Merge branch 'AUTOMATIC1111:dev' into mac-builds-experimental

javascript/ui.js

@@ -19,28 +19,11 @@ function all_gallery_buttons() {
 }
 
 function selected_gallery_button() {
-    var allCurrentButtons = gradioApp().querySelectorAll('[style="display: block;"].tabitem div[id$=_gallery].gradio-gallery .thumbnail-item.thumbnail-small.selected');
-    var visibleCurrentButton = null;
-    allCurrentButtons.forEach(function(elem) {
-        if (elem.parentElement.offsetParent) {
-            visibleCurrentButton = elem;
-        }
-    });
-    return visibleCurrentButton;
+    return all_gallery_buttons().find(elem => elem.classList.contains('selected')) ?? null;
 }
 
 function selected_gallery_index() {
-    var buttons = all_gallery_buttons();
-    var button = selected_gallery_button();
-
-    var result = -1;
-    buttons.forEach(function(v, i) {
-        if (v == button) {
-            result = i;
-        }
-    });
-
-    return result;
+    return all_gallery_buttons().findIndex(elem => elem.classList.contains('selected'));
 }
 
 function extract_image_from_gallery(gallery) {

modules/api/api.py

@@ -23,8 +23,7 @@
 from modules.textual_inversion.preprocess import preprocess
 from modules.hypernetworks.hypernetwork import create_hypernetwork, train_hypernetwork
 from PIL import PngImagePlugin,Image
-from modules.sd_models import checkpoints_list, unload_model_weights, reload_model_weights, checkpoint_aliases
-from modules.sd_vae import vae_dict
+from modules.sd_models import unload_model_weights, reload_model_weights, checkpoint_aliases
 from modules.sd_models_config import find_checkpoint_config_near_filename
 from modules.realesrgan_model import get_realesrgan_models
 from modules import devices
@@ -57,6 +56,15 @@ def setUpscalers(req: dict):
 
 
 def decode_base64_to_image(encoding):
+    if encoding.startswith("http://") or encoding.startswith("https://"):
+        import requests
+        response = requests.get(encoding, timeout=30, headers={'user-agent':'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/62.0.3202.94 Safari/537.36'})
+        try:
+            image = Image.open(BytesIO(response.content))
+            return image
+        except Exception as e:
+            raise HTTPException(status_code=500, detail="Invalid image url") from e
+
     if encoding.startswith("data:image/"):
         encoding = encoding.split(";")[1].split(",")[1]
     try:
@@ -567,10 +575,12 @@ def get_latent_upscale_modes(self):
         ]
 
     def get_sd_models(self):
-        return [{"title": x.title, "model_name": x.model_name, "hash": x.shorthash, "sha256": x.sha256, "filename": x.filename, "config": find_checkpoint_config_near_filename(x)} for x in checkpoints_list.values()]
+        import modules.sd_models as sd_models
+        return [{"title": x.title, "model_name": x.model_name, "hash": x.shorthash, "sha256": x.sha256, "filename": x.filename, "config": find_checkpoint_config_near_filename(x)} for x in sd_models.checkpoints_list.values()]
 
     def get_sd_vaes(self):
-        return [{"model_name": x, "filename": vae_dict[x]} for x in vae_dict.keys()]
+        import modules.sd_vae as sd_vae
+        return [{"model_name": x, "filename": sd_vae.vae_dict[x]} for x in sd_vae.vae_dict.keys()]
 
     def get_hypernetworks(self):
         return [{"name": name, "path": shared.hypernetworks[name]} for name in shared.hypernetworks]

modules/cmd_args.py

@@ -81,7 +81,7 @@
 parser.add_argument("--gradio-auth-path", type=str, help='set gradio authentication file path ex. "/path/to/auth/file" same auth format as --gradio-auth', default=None)
 parser.add_argument("--gradio-img2img-tool", type=str, help='does not do anything')
 parser.add_argument("--gradio-inpaint-tool", type=str, help="does not do anything")
-parser.add_argument("--gradio-allowed-path", action='append', help="add path to gradio's allowed_paths, make it possible to serve files from it")
+parser.add_argument("--gradio-allowed-path", action='append', help="add path to gradio's allowed_paths, make it possible to serve files from it", default=[data_path])
 parser.add_argument("--opt-channelslast", action='store_true', help="change memory type for stable diffusion to channels last")
 parser.add_argument("--styles-file", type=str, help="filename to use for styles", default=os.path.join(data_path, 'styles.csv'))
 parser.add_argument("--autolaunch", action='store_true', help="open the webui URL in the system's default browser upon launch", default=False)

