app.py

import os
import gradio as gr
from random import randint
from operator import itemgetter
import bisect
from all_models import tags_plus_models,models,models_plus_tags
from datetime import datetime
from externalmod import gr_Interface_load
import asyncio
import os
from threading import RLock
lock = RLock()
HF_TOKEN = os.environ.get("HF_TOKEN") if os.environ.get("HF_TOKEN") else None # If private or gated models aren't used, ENV setting is unnecessary.


now2 = 0
inference_timeout = 300
MAX_SEED = 2**32-1


nb_rep=2
nb_mod_dif=20
nb_models=nb_mod_dif*nb_rep

cache_image={}
cache_id_image={}
cache_list_task={}


def load_fn(models):
    global models_load
    global num_models
    global default_models
    models_load = {}
    num_models = len(models)
    i=0
    if num_models!=0:
        default_models = models[:num_models]
    else:
        default_models = {}
    for model in models:
        i+=1
        if i%50==0:
            print("\n\n\n-------"+str(i)+'/'+str(len(models))+"-------\n\n\n")
        if model not in models_load.keys():
            try:
                m = gr_Interface_load(f'models/{model}', hf_token=HF_TOKEN)
            except Exception as error:
                m = gr.Interface(lambda txt: None, ['text'], ['image'])
                print(error)
            models_load.update({model: m})


load_fn(models)

tags_plus_models_to_list=[]
list_tags=[]
for tag_plus_m in tags_plus_models:
    list_tags.append(tag_plus_m[0]+f" ({tag_plus_m[1]})")
    
def test_pass(test):
    if test==os.getenv('p'):
        print("ok")
        return gr.Dropdown(label="Lists Tags", show_label=True, choices=list(models_test) , interactive = True)
    else:
        print("nop")
        return gr.Dropdown(label="Lists Tags", show_label=True, choices=list([]) , interactive = True)

def test_pass_aff(test):
    if test==os.getenv('p'):
        return gr.Accordion( open=True, visible=True) ,gr.Row(visible=False)
    else:
        return gr.Accordion( open=True, visible=False) , gr.Row()


# https://huggingface.co/docs/api-inference/detailed_parameters
# https://huggingface.co/docs/huggingface_hub/package_reference/inference_client
async def infer(model_str, prompt, nprompt="", height=None, width=None, steps=None, cfg=None, seed=-1, timeout=inference_timeout):
    from pathlib import Path
    kwargs = {}
    if height is not None and height >= 256: kwargs["height"] = height
    if width is not None and width >= 256: kwargs["width"] = width
    if steps is not None and steps >= 1: kwargs["num_inference_steps"] = steps
    if cfg is not None and cfg > 0: cfg = kwargs["guidance_scale"] = cfg
    noise = ""
    if seed >= 0: kwargs["seed"] = seed
    else:
        rand = randint(1, 500)
        for i in range(rand):
            noise += " "
    task = asyncio.create_task(asyncio.to_thread(models_load[model_str].fn,
                               prompt=f'{prompt} {noise}', negative_prompt=nprompt, **kwargs, token=HF_TOKEN))
    await asyncio.sleep(3)
    try:
        result = await asyncio.wait_for(task, timeout=timeout)
    except (Exception, asyncio.TimeoutError) as e:
        print(e)
        print(f"Task timed out: {model_str}")
        if not task.done(): task.cancel()
        result = None
    if task.done() and result is not None:
        with lock:
            nb_rand1=randint(1, MAX_SEED)
            nb_rand2=randint(1, MAX_SEED)
            nb_rand3=randint(1, MAX_SEED)
            png_path = f"image_{nb_rand1}_{nb_rand2}_{nb_rand3}.png"
            result.save(png_path)
            image = str(Path(png_path).resolve())
        return image
    return None
def gen_fn(model_str, prompt, nprompt="", height=None, width=None, steps=None, cfg=None, seed=-1):
    if model_str == 'NA':
        return None
    try:
        loop = asyncio.new_event_loop()
        result = loop.run_until_complete(infer(model_str, prompt, nprompt,
                                         height, width, steps, cfg, seed, inference_timeout))
    except (Exception, asyncio.CancelledError) as e:
        print(e)
        print(f"Task aborted: {model_str}")
        result = None
    finally:
        loop.close()
    return result

