:,
+          callout_height:
+        )
+
+        add_audio_filters(
+          filters:,
+          audio_mix_inputs:,
+          index:,
+          callout_index:,
+          video_callout:,
+          has_video_audio:,
+          starts_at:,
+          callout_duration:,
+          in_sound:,
+          out_sound:,
+          in_sound_index:,
+          out_sound_index:,
+          animation_filters:
+        )
+
+        current_input_index
+      end
+
+      def find_callout_path(index)
+        callout_path = episode.output_dir.join("callouts").glob(
+          "#{prefix}-#{index}.{png,#{ContentType.video.join(',')}}"
+        ).first
+
+        raise "Callout file not found for #{prefix}-#{index}" unless callout_path
+
+        callout_path
+      end
+
+      def video_callout?(callout_path)
+        ContentType.video.include?(callout_path.extname.delete_prefix("."))
+      end
+
+      def calculate_callout_position(
+        video_callout:,
+        callout_config:,
+        callout_width:,
+        callout_height:
+      )
+        # For video callouts, ignore anchor/margin and position at 0,0
+        # (assumes videos are already properly sized and positioned)
+        if video_callout
+          [0, 0]
+        else
+          calculate_position(
+            anchor: Anchor.new(callout_config[:anchor]),
+            margin: Spacing.new(callout_config[:margin] || 0),
+            width: callout_width,
+            height: callout_height
+          )
+        end
+      end
+
+      def add_callout_inputs(
+        inputs:,
+        current_input_index:,
+        callout_path:,
+        video_callout:,
+        has_video_audio:,
+        duration:
+      )
+        callout_index = current_input_index
+
+        if video_callout
+          # Don't use -t for videos, let them play naturally
+          inputs << "-i" << callout_path
+          current_input_index += 1
+
+          # Add mute audio if video has no audio
+          unless has_video_audio
+            inputs << "-t" << duration << "-i" << episode.mute_sound_path
+            current_input_index += 1
+          end
+        else
+          inputs << "-loop" << "1" << "-t" << duration << "-i" << callout_path
+          current_input_index += 1
+        end
+
+        [callout_index, current_input_index]
+      end
+
+      def add_transition_sound_inputs(
+        inputs:,
+        current_input_index:,
+        video_callout:,
+        in_sound:,
+        out_sound:
+      )
+        # Add transition sounds (only for non-video callouts)
+        return [nil, nil, current_input_index] if video_callout
+
+        in_sound_index = current_input_index
+        inputs << "-i" << in_sound.path
+        current_input_index += 1
+
+        out_sound_index = current_input_index
+        inputs << "-i" << out_sound.path
+        current_input_index += 1
+
+        [in_sound_index, out_sound_index, current_input_index]
+      end
+
+      def add_video_filters(
+        filters:,
+        index:,
+        callout_index:,
+        callout_config:,
+        video_callout:,
+        content_duration:,
+        starts_at:,
+        ends_at:,
+        x:,
+        y:,
+        animation_duration:,
+        callout_width:,
+        callout_height:
+      )
+        input_stream = "v#{index}"
+        output_stream = "v#{index + 1}"
+        animation_method = video_callout ? :video : callout_config[:animation]
+
+        animation_filters = AnimationFilters.new(
+          content_duration:,
+          callout_index:,
+          input_stream:,
+          output_stream:,
+          index:,
+          starts_at:,
+          ends_at:,
+          x:,
+          y:,
+          animation_duration:,
+          image_width: callout_width,
+          image_height: callout_height
+        ).send(animation_method)
+
+        filters.concat(animation_filters[:video])
+
+        animation_filters
+      end
+
+      def add_audio_filters(
+        filters:,
+        audio_mix_inputs:,
+        index:,
+        callout_index:,
+        video_callout:,
+        has_video_audio:,
+        starts_at:,
+        callout_duration:,
+        in_sound:,
+        out_sound:,
+        in_sound_index:,
+        out_sound_index:,
+        animation_filters:
+      )
+        # Mix video audio if present
+        if video_callout && has_video_audio
+          filters <<
+            "[#{callout_index}:a]atrim=end=#{callout_duration}," \
+            "asetpts=PTS-STARTPTS,adelay=#{(starts_at * 1000).to_i}|" \
+            "#{(starts_at * 1000).to_i}[video_audio_#{index}]"
+          audio_mix_inputs << "[video_audio_#{index}]"
+        end
+
+        # For non-video callouts, add transition sounds
+        return if video_callout
+
+        filters << "[#{in_sound_index}:a]volume=#{in_sound.volume}," \
+                   "adelay=#{(starts_at * 1000).to_i}|" \
+                   "#{(starts_at * 1000).to_i}[in_#{index}]"
+        filters << "[#{out_sound_index}:a]volume=#{out_sound.volume}," \
+                   "adelay=#{(animation_filters[:out_start] * 1000).to_i}|" \
+                   "#{(animation_filters[:out_start] * 1000).to_i}" \
+                   "[out_#{index}]"
+
+        audio_mix_inputs << "[in_#{index}]"
+        audio_mix_inputs << "[out_#{index}]"
+      end
     end
   end
 end