modules/images.py

@@ -355,7 +355,9 @@ class FilenameGenerator:
         'date': lambda self: datetime.datetime.now().strftime('%Y-%m-%d'),
         'datetime': lambda self, *args: self.datetime(*args),  # accepts formats: [datetime], [datetime<Format>], [datetime<Format><Time Zone>]
         'job_timestamp': lambda self: getattr(self.p, "job_timestamp", shared.state.job_timestamp),
-        'prompt_hash': lambda self: hashlib.sha256(self.prompt.encode()).hexdigest()[0:8],
+        'prompt_hash': lambda self, *args: self.string_hash(self.prompt, *args),
+        'negative_prompt_hash': lambda self, *args: self.string_hash(self.p.negative_prompt, *args),
+        'full_prompt_hash': lambda self, *args: self.string_hash(f"{self.p.prompt} {self.p.negative_prompt}", *args),  # a space in between to create a unique string
         'prompt': lambda self: sanitize_filename_part(self.prompt),
         'prompt_no_styles': lambda self: self.prompt_no_style(),
         'prompt_spaces': lambda self: sanitize_filename_part(self.prompt, replace_spaces=False),
@@ -368,7 +370,8 @@ class FilenameGenerator:
         'denoising': lambda self: self.p.denoising_strength if self.p and self.p.denoising_strength else NOTHING_AND_SKIP_PREVIOUS_TEXT,
         'user': lambda self: self.p.user,
         'vae_filename': lambda self: self.get_vae_filename(),
-        'none': lambda self: '', # Overrides the default so you can get just the sequence number
+        'none': lambda self: '',  # Overrides the default, so you can get just the sequence number
+        'image_hash': lambda self, *args: self.image_hash(*args)  # accepts formats: [image_hash<length>] default full hash
     }
     default_time_format = '%Y%m%d%H%M%S'
 
@@ -448,6 +451,14 @@ def datetime(self, *args):
 
         return sanitize_filename_part(formatted_time, replace_spaces=False)
 
+    def image_hash(self, *args):
+        length = int(args[0]) if (args and args[0] != "") else None
+        return hashlib.sha256(self.image.tobytes()).hexdigest()[0:length]
+
+    def string_hash(self, text, *args):
+        length = int(args[0]) if (args and args[0] != "") else 8
+        return hashlib.sha256(text.encode()).hexdigest()[0:length]
+
     def apply(self, x):
         res = ''
 

modules/img2img.py

@@ -122,15 +122,14 @@ def img2img(id_task: str, mode: int, prompt: str, negative_prompt: str, prompt_s
     is_batch = mode == 5
 
     if mode == 0:  # img2img
-        image = init_img.convert("RGB")
+        image = init_img
         mask = None
     elif mode == 1:  # img2img sketch
-        image = sketch.convert("RGB")
+        image = sketch
         mask = None
     elif mode == 2:  # inpaint
         image, mask = init_img_with_mask["image"], init_img_with_mask["mask"]
         mask = processing.create_binary_mask(mask)
-        image = image.convert("RGB")
     elif mode == 3:  # inpaint sketch
         image = inpaint_color_sketch
         orig = inpaint_color_sketch_orig or inpaint_color_sketch
@@ -139,7 +138,6 @@ def img2img(id_task: str, mode: int, prompt: str, negative_prompt: str, prompt_s
         mask = ImageEnhance.Brightness(mask).enhance(1 - mask_alpha / 100)
         blur = ImageFilter.GaussianBlur(mask_blur)
         image = Image.composite(image.filter(blur), orig, mask.filter(blur))
-        image = image.convert("RGB")
     elif mode == 4:  # inpaint upload mask
         image = init_img_inpaint
         mask = init_mask_inpaint

modules/launch_utils.py

@@ -246,7 +246,7 @@ def list_extensions(settings_file):
     disabled_extensions = set(settings.get('disabled_extensions', []))
     disable_all_extensions = settings.get('disable_all_extensions', 'none')
 
