i6SSKJr SSKrSSKJrJr SSKJr SSK J r SSK J r J r SSKJrJr SS KJr SS KJr SS KJr SS KJrJr SS KJr Sr"SS5r"SS\5r"SS\5r"SS\5r"SS\ 5rg!\a SSKJr N}f=f)) annotationsN)AnyProtocol)Self)meta) BaseModelValidationError) DefaultCache RenderCache)config)env) AsyncError) ChatMessagechat_message_from_text)generate_canary_word0c\rSrSrSSSSSS.S SjjrSSjr\SSj5r\SSj5r\SSj5r \SS j5r \SS j5r SS jr S r g) BasePromptN)nameversionmetadata canary_word render_cachecJU=(d 0UlU=(d [[R"55UlXlU=(d [ 5Ul[R"U5Ul U=(d [Ul SU=(d [50Ulg)a Prompt constructor. Parameters: text: The template text. name: The name to identify this prompt. version: The version string attached to this prompt. metadata: A key-value set of metadata pairs attached to this prompt. canary_word: The string to use for the `{{canary_word}}` extension. If `None`, a default string will be generated. render_cache: The caching backend to store rendered prompts. If `None`, the default in-memory backend will be used. rN) _metadatastruuiduuid4_name_rawr _render_cacher from_string _templateDEFAULT_VERSION_versionrdefaults)selftextrrrrrs F/home/james-whalen/.local/lib/python3.13/site-packages/banks/prompt.py__init__BasePrompt.__init__sk."R.S.  );\^.2? & (M7K7MN c<Uc UR$XR-$Nr()r)datas r+ _get_contextBasePrompt._get_context<s <== mm##r.cUR$r0)rr)s r+rBasePrompt.metadataAs ~~r.cUR$r0)r!r6s r+rBasePrompt.nameEs zzr.cUR$)z#Returns the raw text of the prompt.)r"r6s r+rawBasePrompt.rawIsyyr.cUR$r0)r'r6s r+rBasePrompt.versionNs }}r.cn[R"UR5n[R"U5$r0)rparser;rfind_undeclared_variables)r)asts r+ variablesBasePrompt.variablesRs%ii!--c22r.c&URSU;$)z^Returns whether the canary word is present in `text`, signalling the prompt might have leaked.rr1)r)r*s r+ canary_leakedBasePrompt.canary_leakedWs}}]+t33r.)rr!r"r#r%r'r()r*rr str | NonerrHrdict[str, Any] | NonerrHrzRenderCache | NonereturnNone)r2z dict | NonerJdict)rJzdict[str, Any])rJr)rJrH)rJzset[str])r*rrJbool)__name__ __module__ __qualname____firstlineno__r,r3propertyrrr;rrCrF__static_attributes__r.r+rrs  "*."&+/OO O  O ( O O)O O@$ 334r.rc4\rSrSrSrSSSjjrSS SjjrSrg) Prompt\a The `Prompt` class is the only thing you need to know about Banks on the Python side. The class can be initialized with a string variable containing the prompt template text, then you can invoke the method `text` on your instance to pass the data needed to render the template and get back the final prompt. A quick example: ```py from banks import Prompt p = Prompt("Write a 500-word blog post on {{ topic }}.") my_topic = "retrogame computing" print(p.text({"topic": my_topic})) ``` NcURU5nURRU5nU(aU$URR U5nURR X5 U$){ Render the prompt using variables present in `data` Parameters: data: A dictionary containing the context variables. )r3r#getr%rendersetr)r2cachedrendereds r+r* Prompt.textms_  &##''- M--d3 t.r.cURU5nURRU5nU(d6URR U5nURR X5 /nUR 5RS5H)nUR[R"U55 M+ U(dUR[SUS95 U$![a M\f=f)rY user)rolecontent) r3r#rZr%r[r\stripsplitappendrmodel_validate_jsonr r)r)r2r_messageslines r+ chat_messagesPrompt.chat_messages}s  &%%))$/~~,,T2H    " "4 2&(NN$**40D  ? ? EF1 OO2Q R#  s%C C('C(rTr0r2rIrJr)r2rIrJzlist[ChatMessage])rNrOrPrQ__doc__r*rlrSrTr.r+rVrV\s  r.rVc>^\rSrSrSrSU4SjjrSSSjjrSrU=r$) AsyncPrompta Banks provides async support through the machinery [provided by Jinja](https://jinja.palletsprojects.com/en/3.0.x/api/#async-support) Since the Jinja environment is a global state in banks, the library can work either with or without async support, and this must be known before importing anything. If the application using banks runs within an `asyncio` loop, you can do two things to optimize banks' execution: 1. Set the environment variable `BANKS_ASYNC_ENABLED=true`. 2. Use the `AsyncPrompt` class that has an awaitable `run` method. For example, let's render a prompt that contains some calls to the `generate` extension. Those calls will be heavily I/O bound, so other tasks can take advantage and advance while the prompt is being rendered. Example: ```python # Enable async support before importing from banks import os os.environ["BANKS_ASYNC_ENABLED"] = "true" # Show logs to see what's happening at runtime import logging logging.basicConfig(level=logging.INFO) import asyncio from banks import AsyncPrompt prompt_template = """ Generate a tweet about the topic '{{ topic }}' with a positive sentiment. """ async def task(task_id: int, sleep_time: int): logging.info(f"Task {task_id} is running.") await asyncio.sleep(sleep_time) logging.info(f"Task {task_id} done.") async def main(): p = AsyncPrompt(prompt_template) # Schedule the prompt rendering along with two executions of 'task', one sleeping for 10 seconds # and one sleeping for 1 second results = await asyncio.gather(p.text({"topic": "AI frameworks"}), task(1, 10), task(2, 1)) print("All tasks done, rendered prompt:") print(results[0]) asyncio.run(main()) ``` cj>[TU]"U0UD6 [R(d Sn[ U5eg)NzaAsync is not enabled. Please set the environment variable 'BANKS_ASYNC_ENABLED=on' and try again.)superr,r ASYNC_ENABLEDr)r)argskwargsmsg __class__s r+r,AsyncPrompt.__init__s2 $)&)##uCS/ !$r.c# URU5nURRU5nU(aU$URR U5IShvN nURR X5 U$N!7f)rYN)r3r#rZr% render_asyncr\r]s r+r*AsyncPrompt.textsk  &##''- M"nn99$?? t.@sAA:A8"A:rT)rvrrwrrJrKr0rn) rNrOrPrQror,r*rS __classcell__)rys@r+rqrqs4l"r.rqc<\rSrSrSrSS.S SjjrSS.S SjjrS rg) PromptRegistryz:Interface to be implemented by concrete prompt registries.N)rcgr0rT)r)rrs r+rZPromptRegistry.getsr.F) overwritecgr0rT)r)promptrs r+r\PromptRegistry.setrr.rT)rrrrHrJrV)rrVrrMrJrK)rNrOrPrQrorZr\rSrTr.r+rrsD6:J7<JJr.rcb\rSrSr%SrS\S'SrS\S'SrS\S'SrS \S '\ S S j5r S r g) PromptModelz(Serializable representation of a Prompt.rr*NrHrrrIrcdU"URURURURS9$)N)r*rrr)r;rrr)clsrs r+ from_promptPromptModel.from_prompts& fnnW]WfWfggr.rT)rz type[Self]rrVrJr) rNrOrPrQro__annotations__rrr classmethodrrSrTr.r+rrs=2 ID*GZ&*H#*hhr.r) __future__rrtypingrrr ImportErrortyping_extensionsjinja2rpydanticrr cacher r r rerrorsrtypesrrutilsrr&rrVrqrrrTr.r+rs# '/,6'>4>4B;Z;|L*L^KXK h) hM'&'sB BB