data/lib/screenkit/generators/episode/callouts/001.yml

@@ -0,0 +1,12 @@
+---
+- type: default
+  title: ScreenKit
+  body: https://github.com/fnando/screenkit
+  duration: 5s
+  starts_at: 00:00:02
+  width: 800
+
+- type: inline_block
+  text: gem install screenkit
+  duration: 5s
+  starts_at: 00:00:02

data/lib/screenkit/generators/episode/config.yml.erb

@@ -1,5 +1,5 @@
 ---
-# yaml-language-server: $schema=../../../screenkit/lib/screenkit/schemas/episode.json
+# yaml-language-server: $schema=https://screenkit.dev/schemas/episode.json
 
 # The episode title. When there are no linebreaks, they will be inferred based
 # on the approximate text width.
@@ -10,7 +10,7 @@ title: "CREATING SCREENCASTS WITH SCREENKIT"
 #
 # The project's configuration can define the callout structure, while the
 # episode configuration can define the content and timing.
-callouts:
+callouts_styles:
   - type: info
     title: "ScreenKit"
     body: "Visit https://github.com/fnando/screenkit to learn more."
@@ -45,7 +45,7 @@ callouts:
 #     title:
 #       x: 100
 #       y: 300
-#       font_path: lib/screenkit/generators/project/resources/fonts/opensans/OpenSans-ExtraBold.ttf
+#       font_path: open-sans/OpenSans-ExtraBold.ttf
 #       size: 144
 #       color: "#ffffff"
 #
@@ -53,7 +53,7 @@ callouts:
 #     # Works best with a transparent PNG file in high resolution (at least 2x the
 #     # size it'll be displayed).
 #     logo:
-#       path: lib/screenkit/generators/project/resources/images/logo.png
+#       path: images/logo.png
 #       x: 100
 #       y: 200
 #       width: 200
@@ -62,7 +62,7 @@ callouts:
 #     # The sound to be played along with the logo in the intro.
 #     # This must correspond to a sound file in the sounds directory, with any
 #     # audio extension (e.g. mp3, m4a, wav).
-#     sound: lib/screenkit/generators/project/resources/sounds/chime.mp3
+#     sound: sounds/chime.mp3
 #
 #   outro:
 #     # The duration of the outro scene in seconds.
@@ -78,11 +78,11 @@ callouts:
 #     background: "#100f50"
 #
 #     # The sound to be played along with the logo in the outro.
-#     sound: lib/screenkit/generators/project/resources/sounds/chime.mp3
+#     sound: sounds/chime.mp3
 #
 #     # The logo to be displayed in the video outro.
 #     logo:
-#       path: lib/screenkit/generators/project/resources/images/logo.png
+#       path: images/logo.png
 #       x: center
 #       y: center
 #       width: 500
@@ -95,7 +95,7 @@ callouts:
 # When present, this will have higher precedence over the project's
 # configuration, so you can have different intro/outro segments for each
 # episode.
-# voice:
+# tts:
 #   engine: elevenlabs
 #   voice_id: 56AoDkrOh6qfVPDXZ7Pt
 #   language_code: en