-    if disable_all_extensions != 'none' or args.disable_extra_extensions or args.disable_all_extensions:
+    if disable_all_extensions != 'none' or args.disable_extra_extensions or args.disable_all_extensions or not os.path.isdir(extensions_dir):
         return []
 
     return [x for x in os.listdir(extensions_dir) if x not in disabled_extensions]

modules/processing.py

@@ -386,14 +386,14 @@ def get_token_merging_ratio(self, for_hr=False):
         return self.token_merging_ratio or opts.token_merging_ratio
 
     def setup_prompts(self):
-        if type(self.prompt) == list:
+        if isinstance(self.prompt,list):
             self.all_prompts = self.prompt
-        elif type(self.negative_prompt) == list:
+        elif isinstance(self.negative_prompt, list):
             self.all_prompts = [self.prompt] * len(self.negative_prompt)
         else:
             self.all_prompts = self.batch_size * self.n_iter * [self.prompt]
 
-        if type(self.negative_prompt) == list:
+        if isinstance(self.negative_prompt, list):
             self.all_negative_prompts = self.negative_prompt
         else:
             self.all_negative_prompts = [self.negative_prompt] * len(self.all_prompts)
@@ -512,10 +512,10 @@ def __init__(self, p: StableDiffusionProcessing, images_list, seed=-1, info="",
         self.s_noise = p.s_noise
         self.s_min_uncond = p.s_min_uncond
         self.sampler_noise_scheduler_override = p.sampler_noise_scheduler_override
-        self.prompt = self.prompt if type(self.prompt) != list else self.prompt[0]
-        self.negative_prompt = self.negative_prompt if type(self.negative_prompt) != list else self.negative_prompt[0]
-        self.seed = int(self.seed if type(self.seed) != list else self.seed[0]) if self.seed is not None else -1
-        self.subseed = int(self.subseed if type(self.subseed) != list else self.subseed[0]) if self.subseed is not None else -1
+        self.prompt = self.prompt if not isinstance(self.prompt, list) else self.prompt[0]
+        self.negative_prompt = self.negative_prompt if not isinstance(self.negative_prompt, list) else self.negative_prompt[0]
+        self.seed = int(self.seed if not isinstance(self.seed, list) else self.seed[0]) if self.seed is not None else -1
+        self.subseed = int(self.subseed if not isinstance(self.subseed, list) else self.subseed[0]) if self.subseed is not None else -1
         self.is_using_inpainting_conditioning = p.is_using_inpainting_conditioning
 
         self.all_prompts = all_prompts or p.all_prompts or [self.prompt]
@@ -702,11 +702,8 @@ def process_images(p: StableDiffusionProcessing) -> Processed:
     stored_opts = {k: opts.data[k] for k in p.override_settings.keys()}
 
     try:
-        # after running refiner, the refiner model is not unloaded - webui swaps back to main model here
-        if shared.sd_model.sd_checkpoint_info.title != opts.sd_model_checkpoint:
-            sd_models.reload_model_weights()
-
         # if no checkpoint override or the override checkpoint can't be found, remove override entry and load opts checkpoint
+        # and if after running refiner, the refiner model is not unloaded - webui swaps back to main model here, if model over is present it will be reloaded afterwards
         if sd_models.checkpoint_aliases.get(p.override_settings.get('sd_model_checkpoint')) is None:
             p.override_settings.pop('sd_model_checkpoint', None)
             sd_models.reload_model_weights()
@@ -741,7 +738,7 @@ def process_images(p: StableDiffusionProcessing) -> Processed:
 def process_images_inner(p: StableDiffusionProcessing) -> Processed:
     """this is the main loop that both txt2img and img2img use; it calls func_init once inside all the scopes and func_sample once per batch"""
 
-    if type(p.prompt) == list:
+    if isinstance(p.prompt, list):
         assert(len(p.prompt) > 0)
     else:
         assert p.prompt is not None
@@ -772,12 +769,12 @@ def process_images_inner(p: StableDiffusionProcessing) -> Processed:
 
     p.setup_prompts()
 