   
def add_gallery(image, model_str, gallery):
    if gallery is None: gallery = []
    #with lock:
    if image is not None: gallery.append((image, model_str))
    return gallery

def reset_gallery(gallery):
    return add_gallery(None,"",[])

def load_gallery(gallery,id):
    gallery = reset_gallery(gallery)
    for c in cache_image[f"{id}"]:
        gallery=add_gallery(c[0],c[1],gallery)
    return gallery
def load_gallery_sorted(gallery,id):
    gallery = reset_gallery(gallery)
    for c in sorted(cache_image[f"{id}"], key=itemgetter(1)):
        gallery=add_gallery(c[0],c[1],gallery)
    return gallery

def add_cache_image(image, model_str,id,cache_image=cache_image):
    if image is not None:
        cache_image[f"{id}"].append((image,model_str))
    #cache_image=sorted(cache_image, key=itemgetter(1))
    return 
def reset_cache_image(id,cache_image=cache_image):
    cache_image[f"{id}"].clear()
    return 
def reset_cache_image_all_sessions(cache_image=cache_image):
    for key, listT in cache_image.items():
        listT.clear()
    return 
    
def set_session(id):
    if id==0:
        randTemp=randint(1,MAX_SEED)
        cache_image[f"{randTemp}"]=[]
        cache_id_image[f"{randTemp}"]=[]
        
        cache_list_task[f"{randTemp}"]=[]
        return gr.Number(visible=False,value=randTemp)
    else :
        return id
def print_info_sessions():
    lenTot=0
    print("###################################")
    print("number of sessions : "+str(len(cache_image)))
    for key, listT in cache_image.items():
        print("session "+key+" : "+str(len(listT)))
        lenTot+=len(listT)
    print("images total = "+str(lenTot))
    print("###################################")
    return

def disp_models(group_model_choice,nb_rep=nb_rep):
    listTemp=[]
    strTemp='\n'
    i=0
    for m in group_model_choice:
        if m not in listTemp:
            listTemp.append(m)
    for m in listTemp:
        i+=1
        strTemp+="\"" + m + "\",\n"
        if i%(8/nb_rep)==0:
            strTemp+="\n"
    return gr.Textbox(label="models",value=strTemp)

def search_models(str_search,tags_plus_models=tags_plus_models):
    output1="\n"
    output2=""
    for m in tags_plus_models[0][2]:
        if m.find(str_search)!=-1:
            output1+="\"" + m + "\",\n"
    outputPlus="\n From tags : \n\n"
    for tag_plus_models in tags_plus_models:
        if str_search.lower() == tag_plus_models[0].lower() and str_search!="":
            for m in tag_plus_models[2]:
                output2+="\"" + m + "\",\n"
    if output2 != "":
        output=output1+outputPlus+output2
    else :
        output=output1
    return gr.Textbox(label="out",value=output)

def search_info(txt_search_info,models_plus_tags=models_plus_tags):
    outputList=[]
    if txt_search_info.find("\"")!=-1:
        start=txt_search_info.find("\"")+1
        end=txt_search_info.find("\"",start)
        m_name=cutStrg(txt_search_info,start,end)
    else :
        m_name = txt_search_info
    for m in models_plus_tags:
        if m_name == m[0]:
            outputList=m[1]
    if len(outputList)==0:
        outputList.append("Model Not Find")
    return gr.Textbox(label="out",value=outputList)


def ratio_chosen(choice_ratio,width,height):
    if choice_ratio == [None,None]:
        return width , height
    else :
        return gr.Slider(label="Width", info="If 0, the default value is used.", maximum=2024, step=32, value=choice_ratio[0]), gr.Slider(label="Height", info="If 0, the default value is used.", maximum=2024, step=32, value=choice_ratio[1])

list_ratios=[["None",[None,None]],
             ["4:1 (2048 x 512)",[2048,512]],
             ["12:5 (1536 x 640)",[1536,640]],
             ["~16:9 (1344 x 768)",[1344,768]],
             ["~3:2 (1216 x 832)",[1216,832]],
             ["~4:3 (1152 x 896)",[1152,896]],
             ["1:1 (1024 x 1024)",[1024,1024]],
             ["~3:4 (896 x 1152)",[896,1152]],
             ["~2:3 (832 x 1216)",[832,1216]],
             ["~9:16 (768 x 1344)",[768,1344]],
             ["5:12 (640 x 1536)",[640,1536]],
             ["1:4 (512 x 2048)",[512,2048]]]