data/lib/screenkit/generators/project/screenkit.yml

@@ -1,5 +1,6 @@
 ---
 # yaml-language-server: $schema=../../schemas/project.json
+## yaml-language-server: $schema=https://screenkit.dev/schemas/project.json
 
 # ScreenKit project configuration file.
 schema: 1
@@ -76,19 +77,35 @@ watermark:
   margin: 100
 
 # TTS configuration.
-# To disable voiceover entirely, set `tts: false`.
-# tts:
-#   engine: elevenlabs
-#   voice_id: 56AoDkrOh6qfVPDXZ7Pt
-#   language_code: en
-#   voice_settings:
-#     speed: 0.9
-#     stability: 0.5
-#     similarity: 0.75
-#     style: 0.0
+# To disable voiceover entirely, set `tts: false` or disable all engines
+# individually.
+#
+# You can set up multiple TTS engines to be used as fallbacks in case one fails.
+# Each TTS engine has its own detection mechanism. For instance, say and espeak
+# checks for a binary with the same name. ElevenLabs checks for the presence of
+# `--tts-api-key`.
 tts:
-  engine: say
-  rate: 150
+  # Apple Say TTS engine configuration.
+  - engine: say
+    rate: 150
+    enabled: true
+
+  # eSpeak TTS engine configuration.
+  - engine: espeak
+    rate: 150
+    voice: en-us
+    enabled: true
+
+  # Eleven Labs TTS engine configuration.
+  - engine: eleven_labs
+    enabled: true
+    voice_id: 56AoDkrOh6qfVPDXZ7Pt
+    language_code: en
+    voice_settings:
+      speed: 0.9
+      stability: 0.5
+      similarity: 0.75
+      style: 0.0
 
 # Scene configurations define how different scenes in the video are rendered.
 # There are three types of scenes: intro, outro, and segment.
@@ -98,13 +115,13 @@ tts:
 scenes:
   intro:
     # The duration of the intro scene in seconds.
-    duration: 5.5
+    duration: 5.5s
 
     # The fade-in duration in seconds.
-    fade_in: 0
+    fade_in: 0s
 
     # The fade-out duration in seconds.
-    fade_out: 0.5
+    fade_out: 0.5s
 
     # The background color/image of the intro scene.
     background: "#100f50"
@@ -113,7 +130,7 @@ scenes:
     title:
       x: 100
       y: 300
-      font_path: opensans/OpenSans-ExtraBold.ttf
+      font_path: open-sans/OpenSans-ExtraBold.ttf
       size: 144
       color: "#ffffff"
 
@@ -133,13 +150,13 @@ scenes:
 
   outro:
     # The duration of the outro scene in seconds.
-    duration: 5.5
+    duration: 5.5s
 