-    if type(seed) == list:
+    if isinstance(seed, list):
         p.all_seeds = seed
     else:
         p.all_seeds = [int(seed) + (x if p.subseed_strength == 0 else 0) for x in range(len(p.all_prompts))]
 
-    if type(subseed) == list:
+    if isinstance(subseed, list):
         p.all_subseeds = subseed
     else:
         p.all_subseeds = [int(subseed) + x for x in range(len(p.all_prompts))]
@@ -1155,6 +1152,9 @@ def sample(self, conditioning, unconditional_conditioning, seeds, subseeds, subs
             devices.torch_gc()
 
     def sample_hr_pass(self, samples, decoded_samples, seeds, subseeds, subseed_strength, prompts):
+        if shared.state.interrupted:
+            return samples
+
         self.is_hr_pass = True
 
         target_width = self.hr_upscale_to_x
@@ -1268,12 +1268,12 @@ def setup_prompts(self):
         if self.hr_negative_prompt == '':
             self.hr_negative_prompt = self.negative_prompt
 
-        if type(self.hr_prompt) == list:
+        if isinstance(self.hr_prompt, list):
             self.all_hr_prompts = self.hr_prompt
         else:
             self.all_hr_prompts = self.batch_size * self.n_iter * [self.hr_prompt]
 
-        if type(self.hr_negative_prompt) == list:
+        if isinstance(self.hr_negative_prompt, list):
             self.all_hr_negative_prompts = self.hr_negative_prompt
         else:
             self.all_hr_negative_prompts = self.batch_size * self.n_iter * [self.hr_negative_prompt]

modules/prompt_parser.py

@@ -86,7 +86,7 @@ def alternate(self, args):
                 yield args[(step - 1) % len(args)]
             def start(self, args):
                 def flatten(x):
-                    if type(x) == str:
+                    if isinstance(x, str):
                         yield x
                     else:
                         for gen in x:

modules/script_callbacks.py

@@ -28,6 +28,15 @@ def __init__(self, image, p, filename, pnginfo):
         """dictionary with parameters for image's PNG info data; infotext will have the key 'parameters'"""
 
 
+class ExtraNoiseParams:
+    def __init__(self, noise, x):
+        self.noise = noise
+        """Random noise generated by the seed"""
+
+        self.x = x
+        """Latent image representation of the image"""
+
+
 class CFGDenoiserParams:
     def __init__(self, x, image_cond, sigma, sampling_step, total_sampling_steps, text_cond, text_uncond):
         self.x = x
@@ -100,6 +109,7 @@ def __init__(self, imgs, cols, rows):
     callbacks_ui_settings=[],
     callbacks_before_image_saved=[],
     callbacks_image_saved=[],
+    callbacks_extra_noise=[],
     callbacks_cfg_denoiser=[],
     callbacks_cfg_denoised=[],
     callbacks_cfg_after_cfg=[],
@@ -189,6 +199,14 @@ def image_saved_callback(params: ImageSaveParams):
             report_exception(c, 'image_saved_callback')
 
 
+def extra_noise_callback(params: ExtraNoiseParams):
+    for c in callback_map['callbacks_extra_noise']:
+        try:
+            c.callback(params)
+        except Exception:
+            report_exception(c, 'callbacks_extra_noise')
+
+
 def cfg_denoiser_callback(params: CFGDenoiserParams):
     for c in callback_map['callbacks_cfg_denoiser']:
         try:
@@ -367,6 +385,14 @@ def on_image_saved(callback):
     add_callback(callback_map['callbacks_image_saved'], callback)
 