def fonc_add_param(lp,txt_input,neg_input,width,height,steps,cfg,seed):
    if lp == [["","",0,0,0,0,-1]]:
        lp.remove(["","",0,0,0,0,-1])
    lp.append([txt_input,neg_input,width,height,steps,cfg,seed])
    return gr.Dataset(samples=lp) , gr.Dropdown(choices=[["a",lp]], value=lp)
def fonc_del_param(lp,txt_input,neg_input,width,height,steps,cfg,seed):
    if [txt_input,neg_input,width,height,steps,cfg,seed] in lp :
        lp.remove([txt_input,neg_input,width,height,steps,cfg,seed])
    if lp == []:
        lp.append(["","",0,0,0,0,-1])
    return gr.Dataset(samples=lp) , gr.Dropdown(choices=[["a",lp]], value=lp)

def fonc_load_info(nb_of_models_to_gen,index_tag,index_first_model):
    str_temp=""
    list_models_temp=[]
    if index_first_model+nb_of_models_to_gen>len(tags_plus_models[index_tag][2]):
        if nb_of_models_to_gen>len(tags_plus_models[index_tag][2]):
            str_temp+="warning : to many model chosen"
        else:
            str_temp+="warning : first model to close to the last model"
        nb_of_models_to_gen= len(tags_plus_models[index_tag][2])-index_first_model
        str_temp+=f" - only {nb_of_models_to_gen} will be use\n\n"
    str_temp+="list of models use (from "
    str_temp+=f"{index_first_model+1}/{len(tags_plus_models[index_tag][2])} to {index_first_model+nb_of_models_to_gen}/{len(tags_plus_models[index_tag][2])}) :\n\n"
    for i in range(nb_of_models_to_gen):
        list_models_temp.append(tags_plus_models[index_tag][2][i+index_first_model])
        str_temp+=f"\"{tags_plus_models[index_tag][2][i+index_first_model]}\",\n"
    return nb_of_models_to_gen,gr.Textbox(str_temp),gr.Dropdown(choices=[["",list_models_temp]], value=list_models_temp )
def fonc_load_info_custom(nb_of_models_to_gen,list_model_custom,index_first_model):
    str_temp=""
    list_models_temp=[]
    if index_first_model+nb_of_models_to_gen>len(list_model_custom):
        if nb_of_models_to_gen>len(list_model_custom):
            str_temp+="warning : to many model chosen"
        else:
            str_temp+="warning : first model to close to the last model"
        nb_of_models_to_gen= len(list_model_custom)-index_first_model
        str_temp+=f" - only {nb_of_models_to_gen} will be use\n\n"
    str_temp+="list of models CUSTOM use (from "
    str_temp+=f"{index_first_model+1}/{len(list_model_custom)} to {index_first_model+nb_of_models_to_gen}/{len(list_model_custom)}) :\n\n"
    for i in range(nb_of_models_to_gen):
        list_models_temp.append(list_model_custom[i+index_first_model])
        str_temp+=f"\"{list_model_custom[i+index_first_model]}\",\n"
    return nb_of_models_to_gen,gr.Textbox(str_temp),gr.Dropdown(choices=[["",list_models_temp]], value=list_models_temp )

def crea_list_task(id_session,list_param,list_models_to_gen,nb_images_by_prompt):
    cache_list_task[f"{id_session}"]=[]
    dict_temp={}
    for m in list_models_to_gen:
        dict_temp={}
        dict_temp["model"]=m
        dict_temp["id_module"]=-1
        dict_temp["task"]=[]
        for p in list_param:
            for i in range(nb_images_by_prompt):
                dict_temp["task"].append({"prompt":p[0],"nprompt":p[1],"width":p[2],"height":p[3],"steps":p[4],"cfg":p[5],"seed":p[6]})
        cache_list_task[f"{id_session}"].append(dict_temp)