     # The fade-in duration in seconds.
-    fade_in: 0.5
+    fade_in: 0.5s
 
     # The fade-out duration in seconds.
-    fade_out: 0.5
+    fade_out: 0.5s
 
     # The background color/image of the outro scene.
     background: "#100f50"
@@ -156,34 +173,113 @@ scenes:
 
   segment:
     # The duration of the crossover transition between segments in seconds.
-    crossfade_duration: 0.5
+    crossfade_duration: 0.5s
+
+# Configure [demotape](https://github.com/fnando/demotape).
+demotape:
+  # Global speed for typing
+  typing_speed: 50ms
+
+  # Whether to add variability to typing speed
+  variable_typing: 0.25
+
+  # Shell to use for the terminal session
+  shell: zsh
+
+  # Path to a shell rc file to source on startup
+  # rc_file:
+
+  # Do not load the default rc file
+  no_rc: false
+
+  # A built-in theme name or a path to a custom theme JSON file
+  theme: default_dark
+
+  # Font size for the terminal session
+  font_size: 32
+
+  # Font family for the terminal session
+  font_family: "'JetBrainsMono Nerd Font Propo', monospace"
+
+  # Line height for the terminal session
+  line_height: 1.2
+
+  # Whether the cursor blinks
+  cursor_blink: true
+
+  # The width of the cursor when in 'bar' style
+  cursor_width: 2
+
+  # The style of the cursor when the terminal is focused.
+  cursor_style: block
+
+  # Letter spacing for the terminal session
+  letter_spacing: 0
+
+  # Padding around the terminal content. It can be 1, 2, or 4 numbers.
+  padding: 50
+
+  # Loop the GIF infinitely
+  loop: true
+
+  # Delay time (in seconds) before restarting the GIF animation
+  loop_delay: 0
+
+  # Margin around the terminal content. It can be 1, 2, or 4 numbers.
+  margin: 0
+
+  # Color or image to use for the margin around the terminal content
+  margin_fill: "#000000"
+
+  # Border radius for the terminal canvas
+  border_radius: 0
+
+  # Delay (in seconds) before pressing Enter
+  run_enter_delay: 1s
+
+  # Delay (in seconds) after running each command
+  run_sleep: 3s
 
 # Callout configurations define the appearance and behavior of callouts used
 # in the video. Callouts are visual elements that highlight important
 # information. Each callout type can have its own settings for icon, background,
 # text styles, animations, and sounds.
-callouts:
-  info:
-    # yaml-language-server: $schema=../../schemas/callouts/default.json
+callout_styles:
+  shadow_block:
+    # yaml-language-server: $schema=https://screenkit.dev/schemas/callout_styles/default.json
+    style: shadow_block
     background_color: "#ffff00"
     shadow: "#2242d3"
     title_style:
       color: "#000000"
       size: 40
-      font_path: opensans/OpenSans-ExtraBold.ttf
+      font_path: open-sans/OpenSans-ExtraBold.ttf
     body_style:
       color: "#000000"
       size: 32
-      font_path: opensans/OpenSans-Semibold.ttf
+      font_path: open-sans/OpenSans-Semibold.ttf
     margin: 100
     padding: 50
     anchor: [left, bottom]
     animation: fade
     in_transition:
-      duration: 0.4
+      duration: 0.4s
       sound: pop.mp3
     out_transition:
-      duration: 0.3
+      duration: 0.3s
       sound:
         path: pop.mp3
         volume: 0.7
+
+  inline_block:
+    # yaml-language-server: $schema=https://screenkit.dev/schemas/callout_styles/inline_block.json
+    style: inline_block
+    anchor: [left, bottom]
+    margin: 100
+    animation: fade
+    in_transition:
+      duration: 0.4s
+      sound: pop.mp3
+    out_transition:
+      duration: 0.4s
+      sound: pop.mp3

data/lib/screenkit/schema_validator.rb

@@ -6,7 +6,7 @@ module ScreenKit
       errors = JSON::Validator
                .fully_validate("file://#{schema_path}", attributes)
 
-      return if errors.empty?
+      return true if errors.empty?
 
       raise ArgumentError, "Invalid attributes: #{errors.first}"
     end

data/lib/screenkit/schemas/callout_styles/file_copy.json

@@ -0,0 +1,8 @@
+{
+  "$schema": "http://json-schema.org/draft-04/schema#",
+  "$id": "https://screenkit.dev/schemas/callout_styles/file_copy.json",
+  "title": "File Copy Callout",
+  "type": "object",
+  "required": ["file_path"],
+  "properties": { "file_path": { "type": "string", "format": "uri" } }
+}