 
+def on_extra_noise(callback):
+    """register a function to be called before adding extra noise in img2img or hires fix;
+    The callback is called with one argument:
+        - params: ExtraNoiseParams - contains noise determined by seed and latent representation of image
+    """
+    add_callback(callback_map['callbacks_extra_noise'], callback)
+
+
 def on_cfg_denoiser(callback):
     """register a function to be called in the kdiffussion cfg_denoiser method after building the inner model inputs.
     The callback is called with one argument:

modules/scripts.py

@@ -269,7 +269,7 @@ def elem_id(self, item_id):
         """helper function to generate id for a HTML element, constructs final id out of script name, tab and user-supplied item_id"""
 
         need_tabname = self.show(True) == self.show(False)
-        tabkind = 'img2img' if self.is_img2img else 'txt2txt'
+        tabkind = 'img2img' if self.is_img2img else 'txt2img'
         tabname = f"{tabkind}_" if need_tabname else ""
         title = re.sub(r'[^a-z_0-9]', '', re.sub(r'\s', '_', self.title().lower()))
 
@@ -289,7 +289,7 @@ def elem_id(self, item_id):
         """helper function to generate id for a HTML element, constructs final id out of tab and user-supplied item_id"""
 
         need_tabname = self.show(True) == self.show(False)
-        tabname = ('img2img' if self.is_img2img else 'txt2txt') + "_" if need_tabname else ""
+        tabname = ('img2img' if self.is_img2img else 'txt2img') + "_" if need_tabname else ""
 
         return f'{tabname}{item_id}'
 

modules/sd_hijack.py

@@ -245,7 +245,21 @@ def flatten(el):
         ldm.modules.diffusionmodules.openaimodel.UNetModel.forward = sd_unet.UNetModel_forward
 
     def undo_hijack(self, m):
-        if type(m.cond_stage_model) == sd_hijack_xlmr.FrozenXLMREmbedderWithCustomWords:
+        conditioner = getattr(m, 'conditioner', None)
+        if conditioner:
+            for i in range(len(conditioner.embedders)):
+                embedder = conditioner.embedders[i]
+                if isinstance(embedder, (sd_hijack_open_clip.FrozenOpenCLIPEmbedderWithCustomWords, sd_hijack_open_clip.FrozenOpenCLIPEmbedder2WithCustomWords)):
+                    embedder.wrapped.model.token_embedding = embedder.wrapped.model.token_embedding.wrapped
+                    conditioner.embedders[i] = embedder.wrapped
+                if isinstance(embedder, sd_hijack_clip.FrozenCLIPEmbedderForSDXLWithCustomWords):
+                    embedder.wrapped.transformer.text_model.embeddings.token_embedding = embedder.wrapped.transformer.text_model.embeddings.token_embedding.wrapped
+                    conditioner.embedders[i] = embedder.wrapped
+
+            if hasattr(m, 'cond_stage_model'):
+                delattr(m, 'cond_stage_model')
+
+        elif type(m.cond_stage_model) == sd_hijack_xlmr.FrozenXLMREmbedderWithCustomWords:
             m.cond_stage_model = m.cond_stage_model.wrapped
 
         elif type(m.cond_stage_model) == sd_hijack_clip.FrozenCLIPEmbedderWithCustomWords:

modules/sd_models.py

@@ -343,7 +343,10 @@ def load_model_weights(model, checkpoint_info: CheckpointInfo, state_dict, timer
         model.to(memory_format=torch.channels_last)
         timer.record("apply channels_last")
 
-    if not shared.cmd_opts.no_half:
+    if shared.cmd_opts.no_half:
+        model.float()
+        timer.record("apply float()")
+    else:
         vae = model.first_stage_model
         depth_model = getattr(model, 'depth_model', None)
 

modules/sd_samplers_common.py

@@ -36,7 +36,7 @@ def setup_img2img_steps(p, steps=None):
 
 def samples_to_images_tensor(sample, approximation=None, model=None):
     '''latents -> images [-1, 1]'''
-    if approximation is None:
+    if approximation is None or (shared.state.interrupted and opts.live_preview_fast_interrupt):
         approximation = approximation_indexes.get(opts.show_progress_type, 0)
 
     if approximation == 2:
@@ -49,7 +49,8 @@ def samples_to_images_tensor(sample, approximation=None, model=None):
     else:
         if model is None:
             model = shared.sd_model
-        x_sample = model.decode_first_stage(sample.to(model.first_stage_model.dtype))
+        with devices.without_autocast(): # fixes an issue with unstable VAEs that are flaky even in fp32
+            x_sample = model.decode_first_stage(sample.to(model.first_stage_model.dtype))
 
     return x_sample