    
def cutStrg(longStrg,start,end):
    shortStrg=''
    for i in range(end-start):
        shortStrg+=longStrg[start+i]
    return shortStrg
def aff_models_perso(txt_list_perso,models=models):
    list_perso=[]
    t1=True
    start=txt_list_perso.find('\"')
    if start!=-1:
        while t1:
            start+=1
            end=txt_list_perso.find('\"',start)
            if end != -1:
                txtTemp=cutStrg(txt_list_perso,start,end)
                if txtTemp in models:
                    list_perso.append(cutStrg(txt_list_perso,start,end))
            else :
                t1=False
            start=txt_list_perso.find('\"',end+1)
            if start==-1:
                t1=False
    return gr.Dropdown(choices=[["",list_perso]], value=list_perso )

def add_gallery(image, model_str, gallery):
    if gallery is None: gallery = []
    #with lock:
    if image is not None: gallery.append((image, model_str))
    return gallery

def reset_gallery(gallery):
    return add_gallery(None,"",[])
def fonc_load_gallery(id_session,gallery):
    gallery = reset_gallery(gallery)
    for i in range(len(cache_image[f"{id_session}"])):
        gallery=add_gallery(cache_image[f"{id_session}"][i]["image"],cache_image[f"{id_session}"][i]["model"],gallery)
    return gallery
def fonc_load_gallery_by_model(id_session,gallery,models,index):
    gallery = reset_gallery(gallery)
    for i in range(len(cache_image[f"{id_session}"])):
        if cache_image[f"{id_session}"][i]["model"]==models[index]:
            gallery=add_gallery(cache_image[f"{id_session}"][i]["image"],cache_image[f"{id_session}"][i]["model"],gallery)
    return gallery

def fonc_start(id_session,id_module,s,cont):
    if cont==False:
        print("manual stop")
        return None,gr.Textbox(s)
    task_actu={}
    model_actu=""
    print(f"in fonc : id module={id_module}\n")
    for model_plus_tasks in cache_list_task[f"{id_session}"]:
        if model_plus_tasks["id_module"]==id_module:
            model_actu=model_plus_tasks["model"]
            task_actu=model_plus_tasks["task"][0]
            
            print(f"find model : {model_actu}\n")
            
    if model_actu=="":
        for model_plus_tasks in cache_list_task[f"{id_session}"]:
            if model_plus_tasks["id_module"]==-1 :
                if model_actu=="":
                    model_plus_tasks["id_module"]=id_module
                    model_actu=model_plus_tasks["model"]
                    task_actu=model_plus_tasks["task"][0]
                    print(f"module num {id_module} take {model_actu}\n")
    if model_actu=="":#crash
        print("Stop with :"+s+"\n")
        return None,gr.Textbox(s)
    print("begin gen image:")
    print(model_actu)
    print(task_actu)
    result=gen_fn(model_actu, task_actu["prompt"], task_actu["nprompt"], task_actu["height"], task_actu["width"], task_actu["steps"], task_actu["cfg"], task_actu["seed"])
    print("reception")
    if result!=None:
        #result=gr.Image(result)
        id_image=len(cache_image[f"{id_session}"])
        for model_plus_tasks in cache_list_task[f"{id_session}"]:
            if model_plus_tasks["id_module"]==id_module:
                model_plus_tasks["task"].remove(task_actu)
            if len(model_plus_tasks["task"])==0:
                cache_list_task[f"{id_session}"].remove(model_plus_tasks)
            
        task_actu["id_image"]=id_image
        task_actu["model"]=model_actu
        task_actu["image"]=result
        #cache_image[f"{id_session}"].append(result)
        #cache_id_image[f"{id_session}"].append(task_actu)
        cache_image[f"{id_session}"].append(task_actu)
        print("image saved\n")
    else:
        print("fail to generate\n")
    num_task_undo=0
    for model_plus_tasks in cache_list_task[f"{id_session}"]:
        for task in model_plus_tasks["task"]:
            num_task_undo+=1
    print(f"\n {num_task_undo} tasks undo\n")
    return result , gr.Textbox(s+"1")

def fonc_init(s):
    return gr.Textbox(s+"1")

def index_gallery_first():
    return gr.Number(0)
def index_gallery_next(i,list_models):
    iT=i+1
    return gr.Number(iT%len(list_models))
def index_gallery_prev(i,list_models):
    iT=i-1
    return gr.Number(iT%len(list_models))

def make_me():
        with gr.Column():
            id_session=gr.Number(visible=False,value=0)
            with gr.Group():
                with gr.Row():
                    with gr.Column(scale=4):
                        txt_input = gr.Textbox(label='Your prompt:', lines=4, interactive = True)
                        neg_input = gr.Textbox(label='Negative prompt:', lines=4, interactive = True)
                    with gr.Column(scale=4):
                        with gr.Row():
                            width = gr.Slider(label="Width", info="If 0, the default value is used.", maximum=2024, step=32, value=0, interactive = True)
                            height = gr.Slider(label="Height", info="If 0, the default value is used.", maximum=2024, step=32, value=0, interactive = True)
                        with gr.Row():
                            choice_ratio = gr.Dropdown(label="Ratio Width/Height", 
                                                info="OverWrite Width and Height (W*H<1024*1024)", 
                                                show_label=True, choices=list(list_ratios) , interactive = True, value=list_ratios[0][1])
                            choice_ratio.change(ratio_chosen,[choice_ratio,width,height],[width,height])
                        with gr.Row():
                            steps = gr.Slider(label="Number of inference steps", info="If 0, the default value is used.", maximum=100, step=1, value=0, interactive = True)
                            cfg = gr.Slider(label="Guidance scale", info="If 0, the default value is used.", maximum=30.0, step=0.1, value=0, interactive = True)
                            seed = gr.Slider(label="Seed", info="Randomize Seed if -1.", minimum=-1, maximum=MAX_SEED, step=1, value=-1, interactive = True)
                            
            add_param=gr.Button("Add to the list")
            del_param=gr.Button("Delete to the list")
            
        #gen_button = gr.Button('Generate images', scale=3)
        #stop_button = gr.Button('Stop', variant='secondary', interactive=False, scale=1)
        #gen_button.click(lambda: gr.update(interactive=True), None, stop_button)
            
            list_param=gr.Dropdown(choices=[["a",[["","",0,0,0,0,-1]]]], value=[["","",0,0,0,0,-1]], visible=False)
            disp_param = gr.Examples(
                label="list of prompt",
                examples=list_param.value,
                inputs=[txt_input,neg_input,width,height,steps,cfg,seed],
                outputs=[txt_input,neg_input,width,height,steps,cfg,seed],
            )
            add_param.click(fonc_add_param,[list_param,txt_input,neg_input,width,height,steps,cfg,seed],[disp_param.dataset,list_param])
            add_param.click(set_session,[id_session],[id_session])
            del_param.click(fonc_del_param,[list_param,txt_input,neg_input,width,height,steps,cfg,seed],[disp_param.dataset,list_param])

            with gr.Row():
                list_models_to_gen=gr.Dropdown(choices=[["",[]]], value=[], visible=False)
                disp_info=gr.Textbox(label="Info")
                with gr.Column():
                    with gr.Row():
                        nb_images_by_prompt=gr.Number(2,label="Number of images by prompt:",interactive=True)
                        nb_of_models_to_gen=gr.Number(10,label="Number of Models:",interactive=True)
                    index_tag=gr.Dropdown(label="Tag",choices=list(list_tags),type="index")
                    index_first_model=gr.Dropdown(label="First model",choices=list([]), type="index")
                    index_tag.change(lambda i:gr.Dropdown(choices=list([f"({j+1}/{len(tags_plus_models[i][2])}) {tags_plus_models[i][2][j]}" for j in range(len(tags_plus_models[i][2]))])),
                                     index_tag,index_first_model)
                    load_info=gr.Button("Load Models")
                    load_info.click(fonc_load_info,[nb_of_models_to_gen,index_tag,index_first_model],[nb_of_models_to_gen,disp_info,list_models_to_gen])
                
            with gr.Accordion("Models Custom",open=False) :
                with gr.Row():
                    text_list_model_custom=gr.Textbox(label="List Models Custom")
                    with gr.Column():
                        list_model_custom=gr.Dropdown(choices=[["",[]]], value=[], visible=False)
                        #use_models_custom=gr.Radio("Use Models Custom",value=False)
                        cut_model_custom=gr.Button("Cut Text Models Custom")
                        cut_model_custom.click(aff_models_perso,[text_list_model_custom],[list_model_custom])
                        
                        index_first_model_custom=gr.Dropdown(label="First model",choices=list([]), type="index")
                        list_model_custom.change(lambda li:gr.Dropdown(choices=list([f"({j+1}/{len(li)}) {li[j]}" for j in range(len(li))])),
                                         [list_model_custom],index_first_model_custom)
                        
                        load_model_custom=gr.Button("Load Models Custom")
                        load_model_custom.click(fonc_load_info_custom,[nb_of_models_to_gen,list_model_custom,index_first_model_custom],[nb_of_models_to_gen,disp_info,list_models_to_gen])
            
            list_models_to_gen.change(crea_list_task,[id_session,list_param,list_models_to_gen,nb_images_by_prompt],[])
            

            
            with gr.Column():
                with gr.Row():
                    nb_req_simult=gr.Number(10,"Number simultaneous request")
                    nb_req_simult_max_by_mod=gr.Number(2,"Number max simultaneous request by model")
                
                button_start=gr.Button("START")
                button_stop=gr.Button("STOP")
                cont=gr.Checkbox(True,visible=False)
                button_start.click(lambda:True,[],[cont])
                button_stop.click(lambda:False,[],[cont])
                with gr.Row():
                    outputs=[]
                    id_modules=[]
                    states=[]
                    for i in range(nb_req_simult.value):
                        #outputs.append(gr.Image(None,interactive=False,render=False))
                        #id_modules.append(gr.Number(i,interactive=False,render=False))
                        outputs.append(gr.Image(None,interactive=False,visible=False))
                        id_modules.append(gr.Number(i,interactive=False,visible=False))
                        states.append(gr.Textbox("1",interactive=False,visible=False))
                    print(len(outputs))
                    print(outputs[0])
                    for o,i,s in zip(outputs,id_modules,states):
                        #o.change(fonc_start,[id_session,i],[o])
                        #o.change(test_change,[],[])
                        s.change(fonc_start,[id_session,i,s,cont],[o,s])
                        #button_start.click(lambda : gr.Image(None),[],[o])
                        gen_event = gr.on(triggers=[button_start.click], fn=fonc_init,inputs=[s], outputs=[s])
                        
                    gallery = gr.Gallery(label="Output", show_download_button=True, elem_classes="gallery",
                                    interactive=False, show_share_button=True, container=True, format="png",
                                    preview=True, object_fit="cover",columns=4,rows=4) 
                    with gr.Column():
                        button_load_gallery=gr.Button("Load Gallery All")
                        button_load_gallery.click(fonc_load_gallery,[id_session,gallery],[gallery])

                        index_gallery=gr.Number(0,visible=False)
                        button_load_gallery_first=gr.Button("Load Gallery first model")
                        button_load_gallery_first.click(index_gallery_first,[],[index_gallery])
                        with gr.Row():
                            button_load_gallery_next=gr.Button("Next model")
                            button_load_gallery_next.click(index_gallery_next,[index_gallery,list_models_to_gen],[index_gallery])
                            button_load_gallery_prev=gr.Button("Prev model")
                            button_load_gallery_prev.click(index_gallery_prev,[index_gallery,list_models_to_gen],[index_gallery])
                            index_gallery.change(fonc_load_gallery_by_model,[id_session,gallery,list_models_to_gen,index_gallery],[gallery])
                    
    
        





                
        #gen_button.click(set_session, id_session, id_session)


js_code = """
    
    console.log('ghgh');
"""

with gr.Blocks(theme="Nymbo/Nymbo_Theme", fill_width=True, css="div.float.svelte-1mwvhlq {    position: absolute;    top: var(--block-label-margin);    left: var(--block-label-margin);    background: none;    border: none;}") as demo: 
    gr.Markdown("<script>" + js_code + "</script>")
    make_me()


# https://www.gradio.app/guides/setting-up-a-demo-for-maximum-performance
#demo.queue(concurrency_count=999) # concurrency_count is deprecated in 4.x
demo.queue(default_concurrency_limit=200, max_size=200)
demo.launch(max_threads=400)